Ann Hematol (2011) 90 (Suppl 1):S25–S76 DOI 10.1007/s00277-011-1158-7

ACUTE LEUKEMIAS XIII BIOLOGY AND TREATMENT STRATEGIES February 27–March 2, 2011, Munich, Germany

Satellite Symposia Satellite Symposium III Maintenance Therapy in AML with Emphasis on the Immunotherapeutic Approach JACOB M. ROWE Rambam Health Care Campus and Technion, Israel Institute of Technology, Haifa, Israel To date, relapse from acute myelogenous leukemia (AML) remains the most important barrier to cure in AML. The ideal approach to prevent relapse is the administration of nontoxic targeted or immunotherapeutic maintenance regimen after the conclusion of post-remission therapy. Such therapy needs to be safe and efficacious. (Table 1)

Table 1: Strategies to Prevent Relapse: Post-consolidation Maintenance Therapies

Targeted approach :

“Moderately” myeloablative Less myeloablative Less myeloablative

Immunotherapeutic :

? Non-myeloablative Non-myeloablative

Chemotherapy-based :

maintenance therapy may vary in different diseases. While it is intended as curative therapy in ALL and APL, in AML it may be curative but in most clinical trials this is successful only as far as delaying relapse. The mechanism of action of maintenance therapy is poorly understood. Its effectiveness is counter-intuitive, based on the Goldie-Coldman hypothesis.1 Several hypotheses have been put forward to suggest possible mechanisms, including the difficult postulate that maintenance drugs kill the few surviving leukemia cells2. Another suggestion has been that drugs used as maintenance therapy stimulate the immune system3; also an unlikely postulate given that the cornerstone of maintenance therapy in ALL is 6mercaptopurine and methotraxate, which are both immunosuppressive. A further postulate suggested that maintenance therapy only controls growth of leukemia allowing for normal apoptotic mechanisms to reduce or eliminate tumor burden4. Maintenance therapy is very rarely, if ever, curative in states with detectable minimal residual disease (MRD), while the best results occur in eradicating MRD at levels below the threshold of detection (103–104). In AML there have been several randomized studies of maintenance therapy; many have been widely interpreted as demonstrating no effect, although careful review of the data does not support such a conclusion. For example, one of the earliest studies of maintenance therapy was conducted by the Cancer and Leukemia Group B (CALGB) in 19875. In this trial patients received maintenance therapy for 8 months and were then randomized to a further 36 months of maintenance versus observation (Fig.1).

Maintenance Therapy in Acute Leukemia CALGB Study: Preisler et al, Blood, 1987

Operative definition Maintenance therapy usually refers to therapy that is given for 1–3 years, after successful induction and, usually, consolidative intensification. It is usually not profoundly myelosuppressive. Most importantly, this is given for presumed minimal residual disease—even if this cannot be detected by current molecular technologies. Maintenance with less myeloablative regimens has been convincingly established for childhood acute lymphoblastic leukemia (ALL), and similarly adopted for adult ALL, as well as for acute promyelocytic leukemia (APL). It is important to emphasize that the purpose of

Observation

Induction

Consolidation

Maintenance (8 months)

Sequential

ARA – C 6 TG DNR VCR Prednisone

R

Continue Maintenance (36 months)

Figure 1: Maintenance: Post Remission Therapy (Reproduced with permission from Preisler et al., Blood, 19875)

S26

Ann Hematol (2011) 90 (Suppl 1):S25–S76

Published results indicated that the administration of continued maintenance therapy did not provide any additional benefit for disease-free survival (DFS) (Fig. 2). CALGB Study: Preisler et al, Blood, 1987

EORTC / HOVON

Observation Continued Maintenance

Disease Free Survival (%)

100

LD Ara C

80

No Ara C

60

N = 73

40

P = .006 20

0 0

1

2

3

4

5

6

7

8

9

Years 100

LD Ara-C

0

1

2

3 Years

5

4

80

Survival (%)

Disease Free Survival (%)

100 90 80 70 60 50 40 30 20 10 0

However, there were some responders in the maintenance arm. Among older patients the European Organisation for Research and Treatment of Cancer (EORTC) randomized patients to receive multiple cycles of low-dose cytosine arabinoside versus observation, and showed a benefit in disease-free survival. However, a benefit in overall survival (OS) could not be demonstrated7 (Fig. 5a, 5b).

Figure 2: Maintenance: Post Remission Therapy (Reproduced with permission from from Preisler et al., Blood, 19875)

No Ara-C 60

N = 73

40

P = .29 N = 74

20

0 0

1

2

3

4

5

6

7

8

9

Year

This study has been widely quoted as demonstrating no effect of maintenance therapy. However, this study demonstrates only that maintenance therapy, as given in this study, has no benefit beyond 8 months. The Eastern Cooperative Oncology Group (ECOG) conducted a study in 1983 (E3483) where maintenance therapy was one of the postremission options and was compared with an observation arm (Fig. 3). INDUCTION

CR HLA matched sibling

No sibling

High dose consolidation (Ara C)x 1

Allogeneic BMT

Low dose maintenance (1 year)

Observation only

Figure 3: AML Post-remission Therapy (E3483): Adults 18–60 years This study was terminated early by the National Cancer Institute in the USA when every single patient in the observation arm relapsed6 (Figure 4).

Figure 5: Randomized Trial of Maintenance Therapy in Older Adults— A. Disease-free survival/B. Overall survival (Reproduced with permission from Löwenberg B, et al. JCO 19987) The apparent conclusion is that maintenance therapy is clearly effective as post-remission therapy with some evidence that this has curative potential; however, it is unlikely to be of benefit for younger adults in whom very intensive therapy is the mainstay of post-remission therapy. At the same time, it must be emphasized that even when this was administered as the only post-remission therapy, it has never been studied after modern-day intensive post-remission therapy, including transplantation. Its role in older adults, who cannot withstand intensive therapy, is more compelling and needs to be more completely evaluated. Potential novel or targeted agents that are being studied as maintenance therapy include gemtuzumab ozogamicin, farnesyltransferase inhibitors, hypomethylating agents, lenalidomide, tyrosine kinase inhibitors of FLT3 and histone deacetylase inhibitors. Several studies of gemtuzumab ozogamicin (GO) maintenance have been conducted in the recent years. The Eastern Cooperative Oncology Group (ECOG) reported on their E1900 study that the addition of GO after consolidation therapy and, prior to autologous transplant, has not improved the DFS or OS.8 (Fig. 6a, 6b) GO

100 90

INDUCTION

CONSOLIDATIONx 2

70

No Maintenance n = 26

60

Maintenance n = 66

Overall Survival (%)

Event free survival

AUTOLOGOUS TRANSPLANT

80

50 P = 0.002 40 30 20

No GO n = 132 GO n = 138 P = NS

10 0 0

6

12

18

24

30

36

42

48

54

60

66

72

Weeks

Figure 4: AML Post-remission Therapy (E3483). Every patient in the observation arm relapsed. (Reproduced with permission from Cassileth et al. Blood. 19926)

Months

Figure 6: ECOG Study, E1900, of Gemtuzumab Ozogamicin Post Consolidation Similarly, the Southwest Oncology Group (SWOG) at the 2009 annual meeting of the American Society of Hematology reported

Ann Hematol (2011) 90 (Suppl 1):S25–S76

S27

a similar lack of benefit for maintenance therapy with GO. (Fig. 7)

AML>60 Years

9

Reduced Intensity allo

S0106

OBSERVATION

OBSERVATION

INDUCTION

CONSOLIDATION

3+7 ± GO

HD Ara-C x 3

CONSOLIDATION

INDUCTION

GEMTUZUMAB OZOGAMICIN x3

3+7 vs Clofarabine

HD Ara-C vs Clofarabine

DECITABINE 20 mg/m2 IV days 1–3 q 4 wks x 12

Figure 7: Maintenance in AML <60 years: SWOG Study More recently, the HOVON/SAKK study reported their findings in older patients who were randomized to GO maintenance once again, demonstrating no benefit.10 (Fig. 8, 9) AML

> 60 years

Figure 11: Decitabine as Post-Remission Maintenance—ECOG Study: E2906 A major international study conducted by the CALGB is currently evaluating the role of midostaurin, a tyrosine kinase inhibitor of FLT3 mutations. Results from this study are awaited. (Fig. 12)

MDS Reduced Intensity allo

Newly diagnosed AML <60 years with activating FLT3 mutation

OBSERVATION INDUCTION

INDUCTION

CONSOLIDATION

7+ 3

Ara-C

1–2 cycles Gemtuzumab Ozogamicin 6mg/m2 q 4 wks X3

45 vs 90 mg/m2 n = 237

n = 847

7+3 + Midostaurin

CONSOLIDATION

MAINTENANCE

4 cycles

12 cycles

HD Cytarabine + Midostaurin

vs

Midostaurin vs Placebo

vs

Placebo

Placebo

Figure 8: Gemtuzumab Ozogamicin in Post-Remission—HOVON/ AMLSG/SAKK Study Figure 12: US/Europe AML Intergroup Study [CALGB 10603] [ HOVON / AMLSG / SAKK]

Overall Survival (%) n=119 GO

Other tyrosine kinase inhibitors being tested, for which data are not currently available, include sorafenib and AC220. Thus, at the present time there are no data available to suggest that any of these potential or targeted agents will become incorporated as standard of care as maintenance therapy in AML.

n=113 NO p= .52

Years

Figure 9: GO maintenance in AML (Reproduced with permission from Lowenberg B et al., Blood, 201010) The ECOG has currently completed a study of maintenance with tipifarnib. (Fig. 10) This was well tolerated, but final analysis is still awaited11. The ECOG is currently investigating the role of decitabine as a potential agent in maintenance therapy. (Fig. 11)

IMMUNOTHERAPEUTIC MAINTENANCE THERAPY Several studies of immunotherapeutic maintenance therapy have been conducted. In 2001 the Medical Research Council (MRC) in Britain conducted a large study of maintenance in older patients. (Fig. 13)

α -IFN 3 x 106 units x 3 per week for 1 year

Shortconsolidation ( 1 cycle)

Induction

R

R

ECOG Study (E 2902)

OBSERVATION

Long consolidation (4 cycles)

Observation

CR Post-consolidation Tipifarnib 300gm PO b.i.d. x 21 days q 28 days

Figure 10: Farnesyltransferase Inhibitor as Maintenance Therapy in AML

Figure 13: MRC AML 11: Maintenance Randomization in Older Adults (Reproduced with permission from Goldstone AH, et al. Blood. 200112) The rationale was based on the known ability of α-interferon to alter cytogenetics in chronic myeloid leukemia (CML). This was a large well-conducted study that, however, unequivocally demonstrated

S28

Ann Hematol (2011) 90 (Suppl 1):S25–S76

the lack of any benefit for interferon compared with observation.12 (Fig. 14) .

IFN

Overall Survival (%)

n = 182

The data from this study, as from other studies if IL-2 maintenance, are disappointing, with no improvement in DFS for patients receiving IL-2 as a single agent. (Fig. 17) 550 patients

n = 180

Median follow up of 3.6 years from second randomization 21% 20%

3 year DFS: IL 2

44%; observation

p =. 57

Years

Figure 14: MRC AML 11: Interferon Maintenance in AML (Reproduced with permission from Goldstone AH, et al. Blood. 200112) IL-2 monotherapy has been studied extensively based on the compelling rationale of the ability of IL-2 to activate the T and NK cells. Randomized studies have been conducted and in none of them did IL-2 show a significant benefit (Fig. 15). IL2 VS CONTROL

IL 2 DOSE/MONTH MIU/M2

Blaise et al

30% vs 36%

120

CALGB 9720

No difference

81

CCG 2961

51% vs 58%

52

0.49

CALGB 19808

56% vs 45%

91

0.11

ALFA 9801

No difference

25

EORTC/GIMEMA

No difference

24

STUDY

P

Figure 17: IL-2 vs observation: EORTC—GIMEMA Study In summary, six randomized studies, with over 1,500 patients, assessed in large groups and over a wide range of IL-2 doses, could not demonstrate any statistically significant benefit for IL-2 alone for any efficiency parameter. Thus, the aggregate data indicate that IL-2 alone, despite the compelling rationale, is insufficient to exert a meaningful response as maintenance therapy in AML.

0.54

Figure 15: Randomized Studies of IL-2 Maintenance The most recent study of IL-2 maintenance in AML was conducted by the EORTC and GIMEMA. This study evaluated over 2,000 patients with AML less than 61 years who received consolidation and an autologous transplant. Five hundred and fifty patients were randomized to receive IL-2 after autologous transplant, versus observation.13 (Fig. 16) Donor

AML < 61 years: Induction

42%

Allogeneic SCT

IL-2 with histamine dihydrochloride (HDC) The addition of histamine dehydrochloride to IL-2 has been investigated given the fact that histamine, though its ability to inhibit the inactivation of T and NK cells by oxygen radicals produced by myeloid cells, is likely to overcome the inhibitory effects when IL-2 is used alone. Based on a phase II study of 39 patients that demonstrated what appeared to be a significant activation of T and NK cells and good compliance and safety of IL-2 with HDC when given as maintenance, a large international prospective phase III study of IL-2 and histamine dihydrochloride (HDC) was conducted and published in 2006. IL 2 / HDC 10 CONSECUTIVE 3-WEEK CYCLES

CR

R

18 MONTHS

Postconsolidation

OBSERVATION

3-year Follow-up

Consolidation

OBSERVATION n=2005 No donor

Autologous SCT

R

IL 2 4.8x106 IU

n=276

x 5 days q . month

n=550

Observation n=274

Figure 16: IL-2 Maintenance in AML—EORTC/GIMEMA AML-12 Study

Figure 18: HDC/IL-2 vs Observation: Multi-institutional Study In this study (Fig. 18) patients were randomized to postconsolidation therapy and received 10 consecutive 3-week cycles of IL-2 with HDC versus observation. Follow-up extended beyond 3 years. Three hundred and twenty patients entered this study with 261 in first complete remission (CR1). The study was

Ann Hematol (2011) 90 (Suppl 1):S25–S76

S29

conducted in 10 countries in 100 sites. Patients were stratified by the CR status.

Leukemia-Free Survival (%)

Overall Population HDC/ IL-2 (n=160) Control (n=160) P = .008

Months

Figure 19: HDC/IL-2 Maintenance in AML (Reproduced with permission from Brune M, et al. Blood. 200614) Figure 19 demonstrates the results from primary endpoint of the study with a significant benefit in leukemia-free survival for the entire population and an even more impressive benefit when the patients in CR1 are considered14 (Fig. 20).

Leukemia-Free Survival (%)

CR1 Population HDC / IL-2 (n=129) Control (n=132)

P= .011

Months

Figure 20: HDC/IL-2 Maintenance in AML: CR1 Population (Reproduced with permission from Brune M, et al. Blood. 2006 14)

Post-relapse Survival (%)

Furthermore, the administration of HDC/IL-2 does not affect the post-relapse survival (Fig. 21). Importantly, the administration of IL-2 with histamine was well tolerated, with no significant morbidity and no mortality and did not compromise the post-relapse survival.

IL-2+ HDC (n= 129) Control (n = 132)

P= 0.810

Months

Figure 21: HDC/IL-2 does not compromise post-relapse survival

Currently, the use of histamine dihydrochloride and IL-2 remains the only maintenance therapy for AML that is approved by the EMEACHMP. This represents a novel approach to post-remission maintenance therapy and given its overall safety and demonstrated efficacy, provides a compelling rationale for AML patients in remission. Conflict of interest None References 1. Goldie JH, Coldman AJ. A mathematic model for relating the drug sensitivity of tumors to their spontaneous mutation rate. Cancer Treat Rep. 1979;63:1727–1733. 2. Pinkel D. Curing children of leukemia. Cancer. 1987;59:1683– 1691. 3. Mokyr MB, Dray S. Interplay between the toxic effects of anticancer drugs and host antitumor immunity in cancer therapy. Cancer Invest. 1987;5:31–38. 4. Gale RP, Butturini A. Maintenance chemotherapy and cure of childhood acute lymphoblastic leukaemia. Lancet. 1991;338: 1315–1318. 5. Preisler H, Davis RB, Kirshner J, et al. Comparison of three remission induction regimens and two postinduction strategies for the treatment of acute nonlymphocytic leukemia: a cancer and leukemia group B study. Blood. 1987;69:1441– 1449. 6. Cassileth PA, Lynch E, Hines JD, et al. Varying intensity of postremission therapy in acute myeloid leukemia. Blood. 1992;79: 1924–1930. 7. Lowenberg B, Suciu S, Archimbaud E, et al. Mitoxantrone versus daunorubicin in induction-consolidation chemotherapy— the value of low-dose cytarabine for maintenance of remission, and an assessment of prognostic factors in acute myeloid leukemia in the elderly: final report. European Organization for the Research and Treatment of Cancer and the Dutch-Belgian Hemato-Oncology Cooperative Hovon Group. J Clin Oncol. 1998;16:872–881. 8. Fernandez HF, Sun Z, Bennett JM, et al. A single dose of gemtuzumab ozogamicin (GO) in consolidation prior to autologous transplant for younger patients with newly diagnosed acute myeloid leukemia (AML) is safe but has no effect on disease-free survival: interim results of the Eastern Cooperative Oncology Group study (E1900). Biology of Blood and Marrow Transplantation 2008;14 52a. 9. Petersdorf S, Kopecky K, Stuart RK, et al. Preliminary Results of Southwest Oncology Group Study S0106: An International Intergroup Phase 3 Randomized Trial Comparing the Addition of Gemtuzumab Ozogamicin to Standard Induction Therapy Versus Standard Induction Therapy Followed by a Second Randomization to Post-Consolidation Gemtuzumab Ozogamicin Versus No Additional Therapy for Previously Untreated Acute Myeloid Leukemia. Blood. 2009;114:790a. 10. Lowenberg B, Beck J, Graux C, et al. Gemtuzumab ozogamicin as postremission treatment in AML at 60 years of age or more:

S30 results of a multicenter phase 3 study. Blood. 2010;115:2586– 2591. 11. Luger S, Yao X, Paietta E, et al. Tipifarnib Is Well Tolerated as Maintenance Therapy In Acute Myeloid Leukemia (AML). Significant, but Non-Fatal, Hematologic Toxicity Not Ameliorated by Dose Reduction. Preliminary Results of the Phase III Intergroup Trial E2902. Blood. 2010;116: 3315a. 12. Goldstone AH, Burnett AK, Wheatley K, Smith AG, Hutchinson RM, Clark RE. Attempts to improve treatment outcomes in acute myeloid leukemia (AML) in older patients: the results of the United Kingdom Medical Research Council AML11 trial. Blood. 2001;98:1302– 1311. 13. Willemze R, Suciu S, Mandelli F, et al. Value of Low Dose IL-2 as Maintenance Following Consolidation Treatment or Autologous Transplantation in Acute Myelogenous Leukemia (AML) Patients Aged 15–60 Years Who Reached CR After High Dose (HD-AraC) Vs Standard Dose (SD-AraC) Cytosine Arabinoside During Induction: Results of the AML-12 Trial of EORTC and GIMEMA Leukemia Groups. Blood. 2009; 114:791a. 14. Brune M, Castaigne S, Catalano J, et al. Improved leukemiafree survival after postconsolidation immunotherapy with histamine dihydrochloride and interleukin-2 in acute myeloid leukemia: results of a randomized phase 3 trial. Blood. 2006; 108:88–96.

Age-Related Efficacy of Immunotherapy with Histamine Dihydrochloride and Interleukin-2 for Relapse Prevention in Acute Myeloid Leukemia K. HELLSTRAND1, F.B. THORÉN1,2, A. MARTNER1, J. SÖDERHOLM1, W.K HOFMANN, J.M ROWE, and M. BRUNE2 Departments of 1Infectious Diseases and 2Hematology, University of Gothenburg, Sweden

Abstract Recurrence of leukemia after the completion of induction and consolidation chemotherapy is a significant clinical concern in acute myeloid leukemia (AML). Apart from allogeneic bone marrow transplantation there is no consensus about effective relapse-protective therapy beyond the consolidation phase, and the standard of care for the majority of patients in complete remission (CR) hence is no treatment. Here we present updated results from a phase 3 trial (n = 320) evaluating the prevention of relapse in AML patients receiving immunotherapy with histamine dihydrochloride (HDC) and low-dose interleukin-2 (IL-2). This trial was previously reported to meet the primary endpoint of improved leukemia-free survival (LFS) in the primary population of all randomized patients. Our results imply that treatment with HDC/IL-2 prevents relapse in patients 40–70 years old in first CR (p = 0.008, leukemia-free survival (LFS), n = 190,

Ann Hematol (2011) 90 (Suppl 1):S25–S76 log rank test) with a more than 80% relative increase in the likelihood of LFS at 3 years. HDC/IL-2 was not significantly efficacious in young patients (<40 years old). Further studies are underway to define the impact of HDC/IL-2 on immune function and the putative efficacy of therapy in genetic subgroups of AML.

Introduction The high rate of leukemic relapse after the completion of induction and consolidation chemotherapy is a significant clinical concern in AML. Overall, two thirds of adult AML patients who have achieved complete remission (CR) will relapse within 12– 18 months, and the relapse risk is clearly higher in elderly patients[1–2]. The increased risk of disease recurrence in older patients is multi-factorial and explained, in part, by a reduced tolerability to induction/consolidation therapies and a higher incidence of adverse chromosomal and sub-chromosomal aberrations in leukemic cells[3]. With the exception of allogeneic hematopoietic stem cell transplantation (allo-SCT), which is feasible in a fraction of younger patients, no relapse-preventive therapy has been consensually accepted[4], and the standard of care hence is no treatment. The median overall survival after a relapse is approximately 6 months with few long-term survivors, thus emphasizing an urgent need for novel strategies to prevent relapse. AML patients in remission display significant immune dysfunction, which impacts on the relapse risk[5]. Previous attempts to prevent relapse by pharmacological activation of anti-leukemic effector cells, mostly using the T cell- and NK cell-activating cytokine interleukin-2 (IL-2), have met with disappointment[6–10]. Histamine dihydrochloride (HDC; Ceplene ® ) has been developed to counteract leukemia-related immunosuppression in AML and hence to improve the efficiency of IL-2 therapy[11–12]. The combination of HDC and IL-2 strongly activates anti-leukemic functions of T cells and natural killer (NK) cells in the presence of normal and malignant myeloid cells; in this environment, IL-2 alone is inefficacious [13]. These findings have formed the basis for the use of HDC/IL-2 as a relapse-preventing strategy in AML. A phase 3 trial with 320 AML patients in CR demonstrated a significant reduction of relapse risk in patients receiving post-consolidation immunotherapy with HDC/IL-2[14]. Here we report the results of post-hoc analyses to define the impact of age for the relapse prevention by HDC/IL-2 in AML.

Patients, Materials and Methods 320 AML patients were enrolled in an open-label, randomized, multicenter phase 3 study after completion of induction and consolidation therapies. Patients were randomly assigned to either a treatment arm (HDC/IL-2) or a control arm (standard of care, no treatment). Patients in this trial were not candidates for allo-SCT.

Ann Hematol (2011) 90 (Suppl 1):S25–S76

S31 treatment with HDC/IL-2 doubled the likelihood of LFS at 3 years with a trend towards improved OS (fig 1a and 1c). As reported previously, the study did not show a benefit of HDC/ IL-2 among all randomized patients above the age of 60 [5]. However, a non-significant trend towards improved LFS was observed in the subgroup of patients between 60 and 70 years old (p=0.21, HR=0.72, n=77). In addition, HDC/IL-2 significantly improved LFS within the subgroup of patients 40–70 years old. In this patient group the relative increase in LFS exceeded 80% at 3 years with a trend towards an overall survival benefit (fig 1b and 1d). Thus, with the precaution that the presented analyses were based on small numbers of patients, our results imply that AML patients in intermediate age groups (40–70 years) are likely to benefit from HDC/ IL-2 therapy. The reason for the age-related efficacy of HDC/IL-2 should be the focus of further study. An ongoing phase IV trial (Re: Mission Trial) monitors immune response markers during HDC/IL-2 immunotherapy and the impact of therapy on minimal residual disease. Additional udies are underway aiming to clarify the role of HDC/IL-2 in genetically defined forms of AML. Conflict of interest KH is a consultant of EpiCept Corporation. KH holds patent rights to the use of histamine dihydrochloride in Leukemia.

Figure 1. Kaplan-Meier plots of leukemia-free survival (LFS; A and B) and overall survival (OS; C and D) of CR1 patients in defined age groups. Immunotherapy with HDC/IL-2 significantly improved LFS in patients between 40 and 60 years of age (B; p=0.008, HR=0.50, n=113) and in patients between 40 and 70 years (B; p=0.008, HR=0.62, n=190). There was a trend towards increased overall survival in the group of patients between 40 and 60 years of age (C; p=0.09) and in the group of patients 40–70 years old (D; p=0.07). The median follow-up time for all randomized patients was 47 months. All analyses of efficacy were performed according to ITT. In the treatment arm, patients received ten 3-week cycles of HDC/ IL-2, while patients in the control arm received no treatment. Therapy continued for 18 months or until patients relapsed, died, discontinued because of adverse events, withdrew consent, or became lost to follow-up. Results and Discussion We have earlier reported that immunotherapy with HDC/IL-2 significantly prevents relapse in phase 3 trial with 320 AML patients in first or subsequent CR (primary endpoint LFS, p=0.008, HR 0.71, n=320)[14] and that the benefit of HDC/IL-2 is superior in patients in first CR below the age of 60[5]. In a further analysis of the impact of age for the response to HDC/IL-2, it was found that younger patients (<40) did not significantly benefit from treatment (LFS, p>0.5, HR 0.77, n=51). In contrast, in patients between 40 and 60 years old,

References 1. Juliusson G, Billstrom R, Gruber A, Hellstrom-Lindberg E, Hoglunds M, Karlsson K, et al. Attitude towards remission induction for elderly patients with acute myeloid leukemia influences survival. Leukemia 2006; 20:42–7. 2. Wahlin A, Billstrom R, Bjor O, Ahlgren T, Hedenus M, Hoglund M, et al. Results of risk-adapted therapy in acute myeloid leukaemia. A long-term population-based follow-up study. Eur J Haematol 2009; 83:99–107. 3. Appelbaum FR, Gundacker H, Head DR, Slovak ML, Willman CL, Godwin JE, et al. Age and acute myeloid leukemia. Blood 2006; 107:3481–5. 4. Dohner H, Estey EH, Amadori S, Appelbaum FR, Buchner T, Burnett AK, et al. Diagnosis and management of acute myeloid leukemia in adults: recommendations from an international expert panel, on behalf of the European LeukemiaNet. Blood 2010; 115:453–74. 5. Martner A, Thoren FB, Aurelius J, Soderholm J, Brune M, Hellstrand K. Immunotherapy with histamine dihydrochloride for the prevention of relapse in acute myeloid leukemia. Expert Rev Hematol 2010; 3:381–91. 6. Kolitz JE, Hars V, DeAngelo DJ, Allen SL, Shea TC, Vij R, et al. Phase III Trial of Immunotherapy with Recombinant Interleukin-2 (rIL-2) Versus Observation in Patients <60 Years with Acute Myeloid Leukemia (AML) in First Remission (CR1): Preliminary Results from Cancer and Leukemia Group B (CALGB) 19808. ASH Annual Meeting Abstracts 2007; 110:157. 7. Lange BJ, Smith FO, Feusner J, Barnard DR, Dinndorf P, Feig S, et al. Outcomes in CCG-2961, a children’s oncology group phase 3 trial for untreated pediatric acute myeloid leukemia: a report from the children’s oncology group. Blood 2008; 111:1044–53.

S32 8. Pautas C, Merabet F, Thomas X, Raffoux E, Gardin C, Corm S, et al. Randomized study of intensified anthracycline doses for induction and recombinant interleukin-2 for maintenance in patients with acute myeloid leukemia age 50–70 years: results of the ALFA-9801 study. J Clin Oncol 2010; 28:808–14. 9. Baer MR, George SL, Caligiuri MA, Sanford BL, Bothun SM, Mrozek K, et al. Low-Dose Interleukin-2 Immunotherapy Does Not Improve Outcome of Patients Age 60 Years and Older With Acute Myeloid Leukemia in First Complete Remission: Cancer and Leukemia Group B Study 9720. J Clin Oncol 2008. 10. Blaise D, Attal M, Reiffers J, Michallet M, Bellanger C, Pico JL, et al. Randomized study of recombinant interleukin-2 after autologous bone marrow transplantation for acute leukemia in first complete remission. Eur Cytokine Netw 2000; 11:91–8. 11. Hellstrand K, Brune M, Dahlgren C, Hansson M, Hermodsson S, Lindner P, et al. Alleviating oxidative stress in cancer immunotherapy: a role for histamine? Med Oncol 2000; 17:258–69. 12. Hellstrand K. Histamine in cancer immunotherapy: a preclinical background. Semin Oncol 2002; 29:35–40. 13. Hellstrand K, Asea A, Dahlgren C, Hermodsson S. Histaminergic regulation of NK cells. Role of monocyte-derived reactive oxygen metabolites. J Immunol 1994; 153:4940–7. 14. Brune M, Castaigne S, Catalano J, Gehlsen K, Ho AD, Hofmann WK, et al. Improved leukemia-free survival after postconsolidation immunotherapy with histamine dihydrochloride and interleukin-2 in acute myeloid leukemia: results of a randomized phase 3 trial. Blood 2006; 108:88–96.

SYMPOSIUM ARBEITSGEMEINSCHAFT INFEKTIONEN IN DER HÄMATOLOGIE UND ONKOLOGIE (AGIHO)

Ann Hematol (2011) 90 (Suppl 1):S25–S76 patient populations (typically acute myeloid leukemia and allogeneic stem cell transplant patients) (2) knowledge of local incidences and fatality rates due to IFI, (3) use of non-culture based techniques such as Aspergillus galactomannan enzyme immunoassay or beta-D-glucan monitoring from blood samples and (4) straight forward use of imaging procedures, preferably early thoracic computed tomography scan. By utilizing these instruments, non-selective empirical antifungal therapy just for the reason of fever during neutropenia unresponsive to broad-spectrum antibacterials may be replaced by pre-emptive antifungal treatment restricted to patients who have additional clinical, laboratory and/ or imaging findingd indicative of IFI. Prospective studies to compare empirical to pre-emptive antifungal treatment (Maertens et al. 2005, Cordonnier et al. 2009) have shown that withholding systemic antifungal treatment until more than just neutropenic fever is present logically results in a higher number of probable and proven IFI, however, this did not translate into significantly higher fatality rates. Limitations of this pre-emptive approach are the need for prompt availability of diagnostic procedures and the poor positive predictive value of serial galactomannan monitoring results. The latter might be resolved by highly sensitive polymerase chain reaction (PCR)based monitoring, the latter being in process to become validated and standardized internationally. A pragmatic conclusion could be that systemic antifungal chemoprophylaxis might be preferred by centers with an unacceptably high rate of (fatal) IFI, empirical antifungal intervention preferred by centers with limited access to diagnostic instruments to early identify candidates for pre-emptive antifungal treatment, while the latter should be favored by centers with a low or intermediate rate of IFI and have the capability to obtain prompt thoracic CT scans in febrile neutropenic patients, perform bronchoscopy and bronchoalveolar lavage and have access to serial galactomannan or beta-D-glucan monitoring. (Maschmeyer et al. 2009)

Is it Time to Switch from Empirical to Preemptive Treatment of IFI? G. MASCHMEYER Dept. of Hematology, Oncology and Palliative Care, Klinikum Ernst von Bergmann, D-14467 Potsdam, Germany

Conflict of interest Within the past 5 years, George Maschmeyer has served as a consultant for Gilead Sciences, MSD and Pfizer and has been on the Speakers’ Bureau for Gilead Sciences, MSD, Pfizer and Cephalon.

The successful management of invasive fungal infections (IFI) is essential for enabling patients undergoing intensive chemotherapy or allogeneic stem cell transplantation for aggressive haematological malignancies. Infection control measures, systemic antifungal chemoprophylaxis, early empirical antifungal therapy and the availability of new, effective and generally well-tolerated antifungal agents have become the cornerstones of strategies to minimize fatal outcomes due to IFIs. However, the thread of resistance among fungal pathogens to broad-spectrum antifungal azoles, the relevant number of adverse reactions to antifungal drugs, the potential for important drug-drug interactions in patients receiving azole antifungals, and pharmacoeconomic considerations prompt us to identify more precisely which patient definitely needs antifungal therapy, rather than administering these drugs to an unselected cohort of patients only categorized as patients-at-risk for IFI by their underlying disease or the treatment procedure they undergo. Instruments to promote a more selective approach to systemic antifungal treatment are (1) proper identification of high-risk

References Maertens J, Theunissen K, Verhoef G, Verschakelen J, Lagrou K, Verbeken E, Wilmer A, Verhaegen J, Boogaerts M, Van Eldere J. Galactomannan and computed tomography-based preemptive antifungal therapy in neutropenic patients at high risk for invasive fungal infection: a prospective feasibility study. Clin Infect Dis 2005;41: 1242–50 Cordonnier C, Pautas C, Maury S, Vekhoff A, Farhat H, Suarez F, Dhédin N, Isnard F, Ades L, Kuhnowski F, Foulet F, Kuentz M, Maison P, Bretagne S, Schwarzinger M. Empirical versus preemptive antifungal therapy for high-risk, febrile, neutropenic patients: a randomized, controlled trial. Clin Infect Dis 2009;48:1042–51 Maschmeyer G, Beinert T, Buchheidt D, Cornely OA, Einsele H, Heinz W, Heussel CP, Kahl C, Kiehl M, Lorenz J, Hof H, Mattiuzzi G. Diagnosis and antimicrobial therapy of lung infiltrates in febrile neutropenic patients: Guidelines of the infectious diseases working party of the German Society of Haematology and Oncology. Eur J Cancer 2009;45:2462–72

Ann Hematol (2011) 90 (Suppl 1):S25–S76 Epidemiology and Resistance of Fungi: Is the Enemy Changing his Face? P.-M. RATH Institute of Medical Microbiology, University Hospital Essen, Hufelandstraße 55, D-45122 Essen, Germany The true epidemiology of invasive mould infections is not known, although internationally accepted definitions (EORTC/MSG), which might be helpful not only for scientific purposes but also in the clinical routine, have been established and adapted recently. One corner stone of diagnosis of invasive infections is the laboratory diagnosis. For the most common species, Aspergillus fumigatus, the laboratory diagnosis has improved in the last years. For example, the lowered cut-off value of the galactomannan antigen assay in serum samples has increased sensitivity without considerable altering specificity. Testing galactomannan in bronchoalveolar lavage fluids further improves sensitivity. For PCR-based techniques for detecting DNA in blood and bronchoalveolar lavage fluid international approaches for standardization are under way. Furthermore, PCR assays are now commercially available. However, robust data on sensitivity and specificity are lacking up to now. These methods can contribute to our knowledge of the prevalence of invasive aspergillosis. Due to antifungal prophylaxis, the earlier laboratory diagnosis and aggressive treatment with antifungals, the incidence and mortality of invasive aspergillosis has decreased over the last years. However, infections with non-Aspergillus species (i.e. Zygomycetes, Scedosporium/Pseudallescheria spp, Fusarium spp), which are resistant to some commonly used antifungals, have been described in increasing frequency. Furthermore, azole-resistant Aspergillus species have been reported and some data indicate that Aspergillus fumigatus show increased frequency of azole-resistance at least in some countries. In infections with non-Aspergillus species, the diagnosis is more difficult. Antigen assays or DNA-based methods are experimental and therefore not commonly available. Cultural detection and pathohistological examination of biopsies are still the basis of laboratory diagnosis. Lack of experience in culturing and species identification of these up to now rare fungi may result in false-negative results or misidentification of species. Since susceptibility testing is not standardized for these species, it is not established in many laboratories. This might result not only in an inadequate therapy but also in an inaccurate epidemiological data set. If infection with non-Aspergillus moulds or failure of standard therapy in a patient with aspergillosis is suspected, an early and aggressive approach to verify the diagnosis is recommended. A laboratory with experience in the culture, identification and susceptibility testing of non-Aspergillus species should be involved. Furthermore, patient data should be included in international register. Conflict of interest None

SPECIAL LECTURE Genetic Architecture of Leukaemia M. GREAVES Section of Haemato-Oncology, The Institute of Cancer Research, Sutton, United Kingdom Cancer clone expansion is a Darwinian, evolutionary process of genetic diversification and selection—in somatic cells inhabiting a

S33 tissue ecosystem. Childhood acute lymphoblastic leukaemia (ALL), though an intrinsically lethal malignancy, develops over a short time frame and has only modest genetic complexity. It is therefore amenable to interrogation of its pre-clinical evolutionary or natural history, i.e. the timing and sequence of critical mutational events and, ultimately, the causal basis of this process. For the common variant, B cell precursor ALL, we have determined the temporal sequence of mutations. ETV6-RUNX1 fusion (and hyperdiploidy) are commonly pre-natal and presumed initiating events followed by a modest set of secondary copy number alterations (CNA) or sequence-based mutations more proximal to diagnosis. Current whole genome sequencing screens should provide a complete audit of ‘driver’ mutations for ALL. Genetic studies on monozygotic twins with concordant and discordant ALL have been invaluable in these studies as has been modelling studies with murine and human cells. Single cell analysis with multiplexed probes for ETV6-RUNX1 fusion gene and CNA has enabled us to investigate the detailed genetic architecture of clones in ALL. This reveals that the evolutionary trajectory of this cancer (and we believe most others) is non-linear, and with a branching sub-clonal architecture of genetically distinct sub-clones, as anticipated on Darwinian principles. We have extended this analysis to show that the ‘stem’ or propagating cells driving this process are themselves genetically diverse or variegated in individual patients. Current pan-genomic snapshots (SNP arrays or sequencing) of ALL and other cancer cells are largely blind to this underlying heterogeneity. These differing perspectives of genetic architecture in cancer are of some clinical consequence. Conflict of interest None Selected references: Greaves MF, Maia AT, Wiemels JL, Ford AM (2003) Leukemia in twins: lessons in natural history. Blood, 102: 2321–2333. Greaves MF, Wiemels J (2003) Origins of chromosome translocations in childhood leukaemia. Nature Rev Cancer, 3: 639–649. Greaves M (2006) Infection, immune responses and the aetiology of childhood leukaemia. Nat Rev Cancer, 6: 193–203. Hong D, Gupta R, Ancliff P, Atzberger A, Brown J, Soneji S, Green J, Colman S, Piacibello W, Buckle V, Tsuzuki S, Greaves M, Enver T (2008) Initiating and cancer-propagating cells in TEL-AML1associated childhood leukemia. Science, 319: 336–339. Ford AM, Palmi C, Bueno C, Hong D, Cardus P, Knight D, Cazzaniga G, Enver T, Greaves M (2009) The TEL-AML1 leukemia fusion gene dysregulates the TGFβ pathway in early B lineage progenitor cells. J Clin Invest, 119: 826–836. Papaemmanuil E, Hosking FJ, Vijayakrishnan J, Price A, Olver B, Sheridan E, Kinsey SE, Lightfoot T, Roman E, Irving JAE, Allan JM, Tomlinson IP, Taylor M, Greaves M, Houlston RS (2009) Loci on 7p12.2, 10q21.2 and 14q11.2 are associated with risk of childhood acute lymphoblastic leukemia. Nature Genet, 41: 1006–1010. Bateman CM, Colman SM, Chaplin T, Young BD, Eden TO, Bhakta M, Gratias EJ, van Wering ER, Cazzaniga G, Harrison CJ, Hain R, Ancliff P, Ford AM, Kearney L, Greaves M (2010) Acquisition of genome-wide copy number alterations in monozygotic twins with acute lymphoblastic leukemia. Blood, 115: 3553–3558. Greaves M (2010) Cancer stem cells: back to Darwin? Sem Cancer Biol, 20: 65–70.

S34 Anderson K, Lutz C, van Delft FW, Bateman CM, Guo Y, Colman SM, Kempski H, Moorman AV, Titley I, Swansbury J, Kearney L, Enver T, Greaves M. Genetic variegation of clonal architecture and propagating cells in leukaemia. Nature, in press.

Main Sessions (authors in alphabetical order) Molecular Markers in Cytogenetically Normal Elderly Acute Myeloid Leukemia C.D. BLOOMFIELD The Ohio State University Comprehensive Cancer Center, Columbus, Ohio, U.S.A Approximately two-thirds of patients with acute myeloid leukemia (AML) are 60 years or older at the time of diagnosis. Even among de novo older AML the outcome remains poor with only 5–15% achieving long-term survival. Among adults under the age of 60 the cytogenetic and molecular heterogeneity of AML has been extensively studied, particularly in cytogenetically normal (CN) AML, and resulted in strikingly different prognostic groups for whom different therapies are utilized. Even though 45–50% of AML patients over the age of 60 years have CN-AML, relatively few studies have been published on the impact of molecular markers in these patients. Thus we have undertaken a series of investigations of the clinical and biologic significance of molecular markers shown to be important in younger adults with CN-AML in CN-AML patients over the age of 60 years [1–7]. Utilizing pretreatment bone marrow and blood, we have studied a number of molecular markers previously shown to be prognostic in younger adults in relatively large numbers of older CN-AML patients treated with curative intent on Cancer and Leukemia Group B (CALGB) protocols. Herein we will focus on mutations in NPM1, FLT3 and CEBPA and expression of BAALC and ERG. In addition to evaluating the associations of these markers with clinical pretreatment features, molecular characteristics and prognosis, we have performed genome-wide gene and microRNA expression profiling to gain insights into the biology of older CN-AML. Among CN-AML patients over the age of 60 years 56% had NPM1 mutations [1]. In multivariable analyses. NPM1 mutations were independent predictors for higher complete remission (CR) rates (P <.001), longer disease-free survival (DFS, P =.004) and longer overall survival (OS, P <.001). For unknown reasons, the prognostic impact of NPM1 mutations was observed in patients over the age of 70 years (70–83 years) but not in those 60–69 years. Gene- and microRNA-expression profiles associated with NPM1 mutations were similar in both older patient age groups and similar to those in younger (<60 year) adults with CN-AML. These profiles were characterized by upregulation of HOX genes and their embedded microRNAs and downregulation of the prognostically adverse MN1, BAALC, and ERG genes. The data suggest that all CN-AML adult patients with NPM1 mutations regardless of age might benefit from similar therapies. Among CN-AML patients over the age of 60 years 30% had FLT3ITD mutations [2]. As in younger adults in CALGB, FLT3-ITD mutations were not predictive of achievement of CR. However, FLT3-

Ann Hematol (2011) 90 (Suppl 1):S25–S76 ITD mutations were independent predictors for shorter DFS (P <.001) and overall survival (P <.001). In contrast to NPM1 mutations, the prognostic impact of FLT3-ITD mutations was observed in patients 60–69 years but not in those over the age of 70 years. A FLT3-ITDassociated gene-expression signature revealed overexpression of FLT3, homeobox genes (MEIS1, PBX3, HOXB3), and immunotherapeutic targets (WT1, CD33) and underexpression of leukemiaassociated (MLLT3, TAL1) and erythropoiesis-associated (GATA3, EPOR, ANK1, HEMGN) genes. A FLT3-ITD-associated microRNAexpression signature included overexpressed miR-155 and underexpressed miR-144 and miR-451. FLT3-ITD identifies older CN-AML patients with molecular high-risk and is associated with gene- and microRNA-expression signatures that provide biologic insights for novel therapeutic approaches. Twelve percent of older CN-AML patients harbored CEBPA mutations; 3.7% had bi-allelic CEBPA mutations (CEBPA-2mut), and all of these were NPM1 wild-type (NPM1-wt) [3]. No significant differences were observed for outcome among CEBPA-2mut, CEBPA1mut, and CEBPA-wt patients. When restricting analyses to NPM1-wt patients, we still observed no outcome differences between CEBPA1mut and CEBPA-wt patients. However, CEBPA-2mut patients tended to have longer DFS (P=.06) and had a significantly longer OS (P= .03). Expression profiling distinguished CEBPA-2mut from CEBPA-wt patients. Prominent signature features were up-regulation of CEBPA and the hematopoietic progenitor markers CD34, CD38, CD7 and CD96, and down-regulation of the leukemia-associated RUNX1, HOXA and HOXB genes. CEBPA-2mut could also be distinguished from CEBPA-1mut patients. Consistent with the HOX down-regulation, CEBPA-2mut compared to CEBPA-wt and to CEBPA-1mut patients had significant down-regulation of miR-10a, a microRNA embedded in the HOXB gene cluster. Our data suggest that as in younger CN-AML patients bi-allelic CEBPA mutations have a more favorable outcome and represent a distinct genetic entity within CN-AML with NPM1-wt. Among CN-AML patients over the age of 60 years BAALC expression was strongly associated with outcome [4]. In multivariable analyses, low BAALC expression independently predicted for higher CR rates (P<.001), longer DFS (P=.03) and longer OS (P<.001). In multivariable analyses, low ERG expression also independently predicted for longer OS (P<.03)[4]. As noted above, we found a strong prognostic impact of NPM1 mutations on CR rates, DFS and OS in older CN-AML patients [1]. However, in the set of patients investigated here, NPM1 mutation status was retained only in our model for CR. Therefore, we compared different multivariable models for CR attainment, DFS and OS using the Akaike information criterion. We found that the multivariable model for DFS that included BAALC expresser status, FLT3-ITD mutation status and age, excluding NPM1 mutation status, appeared to be better than other evaluated models that included NPM1 mutation status. For OS, we found that a model including only the BAALC and ERG expresser status was better than two other evaluated models that contained NPM1 mutation status. To gain biological insights we derived gene-expression signatures associated with BAALC- and ERG-expression in older CN-AML patients [4]. Furthermore, we derived the first microRNA-expression signatures associated with the expression of these two genes. In low BAALC expressers, genes associated with undifferentiated hematopoietic precursors and unfavorable outcome predictors were downregulated, while HOX-genes and HOX-gene-embedded-microRNAs

Ann Hematol (2011) 90 (Suppl 1):S25–S76 were upregulated. Low ERG expressers presented with downregulation of genes involved in the DNA-methylation-machinery, and upregulation of miR-148a, which targets DNMT3B. We conclude that in older CN-AML patients, low BAALC and ERG expression associates with better outcome and distinct gene- and microRNAexpression signatures that could aid in identifying new targets and novel therapeutic strategies for older patients. Conflict of interest None References 1. Becker H, Marcucci G, Maharry K, Radmacher MD, Mrózek K, Margeson D, Whitman SP, Wu YZ, Schwind S, Paschka P, Powell BL, Carter TH, Kolitz JE, Wetzler M, Carroll AJ, Baer MR, Caligiuri MA, Larson RA, Bloomfield CD(2010) Favorable prognostic impact of NPM1 mutations in older patients with cytogenetically normal de novo acute myeloid leukemia and associated gene- and microRNA-expression signatures: a Cancer and Leukemia Group B study. J Clin Oncol 28(4):596–604. PMID: 20026798. http://jco.ascopubs.org/content/28/4/596.full 2. Whitman SP, Maharry K, Radmacher MD, Becker H, Mrózek K, Margeson D, Holland KB, Wu YZ, Schwind S, Metzeler KH, Wen J, Baer MR, Powell BL, Carter TH, Kolitz JE, Wetzler M, Moore JO, Stone RM, Carroll AJ, Larson RA, Caligiuri MA, Marcucci G, Bloomfield CD (2010) FLT3 internal tandem duplication associates with adverse outcome and gene- and microRNA-expression signatures in patients 60 years of age or older with primary cytogenetically normal acute myeloid leukemia: a Cancer and Leukemia Group B study. Blood 116(18):3622–3626. PMID: 20656931. http://bloodjournal.hematologylibrary.org/cgi/content/ full/116/18/3622 3. Becker H, Marcucci G, Maharry K, Radmacher MD, Wu Y-Z, Mrózek K, Whitman SP, Margeson D, Holland KB, Schwind S, Metzeler KH, Powell BL, Carter TH, Kolitz JE, Wetzler M, Carroll AJ, Baer MR, Moore JO, Caligiuri MA, Larson RA, Bloomfield CD (2010) CEBPA double mutations impact favorably on the outcome of older adults with wild-type NPM1 cytogenetically normal acute myeloid leukemia and are associated with distinct gene and microRNA expression. Fifteenth Congress of the European Hematology Association, Barcelona, Spain, June 10– 13, 2010. Haematologica, 95(suppl. 2):247–248 (abstract #593). http://online.haematologica.org/EHA15/browserecord.php?action=browse&-recid=6663 4. Schwind S, Marcucci G, Maharry K, Radmacher MD, Mrózek K, Holland KB, Margeson D, Becker H, Whitman SP, Wu YZ, Metzeler KH, Powell BL, Kolitz JE, Carter TH, Moore JO, Baer MR, Carroll AJ, Caligiuri MA, Larson RA, Bloomfield CD (2010) BAALC and ERG expression levels are associated with outcome and distinct gene- and microRNA-expression profiles in older patients with de novo cytogenetically normal acute myeloid leukemia: a Cancer and Leukemia Group B study. Blood 116 (25):5660–5669. PMID: 20841507. http://bloodjournal.hematolo gylibrary.org/cgi/reprint/blood-2010-06-290536v1 5. Schwind S, Marcucci G, Holland KB, Mrózek K, Radmacher MD, Maharry K, Becker H, Whitman SP, Wu Y-Z, Carter TH, Powell BL, Caligiuri MA, Baer MR, Larson RA, Bloomfield CD. Higher MN1 expression is an unfavorable prognosticator in older patients (Pts) with cytogenetically normal acute myeloid leukemia (CN-AML): A CALGB study. One-Hundred-First Annual Meeting

S35 of the American Association for Cancer Research, Washington, DC, April 17–21, 2010. Proc Am Assoc Cancer Res, 51:660 (abstract # 2717). http://www.abstractsonline.com/Plan/ViewAb stract.aspx?sKey=3a98c08e-5100-427d-b077-f6c5984fe483&cK ey=c8810c49-daee-4de8-8e66-da436a046dd0&mKey=% 7b0591FA3B-AFEF-49D2-8E65-55F41EE8117E%7d 6. Becker H, Marcucci G, Maharry K, Radmacher MD, Mrózek K, Margeson D, Whitman SP, Paschka P, Holland KB, Schwind S, Wu YZ, Powell BL, Carter TH, Kolitz JE, Wetzler M, Carroll AJ, Baer MR, Moore JO, Caligiuri MA, Larson RA, Bloomfield CD (2010) Mutations of the Wilms tumor 1 gene (WT1) in older patients with primary cytogenetically normal acute myeloid leukemia: a Cancer and Leukemia Group B study. Blood 116 (5):788–792. PMID: 20442368. http://bloodjournal.hematologyli brary.org/cgi/content/full/116/5/788 7. Marcucci G, Maharry K, Wu Y-Z, Radmacher MD, Mrózek K, Margeson D, Holland KB, Whitman SP, Becker H, Schwind S, Metzeler KH, Powell BL, Carter TH, Kolitz JE, Wetzler M, Carroll AJ, Baer MR, Caligiuri MA, Larson RA, Bloomfield CD (2010) IDH1 and IDH2 gene mutations identify novel molecular subsets within de novo cytogenetically normal acute myeloid leukemia: a Cancer and Leukemia Group B study. J Clin Oncol 28(14):2348–2355. PMID: 20368543. http://jco.ascopubs.org/content/28/14/2348.full. Characterization of CALM/AF10 Leukemias ST.K. BOHLANDER Department of Medicine III, University of Munich Hospital, Campus Grosshadern, Marchioninistr. 15, 81377 Munich, Germany, Abstract The CALM/AF10 fusion gene is the result of a balanced t(10;11)(p13; q14) chromosomal translocation, which is found in acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL) and also in malignant lymphoma. The study of the CALM/AF10 fusion protein and its protein interactors has led to a number of important insights into leukemogenesis in general. Overview The CALM/AF10 fusion gene was initially identified and cloned from the myelomonocytic cell line U937. The t(10;11)(p13;q14) translocation in U937 results in the fusion of the putative transcription factor AF10 located at 10p13 to the clathrin-assembly myeloid-lymphoid leukemia gene CALM at 11q14. The CALM/AF10 fusion protein from U937 comprises 648 of the 652 amino acids of CALM and all but the first 81 amino acids of the 1027 amino acids of AF10 (1). The CALM/AF10 fusion is present in a small percentage (<1%) of AML cases. There is no preference for a specific FAB subtype but there is a tendency for the CALM/AF10 fusion to occur in more immature myeloid leukemias. The CALM/AF10 fusion could also be identified in T-ALL and malignant lymphoma (2–4). Interestingly, in T-ALL cases with a T cell receptor gamma/delta rearrangment a high frequency of CALM/AF10 rearrangements was found (5). There is a slight variability in the location of the breakpoints of the CALM/AF10 fusion gene especially within the AF10 gene with some breakpoints as far Nterminal as amino acid 266 of AF10. The N-terminal plant-homeo domain (PHD) of AF10 is absent in all CALM/AF10 fusion proteins. Expression of the CALM/AF10 fusion protein in primary murine bone marrow cells leads to the rapid development of an aggressive

S36 acute leukemia in a mouse bone marrow transplant model within 9–15 weeks post transplant. This CALM/AF10 leukemia model is characterized by a leukemia propagating cell with lymphoid surface markers and a leukemic bulk with myeloid markers (6). Interestingly, transgenic mice which express CALM/AF10 under the control of various hematopoietic specific promoters either do not develop leukemia (immunoglobulin heavy chain enhancer/promoter or the proximal Lck promoter) or they develop leukemia after a median latency of 12 months with a 40–50% penetrance (vav promoter) (7). This implies that it is either critical that CALM/AF10 is expressed at the correct differentiation stage and/or in the correct cell type to cause leukemia or that additional genetic lesions are required for the development of leukemia. Structure function analysis of the MLL/AF10 fusion protein demonstrated that the octapeptide motif/leucine zipper motif (OM/LZ) of the AF10 protein is critical for malignant transformation (8). This motif was also found critical for the transforming abilities of the CALM/AF10 fusion protein. The OM/LZ motif interacts with many proteins including the trancriptional regulator Ikaros (9) which is required for lymphoid differentiation and the histone methyl transferase DOT1L (10). DOT1L specifically methylates lysine 79 of histone H3. We recently discovered that CALM/AF10 expression leads to a global decrease in histone H3 lysine 79 methylation (H3K79). This global H3K79 hypomethylation was observed both after expression of CALM/ AF10 in cell lines and in primary leukemic cells harboring the CALM/ AF10 fusion gene. Furthermore, this gobal H3K79 hypomethylation was associated with an increased genomic instability in cell line models, and patients with the CALM/AF10 rearrangement exhibited an increased rate of additional chromosomal abnormalities. The global H3K79 hypomethylation phenotype upon CALM/AF10 expression was dependent on the presence of the OM/LZ motif in the fusion. These observations suggests that the expression of CALM/AF10 might actively promote the accumulation of secondary, leukemia causing genetic lesions (11). This would also be an explanation for the long latency to leukemia development in transgenic CALM/AF10 models. An increase in genomic instability caused by leukemic fusion genes like activated tyrosine kinases which lead to an increase in reactive oxygen species production (e.g. in BCR/ABL positive cells) or as in the case of CALM/AF10 through genome-wide changes in an epigenetic mark (H3K79 hypomethylation) is emerging as a recurring theme in leukemogensis (12). Conflict of interest None References 1. Dreyling MH, Martinez-Climent JA, Zheng M, Mao J, Rowley JD, Bohlander SK. The t(10;11)(p13;q14) in the U937 cell line results in the fusion of the AF10 gene and CALM, encoding a new member of the AP-3 clathrin assembly protein family. Proc Natl Acad Sci U S A. 1996;93:4804–4809. 2. Dreyling MH, Schrader K, Fonatsch C, Schlegelberger B, Haase D, Schoch C, Ludwig W, Löffler H, Büchner T, Wörmann B, Hiddemann W, Bohlander SK. MLL and CALM are fused to AF10 in morphologically distinct subsets of acute leukemia with translocation t(10;11): both rearrangements are associated with a poor prognosis. Blood. 1998;91:4662–4667. 3. Bohlander SK, Muschinsky V, Schrader K, Siebert R, Schlegelberger B, Harder L, Schemmel V, Fonatsch C, Ludwig WD, Hiddemann W, Dreyling MH. Molecular analysis of the CALM/AF10 fusion:

Ann Hematol (2011) 90 (Suppl 1):S25–S76 identical rearrangements in acute myeloid leukemia, acute lymphoblastic leukemia and malignant lymphoma patients. Leukemia. 2000;14:93–99. 4. Carlson KM, Vignon C, Bohlander S, Martinez-Climent JA, Le Beau MM, Rowley JD. Identification and molecular characterization of CALM/AF10fusion products in T cell acute lymphoblastic leukemia and acute myeloid leukemia. Leukemia. 2000;14:100– 104. 5. Asnafi V, Radford-Weiss I, Dastugue N, Bayle C, Leboeuf D, Charrin C, Garand R, Lafage-Pochitaloff M, Delabesse E, Buzyn A, Troussard X, Macintyre E. CALM-AF10 is a common fusion transcript in T-ALL and is specific to the TCRgammadelta lineage. Blood. 2003;102:1000–1006. 6. Deshpande AJ, Cusan M, Rawat VP, Reuter H, Krause A, Pott C, Quintanilla-Martinez L, Kakadia P, Kuchenbauer F, Ahmed F, Delabesse E, Hahn M, Lichter P, Kneba M, Hiddemann W, Macintyre E, Mecucci C, Ludwig WD, Humphries RK, Bohlander SK, Feuring-Buske M, Buske C. Acute myeloid leukemia is propagated by a leukemic stem cell with lymphoid characteristics in a mouse model of CALM/AF10-positive leukemia. Cancer Cell. 2006;10:363–374. 7. Caudell D, Zhang Z, Chung YJ, Aplan PD. Expression of a CALM-AF10 fusion gene leads to Hoxa cluster overexpression and acute leukemia in transgenic mice. Cancer Res. 2007;67: 8022–8031. 8. DiMartino JF, Ayton PM, Chen EH, Naftzger CC, Young BD, Cleary ML. The AF10 leucine zipper is required for leukemic transformation of myeloid progenitors by MLL-AF10. Blood. 2002;99:3780–3785. 9. Greif PA, Tizazu B, Krause A, Kremmer E, Bohlander SK. The leukemogenic CALM/AF10 fusion protein alters the subcellular localization of the lymphoid regulator Ikaros. Oncogene. 2008;27:2886–2896. 10. Okada Y, Feng Q, Lin Y, Jiang Q, Li Y, Coffield VM, Su L, Xu G, Zhang Y. hDOT1L links histone methylation to leukemogenesis. Cell. 2005;121:167–178. 11. Lin YH, Kakadia PM, Chen Y, Li YQ, Deshpande AJ, Buske C, Zhang KL, Zhang Y, Xu GL, Bohlander SK. Global reduction of the epigenetic H3K79 methylation mark and increased chromosomal instability in CALM-AF10-positive leukemias. Blood. 2009;114:651–658. 12. Popp HD, Bohlander SK. Genetic instability in inherited and sporadic leukemias. Genes Chromosomes Cancer. 2010. Heterogeneity of the AML-Initiating Cell Compartment D. BONNET Haematopoietic Stem Cell Laboratory Cancer Research UK, London Research Institute, UK Key words: Hematopoietic Stem Cell (HSC), Xenotransplantation model, Leukemic Stem Cell (LSC), Cancer Stem Cell (CSC), Selfrenewal, Microenvironment, Niche Abstract An emerging concept in cancer biology is that a subset of cancer cells among the heterogeneous cell mass that constitutes the tumor may drive the growth of the tumor. This so-called “cancer stem cell” (CSC)

Ann Hematol (2011) 90 (Suppl 1):S25–S76 has the ability to self-renewal, differentiate into all the cell types of the original cancer and reinitiate the tumor upon transplantation. Despite of the clear importance of CSCs in the genesis and perpetuation of cancers, little is currently known about the biological and molecular properties that make CSCs distinct from normal stem cells, the developmental and cellular origin of CSCs, and the identification of candidate molecular targets for therapeutic intervention. This report will focus more specifically on the blood-related cancer leukemia, which was the first disease where human CSCs, or leukemic initiating cells (LICs), were isolated. We will summarise our knowledge of LICs notably in Acute Myeloid Leukemia (AML) and discuss different avenues by which we might be able to target these cells. I- Identification and isolation of LICs The existence of such cancer stem cells (CSCs) with self-renewal potential was first documented in leukemias [1,2], but has since then been extended to a wide varieties of solid tumours. The adaptation of xenotransplantation assays to examine the propagation of AML in vivo has allowed the phenotypic identification of the AML-IC. Transplantation of primary AML cells into NOD/ SCID mice led to the finding that only rare cells, termed AMLinitiating cells (AML-IC), are capable of initiating and sustaining growth of the leukemic clone in vivo, and serial transplantation experiments showed that AML-IC possess high self-renewal capacity, and thus can be considered to be the LICs. By LIC we refer to a cell that has self-renewal and differentiation potential and is able to reinitiate the leukemia when transplanted into NOD/SCID. Importantly, AML-IC can be prospectively identified and purified as CD34+/CD38- cells in AML patient samples, regardless of the phenotype of the bulk blast population, and represented the only AML cells capable of self-renewal [1,2]. However, subsequently, considerable heterogeneity has been revealed. Using lentiviral gene marking to track the behaviour of individual LSCs, following serial transplantation, has revealed heterogeneity in their ability to self-renew, similar to what is seen in the normal HSC compartment [3]. Recent evidence from our laboratory suggests that in some patients, AML-ICs could have a progenitor phenotype (CD34+/CD38+) and in patients with NPM mutation, the AML-ICs could be found in the CD34- fraction [4,5]. Based on the heterogeneity of AML in terms of karyotype, differentiation stage of the blasts and clinical outcome, it is not surprising that AML-ICs could be heterogeneous too. Thus the phenotype of AML-ICs is more complex than previously thought and can varied from patient to patient and also probably in the same patients depending on the stage of the disease [4,5]. Indeed, different cells or acquisition of other mutations in the original LSC might occur during the progression of the leukemia. This represents an even more important challenge in the development of LSC specific targeting therapy as LSC, even in the same patient, might represent a moving target.

S37 morphology of the bulk of AML cells is very similar to various stages of hematopoietic progenitor development. However, the LIC reportedly has a phenotype that is very similar to a subset of non-malignant HSCs. This is indirect evidence that relies in the premise that the phenotype of the AML-IC does not change significantly during leukemogenesis. Some reported observations support this hypothesis. For instance, the immunophenotype of primitive haematopoietic cell does not change significantly when transformed with the AML-ETO fusion protein [6]. However, other studies have described profound alterations in immunophenotype upon MLL-ENL and MLL-GAS7 transformations of HSCs [7,8]. One consideration however, is that only bulk immuno-phenotying has been reported, it may be that the LSC in these MLL generated leukemias has a different immunophenotype to the majority of the leukemia. Furthermore, there are more similarities between LIC and HSC than between LIC and progenitors. Hallmark features of both cell types are the ability to self-renew and proliferate extensively. The cellular properties of HSCs are thus very close to the behaviour of LIC and fewer changes are required to transform an HSC into an LSC. Another aspect of the multi-hit hypothesis is that the original hit (potential pre-leukemic stem cell) may occur in one cell type, but the final leukemogenic event may occur in a more differentiated progeny of the original cell. In this situation, the first hit could predispose in which cell the second hit may be leukemogenic. Knowing the nature of cells from where the LIC originates might represent thus an important question on to how to treat leukemia but might not be easily achievable as patients at diagnostic have already a full blown leukemia.

II- Considerations in targeting LIC

3-2 Targeting cell surface markers Cell surface antigens have been used to isolate CSCs in a range of conditions. These surface antigens can be used as potential targets for therapy. CD123 and CD33 are both expressed on AML LSCs and therapies that target these antigens are undergoing clinical trials. One therapy is comprised of diphtheria toxin fused to interleukin 3. Interleukin 3 binds to its receptor (CD123), which is expressed on AML LSCs but not on most normal bone marrow HSCs. This compound inhibits growth of AML in vivo while having a limited effect on normal hematopoietic cells [9]. Monoclonal antibody against CD33 has been conjugated to a cytotoxic agent to form gemtuzumab ozagamicin (GO). Antibody to CD33 is internalised on binding to the cell surface and so brings the cytotoxic agent into the cell. We demonstrate that CD33, CD13 and CD123 are indeed present on most AML-LICs [10]. Nevertheless, these markers thought to be myeloid specific are also expressed on the majority of normal HSCs indicating that careful assessment of antigen expression on both normal HSCs and LSCs will be needed when designing new targeted-therapies to ensure they are selective for LICs. For some diseases the CSCs share a number of features in common with normal HSCs and thus the task of identifying antigens specific for the LIC might be difficult to achieve.

3.1 The cell of origin in leukemia An important question for a more complete understanding of leukemogenesis is the cellular origin of the leukemic stem cell. In human samples, the study of the cellular origin of the leukemia transformation has until now relied in large part on the immunophenotyping of the LICs. It is clear that the immunophenotype and

3-3 Signalling pathways regulating the self-renewal of LICs When a hematopoietic stem cell has entered the cell cycle, two fates are possible: self-renewal and differentiation with associated proliferation. The control of self-renewal is under the control of various regulators. Deregulated self-renewal would involve an expansion of primitive cells and is probably very important in leukemogenesis. The

S38 hedgehog, Wnt and Notch pathways have been implicated in promoting normal stem cell self-renewal [11]. The molecular mechanisms by which all these factors interact to regulate stem and progenitor cell self-renewal remain to be elucidated. The question of whether these signalling molecules play an important role in the regulation of LICs remains mainly unknown. Nevertheless, Bmi-1 has been reported to play a key role in selfrenewal determination in both normal and leukemic murine stem cells [12]. Similarly, the role of Wnt pathway for survival, proliferation and self-renewal of normal HSCs has raised the hypothesis that aberrant Wnt signalling might also contribute to the pathogenesis of AML/CML. Recently, aberrant activation of Wnt signalling and downstream effectors has been demonstrated in AML as well as in CML patients [13]. Thus, it seems that the same molecular pathways that govern the self-renewal in normal stem cell are being usurped by LSCs and other CSCs. Thus, it remains to be seen whether any of this pathway might be used for targeting LSCs. 3-4 LSC and their microenvironment The hematopoietic microenvironment consists of bone cells, stromal cells, macrophages, fat cells and extracellular matrix. Stem cells and their immediate progeny interact with the hematopoietic microenvironment. Stem/progenitor cells adhere to stromal cells via adhesion molecules. Cytokines and chemokines are also retained in the microenvironment; they bind to extracellular matrix and some are presented on the surface of stromal cells. Both cellular as well as extracellular matrix components of the stem cell microenvironment or niche are critical in stem cell regulation. By labelling stem cells and re-infusing them, it has been shown that HSCs reside in close proximity to the endosteal bone surface [14]. Overall, it appears that the regulation of hematopoiesis is the result of multiple processes involving cell-cell and cell-extracellular matrix interactions, the action of specific growth factors and others cytokines as well as intrinsic modulators of haematopoietic development. Currently, the question of whether leukemic stem cells depend on their niche for self-renewal remains unresolved. Both normal and LICs depend on SDF-1/CXCR4 axis for homing [15]. In an AML study, in vivo administration of anti-CD44 antibody to NOD/SCID mice transplanted with human AML significantly reduced leukemic engraftment. Absence of leukemia in serially transplanted mice demonstrated that AML stem cells have been targeted. The mechanisms underlying this eradication included interference with the homing of these LSC to their niche and alteration of LSC fate. This report thus suggests that LSC require interaction with a niche to maintain their stem cell properties [16]. Xenograft experiments provide additional evidence for a supportive LSC niche; few primary leukemias seed organs other than the bone marrow and related hematopoieitc organs (eg spleen). In immunodeficient mice, LICs home to the endosteal region of bone marrow, and leukemia cells spread from here to the rest of the marrow. The LSCs within the endosteal region survive treatment by cytarabine chemotherapy, while cells within the central cavity undergo apoptosis [17]. Thus, building evidences suggest an involvement of a niche for LICs development. It is nevertheless still unclear whether LSCs

Ann Hematol (2011) 90 (Suppl 1):S25–S76 depend on the same niche as HSCs. This has not been proven yet and remains an area of active investigation. It has also been well-documented especially for solid tumor that aberrations could occur in the functional interactions between cancer cells and cells of the microenvironment, a scenario whereby the normal stem cell niche is replaced by the “tumor microenvironment” [18]. Thus, it is important to understand how cancer stem cells are regulated within the context of their natural tumor microenvironment. In the context of LSC it is unclear to what extend the stem cell niche is or not perturbed by leukemic burden but it is extremely worthwhile to explore this issue in future study. Conflict of interest None

References 1. Lapidot T.,et al. (1994) A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature, 367, 645–8. 2. Bonnet D & Dick JE (1997) Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med, 414, 105–111. 3. Hope KJ, Jin L & Dick JE. (2004), Acute myeloid leukemia originates from a hierarchy of leukemic stem cell classes that differ in self-renewal capacity. Nat Immunol, 5, 738–43. 4. Taussig DC, et al. (2008) Anti-CD38 antibody-mediated clearance of human repopulating cells masks the heterogeneity of leukemia initiating cells. Blood, 112, 568–575. 5. Taussig DC, et al. (2010) Leukemia-initiating cells from some acute myeloid leukemia patients with mutated nucleophosmin reside in the CD34-fraction. Blood, 115,1976–84. 6. Delaney C, Bernstein, I.D. (2004). Establishment of a pluripotent preleukemic stem cell line by expression of the AML1-ETO fusion protein in Notch1 immortalized HSCN1cl10 cells. J Haematol.125,353–7. 7. Cozzio A et al. (2003) Similar MLL-associated leukemias arising from self-renewing stem cells and short-lived myeloid progenitors. Genes Dev, 17, 3029–35. 8. So CW, et al. (2003) MLL-GAS7 transforms multipotent hematopoietic progenitors and induces mixed lineage leukemias in mice. Cancer Cell, 3, 161–71. 9. Yalcintepe L et al. (2006) Expression of interleukin-3 receptor subunits on defined subpopulations of acute myeloid leukemia blasts predicts the cytotoxicity of diphtheria toxin interleukin-3 fusion protein against malignant progenitors that engraft in immunodeficient mice. Blood, 108, 3530–3537. 10. Taussig DC et al. (2005) Hematopoietic stem cells express multiple myeloid markers: implications for the origin and targeted therapy of acute myeloid leukemia.106, 4086–92. 11. Zon LI (2008) Intrinsic and extrinsic control of haematopoietic stem-cell self-renewal. Nature, 453:306–13. 12. Lessard J & Sauvageau G. (2003), Bmi-1 determines the proliferative capacity of normal and leukemic stem cells. Nature, 423, 255–260. 13. Jamieson CH et al. (2004) Granulocyte-macrophage progenitors as candidate leukemic stem cells in blast-crisis CML. N Engl J Med, 351, 657–667.

Ann Hematol (2011) 90 (Suppl 1):S25–S76 14. Lo Celso C et al. (2009) Live–animal tracking of individual haematopoietic stem/progenitor cells in their niche. Nature 457, 92–96. 15. Adams GB et al., Stem cell engraftment at the endosteal niche is (2002). The essential roles of the chemokine SDF-1 and its receptor CXCR4 in human stem cell homing and repopulation of transplanted immune-deficient NOD/SCID and NOD/SCID/B2m (null) mice. Leukemia, 16: 1992–2003. 16. Jin L et al. (2006) Targeting of CD44 eradicates human acute myeloid leukemic stem cells. Nat Med, 12, 1167–1174. 17. Ishikawa F et al. (2007) Chemotherapy-resistant human AML stem cells home to and engraft within the bone-marrow endosteal region. Nat Biotech. 25,1315–1321. 18. Clarke MF and Fuller M. (2006) Stem cells and cancer: two faces of eve. Cell, 124,111–1115.

The Relevance of Dose and Dose-Density in Induction Chemotherapy Exemplified by High-Dose Cytarabine and Mitoxantrone containing Regimens in Either Standard Double Induction or Dose-Dense Induction - An Analysis of the AML-CG – J. BRAESS1, K. SPIEKERMANN1, P. STAIB3, K. A. KREUZER2, A. GRÜNEISEN4, B. WÖRMANN5, W.-D. LUDWIG6, H. SERVE12, A. REICHLE8, R. PECENY9, D. ORUZIO10, CH. SCHMID10, X. SCHIEL11, M. HENTRICH11, CH. SAUERLAND16, M. UNTERHALT1, M. FIEGL1, W. KERN13, CH. BUSKE1, ST. BOHLANDER1, A. HEINECKE16, H. BAURMANN14, D.W. BEELEN15, W.E. BERDEL7, TH. BÜCHNER7, AND W. HIDDEMANN1 for the AML-cooperative study group (AML-CG) 1 Klinikum Großhadern der LMU München, Marchioninistrasse 15, 81377 München; 2Universitätsklinikum Köln, Kerpener Straße 62, 50924 Köln; 3St.-Antonius-Hospital Eschweiler, Dechant-DeckersStrasse 8, 52249 Eschweiler; 4Vivantes Klinikum Berlin-Neukölln, Rudower Straße 48, 12313 Berlin; 5Städtisches Klinikum Braunschweig, Celler Straße 38, 38114 Braunschweig; 6Robert-Rössle-Klinik, Campus Berlin-Buch, Lindenberger Weg 80, 13122 Berlin; 7Universitätsklinikum Münster, Albert-Schweitzer-Str. 33, 48129 Münster; 8 Klinikum der Universität Regensburg, Franz-Josef-Strauß-Allee 11, 93053 Regensburg; 9Klinikum Osnabrück, Am Finkenhügel 1, 49028 Osnabrück; 10Zentralklinikum Augsburg, II. Medizinische Klinik, Stenglinstraße, 86156 Augsburg; 11Städtisches Klinikum München, Klinikum Harlaching, Sanatoriumsplatz 2, 81545 München; 12Klinikum der J.W. Goethe-Universität, Theodor Stern Kai 7, 60590 Frankfurt/M; 13MLL Münchner Leukämielabor, Max Lebsche Platz, 81377 München; 14Deutsche Klinik für Diagnostik, Aukammallee 33, 65191 Wiesbaden; 15Universitätsklinikum Essen, Hufelandstrasse 55, 45122 Essen; 16 Institut für Medizinische Informatik und Biomathematik, Universität Münster, 48129 Münster Abstract The overall prognosis of patients suffering from AML has steadily improved over the last three decades. Nowadays, complete remissions are achieved in 60–70% of all patients with long-term disease free survival and potential cure in 25–40% of cases. A more detailed analysis indicates that this progress has mainly been achieved in patients <60 years of age while in older patients little improvements have been obtained (Pulte 2008).

S39 When analyzing the approaches that underlie the progress in AML therapy two major developments appear essential: the intensification of therapy and the improvement of supportive care. The standard induction therapy for AML is still a “3+7” type regimen comprising 3 days of Daunorubicin and 7 days of standarddose Cytosine Arabinoside (AraC) as continuous infusion (Hann 1997; Preisler 1987). Based on cell biologic data, the German AML-CG modified the 3+7 regimen and established the TAD-9 regimen which is the combination of Thioguanine, AraC, and Daunorubicin (Buchner 1982). The TAD-9 regimen resulted in a high CR rate and has been part of the induction strategies of the German AML-CG since 1979. For the dosing of Daunorubicin in particular an improvement in overall survival was demonstrated for full dosing as compared to reduced dosing in patients younger than 60 years (Fernandez 2009) and (but less so) in 60 + patients (Rees 1996; Wiernik 1976; Löwenberg 2009). In an attempt to improve the long-term prognosis of patients with AML, the AML-CG introduced the concept of “double induction”. This strategy is primarily focussed on patients<60 years of age. It consists of two courses of chemotherapy irrespective of the degree of cytoreduction in the bone marrow after the first course with the second course starting on day 21 unless severe complications prohibit its application. This strategy resulted in a significantly longer remission duration and overall survival as compared to standard induction (Buchner 1991). In order to further improve on these results double induction with two courses of TAD 9 was compared to a first course of TAD 9 followed by high dose AraC (HD-AraC) plus Mitoxantrone (HAM) as second course. While no significant differences in outcome were observed for the overall group of patients a favourable effect of HAM was seen in the subgroup of high-risk patients as defined by unfavourable karyotype and/or elevated LDH level and/or residual day 16 bone marrow blasts with an OS at 5 years of 25% vs. 18% (p=0.0118) (Buchner 1999). The subsequently performed comparison of two courses of HAM (HAM/HAM) versus the TAD 9/HAM sequence, however, showed no significant differences between HAMHAM and TAD-HAM in terms of CR rate (71% vs. 65%), RFS at 5 years (35% vs. 29%), and OS at 5 years (32% vs. 30%) (Buchner 2006). While the escalation of drug doses thus obviously has reached a limit, further intensification of therapy by shortening the time interval between induction cycles appeared as a promising new approach. This strategy was first evaluated in patients with relapsed and refractory AML. Based on prior studies by Burke et al. and Archimbaud et al. (Burke 1977; Archimbaud 1991) the HAM regimen was modified into a sequential application of two HAM courses (S-HAM). S-HAM comprises HD-AraC bid on days 1, 2, and Mitoxantrone on days 3, 4; after a rest period of only 3 days the identical sequence is repeated on days 8 and 9 (HD-AraC) and 10 and 11 (Mitoxantrone), respectively. The S-HAM protocol was highly effective in patients with advanced disease (primary refractory or relapsed AML) with a CR rate of more than 50% but was complicated by a high early death rate from infections (Kern 1997). Subsequent supportive therapy with G-CSF, however, reduced the duration of critical neutropenia from 40 to 36 days (p=0.008) and the ED rate from 30% to 21% (not significant) (Kern 1998). First results of dose dense therapy in first line therapy of de-novo AML were gained by a prospective randomised comparison of conventional versus dose dense therapy in children with AML. In the COG (Children Oncology Group) study 2891 dose dense therapy comprising Dexamethasone, Cytarabine, Thioguanine, Etoposide and Rubidomycin (DCTER) given on days

S40 0–4 and 10–14 regardless of response was compared to the standard DCTER regimen given on days 0–4 and 14–18 or later, depending on response. Dose dense treatment resulted in a significantly longer diseasefree and overall survival after 3 years of 55% versus 37% (p=0.0002) (DFS) and 52±6% versus 42±6% (OS), respectively (Woods 1996). In adult patients a French study showed that a similar sequential approach (however not involving high-dose AraC) resulted in a surprisingly low hematological toxicity and a lower cumulative incidence of relapse as compared to conventional induction (Castaigne 2004). These results prompted the AMLCG to assess the efficacy and feasibility of dose-dense therapy with S-HAM in newly diagnosed de novo AML in a phase II study in which two induction cycles are applied over 11–12 days instead of 25–29 days as used in conventional double induction thereby increasing dose density twofold. Supportive therapy with pegfilgrastim was mandatory (Fiegl 2008). In addition, a three step escalation of treatment days was planned to obtain total doses of AraC and Mitoxantrone equivalent to the HAM-HAM arm of double induction therapy (Braess 2009). At the present analysis of 254 de-novo AML patients (excluding acute promyelocytic leukemia) 65% reached a complete remission, 16% a complete remission with incomplete peripheral recovery, 8% had persistent leukemia, 11% succumbed to early death—resulting in an overall response rate of 81%. Kaplan Meier estimated survival at 2 years was 75% for the whole group after S-HAM treatment. Importantly the compression of the two induction cycles into the first 11–12 days of treatment was beneficial for normal hematopoesis as demonstrated by a significantly shortened duration of critical neutropenia of 30 days as compared to 46 days after conventionally timed double induction. The antileukemic efficacy was especially pronounced in the rate of early blast clearance (EBC), because in the S-HAM setting 90% of patients had less than 10% residual blast in the bone marrow on day 18 or 19 which contrasts favourably with the situation of conventional double induction as the historical control where this was only the case in 63% of patients on day 16. In conclusion the dose-dense intensive S-HAM regimen is a highly effective treatment regimen with a response rate of 81% and a low early death rate of 11% in the first 65 days which is most probably due to the short duration of critical neutropenia. In spite of these promising results a prospective randomized comparison of such a dose-reduced but intensely-timed therapy versus conventional double induction is required to prove the potential superiority of the new approach. This randomized trial is ongoing within the next generation of the AMLCG studies (AML-CG 2008). Conflict of interest Research Grant from AMGEN

AML in the Elderly—The AMLCG Experience T BÜCHNER1, WE BERDEL1, B WÖRMANN2, A HEINECKE3, MC SAUERLAND3, and W HIDDEMANN4 1 Dept. of Medicine A, Hematology and Oncology, University of Münster; 2 Dept. of Medicine II, Municipal Hospital Braunschweig; 3Dept. of Medical Informatics and Biomathematics, University of Münster; 4Dept of Internal Medicine III, University of Munich, Germany Introduction In order to quantify the influence of major treatment variables and risk factors we conducted a multicenter trial nationwide with 56 centers in Germany participating.

Ann Hematol (2011) 90 (Suppl 1):S25–S76 Methods Patients at all ages with de-novo or secondary AML were upfront randomized to two different intensity induction regimens containing standard dose TAD followed by high-dose HAM versus two courses of HAM. In the same step patients were randomized to G-CSF priming versus no G-CSF. Also, post-remission therapy by prolonged maintenance chemotherapy versus autologous stem cell transplantation was upfront randomly assigned. Results Table 1 shows the numbers of patients in two age groups, their therapeutic outcome and their profile of risk factors. Table 1: AMLCG 99

Age within age groups Favorable Karyotype Unfavorable Karyotype NPM1/FLT3-ITD Mutation Status

Hazard Ratio Age 16–60Y

(P value) Age 60–85Y

0.6 0.4 1.6 0.3

0.8 0.8 2.2 0.5

(0.005) (< 0.001) (< 0.001) (< 0.001)

(0.009) (0.009) (< 0.001) (< 0.001)

Table 2 contains data of a multivariable analysis for overall survival restricted to risk factors showing significance in younger as well as in older patients.

Table 2: AMLCG 99 Multivariate Analysis of Risk Factors for 5 year Overall Survival

Patients de-novo AML Secondary AML High-risk MDS Favorable Karyotype Intermediate Karyotype Unfavorable Karyotype Normal Karyotype and and NPM1mut/FLT3 ITD +/− other WBC/μl Median LDH U/L Median CR OS at 5Y RR at 5Y

Age 16–60Y 1223 81% 16% 3% 12% 66% 22%

Age 60–85Y 1470 68% 29% 4% 4% 67% 29%

P

< .001

< .001

36% 64% 13.350 420 70% 41% 49%

26% 74% 7.565 342 54% 13% 72%

< .001 < .001 < .001 < .001

In the same setting the treatment variables tested did not result as risk factors for patients overall survival. By the strictly prospective design of the trial a selection of patients by dropouts at the occasion of later randomizations was avoided and a reliable intent-to-treat evaluation was guaranteed.

Ann Hematol (2011) 90 (Suppl 1):S25–S76 Summary and Conclusion We conclude that the outcome of patients with AML is mainly determined by their risk factors representing their individual disease biology and not by current treatment variables. In particular, age emerges as a strong and independent risk factor (1). The setting of early assessable risk factors even allowed to predict the chance of remissions and the risk of death in older patients with AML (2). Conflict of interest None References 1. Age-Related Risk Profile and Chemotherapy Dose Response in Acute Myeloid Leukemia: A Study by the German Acute Myeloid Leukemia Cooperative Group Büchner T, Berdel WE, Haferlach C, et. al. J Clin Oncol 2009; 27:61–69 2. Complete remission and early death after intensive chemotherapy in patients aged 60 years or older with acute myeloid leukaemia: a web-based application for prediction of outcomes Krug U, Röllig C, Koschmieder A, et al. Lancet 2010; 376: 2000–08 (Online Dec 4, 2010)

The MRC Experience in Older Patients with AML A.K BURNETT Cardiff University, United Kingdom In recent years the UK MRC/NCRI Group has developed two treatment strategies for older patients with AML and high risk MDS (>10% marrow blasts) who are older than 60 years. Patients who are considered fit for an intensive approach by their physician enter randomisations of conventional treatment comparisons, while those thought to be unfit enter less intensive randomisations. The age limit of 60 years is not fixed and about 10% of patients enter the contemporary trial for younger patients. Epidemiological studies indicate that only 30–40% of older patients receive chemotherapy, but the indentification of patients who are “unfit” remains controversial. A number of retrospective studies have delineated demographic characteristics associated with a poor response to intensive treatment, but in general these have not been prospectively validated, but more importantly were derived from datasets of patients treated intensively. A similar exercise was undertaken in our AML11 trial involving conventional chemotherapy (Wheatley et al. Brit. J. Haem.). Based on a multivariate analysis a weighted score patients were divided into tertiles with predicted 12 month survivals. This was prospectively validated on the AML14 trial dataset, which1273 were treated intensively, and 217 non-intensively treated patients where the 3 patient group survivals at 12 months were 65%: 40% and 25% for intensive and 40%: 35% and 12% for non-intensively treated patients. While this is informative about a patient treated by one or other strategy, it is not valid to conclude, for example, that poor risk patients should always be treated intensively because randomisation was not balancing for unknown/unmeasured characteristics of the patients. The AML14 trial attempted to tackle this issue by randomising patients where there was uncertainty, to and intensive versus a non-intensive approach. Of 1490 patients entered, only 10 were randomised. When an analysis was done to explore the factors

S41 which determined which strategy was chosen the 3 more influential factors were age, cytogenetics (which was not known at randomisation), and physician. This testifies to the importance of physicians’ perspective of treatment options in the patient population, and the need to document the reasons why patients are considered “unfit”. AML 14 Trial This trial had two treatment streams. For intensive xx questions were addressed i) does MDR modulation with PSC-833 improve outcome ii) is Dauno 50 mg superior to 35 mg dose level iii) is augmented Ara-C dose (200 mg vs 400 mg) beneficial and iv) is a total of 4 courses better than 3. None of these comparisons showed significant difference in survival, and the overall response rate was 62% (54%CR: 8%CRi), and OS was 12% at 5 years. The OS of the CRi subpopulation was inferior to CR (10% vs 20%) with the main reason being an excess deaths in remission. In the non-intensive part of the trial Low Dose Ara-C (LDAC) was compared with Hydroxyurea+ best supportive care (BSC) each +/− ATRA. In this comparison the DEMC closed the because LDAC was significantly superior. This established a baseline standard of care which could be carried forward in future comparisons, however the benefit was limited to the 18% of patients who achieved CR who enjoyed a median of 15 months OS from CR. In the cohort of 17 patients with poor risk cytogenetics, there were no CRs. The conclusion was that in order to gain survival benefit, CR remained a requirement and could act as a useful surrogate endpoint for future comparisons. AML16 Trial The ongoing AML 16 trial persisted with offering an intensive and non-intensive option, with the investigator deciding in consultation with the patient to opt for one or other approach. Based on our preliminary observation of benefit of adding Gemtuzumab Ozogamicin (GO) to conventional chemotherapy in MRC AML15, in which 10% of patients were >60, we decided to pose the same question in induction in older patients. Second we had established that clofarabine was an effective drug that could be combined with both daunorubicin and GO, so the induction question has been DA vs DClo each +/− GO 3 mg/m2. Since to appropriate number of courses in not known, we randomise patients to a total of 3 vs 2. Patients are permitted to undergo RIC transplantation, but the others are randomised to maintenance with Azacitidine or not. To date 1298 patients have been randomised, and the induction questions are complete. The non-intensive option incorporates a “pick a winner” design whereby new treatments are compared with “standard of care”, which is currently LDAC. In stage 1 of each assessment 100 patients are randomised with the aspiration to double the CR rate to >30% as a surrogate for survival benefit. The first 4 new treatments to be compared were: LDAC + Arsenic trioxide/LDAC + Tipifarnib/LDAC + GO/low dose Clofarabine (20 mg/m2 X5). The first two options failed to pass the stage 1 assessment and were stopped. The second two options passed stage 1 ans continues to a phase 3 evaluation with OS as the endpoint. The Clofarabine comparison will conclude in Q2 2011, but the addition of GO, although doubling the remission rate, did not improve overall survival. Within the pick a winner design, there is the option to introduce “drug X” whereby new treatments can be introduced. Recently spacitibine has been introduces and in Q2 2011, voreloxin and LDAC + voreloxin will be added.

S42 Conclusion The majority of patients with AML are older, and represent a formidable therapeutic challange, the first of which is to decide whether it is in a patients interests to be treated intensively with a risk of early death and investment of valuable time with a modest likelihood of durable benefit, or non-intensively with less hospitalisation, treatment related mortality and morbidity, but with al low chance of success. While both strategies are legitimate there is an urgent need for improvement which needs new agents to be assessed rapidly, and therefore novel approaches to trial design, with randomisation remaining the cornerstone, albeit with higher aspirations. Conflict of interest Receipt of research support from Pfizer. Consultancy with Pfizer. Selected References Wheatley K, Brookes C L, Howman A J et al. Brit J. Haem. (2009) 145: 598–605 Burnett A K, Milligan D, Wheatley K et al. Brit. J. Haem. (2009) 145: 318–332 Burnett A K, Milligan D, Prentice A G Cancer (2007) 109: (6) 1114– 1124 Burnett A K, Hills R K, Wheatley K et al. Blood (2010) 108: (11) 10a. Promiscuity of Leukemic Stem Cells NAIDU. VEGI AND C. BUSKE Institute of Experimental Cancer Research, Comprehensive Cancer Center Ulm, University Hospital Acute myeloid leukemia (AML) is a heterogenous disease with a wide variety of morphological and immunophenotypic characteristics. One of the most important determining factors for this heterogeneity is the genotype, which is also one of the strongest prognostic factors in this disease. In the last years experimental evidence has clearly shown that beside genetic alterations the nature of the leukemic stem cell (LSC) has a major impact on the biology of AML (Bonnet and Dick, 1997) (Guan et al., 2003)(Cozzio et al., 2003; Huntly and Gilliland, 2004). In this respect it is intriguing to speculate that the morphological and immunophenotypic diversity observed in AML patients might be at least partly based on the biology of the LSC. For example it is known for a long time that many AML cases express lymphoid antigens. This is interpreted as an aberrant expression of a lymphoid antigen on a malignant transformed myeloid cell. However, it is conceivable that this lymphoid antigen expression just reflects the biology of the LSC, which itself might have lymphoid characteristics or at least the capacity to develop into a myeloid bulk with lymphoid features in this given AML case. In our recent work we already showed in a murine model of CALM/AF10 positive AML, that a LSC with lymphoid characteristics can robustly generate AML in transplanted mice (Deshpande et al., 2006): In this model mice transplanted with BM expressing CALMAF10 developed AML. These AML cases were characterized by a leukemic bulk population expressing only myeloid markers, a subpopulation expressing both myeloid markers and the lymphoidassociated marker B220, but also by a minor population of cells

Ann Hematol (2011) 90 (Suppl 1):S25–S76 expressing B220 alone. All AML cases displayed IgH genomic rearrangements and showed promiscuous myeloid/lymphoid factor expression such as EBF, E2A and MPO. By applying limiting dilution transplantation and single cell assays we could prove that the LSC in this model is B220 positive and negative for myeloid markers. These findings would have clinical implications if the situation would be similar in patients with AML as such a LSC could be targeted by antilymphoid antibodies, which would not harm the normal stem cell pool. CALM-AF10 leukemias are rare. However, more frequent AML subtypes such as the AML1-ETO positive AML are also positive for lymphoid antigens in the majority of cases: in our series at least 9 of 10 patients with AML1-ETO positive AML expressed at least one lymphoid antigen (Schessl et al., 2005). These results were well supported by findings from other groups that showed overexpression of the lymphoid transcription factor Pax5 in AML1-ETO positive AML (Tiacci et al., 2004), which explains the expression of CD19 in these cases, a known downstream target of Pax5 (Busslinger, 2004). Of note, in one of our mouse models testing the collaboration of AML1-ETO with the FLT3 length mutation, we consistently observed the development of acute lymphoid leukemias beside the development of AML (Schessl et al., 2005). Another example for AML cases with lymphoid characteristics is the subgroup of AML patients with silenced CEBPA (n=6) which typically express T-cell markers like CD7 and CD3 (Wouters et al., 2007). In MLL translocated AML clonal TCRd, TCRg and IgH DJ rearrangements were observed in 32% of the cases (Dupret et al., 2005). In acute promyelocytic leukemia 11% of patients showed aberrant expression of the B-cell marker CD19 (Guglielmi et al., 1998). When 36 APL patients (including 15 M3v) were analysed for T-cell rearrangements (TCR), the majority of M3v APLs expressed PTCRA transcripts or immature TCR rearrangements in parallel to the expression of T-cell markers such as CD2 and CD3 (Chapiro et al., 2006). Future experiments have to show whether we are able to prove that in patients with AML LSCs with lymphoid characteristics exist and are responsible for the lymphoid characteristics of the leukemic bulk. If so this would open a door for strategies targeting lymphoid antigens on AML LSC. Conflict of interest None References Bonnet, D., and Dick, J.E. (1997). Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med 3, 730–737. Busslinger, M. (2004). Transcriptional control of early B cell development. Annu Rev Immunol 22, 55–79. Chapiro, E., Delabesse, E., Asnafi, V., Millien, C., Davi, F., Nugent, E., Beldjord, K., Haferlach, T., Grimwade, D., and Macintyre, E.A. (2006). Expression of T-lineage-affiliated transcripts and TCR rearrangements in acute promyelocytic leukemia: implications for the cellular target of t(15;17). Blood 108, 3484–3493. Cozzio, A., Passegue, E., Ayton, P.M., Karsunky, H., Cleary, M.L., and Weissman, I.L. (2003). Similar MLL-associated leukemias

Ann Hematol (2011) 90 (Suppl 1):S25–S76 arising from self-renewing stem cells and short-lived myeloid progenitors. Genes Dev 17, 3029–3035. Deshpande, A.J., Cusan, M., Rawat, V.P., Reuter, H., Krause, A., Pott, C., Quintanilla-Martinez, L., Kakadia, P., Kuchenbauer, F., Ahmed, F., et al. (2006). Acute myeloid leukemia is propagated by a leukemic stem cell with lymphoid characteristics in a mouse model of CALM/AF10-positive leukemia. Cancer Cell 10, 363–374. Dupret, C., Asnafi, V., Leboeuf, D., Millien, C., Ben Abdelali, R., Preudhomme, C., Beldjord, K., Delabesse, E., and Macintyre, E. (2005). IgH/TCR rearrangements are common in MLL translocated adult AML and suggest an early T/myeloid or B/myeloid maturation arrest, which correlates with the MLL partner. Leukemia 19, 2337–2338. Guan, Y., Gerhard, B., and Hogge, D.E. (2003). Detection, isolation, and stimulation of quiescent primitive leukemic progenitor cells from patients with acute myeloid leukemia (AML). Blood 101, 3142–3149. Guglielmi, C., Martelli, M.P., Diverio, D., Fenu, S., Vegna, M.L., Cantu-Rajnoldi, A., Biondi, A., Cocito, M.G., Del Vecchio, L., Tabilio, A., et al. (1998). Immunophenotype of adult and childhood acute promyelocytic leukaemia: correlation with morphology, type of PML gene breakpoint and clinical outcome. A cooperative Italian study on 196 cases. Br J Haematol 102, 1035– 1041. Huntly, B.J., and Gilliland, D.G. (2004). Blasts from the past: new lessons in stem cell biology from chronic myelogenous leukemia. Cancer Cell 6, 199–201. Huntly, B.J., and Gilliland, D.G. (2005). Leukaemia stem cells and the evolution of cancer-stem-cell research. Nat Rev Cancer 5, 311–321. Miyamoto, T., Iwasaki, H., Reizis, B., Ye, M., Graf, T., Weissman, I. L., and Akashi, K. (2002). Myeloid or lymphoid promiscuity as a critical step in hematopoietic lineage commitment. Dev Cell 3, 137– 147. Schessl, C., Rawat, V.P., Cusan, M., Deshpande, A., Kohl, T.M., Rosten, P.M., Spiekermann, K., Humphries, R.K., Schnittger, S., Kern, W., et al. (2005). The AML1-ETO fusion gene and the FLT3 length mutation collaborate in inducing acute leukemia in mice. J Clin Invest 115, 2159–2168. Tiacci, E., Pileri, S., Orleth, A., Pacini, R., Tabarrini, A., Frenguelli, F., Liso, A., Diverio, D., Lo-Coco, F., and Falini, B. (2004). PAX5 expression in acute leukemias: higher B-lineage specificity than CD79a and selective association with t(8;21)-acute myelogenous leukemia. Cancer Res 64, 7399–7404. Wouters, B.J., Jorda, M.A., Keeshan, K., Louwers, I., ErpelinckVerschueren, C.A., Tielemans, D., Langerak, A.W., He, Y., YashiroOhtani, Y., Zhang, P., et al. (2007). Distinct gene expression profiles of acute myeloid/T-lymphoid leukemia with silenced CEBPA and mutations in NOTCH1. Blood 110, 3706–3714.

Results of Study AML-BFM 2004 Support the Need for Further Risk-Adapted Treatment Options U. CREUTZIG1 AND D. REINHARDT2 Dept. of Pediatric Hematology/Oncology University of Muenster1 and Medical School Hannover², Germany

S43 Abstract Acute myeloid leukaemia (AML) in children has become a curable disease with survival rates in the range of 60%(Kaspers and Creutzig 2005). This success was only possible by a stepwise intensification of chemotherapy from study to study. Results of Study AML-BFM 2004 improved compared to the previous study (overall survival 73%) due to therapy intensification, better supportive care, and further due to improved treatment of patients with relapse or nonresponse. Future treatment concepts will consider even more individual risk factors. Therefore, new, highly sensitive diagnostic methods including immunphenotyping, cytogenetics and molecular genetics are necessary.

Introduction The main aim in pediatric AML is to improve outcome without increasing acute and longterm toxicity. Since 1978, the AML-BFM study group has performed six consecutive multicenter studies— initially in Germany and later on including Austria, Switzeland and the Czech Rebublic(Creutzig et al. 2005). During that period, 5-year overall survival (OS) improved from 41% to 73%.

Purpose and Methods The main aim of study AML-BFM 2004 is to improve prognosis in children and adolescents with AML (<18 years) by intensification of chemotherapy by (1) the randomized introduction of liposomal daunorubicin (L-DNR) in a higher equivalent dose than idarubicin during induction (L-DNR 80 mg/m²/day/3x) in comparison to the standard induction with idarubicin 12 mg/m²/day/3x (equivalent dose 60 mg/m²), each combined with cytarabine and etoposide (L-DNR offers the possibility to increase cumulative dosages of anthracyclines with lower cardiotoxicity) and (2) randomised introduction of 2-chloro-2-deoxyadenosine (2-CDA, 2x6mg/ m²) as intensification during the cytarabine/idarubicin (AI) consolidation for high-risk (HR) patients. (3) The 3rd aim was to verify the equivalence of a prophylactic CNS irradiation by randomizing a total dose of 18 Gy vs. 12 Gy (randomization over two study periods). Patients were stratified into a standard- (SR) or high-risk (HR) group according to morphology, cyto-/molecular genetics, and therapy response at day 15(1)(Creutzig et al. 1999). Later on, FLT3-ITD positivity was defined as a stratification criterion to the HR group.

Results Results improved compared to the previous study AML-BFM 98: Overall survival estimates at 5 year (pOS) in 582 patients (excluding myeloid leukemia in Down syndrome) were 73%±3% vs. 65%±2% (AML-BFM 04 n=582 vs. AML-BFM 98 n=472), plogrank =0.002 (Figure 1a); the 5-year event-free survival (pEFS) was 56%±2% vs. 50%±2%, plogrank =.14 (Figure 1b). Results in the 202 SR patients (35% of the patients) were excellent: pOS 90%±2% vs. 79%±3% (n=182), plogrank =.003, EFS 72%, ± 3% vs. 65%±4%, plogrank =.16. Results in the 380 HR patients also improved: pOS 63%±3% vs. 56%±3% (n=290), plogrank =.02, EFS 47%, ± 3% vs. 41%±3%, plogrank = .23. OS improvement was partly due to better results after treatment of relapse or nonresponse(Sander et al. 2010) (3-year pOS after nonresponse/relapse in 192 patients of study 2004=42%±5% vs. 32%±3% in 198 patients in AML-BFM 98, plogrank =.006).

S44

Ann Hematol (2011) 90 (Suppl 1):S25–S76

a) 1.0 0.9 0.8

0.73, SE=0.02

0.7 0.6 P

0.65, SE=0.02

0.5

aml04_rand1.tab 17DEC10

0.4 0.3 0.2 0.1 Log-Rank p = .0022

0.0 0

1

2

3

4

5

6

7

8

9

10

11

12 13 years

AML-BFM 98 (N=472, 173 events) AML-BFM 04 (N=582, 125 events)

b) 1.0 0.9 0.8 0.7 0.6 P

0.56, SE=0.02

0.5 0.50, SE=0.02

aml04_rand1.tab 17DEC10

0.4

rates by adding L-DNR during induction and 2-CDA during HR consolidation, we will continue to use these agents in the forthcoming AML-BFM study. Cranial irradiation (RT): As significantly less relapses occurred after randomly assigned cranial irradiation in study AML-BFM 87, RT was continued in the following AML-BFM studies. However, results improved in general considerably irrespective of RT since study −87 due to intensification of chemotherapy. Since the results of other international pediatric AML studies indicate that prophylactic CNSRT does not improve outcome in the context of current chemotherapy regimens and because intensified chemotherapy elements with improved CNS efficiency such as high dose cytarabine and liposomal daunorubicin have been introduced in the AML-BFM treatment schedule after study −87, the next trial will replace CNS-RT by a triple intrathecal therapy in order to reduce irradiation related longterm sequelae(Creutzig et al. 2010). The improvement of overall results is not only due to more intensive chemotherapy, but also based on a high standard of experience regarding both the treatment itself and supportive care in the participating hospitals. Intensity of chemotherapy can hardly be increased in the future. Therefore, new therapy options including targeted therapy will be introduced in the forthcoming study. Consequently, the new treatment concepts are more geared to individual risk factors which implicates that results of immunphenotyping, cytogenetics and molecular genetics have to be available for all patients shortly after diagnosis. *Supported by the Deutsche Krebshilfe e.V.

0.3 0.2

Conflict of interest None

0.1 Log-Rank p = .14

0.0 0

1

2

3

4

5

6

7

8

9

10

11

12 13 years

AML-BFM 98 (N=472, 240 events) AML-BFM 04 (N=582, 229 events)

Figure 1: Estimated probability of (a) survival (pOS), (b) event-free survival (pEFS) in patients of studies AML-BFM 98 and-2004, SE = standard error. Results for the 1st randomization L-DNR vs. idarubicin during induction were similar (pOS 76%±4% vs. 73%±4%, plogrank =.27, pEFS 60%±4% vs. 55%±3%, plogrank =.23). There were less early deaths (3 vs. 7 patients) and less treatment related deaths in remission in the L-DNR group (1 vs. 4 patients). The rate of severe infection was slightly lower with L-DNR (pFisher 0.14). Two L-DNR vs. 5 idarubicin patients showed grade III/IV cardiotoxicity after induction. Results of the 2nd randomization in HR patients (AI/2-CDA vs. AI) were also similar: pEFS 52%±4% vs. 52%±4%, plogrank =.96. Toxicity rates of the intensification with 2-CDA were tolerable. Results of the 3rd randomization showed that that there was no disadvantage for patients irradiated with a reduced CNS dose of 12 Gy regarding OS, EFS and rate of relapse(Creutzig et al. 2010). Conclusion Results of study AML-BFM 2004 show notable improvement compared to the previous study AML-BFM 98. This may be attributable to therapy intensification/increased anthracycline doses by introduction of liposomal daunorubicin, better supportive care and improved treatment of patients with relapse or nonresponse. Given the reduced toxicity of L-DNR and slightly better survival

References Creutzig U, Zimmermann M, Bourquin JP, Dworzak M, Fleischhack G, von Neuhoff C, Sander A, Schrauder A, von Stackelberg A, Ritter J, Stary J, Reinhardt D (2010) Preventive CNS Irradiation in Pediatric Acute Myleoid Leukemia: Equal Results by 12 Gy or 18 Gy in Studies AML-BFM 98 and 2004. Pediatr.Blood Cancer in press. Creutzig U, Zimmermann M, Ritter J, Henze G, Graf N, Löffler H, Schellong G (1999) Definition of a standard-risk group in children with AML. Br.J.Haematol. 104, 630–639. Creutzig U, Zimmermann M, Ritter J, Reinhardt D, Hermann J, Henze G, Jurgens H, Kabisch H, Reiter A, Riehm H, Gadner H, Schellong G (2005) Treatment strategies and long-term results in paediatric patients treated in four consecutive AML-BFM trials. Leukemia 19, 2030–2042. Kaspers GJ, Creutzig U (2005) Pediatric acute myeloid leukemia: international progress and future directions. Leukemia 19, 2025–2029. Sander A, Zimmermann M, Dworzak M, Fleischhack G, von Neuhoff C, Reinhardt D, Kaspers GJ, Creutzig U (2010) Consequent and intensified relapse therapy improved survival in pediatric AML: results of relapse treatment in 379 patients of three consecutive AML-BFM trials. Leukemia 24, 1422–1428. Monitoring of NPM1 Mutations in AML Therapy K. DÖHNER and J. KRÖNKE Department of Internal Medicine III, University of Ulm, Ulm, Germany Background Mutations in the nucleophosmin 1 (NPM1) gene represent the most frequent gene mutations in acute myeloid leukemia (AML) with an

Ann Hematol (2011) 90 (Suppl 1):S25–S76 overall incidence of about 25–35%. Cytogenetic subgroup analysis revealed that NPM1 mutations mainly occur in cytogenetically normal (CN)-AML (45–62%). Moreover, NPM1 mutations are significantly associated with FLT3 internal tandem duplication (ITD) mutations; approximately 40% of the mutated cases concurrently carry FLT3ITDs. Clinically, NPM1 mutations are correlated with features like myelomonocytic or monocytic morphology, high CD33 and low or absent CD34 expression. Beside these aspects, NPM1 mutations are of major prognostic relevance since several studies have identified the NPM1mut/FLT3-ITDnegative genotype to be associated with a relatively favourable prognosis. Due to its distinct biological and clinical characteristics, NPM1 mutated AML has been included in the recent WHO classification as a provisional entity. Based on these aspects, NPM1 mutations represent an excellent marker for monitoring minimal residual disease (MRD). In recent years, several groups have shown the applicability of an RQ-PCR based MRD assay for quantification of NPM1mut RNA or DNA (Gorello P et al., 2006; Chou WC et al., 2007; Papadaki C et al., 2009; Schnittger S et al., 2009). So far, there are only scarce data on the prognostic value of NPM1mut MRD levels and none of the studies was performed in controlled AML patient cohorts treated on prospective clinical trials. Aims Within our AMLSG studies we aimed to evaluate the prognostic value of NPM1mut MRD levels in younger (16–60 years) AML patients harbouring NPM1 mutations type A, B or D, and to assess the influence of concurrent FLT3 internal tandem duplications (ITD). Methods All patients were enrolled in the prospective AMLSG 07-04 and AML HD98A treatment trials. Treatment comprised double induction therapy with ICE (idarubicin, cytarabine, etoposide) followed by high-dose cytarabine-based consolidation, autologous or allogeneic stem cell transplantation. Levels of NPM1mut expression ratios, defined as NPM1mut copies per 104 ABL copies, were determined by RQ-PCR using TaqMan technology. Dilution series showed a maximum sensitivity of 10−6 and high specificity as no wildtype NPM1 could be detected. Results A total of 1079 samples, [bone marrow (BM), n=1062; peripheral blood, n=17) from 212 patients were analyzed at diagnosis, after each treatment cycle, during follow-up and at relapse (median number of samples per pt, n=4; range, 1–16). NPM1mut expression ratios at diagnosis varied between 1.1×104 and 10.4×106 (median, 6.9x105). Pretreatment transcript levels were not associated with clinical characteristics (e.g., age, white cell counts, BM blasts) and did not impact on relapse-free (RFS) and overall survival (OS). Following the first induction cycle, the median decrease of the MRD level ratio normalized to pretreatment levels was 4.21×10−3, independent of the presence of concurrent FLT3-ITD (p=0.39). After the 2nd induction cycle, the median reduction of MRD levels was significantly stronger in the FLT3-ITDnegative group (6.75×10−5) compared with the FLT3ITDpositive group (4.19×10−4) (p=0.003) and this differential effect was observed throughout consolidation therapy. For evaluation of the prognostic impact of NPM1mut MRD levels, we compared patients achieving PCR-negativity with those with positive values at different checkpoints. The first reliable checkpoint was after double-induction therapy: the cumulative incidence of

S45 relapse (CIR) at 4 years of PCR-negative patients (n=27) was 0% compared with 48% (SE, 4.4%, p<0.00001) for PCR-positive patients (n=105). This translated into a significant better OS (p=0.0005). The second checkpoint was after completion of consolidation therapy (first measurement during follow-up period). Again, 4-year CIR was significantly (p<.00001) lower in the PCR-negative group with 11% (SE, 6.5%) compared with 51% (SE, 4.8%) in PCR-positive patients, again translating in a significantly better OS (p<.00001). In addition, the level of NPM1mut expression ratio at any time point examined after completion of therapy correlated with the risk of relapse, since 20 of 22 pts with a value above 1000 NPM1mut/104 ABL copies relapsed after a median interval of 90 days (range, 11–709 days). The remaining 2 pts had increasing levels at last follow-up but are still in continuous complete remission (CR). Of note, in a few cases relapse prediction appeared to be limited due to inadequate increase of NPM1mut expression levels or to loss of NPM1 mutation at the time of relapse (n=5). On the other hand, we observed a number of pts (n=17) in continuous CR who had intermittent low (<1000 NPM1mut/104 ABL copies) NPM1mut expression ratios. Conclusions The levels of NPM1mut expression at two distinct checkpoints, after double induction therapy and after completion of consolidation therapy, can be used as a prognostic marker in NPM1mut AML pts. The adverse outcome of patients carrying a concurrent FLT3-ITD is reflected by a significant lower reduction of tumor burden. MRD assessment in NPM1mut AML patients should be incorporated in future clinical trials to determine the value of pre-emptive treatment strategies. Conflict of interest None

Recent Results of the French Belgian Swiss Acute Promyelocytic Leukemia (APL) Group L. ADES AND P. FENAUX Service d’Hématologie Clinique, Hôpital Avicenne, Assistance Publique Hôpitaux de Paris, and Université Paris 13, France I) Is AraC Required in the Treatment of standard risk APL? Long term Results of a Randomized Trial (APL 2000) The combination of ATRA and anthracycline based chemotherapy (CT) is the reference induction and consolidation treatment of newly diagnosed APL 1 2. Whereas in high risk pts (ie with baseline WBC> 10 G/L), AraC is often considered useful in combination with an anthracycline to prevent relapse 3,4, CT with idarubicin alone appears sufficient to yield very low relapse rates in standard risk APL (with WBC<10 G/L) 5. On the other hand our APL2000 trial, where standard risk pts were randomized between ATRA with DNR + AraC and ATRA with DNR without AraC, was prematurely terminated after the first interim analysis due to significantly more relapses and shorter survival in the arm without AraC 4. We reevaluated those results, 6 years after the last patient inclusion. In APL 2000 trial newly diagnosed APL pts<60 years with WBC <10,G/L were randomized between the AraC+ group: induction : ATRA 45 mg/m2/d until CR with DNR 60 mg/m2/dx3 and AraC 200 mg/m2/dx7 started on day 3; first consolidation with the same CT course, second consolidation with DNR 45 mg/m2/dx3 and

S46 AraC 1 g/m2/12 h x4d; maintenance during two years with intermittent ATRA (15 d/3 months)and continuous 6 MP + MTX, and the AraC- group (same treatment, but without AraC). Other pts: Pts<60 years with WBC >10 G/l (high WBC Group) were not randomized but received the AraC + group treatment, but with higher AraC dose during the second consolidation 4. The current analysis was made at the reference date of January 2010, 72 months after inclusion of the last pt. Overall, 340 pts entered APL 2000 trial between July 2000 and Feb, 2004. The AraC+ and AraC- groups (95 and 101 pts, resp) were well balanced for all pretreatment characteristics. TheCR rate was similar in both groups. The 5-year cumulative incidence of relapse was significantly higher, EFS significantly lower in the AraC— than in the AraC+ group, while there was a trend for lower OS in the AraC—group. In the high WBC group (where there was no randomization and all pts received AraC), outcome appeared paradoxically superior to that of standard risk pts treated without AraC 4. Thus, with longer follow up, our findings suggest that, in standard risk APL, avoiding AraC for chemotherapy may lead to an increased risk of relapse, at least when the anthracycline used is DNR. Our results caution against the use in standard risk APL of very effective treatment regimens without AraC like the PETHEMA 99 trial, but where DNR would be substituted for idarubicin 5. II) Arsenic trioxide (ATO) in the consolidation treatment of Newly Diagnosed APL First interim analysis of a randomized Trial (APL 2006) ATRA combined to anthracycline based chemotherapy (CT) is the reference treatment of newly diagnosed APL, but this treatment is myelosuppressive and may be associated with long term cardiac toxicity. The use of ATO may allow reduction of the amount of CT (and in particular avoid ara C), and further diminish the relapse risk, especially when used in consolidation treatment (US intergroup study,Powell, ASCO 2008). In a randomized trial, we used ATO for consolidation treatment instead of ara C in standard risk APL (baseline WBC< 10 G/L), and in addition to AraC in high risk patients (baseline WBC>10 G/L). ATRA, suggested to reduce the relapse risk when used during consolidation was also tested in standard risk pts6. In APL 2006 trial (started in Nov, 2006) newly diagnosed APL patients (pts)< 70 years with WBC < 10 G/L were randomized between: group A1(standard group) (induction : ATRA 45 mg/m2/d until CR with Ida 12 mg/m2/dx3 and AraC 200 mg/m2/dx7 started on day 3; first consolidation with the same CT course, second consolidation with Ida 12 mg/m2/dx3 and AraC 1 g/m2/12 h x4d; maintenance during 2 years: intermittent ATRA 15d/3 months and continuous 6 MP + MTX,); Group A2 :same treatment as group A1, but AraC replaced by ATO 0.15 mg/Kg/D D1 to 25 days of both consolidation courses; Group A3 : same treatment as group A1, but AraC replaced by ATRA 45 mg/m2 d1 to 15 of both consolidation courses,. Pts aged<70 with WBC>10 G/L (Group C) were randomized between: group C1 (standard group): same treatment as Group A1; group C2: same as C1, but with addition of ATO 0.15 mg/Kg/d d1 to 25 of both consolidation courses, at d1. Pts with WBC>10 G/L

Ann Hematol (2011) 90 (Suppl 1):S25–S76 all received intrathecal CT for CNS prophylaxis. This first interim analysis was made at the reference date of 1 Jan 2010, in 186 pts included before 2009 (141 pts in group A (45/45/51 pts in A1/A2/A3 arms), 45 pts in group C (24/21 pts in C1/21 arms)). In standard risk patients 99.3% patients achieved CR. Only 2 patients had relapsed... In high risk pts (group C) 45 (100%) patients achieved CR. Only 2 had relapsed . Results of this first interim show that very high CR rates (>95%) can be observed in very multicenter trials in APL, by combining ATRA and anthracycline based CT, while the relapse rate with consolidation and maintenance was very low in all treatments arms, including in pts with WBC>10 G/L. Nevertheless ATO, when combined to high dose CT during consolidation cycles, increased myelosuppression. An amendment further reducing CT in pts receiving ATO is thus being implemented in the trial. III) Outcome of APL in children and adolescents APL is rare in children (approximately 10% of childhood AML, and 10% of APL) and is characterized by a higher incidence of WBC> 10 G/L, microgranular M3 variant (M3v),and bcr2-3 isoforms of PML-RAR rearrangement than in adults. ATRA combined to chemotherapy (CT) is generally considered to give similar overall results in children and adults7,8,9. However, whether children age has an impact on APL outcome is unknown. We compared disease characteristics and outcome of children aged <=12, adolescents (ados), aged 13–18 and adults, aged > 18 included in 2 multicenter trials of the European APL group (APL 93 and 2000) with a median follow up of 10 and 5 years, respectively (resp). In both trials, pts received induction therapy with ATRA (45 mg/m2/d until CR) and DNR (60 mg/m2/d x3d)+/− AraC (200 mg/m2/dx7) followed by consolidation with a similar course and a final DNR (45 mg/m2/d x 3) +/− AraC (1–2 g/m2/12 h x 8) course with or without maintenance with intermittent ATRA and/or continuous low dose 6MP+MTX, during 2 years. ATRA dose during induction was reduced in children to 25 mg/m2/d only in case of side effects. In APL93 trial, pts with WBC<5 G/l were randomized to receive ATRA + CT or ATRA followed by CT during induction, and maintenance was also randomized. In APL2000 trial, pts with WBC<10 G/l were randomized to receive or not Ara C during induction and consolidation courses, while all pts received maintenance and pts with WBC>10 G/L received intrathecal CNS prophylaxis. Of the 833 consecutive newly diagnosed APL pts aged <60y included in the 2 trials, 3%, 7% and 90% were aged <=12 (children), 13-18 (ados), and>18 (adults), resp. Children had significantly higher WBC counts and incidence of M3v. CR rates and incidence of differentiation syndrome did not differ according to. The 5 year cumulative incidence of relapse (CIR) was significantly higher in children.The higher incidence of relapse in children almost exclusively concerned those aged 4 years or less, who could however all be durably salvaged by auto or allogeneic SCT Our results suggest that adolescents with APL have an outcome at least as favorable as that of adults. Children appear to have more relapse, however mainly observed in children <4 years, and correlated with pre-treatment risk factors (especially high WBC counts). Young children may require reinforcement of first line treatment, including possible addition of ATO .

Ann Hematol (2011) 90 (Suppl 1):S25–S76 IV) Therapy related APL (tAPL). Prospective analysis of etiological factors in recent cases, and comparison with de novo cases Our group 10 and others have reported tAPL to occur at increased frequency (from 5% of APL between 1984 and 1993 to 22% between 1994 and 2000 in our experience), generally less than 3 years after a primary neoplasm (mainly breast carcinoma) treated with the combination of anthracyclines (anthr) and cyclophosphamide (CY), radiotherapy (RT) and less often VP 16, but our last published series analyzed patients (pts) diagnosed before 200111,12. In addition, characteristics and outcome of tAPL are considered similar to those of de novo APL, but all reported comparisons were retrospective. We took advantage of our current multicenter APL 2006 trial that includes both de novo and tAPL, to analyze possible recent changes in etiological factors of tAPL cases, and to prospectively compare them with de novo APL. In APL 2006 trial, pts <70 years with WBC<10 G/l, received ATRA + Idarubicin + AraC for induction, followed by 2 consolidation courses with (randomized) Ida + AraC, ida + ATO or Ida + ATRA. Pts <70 y with WBC>10 G/l, received the same induction treatment, followed by 2 consolidation courses with (randomized) ida + AraC or Ida + AraC + ATO. Pts >70y received the same induction course with reduced dose of Ida and two consolidation courses (Ida + ATO and ATO + ATRA). All patients received 2 y maintenance therapy combining continuous 6MP + MTX and intermittent ATRA. APL 2006 trail included 15% tAPL. By comparison to our tAPL diagnosed before 2001 (JCO 2003, 21, 2123) the primary tumor tended to be less often breast carcinoma and more often prostate carcinoma and other solid tumors. Treatment of the primary tumor, in APL 2006 trial cases was more often a combination of RT and CT included more hormone therapy than in previous cases, and no more included mitoxantrone. In APL 2006 trial, characteristics of tAPL were similar to those of de novo APL, with regards to WBC count and M3 variant. However, tAPL were older and had a higher frequency of bcr3 breakpoint. The CR rate was similar in tAPL and in de novo APL . The 18-month cumulative incidence of relapse and OS were similar in tAPL and de novo APL

V) Central nervous system (CNS) at first relapse in APL. A report on two multicenter trials CNS relapse is rare in APL, and predominates in patients (pts) with increased WBC counts13,14. We report incidence, risk factors and outcome of CNS relapse in 2 multicenter clinical trials of the European APL group (APL 93 and 2000 trials). In both trials, induction treatment consisted of ATRA (45 mg/m2/d until CR) and DNR 60 mg/m2/d x3 + AraC 200 mg/m2/dx7 followed by a first similar consolidation course and a second course with DNR 45 mg/m2/d x3 and AraC 1 g/m2/12hx8 (except in one arm of APL 2000 trial where pts with WBC<10 G/L received no AraC for induction and consolidation). A 2 year maintenance treatment, consisting of intermittent ATRA (15d/3 months) and continuous 6MP + MTX was administered or not (randomized) in APL 93 trial, and systematic in APL 2000 trial. In APL 2000 trial, intrathecal CNS prophylaxis with MTX + AraC (5 injections) was made during

S47 consolidation cycles in all pts with baseline WBC>10 G/L. Median follow up was 10 and 5 years, respectively, in APL 93 and APL 2000 trial. Six first relapses involving the CNS were seen, representing 3% of the relapses. Median age was similar in pts who had or not CNS relapse. Median WBC count was higher in pts with CNS relapse (p=0.07). There was a trend for more CNS relapses in patients who received no AraC, while the prophylactic role of intrathecal CNS prophylaxis was less clear in our experience. The pts with CNS relapse were all salvaged by intrathecal CT, combined to ATRA and systemic CT in APL 93, and ATO in APL 2000 trial. Thus CNS relapse was very rare in APL in our experience was generally associated to concomitant marrow relapse (sometimes only molecular) and there was a possible preventive role of HD AraC. Finally, outcome of CNS relapses was better than in our previous report13, possibly due in part to the recent treatment of relapses by ATO, known to penetrate the blood brain barrier15 and which could also play a role in CNS prophylaxis when administered during first line treatment. Conflict of interest Honoraria for conference and research support: Roche, Cephalon. References 1. Fenaux P, Chastang C, Chevret S, et al. A randomized comparison of all transretinoic acid (ATRA) followed by chemotherapy and ATRA plus chemotherapy and the role of maintenance therapy in newly diagnosed acute promyelocytic leukemia. The European APL Group. Blood. 1999;94:1192–1200. 2. Sanz MA, Grimwade D, Tallman MS, et al. Guidelines on the management of acute promyelocytic leukemia: Recommendations from an expert panel on behalf of the European LeukemiaNet. Blood. 2008. 3. Ades L, Sanz MA, Chevret S, et al. Treatment of newly diagnosed acute promyelocytic leukemia (APL): a comparison of FrenchBelgian-Swiss and PETHEMA results. Blood. 2008;111:1078–1084. 4. Ades L, Chevret S, Raffoux E, et al. Is cytarabine useful in the treatment of acute promyelocytic leukemia? Results of a randomized trial from the European Acute Promyelocytic Leukemia Group. J Clin Oncol. 2006;24:5703–5710. 5. Sanz MA, Montesinos P, Vellenga E, et al. Risk-adapted treatment of acute promyelocytic leukemia with all-trans retinoic acid and anthracycline monochemotherapy: long-term outcome of the LPA 99 multicenter study by the PETHEMA Group. Blood. 2008;112:3130–3134. 6. Sanz MA, Martin G, Gonzalez M, et al. Risk-adapted treatment of acute promyelocytic leukemia with all-trans-retinoic acid and anthracycline monochemotherapy: a multicenter study by the PETHEMA group. Blood. 2004;103:1237–1243. 7. de Botton S, Coiteux V, Chevret S, et al. Outcome of childhood acute promyelocytic leukemia with all-trans-retinoic acid and chemotherapy. J Clin Oncol. 2004;22:1404–1412. 8. Ortega JJ, Madero L, Martin G, et al. Treatment with all-trans retinoic acid and anthracycline monochemotherapy for children with acute promyelocytic leukemia: a multicenter study by the PETHEMA Group. J Clin Oncol. 2005;23:7632–7640. 9. Testi AM, Biondi A, Lo Coco F, et al. GIMEMA-AIEOP AIDA protocol for the treatment of newly diagnosed acute promyelocytic leukemia (APL) in children. Blood. 2005.

S48 10. Beaumont M, Sanz M, Carli PM, et al. Therapy-related acute promyelocytic leukemia. J Clin Oncol. 2003;21:2123–2137. 11. Montesinos P, Gonzalez JD, Gonzalez J, et al. Therapy-related myeloid neoplasms in patients with acute promyelocytic leukemia treated with all-trans-retinoic Acid and anthracycline-based chemotherapy. J Clin Oncol;28:3872–3879. 12. Mistry AR, Felix CA, Whitmarsh RJ, et al. DNA topoisomerase II in therapy-related acute promyelocytic leukemia. N Engl J Med. 2005;352:1529–1538. 13. de Botton S, Sanz M, Chevret S, et al. Extramedullary relapse in acute promyelocytic leukemia treated with all-trans retinoic acid and chemotherapy. Leukemia. 2006;20:35–41. 14. Montesinos P, Diaz-Mediavilla J, Deben G, et al. Central nervous system involvement at first relapse in patients with acute promyelocytic leukemia treated with all-trans retinoic acid and anthracycline monochemotherapy without intrathecal prophylaxis. Haematologica. 2009;94:1242–1249. 15. Au WY, Tam S, Kwong YL. Entry of elemental arsenic into the central nervous system in patients with acute promyelocytic leukemia during arsenic trioxide treatment. Leuk Res. 2008;32:357–358.

Azacitidine Therapy for Elderly Patients with AML C. GARDIN Hématologie Clinique, Hôpital Avicenne (AP-HP), Université Paris 13, France Groupe Francophone des Myélodysplasies (GFM) and ALFA AML Group Background The benefits of intensive chemotherapy (ICTx) remain debated in this older age group. In 2009, the Swedish Acute Leukemia Registry convincingly demonstrated that an intensive chemotherapy approach remains the best currently available option, as it showed that older AML patients treated by intensive chemotherapy in various Swedish regions always had a better outcome and lower early death rates than patients receiving only supportive care (including hydroxyurea or low-dose cytarabine). In this registry study, performance status and age were recorded, but not cytogenetics. What approach should be offered to elderly patients with AML unlikely to tolerate intensive therapy or unlikely to benefit from it, such as patients with adverse cytogenetics, remains unclear. In the UK NCRI non-intensive AML 14 trial, low-dose cytarabine was found of benefit, at least in patients with normal cytogenetics, when compared to best supportive care, including use of hydroxyurea. The hypomethylating agents, azacitidine and decitabine have shown a significant survival benefit in high-risk MDS, including in patients with 20–30% marrow blasts, compared to conventional care, which allowed use of low dose cytarabine in the AZA 001 trial. In the EU, this lead to azacitidine approval for use in patients with high-risk MDS or AML with <30% marrow blasts. As its off-label use is likely to increase, the role of AZA in the current management of older AML patients, especially those with >30% marrow blasts, must be evaluated.

Methods An AZA compassionate program (ATU) was initiated in France in Dec 2004 for higher risk MDS and AML considered not to be candidate or being refractory to intensive chemotherapy. We retrospectively analyzed WHO AML patients having received at least 1 cycle of AZA, until Dec 2008, from 43 centers who provided

Ann Hematol (2011) 90 (Suppl 1):S25–S76 complete reporting of their ATU registered patients, excluding those previously treated by ICTx, allo SCT, low dose cytarabine or a hypomethylating agent. We compared characteristics and outcome of such 124 AML patients aged 65y+, treated upfront with AZA, to those of 403 non CBF-AML patients treated prospectively in our last ICTx ALFA-9803 trial in older AML patients. Risk factors including performance status (PS), age, WBC, prior MDS, and cytogenetics were analyzed, with overall survival (OS) and 3-month death rate as endpoints. We also analysed outcome of AZA treated patients, according to our ALFA risk–index, derived from the analysis of older patients with AML treated with ICTx in the ALFA 9803 trial and validated in an independent AML cohort. Results As expected, patients selected by their physicians to receive AZA as a front-line therapy for AML, had higher median age (75 vs 71 years) (P<.001), lower median WBC count (3.1 vs 5.0 G/L, P=.02), more frequent prior MDS (43% vs 16%, P<.001) and adverse cytogenetics (43% vs 19%, P<.001), than patients who received ICTx, but similar PS (72% having PS score of 0 or 1). Despite these differences, our ICTx-derived risk index, based on adverse cytogenetics, PS, WBC count, and age was also a powerful predictor of OS after AZA (P<.001). In patients with a poor-risk index, the 3-month death rate was 38% in both cohorts, compared to only 18% and 14% in AZA and ICTx patients with a low-risk index, respectively. AZA patients with PS >1, WBC >15 G/L (best cut-off for AZA), and adverse cytogenetics did poorly, while post-MDS AML and age had no prognostic impact in this cohort. After restricting the comparison to patients with WBC≤15 G/L, 18 m-OS was 31% (20–43) in the AZA vs 40% (33–47) in the ICTx cohort (P=.22), with similar early death rates (27% vs 23%). No differences in outcome appeared between AZA and ICTx cohorts when focusing on post-MDS AML, normal cytogenetics (18-m OS, 64% [35–82] vs 49% [39–59], P=0.27), or in patients with adverse cytogenetics, including those with isolated +8, abn(5q) or-7/del7q (18m OS, 10% [3–23] vs 15% [6–28], P=.69). Conclusions This comparison suggests that usual AML and host risk factors defined for ICTx also have a similar impact after AZA. If one excludes patients with WBC >15 G/L who did poorly with AZA, outcomes seem equivalent, even in cytogenetically normal AML. A controlled trial is thus warranted to compare AZA and ICTx in those older patients. Presence or absence of multiple gene mutations have also been shown to be of pronostic value in older AML patients treated with ICTx. Similar studies in the context of frontline treatment with AZA (including mutations of genes involved in epigenetic modifications of AML cells, such as TET2, DNMT3A, EZH2 or possibly IDH1, 2) should be of great interest, as presence of acquired TET2 mutations was recently shown to influence response rate to AZA in MDS. Conflict of interest Celgene, Inc Genzyme, Inc Honoraria and grant support for academic research. References: Juliusson G, Antunovic P, Derolf A et al. Age and acute myeloid leukemia : Real world data on decision to treat and outcomes from the Swedish Acute Leukemia Registry. Blood 2009; 113: 4179–4187.

Ann Hematol (2011) 90 (Suppl 1):S25–S76 Burnett AK, Milligan D, Prentice AG et al. A comparison of low-dose cytarabine and hydroxyurea with or without all-trans retinoic acid for acute myeloid leukemia and high-risk myelodysplastic syndrome in patients not considered fit for intensive treatment. Cancer 2007; 109: 1114–1124. Gardin C, Turlure P, Fagot T et al. Postremission treatment of elderly patients with acute myeloid leukemia in first complete remission after intensive induction chemotherapy: Results of the multicenter randomized Acute Leukemia French Association (ALFA) 9803 trial. Blood 2007; 109: 5129–5135. Fenaux P, Mufti GJ, Hellström-Lindberg E et al. Azacitidine prolongs overall survival compared with conventional care regimens in elderly patients with low bone marrow blast count acute myeloid leukemia. J Clin Oncol 2010; 28: 562–569. Malfuson JV, Etienne A, Turlure P, et al.Risk factors and decision criteria for intensive chemotherapy in older patients with acute myeloid leukemia. Haematologica. 2008 Dec;93(12):1806–13 S. Thépot, R. Itzykson, V. Seegers, et al. Azacytidine (AZA) as First Line Therapy in AML: Results of the French ATU Program. Blood (ASH Annual Meeting Abstracts), Nov 2009; 114: 843. C. Gardin, R. Itzykson, S. Thépot, et al. C. Frontline azacitidine (AZA) or intensive chemotherapy (ICTx) in older AML patients. J Clin Oncol 28:15s, 2010 (suppl; abstr 6530) R. Itzykson, O. Kosmider, T. Cluzeau et al. Presence of TET2 Mutation Predicts A Higher Response Rate to Azacitidine In MDS and AML Post MDS. Blood (ASH Annual Meeting Abstracts), Nov 2010; 116: 439 Disclosures: Grant support and honoraria from Celgene,inc

Discovery of Novel Mutations in Myelodysplastic Syndrome and Acute Myeloid Leukemia by whole Genome Sequencing T. GRAUBERT1,2, L. DING3,4, M. WALTER1,2, D. LARSON4, D. SHEN4, J. DIPERSIO 1,2, E. MARDIS3,4, R. WILSON 3,4, AND T. LEY1,2,3,4. 1 Department of Medicine, 2Siteman Cancer Center, 3 Department of Genetics, 4The Genome Center, Washington University School of Medicine, St. Louis, MO, USA Myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML) are clonal hematopoietic disorders that are caused by mutations arising in hematopoietic stem cells. The mutations necessary for initiation and progression of these tumors are not fully understood. Candidate gene resequencing is neither cost-effective nor sufficiently high throughput for mutation discovery on a genome-wide scale. Our group has utilized whole genome sequencing as an unbiased tool to comprehensively interrogate cancer genomes for novel mutations. We first applied this approach to a patient with normal karyotype AML [1]. To differentiate somatically acquired mutations from inherited polymorphisms, the tumor (bone marrow) and normal (skin) genomes were sequenced in parallel to provide >90% diploid coverage. We identified 10 mutations with predicted translational consequences in protein-coding genes. Two of these were previously described (FLT3, NPM1). The other 8 were novel, although none were recurrent in an additional 187 AML genomes. This first experiment provided proof of concept that whole genome sequencing using primary cells from cancer patients is feasible and can lead to the discovery of novel mutations.

S49 We took a similar approach to characterize the genome of a second patient with AML [2]. A similar number of mutations in proteincoding genes were identified (n=12), including two known AML genes (NRAS, NPM1). In addition, we identified a missense mutation in IDH1 (R132C), the gene encoding isocitrate dehydrogenase 1. This gene was recurrently mutated in 15/187 additional AML cases. These findings have been validated by several other groups and extended to show that mutations of IDH1 and the related family member IDH2 are mutated in 10–20% of AML cases. In the most recent study, we returned to the first AML patient and obtained improved coverage of the tumor and normal genomes using longer, paired-end sequence reads [3]. This led to the identification of a 1 basepair deletion in DNMT3A, the gene encoding DNA methyltransferase 3A. We then surveyed a large panel of AML samples (n=281) and found recurrent DNMT3A mutations in 22.1% of cases. DNMT3A mutations are enriched in cases with intermediate risk AML (33.7%) and are mutually exclusive with favorable risk cytogenetics. In both univariate and multivariate analyses, DNMT3A mutations were associated with poor prognosis (inferior overall and event-free survival) in patients with de novo AML. The genomes of de novo AML samples that have been sequenced so far have a fairly consistent number of mutations in the coding region of genes (~10–20 per genome) and mutations elsewhere in conserved, noncoding and non-repetitive regions of the genome (~500–750 per genome). When these nucleotide positions are sequenced at increased depth to obtain more quantitative measurements of the proportion of mutant:wildtype alleles, the mutant allele frequency in most cases is ~50%, suggesting that these tumors are monoclonal. In contrast, the genomes of 8 MDS-derived secondary AML (20AML) samples studied so far are uniformly oligoclonal. This suggests that the evolution from MDS to 20AML occurs by stepwise accumulation of mutations in single cells that gain clonal dominance over ancestral clones. This may account for the older age of onset and increased rate of chemotherapy resistance that are typical of this subtype of AML. Conclusions Whole genome sequencing of primary cancer specimens is feasible and can yield new clinically-relevant information and insight into the biology of cancer. Most mutations in cancer genomes are not recurrent, and therefore are likely to be background mutations that do not contribute to the phenotype of the tumor. Hundreds of samples from each tumor type will need to be studied in order to learn all the “rules of cancer.” Falling costs for sequence production and increasingly robust and automated analysis tools will help make cancer genome sequencing on this scale a reality in the near future. Conflict of interest None References 1. Ley TJ, Mardis ER, Ding L, Fulton B, McLellan MD, Chen K, Dooling D, Dunford-Shore BH, McGrath S, Hickenbotham M et al.: DNA sequencing of a cytogenetically normal acute myeloid leukaemia genome. Nature 2008, 456(7218):66–72. 2. Mardis ER, Ding L, Dooling DJ, Larson DE, McLellan MD, Chen K, Koboldt DC, Fulton RS, Delehaunty KD, McGrath SD et al.: Recurring Mutations Found by Sequencing an Acute Myeloid Leukemia Genome. N Engl J Med 2009, 361(11):1058–1066.

S50 3. Ley TJ, Ding L, Walter MJ, McLellan MD, Lamprecht T, Larson DE, Kandoth C, Payton JE, Baty J, Welch J et al.: DNMT3A mutations in acute myeloid leukemia. N Engl J Med 2010, 363 (25):2424–2433.

TCR-Directed Immunotherapy of Leukemia PH.D. GREENBERG 1 , 2 , 3 , G. RAGNARRSON 1 , 2 , TH. M. SCHMITT 2,3 , S. OCHSENREITHER 2,3 , R. M. TEAGUE 2,3 , C. CHOU2,3,4, C. FOWLER2,3,4, A. SCHIETINGER2,3, INGUNN M. STROMNES 2,3 , J. N. BLATTMAN 3 , D. H. AGGEN 5 , D.M. KRANZ5, J. KUBALL6 1 Department of Medicine, University of Washington, Seattle WA, USA; 2Program in Immunology, Fred Hutchinson Cancer Research Center, Seattle WA, USA; 3Department of Immunology, University of Washington, Seattle WA, USA; 4Medical Scientist Training Program, University of Washington, Seattle WA, USA; 5Department of Biochemistry, University of Illinois, Urbana IL, USA; 6Department of Hematology, University Medical Center Utrecht, Utrecht, The Netherlands Strategies have been developed to identify candidate leukemia antigens that have the potential to be targeted for attack by T cells. In general, such antigens are identified through molecular methods that distinguish the differential expression of genes in leukemia cells compared to normal hematopoietic cells as well as other normal somatic cells. The ideal target antigen would be one that is unique to leukemic cells, expressed in leukemic stem cells, not found in normal tissues, and essential to the leukemic phenotype so that escape variants would be unlikely to occur. Some of the proteins created by translocations, such as from BCR-Abl in CML, meet these criteria, but offer only a few potential novel epitopes at the site of the gene fusion and have generally proven to be weakly immunogenic. Thus, more recently, efforts have focused on identifying prooncogenic proteins that are not necessarily unique to the leukemia but are over-expressed in leukemic cells, found at only very low levels in normal tissues, and that contribute to leukemogenesis. One such protein is the oncogene, WT1, which is detected in the cells of most patients with leukemia. It is a zinc finger transcription factor that interacts with many cellular targets, has been shown to bind and regulate mRNA and to transactivate or repress a wide range of genes, and is essential during embryogenesis for genitourinary tract development, but after birth expression in normal tissues is limited to low levels in a few cell types, including stromal cells in the kidney, testes, ovary, lung mesothelial cells, and CD34+ hematopoietic stem cells. Studies performed in collaboration with Ravi Majeti and Irv Weissman at Stanford University to assess gene expression patterns have affirmed that WT1 is not only detected in leukemic blasts but also found in relative abundance in leukemic stem cells. Leukemic cells express approximately 10-1000x more WT1 protein than normal CD34+ cells, with purified putative leukemic stem cells expressing the highest levels among the leukemic subpopulations. Most importantly, CD8 T cells can distinguish, based on avidity for complexes of WT1 peptides with MHC Class I molecules, leukemic cells that express higher numbers of such peptide/MHC complexes than normal cells. As a result several groups have pursued strategies to immunologically target WT1. The most straightforward approach has been vaccination to induce CD8 T cell responses to WT1 epitopes, and several dramatic

Ann Hematol (2011) 90 (Suppl 1):S25–S76 responses have been reported, including a complete remission in a patient with progressive leukemia, but most patients have failed to demonstrate clinical benefit. There are many potential reasons for the lack of a clinical response to a vaccine, including poor immunogenicity of the vector and/or the WT1 antigen, or the immuno-compromised state of the patient being vaccinated, that can result in responses of only limited magnitude. To overcome this limitation, our lab has been pursuing adoptive T cell therapy of leukemia patients who have relapsed following an allogeneic hematopoietic cell transplant (HCT). Briefly, WT1specific CD8 T cell clones are being isolated from the HLA-matched sibling HCT donor, expanded to large numbers in vitro, and infused into patients who have relapsed after HCT. This approach makes it possible to achieve very high numbers of WT1-specific T cells in recipients, thus overcoming a critical obstacle in most vaccinated patients. In vivo tracking of the infused CD8 T cells has demonstrated that the cells can traffic to and accumulate in the bone marrow, the primary site of disease, and preliminary results have suggested that the cells can mediate anti-leukemic activity in some patients. However, similar to vaccination, not all patients appear to benefit, and one reason for the failure of T cell therapy, as well as of vaccination, has been the inability to generate a WT1-specific CD8 T cell response of sufficiently high avidity to effectively target the patient’s leukemia. As cloning of T cell responses allows selection and expansion for infusion of the highest avidity WT1-reactive T cells present in the donor repertoire, this result suggests that the low level of expression of WT1 in normal tissues commonly results in the deletion of the highest avidity T cells present in the repertoire. Thus, to uncover the potential efficacy of immunologically targeting WT1, it will likely be essential to identify strategies to increase the avidity of the infused T cells. One approach to reproducibly generate high avidity T cells in every donor is to tranduce CD8 T cells with a TCR specific for WT1 that has been demonstrated to have a high affinity for the WT1 epitope complexed with a Class I molecule. This TCR transfer approach is currently being pursued in other diseases targeting antigens other than WT1, with the largest experience being in the treatment of melanoma. Such studies have revealed several potential problems that need to be resolved to maximize the potential efficacy and safety of this approach. Firstly, the introduced Vα and Vβ TCR chains have the potential to pair not only with each other but with the endogenous Vβ and Vα chains respectively resulting in creation of new receptors of unknown and possibly auto-reactive specificities. Such autoreactivity from mismatced receptor chains has not yet been observed in human clinical trials, but has been well-demonstrated in preclinical mouse models evaluating TCR gene therapy. Our laboratory has described a strategy to greatly reduce the risk of such mismatched pairing of TCR chains by introducing point mutations in the ectodomains of the introduced Vα and Vβ TCR chains to create a new inter-chain disulfide bond [1], and this method is currently being employed in most clinical trails, including the trial we are initiating to treat leukemia with a WT1-specific TCR. Another critical issue that must be addressed is that the transduced T cells need to exhibit high avidity for the leukemic targets. For the clinical trial we are currently designing, we have screened more than 1000 WT1-specific CD8 T cell clones from normal donors to identify the cell with the highest affinity TCR we can detect in normal repertoires. This CD8 T cell clone has much higher avidity than the T

Ann Hematol (2011) 90 (Suppl 1):S25–S76 cell clones that we have infused into patients in our ongoing clinical trial. However, achieving high avidity in transduced T cells requires more than just introduction of a high affinity TCR, but rather also requires stable high level expression of the TCR. Several strategies have been developed to achieve this, including codon optimization of the TCR chain genes to improve translation, introduction of the cysteine mutations described above to enhance proper chain pairing, linking expression of both chains to achieve coordinate expression, and improvements in the vectors and gene constructs used for transduction, including the use of lentiviral vectors that also provide a better safety margin by decreasing the low risk of insertional mutagenesis leading to transformation identified with retroviral vectors. Our preliminary data suggests that, with the use of these combined strategies, the CD8 T cells transduced with our high affinity WT1-specific TCR achieve similar avidity for WT1+ targets as the parental CD8 T cell clone from which the TCR was derived. It still remains possible that a TCR isolated from the normal repertoire will be of inadequate affinity to eliminate leukemia in some patients, particularly those with leukemias that exhibit only slightly increased levels of WT1 expression. Therefore, we have also been pursuing strategies to generate higher affinity TCRs by mutagenesis of the selected TCR chains prior to tranduction. One approach that we have demonstrated to be of potential general use has been the introduction of point mutations in the ectodomains of the Vα and Vβ TCR chains to remove N-glycosylation sites. Such mutations reduce chain glycosylation, which has the potential to affect the flexibility, movement, and interactions of surface molecules, and thus can interfere with TCR aggregation and multimerization. We have shown that such point mutations improve TCR binding, and can increase T cell functional avidity approximately ½ to 1 log10, with demonstrably improved recognition of leukemic cells and with no evidence that this increased avidity results in enhanced recognition of the normal tissues expressing low levels of WT1 [2]. However, it may still be necessary to further increase the affinity of naturally isolated TCRs to assure recognition of leukemic cells. Therefore, we have been collaborating with David Kranz’ lab, which has developed technologies to display TCRs as single chains on the surface of yeast or on immortalized T cell lines. Prior to expression of the TCR chains, saturation mutagenesis of the CDR3 regions of the TCR chains can be performed to create mutated TCR libraries, and the expressed library can be efficiently screened and selected for increases in affinity by increased intensity of binding to multimers of the WT1 epitope and MHC. The most promising TCRs can then be more completely evaluated to quantify the increase in affinity and determine if the increase has occurred at the cost of decreased specificity for the specific epitope/ MHC complex. Using this approach, TCRs with >2 log10 increases in affinity for WT1 have been generated and evaluated in a preclinical mouse model, revealing enhanced recognition of WT1+ targets and no evidence of on-target or off-target toxicity to normal cells, and we are currently pursuing this strategy with human WT1-specific TCRs. In conclusion, clinical trials of TCR-directed immunotherapy are just beginning, and safety as well as efficacy remains to be demonstrated. However, this approach promises to provide means to definitively determine if antigens identified to be over-expressed in leukemic cells can be effectively targeted by T cells, and if such T cells can provide therapeutic benefit without significant toxicity to normal tissues. Moreover, this approach has the potential to overcome the limitations of vaccine strategies by providing not only large functional T cell responses, but also responses that will contain cells of

S51 higher avidity than can be elicited from the normal repertoire. Further development and testing of TCR-directed immunotherapy offers the hope of establishing new modalities that can be incorporated into the treatment of patients with leukemia. Conflict of interest None References 1) Kuball J, Dossett ML, Wolfl M, Ho WY, Voss R-H, Fowler C, and Greenberg PD. (2007) Facilitating matched pairing and expression of TCR chains introduced into human T cells. Blood 109:2331–2338 2) Kuball J, Hauptrock B, Malina V, Antunes E, Voss R-H, Wolfl M, Strong R, Theobald M, and Greenberg PD. (2009) Increasing functional avidity of TCR-redirected T cells by removing defined N -glycosylation sites in the TCR constant domain. J. Exp. Med. 206:463–475 Minimal Residual Disease-Directed Therapy of Acute Promyelocytic Leukaemia D. GRIMWADE Department of Medical & Molecular Genetics, King’s College London School of Medicine, London, UK Abstract Major improvements in outcome for acute promyelocytic leukaemia patients using protocols involving anthracyclines, all-transretinoic acid (ATRA) and arsenic trioxide (ATO) have recently raised questions concerning the role of minimal residual disease (MRD) monitoring. Hitherto, presenting WBC has been the major factor guiding therapy, with additional treatment typically including cytarabine given to patients with high risk disease (WBC>10×109/l) to reduce relapse risk. However, the Medical Research Council (MRC) AML15 trial recently showed that serial MRD monitoring provides the most powerful independent prognostic factor for relapse—far stronger than presenting WBC. In this trial, persistent PCR positivity or molecular relapse served as the trigger to guide pre-emptive administration of ATO. This strategy was associated with a significantly lower frank relapse rate compared to the previous trial (MRC AML12) in which MRD-directed therapy was not used (5% vs 12% at 3y, p = 0.02). While risk-adapted chemotherapy-based protocols achieve high cure rates, a significant proportion of patients are probably over-treated, fostering considerable interest in the evaluation of deintensified protocols involving frontline use of ATRA and ATO. In this setting MRD monitoring has become even more relevant allowing development of a personalised treatment approach, whilst providing a safeguard for rapid identification of patients requiring additional therapy. Introduction The cloning of the t(15;17)(q22;q21) in 1990 paved the way for the development of reverse transcriptase polymerase chain reaction (RTPCR) assays to detect the resulting PML-RARA transcripts and track treatment response in APL patients (reviewed 1). Early studies showed that achievement of molecular remission (CRm) in the bone marrow (BM) using end-point nested RT-PCR assays was a prerequisite for cure, while patients with persistent or recurrent PCR positivity following front-line therapy invariably relapsed1. On this basis,

S52 achievement of CRm using assays achieving a sensitivity of ~1 in 10,000 has been adopted as a critical therapeutic goal in the management of APL2, as recognised in the international European LeukemiaNet disease guidelines3. Given that full blown relapse of APL carries a significant risk of death from bleeding due to the associated coagulopathy, studies were undertaken by the Italian GIMEMA and Spanish PETHEMA groups to investigate the use of serial molecular monitoring to identify patients with evidence of residual disease as a tool to guide pre-emptive therapy to prevent progression to frank relapse4,5. These studies, which were conducted in the pre-ATO era, suggested a significant survival advantage for early treatment intervention4,5. Moreover, pioneering studies performed by the GIMEMA group also highlighted the value of MRD monitoring in informing transplant choices in patients failing first-line therapy6. In particular, autografting used as consolidation following salvage therapy was found to be associated with a low risk of relapse when performed in CRm with a PCR negative graft6. However, autografting was shown to have a high failure rate in patients with evidence of residual disease, who can however be potentially salvaged by allogeneic transplant7,8. While conventional end-point nested RT-PCR assays have been shown to be of clinical value, there are a number of limitations to their use for monitoring MRD. In particular, they cannot reliably distinguish poor quality samples that could potentially give rise to “false negative” results, with the consequence that impending relapses may be missed. Furthermore, in patients with evidence of residual disease they lack the capacity to determine whether leukaemic transcripts are rising or falling. However, these shortcomings have been overcome with the development of real-time quantitative polymerase chain reaction (RQ-PCR) assays, which afford a number of additional advantages, i.e. 1) they are less prone to contamination 2) they lead to improved turnaround times and 3) are more readily standardized compared to conventional end-point assays (reviewed 9). Optimised RQ-PCR assays for detection of PML-RARA transcripts were developed through a multi-laboratory collaboration in the Europe Against Cancer (EAC) program10. These RQ-PCR assays were subject to a rigorous comparison with conventional nested RT-PCR and found to be marginally more sensitive11,12. The standardized EAC RQ-PCR assays were subsequently applied to track treatment response following ATRA and anthracycline-based therapy within the Medical Research Council (MRC) AML15 trial to investigate whether MRD monitoring remains clinically useful in the arsenic era13. Patients and Methods Prospective MRD monitoring was undertaken in a cohort of 406 PML-RARA+ APL patients (median age 42 years, range 0–76 years) treated with standard ATRA and anthracycline-based chemotherapy, diagnosed between 2002 and 2007, including 303 entered in the MRC AML15 trial in which they were randomized to receive MRC combination chemotherapy or a PETHEMA treatment schedule13. There was no significant difference in disease characteristics (age, proportion with high risk disease—defined as presenting WBC >10× 109/l) or rate of relapse between the treatment arms. For MRD monitoring, paired peripheral blood (PB) and bone marrow (BM) samples were requested on regeneration following each course of therapy and then 3-monthly for 36 months. It was recommended that timing of follow-up samples be brought forward if the previous sample afforded suboptimal sensitivity (i.e. not achieving a sensitivity of at least 1 in 104 according to the level of ABL

Ann Hematol (2011) 90 (Suppl 1):S25–S76 control gene transcripts) or yielded equivocal results. Samples were tested with the appropriate standardized EAC RQ-PCR assay for detection of PML-RARA transcripts according to isoform type. Assays were designed for parallel detection of reciprocal RARAPML transcripts, which are evaluable in 70% of patients. In 2% of patients, PML breakpoint locations were atypical and appropriate forward primers were designed to be used in conjunction with the standard EAC probe and reverse primer (both located in RARA) to track PML-RARA transcript levels. Overall, 6,727 samples were analysed (median 10 samples per patient, range 1–36), including 2,276 paired PB and BM samples. Median follow-up was 25 months (range 0–63). Results and Discussion Analysis of diagnostic BM samples derived from 187 patients showed that there is significant variation in fusion transcript expression (~3-log range), which impacts upon the maximal sensitivity with which MRD can be detected in any given patient (1 in 103–105). Using just the PML-RARA assay, MRD could be detected at a sensitivity of at least 1 in 104 in 71% of patients, whereas parallel use of the reciprocal RARA-PML assay increased this proportion to 81%. Accordingly analysis for RARA-PML transcripts in conjunction with the EAC PML-RARA RQ-PCR assay, was observed to increase MRD detection rates. Indeed, RARA-PML transcripts were frequently observed to persist for longer following front-line therapy and in some instances provided earlier warning of impending relapse. In multivariable analysis, considering a number of factors previously reported to be of prognostic relevance, including age, PML-RARA transcript expression level at diagnosis and isoform type, MRD status provided the most powerful predictor of relapse free survival (HR 17.87, 95% CI 6.88–46.41,p<0.0001), far superior to presenting WBC (HR 1.02, CI 1.00–1.03,p=0.02) which is currently widely used to guide therapy. Assessment of parallel PB and BM samples in patients with relapsing disease showed that the latter affords a median of 1.5-log greater sensitivity. Accordingly molecular conversion was observed in the BM a median of 1 month before the PB in 7 of 12 relapsing patients and in 2 cases PB tested PCR negative despite frank relapse in the marrow. These data are in accordance with those reported by the PETHEMA group, in which BM was found to afford greater sensitivity and provide a more reliable predictor of impending relapse12. On this basis, BM is considered the optimal sample type for MRD monitoring in APL3. The MRC study also provided evidence on the kinetics of disease relapse, with PML-RARA transcripts typically found to rise with an increment of 1-log per month. Assay sensitivity and kinetics of disease relapse are critical factors to consider in the design of optimal MRD monitoring schedules and the MRC data suggest that 3-monthly BM assessment is required in order to realistically deliver pre-emptive therapy to prevent disease progression to full-blown relapse. In patients predicted to relapse based on MRD monitoring in the MRC AML15 trial, early treatment intervention with ATO prevented progression to overt relapse in the majority, with 73% relapse free survival 1 year from molecular relapse. Moreover, use of ATO in the context of subclinical disease was associated with less treatmentrelated complications, compared to deployment of this agent in frank relapse. In particular, administration of ATO in molecular relapse did not induce significant hyperleukocytosis (median peak WBC 5.4, 3–18.7×109/l) and no patients experienced differentiation syndrome

Ann Hematol (2011) 90 (Suppl 1):S25–S76 (DS), in contrast to those treated in haematologic relapse (median peak WBC 51.5, 2.4–126×109/l, p=0.008), where 25% developed DS. An interim analysis of the European LeukemiaNet registry for relapsed APL is in accordance with these findings, with no instances of hyperleucocytosis or DS in 25 patients treated with ATO in molecular relapse; whereas 9 of 42 (21%) patients treated in frank relapse suffered this complication14. Moreover, 3 patients treated in frank relapse died of bleeding; whereas no deaths complicated treatment of molecular relapse, in which the coagulation profile of the patients was consistently normal14. Using MRD monitoring to guide early intervention with ATO, 3 year cumulative incidence of clinical relapse was only 5% in the MRC AML15 trial, as compared to 12% (p=0.02) in the previous MRC AML12 trial which involved comparable ATRA and anthracycline-based front-line therapy, but no MRD monitoring. Based on comparison of survival of patients treated with MRC chemotherapy in the successive trials, with RQ-PCR assays costing an average of €4,000 per patient and assuming a life expectancy of 25 years for patients successfully salvaged, MRD monitoring was found to be most cost-effective in high risk patients with a 10% survival benefit at 5 years giving €1,800/quality adjusted life year (QALY) compared to those with WBC<10 (1% survival benefit at 5 years giving €19,300/QALY). To date, presenting WBC has been considered the most important prognostic factor in APL and is widely used for riskstratification targeting more intensive chemotherapy to those with WBC> 10 who are at greatest risk of relapse. Although relapse rates following current standard ATRA and anthracycline-based protocols are relatively low and could potentially fall further with introduction of ATO as part of first-line therapy15, these nevertheless entail prolonged hospital admissions and significant toxicity incurring risk of death in CR, with some patients developing severe cardiac impairment or emergence of secondary myelodysplasia/ AML16,17. The recent MRC study shows that sequential MRD monitoring provides an alternative strategy allowing additional therapy to be targeted specifically to those patients who would otherwise relapse, which is most cost-effective when applied to the group presenting with WBC>10. Due to concerns that many patients are currently over-treated, with the availability of molecularlytargeted agents there has been an increasing trend towards deintensification of APL therapy, with interest in chemotherapy-free schedules based on ATRA and ATO18-23. However, as treatment intensity is reduced, MRD monitoring using optimized schedules assumes greater importance as a safeguard to identify rapidly those patients requiring additional therapy. MRD monitoring is becoming increasingly relevant within clinical trials to provide early evaluation of novel agents and has been widely adopted in chronic myeloid leukaemia to measure response to targeted therapies and identify emergence of resistance prompting treatment modification. The MRC AML15 study has clearly highlighted the challenges involved in successfully delivering MRD-directed therapy in acute leukaemias in which relapse kinetics are typically faster, demanding compliance with optimized monitoring schedules, rapid sample processing using standardized assays and timely treatment intervention. Rigorous sequential RQ-PCR monitoring coupled with pre-emptive therapy can provide a valid strategy to reduce rates of clinical relapse in APL, serving as a model for development of a more individualized approach to management of other molecularly-defined subtypes of acute leukaemia.

S53 Conflict of interest Research support from Cell Therapeutics Inc and Cephalon. Acknowledgments DG gratefully acknowledges the support of Alan Burnett, Nigel Russell, Robert Hills and members of the UK National Cancer Research Institute (NCRI) AML Working Group for their assistance in providing relevant data, as well as Yvonne Morgan and Khalid Tobal of GSTS Pathology for their role in MRD monitoring for APL patients in the NCRI AML17 trial. Assessment of minimal residual disease in the NCRI AML17 trial is in part supported by a programme grant award from the National Institute of Health Research (NIHR). In addition DG is grateful for research funding from Leukaemia and Lymphoma Research of Great Britain, the Guy’s and St. Thomas’ Charity and the MRD (WP12) work-package of the European LeukemiaNet.

References 1. Grimwade D. The significance of minimal residual disease in patients with t(15;17). Best Pract Res Clin Haematol. 2002;15:137–58. 2. Cheson BD, Bennett JM, Kopecky KJ, et al. Revised recommendations of the International Working Group for Diagnosis, Standardization of Response Criteria, Treatment Outcomes, and Reporting Standards for Therapeutic Trials in Acute Myeloid Leukemia. J Clin Oncol 2003; 21:4642–4649. 3. Sanz MA, Grimwade D, Tallman MS, et al. Management of acute promyelocytic leukemia: recommendations from an expert panel on behalf of the European LeukemiaNet. Blood 2009; 113: 1875–1891. 4. Lo Coco F, Diverio D, Avvisati G, et al. Therapy of molecular relapse in acute promyelocytic leukemia. Blood 1999; 94:2225–9. 5. Esteve J, Escoda L, Martín G, et al. Outcome of patients with acute promyelocytic leukemia failing to front-line treatment with all-trans retinoic acid and anthracycline-based chemotherapy (PETHEMA protocols LPA96 and LPA99): benefit of an early intervention. Leukemia 2007; 21:446–452. 6. Meloni G, Diverio D, Vignetti M, et al. Autologous bone marrow transplantation for acute promyelocytic leukemia in second remission: prognostic relevance of pretransplant minimal residual disease assessment by reverse-transcription polymerase chain reaction of the PML/RAR alpha fusion gene. Blood. 1997; 90:1321–5. 7. Lo-Coco F, Romano A, Mengarelli A, et al. Allogeneic stem cell transplantation for advanced acute promyelocytic leukemia: results in patients treated in second molecular remission or with molecularly persistent disease. Leukemia. 2003; 17:1930–3. 8. Kishore B, Stewart A, Jovanovic J, Craddock C, Grimwade D. Arsenic Trioxide and stem cell transplantation is an effective salvage therapy in patients with relapsed APL. Br J Haematol 2010; 149 (suppl 1): 22 (abstract) 9. Freeman SD, Jovanovic JV, Grimwade D. Development of minimal residual disease directed therapy in acute myeloid leukemia. Semin Oncol 2008; 35:388–400. 10. Gabert J, Beillard E, van der Velden VHJ, et al. Standardization and quality control studies of ‘real-time’ quantitative reverse transcriptase polymerase chain reaction of fusion gene transcripts for residual disease detection in leukemia—a Europe Against Cancer Program. Leukemia 2003; 17:2318–2357. 11. Grimwade D, Diverio D, Harrison G, et al. Detection of minimal residual disease (MRD) in APL by ‘real-time’ RT-PCR: Analysis

S54 of cases entered into the UK MRC ATRA trial. Blood 1999; 94 (Suppl. 1): 2778 (abstract). 12. Santamaría C, Chillón MC, Fernández C, et al. Using quantification of the PML-RARα transcript to stratify the risk of relapse in patients with acute promyelocytic leukemia. Haematologica 2007; 92:315–322. 13. Grimwade D, Jovanovic JV, Hills RK, et al. Prospective minimal residual disease monitoring to predict relapse of acute promyelocytic leukemia and to direct pre-emptive arsenic trioxide therapy. J Clin Oncol. 2009;27:3650–8. 14. Lengfelder E, Lo-Coco F, Montesinos P et al. Treatment of molecular and clinical relapse of acute promyelocytic leukemia (APL) with arsenic trioxide: Results of the European Registry of Relapsed APL. Blood 2010 (suppl 1), in press (abstract) 15. Powell BL, Moser B, Stock W, et al. Arsenic trioxide improves event-free and overall survival for adults with acute promyelocytic leukemia: North American Leukemia Intergroup Study C9710. Blood. 2010;116:3751–7. 16. Tallman MS, Rowe JM. Long-term follow-up and potential for cure in acute promyelocytic leukaemia. Best Pract Res Clin Haematol 2003; 16:535–43. 17. Montesinos P, González JD, González J, et al. Therapy-related myeloid neoplasms in patients with acute promyelocytic leukemia treated with all-trans-retinoic Acid and anthracycline-based chemotherapy. J Clin Oncol. 2010;28:3872–9. 18. Estey EH, Giles FJ, Kantarjian H, et al. Molecular remissions induced by liposomal-encapsulated all-trans retinoic acid in newly diagnosed acute promyelocytic leukemia. Blood 1999; 94:2230– 5. 19. Shen, ZX, Shi ZZ, Fang J, et al. All-trans retinoic acid/As2O3 combination yields a high quality remission and survival in newly diagnosed acute promyelocytic leukemia. Proc Natl Acad Sci USA 2004; 101:5328–5335. 20. Mathews V, George B, Lakshmi KM, et al. Single-agent arsenic trioxide in the treatment of newly diagnosed acute promyelocytic leukemia: durable remissions with minimal toxicity. Blood 2006; 107:2627–2632. 21. Estey E, Garcia-Manero G, Ferrajoli A, et al. Use of all-trans retinoic acid plus arsenic trioxide as an alternative to chemotherapy in untreated acute promyelocytic leukemia. Blood 2006; 107:3469–3473. 22. Ghavamzadeh A, Alimoghaddam K, Ghaffari SH, et al. Treatment of acute promyelocytic leukemia with arsenic trioxide without ATRA and/or chemotherapy. Ann Oncol 2006; 17:131–134. 23. Ravandi F, Estey E, Jones D, et al. Effective treatment of acute promyelocytic leukemia with all-trans-retinoic acid, arsenic trioxide, and gemtuzumab ozogamicin. J Clin Oncol. 2009; 27:504–10.

Common Programs in AML Stem Cells B. HUNTLY Cambridge Institute for Medical Research, Cambridge, UK Acute myeloid leukaemia (AML) is an extremely heterogeneous disease at the morphological, biological and molecular level. However, it is characterised by unifying cellular characteristics such as a block in terminal myeloid differentiation and an uncontrolled

Ann Hematol (2011) 90 (Suppl 1):S25–S76 proliferation of immature myeloid progenitor cells1. Hundreds of separate genetic lesions have been described in AML and deep sequencing analyses of AML genomes have demonstrated the occurrence of multiple lesions within individual leukemias2,3. Great improvements have been made in the identification of prognostic factors for AML, such as age, white blood cell count (WBC), cytogenetics and mutational analysis of important genes (such as FLT3 and NPM1). In addition, gene expression studies have defined signatures downstream of many of these genetic lesions, allowing further molecular characterization of AML4,5. However, these findings have only served to reinforce the heterogeneity of AML cases. Knowledge of the molecular lesions associated with specific subtypes of AML has led to the introduction of specifically targeted therapeutics, such as all trans retinoic acid (ATRA) in PML-RARA positive cases and selective FLT3 inhibitors in cases which harbor an internal tandem duplication of the FLT3 gene (FLT3-ITD). However, AML remains a significant clinical problem with over 70% of patients succumbing to the disease, accounting for over 9000 deaths per year alone in the US6, and novel therapeutics with efficacy in the majority of AML patients are urgently required. To this end it is not known whether transformation is mediated by common or overlapping genetic programs downstream of multiple mutations or through the engagement of unique programs downstream of individual mutations. This distinction is important, as the demonstration of common or overlapping pathways may identify common critical molecular targets for the treatment of AML, in all cases or at least in significant subgroups. Over the last decade it has become increasingly apparent that many tumors are wholly dependent upon a sub-population of cells, the socalled cancer stem or initiating cell 7, for their continued growth and propagation. Although initially proposed half a century ago 8, the cancer stem cell hypothesis was only proven a decade ago by John Dick and co-workers in Toronto, using separated populations of blasts from patients with AML9,10. Subsequently, cancer stem cells have been demonstrated in a number of solid organ tumors 11 suggesting that the majority of malignancies are dependent upon such a compartment. Besides informing the cellular biology of cancer, the presence of a cancer stem cell population also has important clinical implications. Many malignancies, including AML, demonstrate significant tumor reduction following initial therapy. However, the majority of cases of AML relapse, often with resistant disease, causing the death of most patients. This suggests that current combination therapies spare the LSC compartment in AML, and that it is this compartment that forms the reservoir for subsequent relapse and resistance. This defines the LSC as the critical target in AML therapy and suggests that LSC selective or specific therapies should improve treatment outcomes in AML. We have sought to identify common and overlapping pathways that regulate leukaemia stem cells and leukaemognesis in AML. Leukaemia–associated oncogenes such as MLL fusions and MOZTIF2 (MYST3-NCOA2) generate aberrant transcriptional programs mediated by their ability to modify chromatin, and confer leukaemia stem cell properties to committed myeloid progenitors12–14. We now demonstrate that the ability to alter self-renewal is a more generalized effect of leukaemia-associated transcription factor fusion oncogenes, further linking transformation to aberrant self-renewal. We have taken further advantage of this restoration of self-renewal properties in vitro, using the same retroviral expression platform to

Ann Hematol (2011) 90 (Suppl 1):S25–S76 show that three disparate AML-associated oncogenes initiate immediate signatures containing common and overlapping genes and genetic programmes implicated in leukaemia and self-renewal. Moreover, elements of these signatures can be detected in established leukaemia stem cells from an animal model of AML. Furthermore, we use cross-species comparisons with large global gene expression datasets of bulk human AML samples to demonstrate their relevance to human disease. When used to classify human AML cases, these genesets strongly predict for the biology of human AML, where they correlate with existing existing prognostic factors such as cytogenetics, mutation status, age and white cell count. Finally, we demonstrate that individual genes from within the immediate signatures, can at least partially phenocopy the leukaemiaassociated oncogenes and alter self-renewal in committed murine progenitors and generate AML when expressed in murine bone marrow. This suggests that common and overlapping transformation and selfrenewal pathways do exist downstream of a variety of leukemiaassociated oncogenes, where they contribute to the induction and maintenance of AML, a finding with important clinical implications. Conflict of interest None References 1 Estey, E. & Dohner, H. Acute myeloid leukaemia. Lancet 368, 1894–1907 (2006). 2 Ley, T. J. et al. DNA sequencing of a cytogenetically normal acute myeloid leukaemia genome. Nature 456, 66-72, 3 Mardis, E. R. et al. Recurring mutations found by sequencing an acute myeloid leukemia genome. N Engl J Med 361, 1058–1066 4 Bullinger, L. et al. Use of gene-expression profiling to identify prognostic subclasses in adult acute myeloid leukemia. N Engl J Med 350, 1605–1616 (2004). 5 Valk, P. J. et al. Prognostically useful gene-expression profiles in acute myeloid leukemia. N Engl J Med 350, 1617–1628 (2004). 6 Jemal, A. et al. Cancer statistics, 2008. CA Cancer J Clin 58, 71–96, 7 Reya, T., Morrison, S., Clarke, M. & Weissman, I. Stem cells, cancer, and cancer stem cells. Nature 414, 105–111 (2001). 8 Chen, J. et al. Constitutively activated FGFR3 mutants signal through PLC{gamma}-dependent and -independent pathways for hematopoietic transformation. Blood (2005). 9 Lapidot, T. et al. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature 367, 645–648 (1994). 10 Bonnet, D. & Dick, J. E. Human acute myeloid leukemia is organized as a hierarchy that originates from a primitive hematopoietic cell. Nat Med 3, 730–737 (1997). 11 Kvinlaug, B. T. & Huntly, B. J. Targeting cancer stem cells. Expert Opin Ther Targets 11, 915–927 (2007). 12 Cozzio, A. et al. Similar MLL-associated leukemias arising from self-renewing stem cells and short-lived myeloid progenitors. Genes Dev 17, 3029–3035 (2003). 13 Huntly, B. J. et al. MOZ-TIF2, but not BCR-ABL, confers properties of leukemic stem cells to committed murine hematopoietic progenitors. Cancer Cell 6, 587–596 (2004). 14 Krivtsov, A. V. et al. Transformation from committed progenitor to leukaemia stem cell initiated by MLL-AF9. Nature 442, 818–822 (2006).

S55 Regulation of Normal and Leukemic Human Stem Cells via Dynamic Interactions of the Nervous and Immune Systems with the Microenvironment. T. LAPIDOT Dept. of Immunology, Weizmann Institute. Rehovot Israel. Hematopoietic stem cells (HSC) are identified based on their functional ability to home to the bone marrow and to durably repopulate it with both myeloid and lymphoid cells, in transplantation assays. Similarly, leukemic stem cells are identified based on their functional ability to home to the bone marrow and to initiate the disease in experimental animal models. The development of functional, preclinical transplantation models, in irradiated immune deficient mice recipients, has enabled identification and characterization of normal and some leukemic human stem cells and their progenitor cell progeny. The chemokine SDF-1 (CXCL12), and its major receptor CXCR4 are essential for maintaining murine HSC quiescence and human and murine SDF-1 are cross reactive. Both normal and leukemic (childhood Pre B ALL and some adult AML FAB subtypes), human stem cell function in transplanted immune deficient mice, is dependent on SDF-1/CXCR4 interactions. This small animal functional model also supports, homing and mobilization of the human cells, which are both dependent on CXCR4 signaling. The nervous system, is the master regulator of the body, including leukocyte production from the bone marrow reservoir of immature and maturing cells. The bones are highly innervated, and the nervous system also regulates bone turnover via the coupled osteoclast/ osteoblast interactions which control bone remodeling. Osteoclasts also regulate HSC migration and development and the stem cell niche via release of TGF-β and Cathepsin K activity. TGF-β recruits Sca-1+ murine stromal precursors to the endosteal region. Osteoclasts also regulate stromal CFU-F levels (fibroblast progenitor cells) which are the progeny of the rare Nestin+/CD45- murine stromal stem cells (MSC), cells that form the HSC niche. In addition, TGF-β also directly regulates murine HSC quiescence and hibernation, and low levels of this cytokine also regulate human CD34+ stem and progenitor cells (HSPC), preventing SDF-1 desensitization. Cathepsin K is essential for bone resorption activity of osteoclasts and also degrades key murine HSC niche components SDF-1, membrane bound kit ligand and osteopontin. Normal human stem cells functionally expressing dopamine and neurotransmitter receptors, which are upregulated by myeloid cytokines such as G-CSF. The nervous system also directly regulates human HSC migration and development via Wnt signaling. Preliminary results reveal that human leukemic stem cells as well, functionally express neuronal receptors and the similarities and differences in their function and regulation are currently studied. Integration of HSC regulation with bone turnover by steroids in the murine system will also be discussed. Finally the roles of ROS in enhanced HSC recruitment, migration and development and its clinical relevance will also be discussed. Conflict of interest None Therapy of Ph+ Acute Lymphoblastic Leukemia (Ph+ ALL) R.A. LARSON University of Chicago, Chicago IL, USA

S56 Abstract Combining a tyrosine kinase inhibitor (TKI) with cytotoxic chemotherapy, followed by hematopoietic stem cell transplant-ation (HCT), whenever possible, has markedly improved the outcomes for newly diagnosed patients with Ph+ ALL. The oncogenic fusion gene BCR/ABL that results from the reciprocal translocation t(9;22)(q34;q11.2) is the single most frequent chromosome abnormality in adult ALL, detected in 11–34% of patients. The prevalence increases with age, affecting ~50% of those over 60 yrs, and is found almost exclusively in CD10+ precursor Bcell ALL. For many years, Ph+ ALL was associated with an unfavorable outcome, low rates of complete remission (CR), and short remission durations. This changed with the advent of imatinib mesylate, a first generation TKI that stabilizes the inactive ATPbinding conformation of Bcr-Abl, inhibiting tyrosine phosphorylation of proteins involved in Bcr-Abl signal transduction.[1] Imatinib induces apoptosis in BCR/ABL+ cells. Imatinib Alone and in Combination By itself, imatinib is not sufficient to induce prolonged remission.[2] Combining imatinib with chemotherapy, either sequen-tially (in order to reduce toxicities) or concurrently results in significant improvements over chemotherapy alone. CR rates exceed 90%. However, the durability of these CRs is still uncertain, and allogeneic (allo)HCT is still recommended for appropriate patients. A major benefit of combined TKI plus chemotherapy has been to achieve a larger proportion of CRs with less toxicity, and thus get more patients to an alloHCT in CR1. AlloHCT has been shown to cure Ph+ ALL in younger adults, but not every patient has a donor. Regimens that did not include any form of HCT (mostly aimed at patients≥55 yrs old) have had inferior disease-free survival (DFS) and increased relapse rates. Also, the DFS of younger patients who did not undergo any form of HCT have declined over time in followup.[3] However, in a small pilot study by the Cancer and Leukemia Group B (CALGB), patients who had molecular CRs after imatinib plus chemotherapy and then underwent autologous (auto)HCT due to lack of matched donors had outcomes not markedly different than those who underwent alloHCT although with short followup.[4] For the foreseeable future, myeloablative HCT remains the best curative option for appropriate patients with Ph+ ALL. Trials for older patients should evaluate the role of reduced-intensity conditioning (RIC) regimens for those who are not eligible for a fully myeloablative approach. Of the 441 Ph+ ALL patients on the UKALLXII/E2993 trial (median, 42 yrs; range, 16–65), 266 were en-rolled and treated before imatinib became available; their CR rate was only 67% compared to 77% among 175 later patients who received imatinib 600 mg/day plus chemotherapy. [3] Overall survival for the pre-imatinib patients was 25% at 3 yrs compared to 42% overall with imatinib (p<0.001), and 48% among the 89 patients who received imatinib during both induction and consolidation. A larger fraction of patients who received imatinib were able to proceed to alloHCT (44% vs 28%), and this translated into a highly significant survival and DFS advantage over chemotherapy alone. The 3yr survival for imatinib-treated patients who received an alloHCT was 59% compared to 28% for those who did not undergo HCT. The German Multicenter ALL study group (GMALL) recently reported long-term results of 335 patients with newly diagnosed Ph +nALL who received imatinib at a single daily dose of 600 mg within 3 successive treatment studies.[5] The median age of all patients was 43 yrs (17–65); only 57 (17%) were 55 yrs or older. CR rate was 86–

Ann Hematol (2011) 90 (Suppl 1):S25–S76 89%. Treatment outcomes improved with earlier initiation and more prolonged administration of imatinib in the 3 successive cohorts. Molecular responses (PCR negativity) after the first consolidation course were observed in 33% in the most recent trial. Overall survival at 4 yrs increased from 31% to 40% to 50%. To date, 219 patients (66%) underwent alloHCT in CR1 (median age, 39.5 yrs). The 3-yr DFS after condition-ing with TBI and cyclophosphamide or etoposide was 72%. DFS was not significantly different between matched sibling and matched un-related donor HCT. For all patients transplanted in CR1, survival was 57% after 3 yrs and 52% after 7 yrs. Only 14% of patients who did not undergo HCT in CR1 were alive after 3 yrs (median, 9.4 months).

Mechanisms of Resistance to Imatinib Four mechanisms of resistance to imatinib have been described to date: mutations at the kinase site, reduced intracellular imatinib concentration, gene amplification, and alternative signaling pathways that compensate for imatinib-sensitive mechanisms. Mutations at the Abl kinase domain impair imatinib’s binding to varying degrees and represent the most common (~50%) mechanism of resistance. In patients with de novo Ph + ALL, kinase domain mutations frequently precede imatinib-based therapy and lead to relapse [6]. Of note, 12 (46%) of 26 patient samples had P-loop (G250E, Q252H, Y253H, E255K/V) mutations and 6 (23%) had gatekeeper (F311I, T315I, F317L) mutations at diagnosis. Remission duration did not differ between patients with or without a detectable early mutation. At relapse, 10 patients with mutations at diagnosis demonstrated the same mutation while one demonstrated a different mutation. These data suggest that, in contrast to CML, it may be useful to consider mutational analysis at diagnosis in Ph + ALL and tailor treatment accordingly. Second generation TKI’s can overcome some kinase domain mutations. Another recently identified mechanism of TKI resistance involves the expression of spliced isoforms of Ikaros (IKZF1).[7] Ikzf1 functions as a critical regulator of normal lymphocyte development and is involved in the rapid development of leukemia in mice expressing non-DNA-binding isoforms. The IK6 isoform, lacking all 4N-terminal zinc fingers responsible for DNA-binding, was detected in 43 of 47 (91%) Ph+ ALL patients resistant to imatinib or dasatinib. [7] The expression level of IK6 correlated with the BCR-ABL transcript level. Hence, restoring Ikzf1 function may provide another approach to combating TKI resistance.

New Drugs to Overcome Resistance: Dasatinib for Ph+ ALL Dasatinib has 325-fold greater potency in vitro compared with imatinib against Abl kinase activity. In addition to Bcr-Abl and Src, it inhibits Kit and PDGFR. It is active against all ABL mutations except gatekeeper mutations. It is not a substrate for P-glycoprotein or OCT1, and it can bind to the kinase domain in the open and closed conformations. The most frequent side effects include diarrhea, nausea, fluid retention, and pleural effusions. Myelosuppression occurs in 60– 70% of Ph+ ALL patients, but is uncommon once a CR is achieved. Dasatinib as a single agent induces rapid hematologic and cytogenetic responses in adult patients with Ph+ ALL with resistance or intolerance to imatinib [8]. Two dosages were tested in patients who failed imatinib: 40 were randomized to 140 mg once daily and 44– 70 mg twice daily. Major hematologic responses were achieved in 38% and 32%, respectively; 50% and 39% had complete cytogenetic responses (CCyR). The lower rate of hematologic responses compared to CCyR is due to cytopenias associated with dasatinib in previously

Ann Hematol (2011) 90 (Suppl 1):S25–S76 treated Ph+ ALL. Median progression-free survival was 4 months and 3 months. Dasatinib was approved by the US Food and Drug Administration for second-line therapy for Ph+ ALL; nilotinib is not yet approved. Recent data demonstrated clonal expansion of T/NK cells during dasatinib treatment for Ph+ ALL [9]. Dasatinib and prednisone were studied as front-line therapy in 48 newly diagnosed Ph+ ALL patients in Italy (median age, 54 yrs; range, 24–76) [10]. All patients achieved CR, most by day 22. After a median follow up of 11 months, 9 had relapsed (more commonly among p210+ cases). Overall survival was >80% at 10 months. Subsequently, investiga-tors at the MD Anderson Cancer Center combined dasatinib with hyper-CVAD, treated 35 patients (median age 53 yrs; range, 21–79), and reported 94% CR [11]. After a median follow-up of 14 months, estimated 2-yr survival was 64%. Only 4 patients proceeded to alloHCT in CR1. Dasatinib (100–140 mg/day) was combined with low intensity chemotherapy in the EWALL-Ph-01 trial for 71 newly diagnosed patients 58–83 yrs old (median, 69).[12] Dexamethasone + vincristine were repeated weekly during induction and methotrexate + Lasparaginase and cytarabine were given during consolidation. The CR rate was 90%; 56% achieved a BCR-ABL/ABL ratio <0.1% at CR and 71% after consolidation. Nineteen patients relapsed after a median of 19 weeks; 13 had mutations in the kinase domain at relapse (12 T315I, 1 F317L). Median DFS was 22 months although the median has not been reached for patients presenting with an isolated Ph; it was 19 months for those with additional cytogenetic abnormalities (p=.03). BCR/ABL transcript level after induction had no effect on DFS. Central Nervous System Prophylaxis The CNS is a sanctuary site for ALL, and relapses may occur first in the cerebrospinal fluid (CSF). CNS prophylaxis with intrathecal (IT) chemotherapy is routine. Imatinib does not cross the blood-brain barrier. Dasatinib has a potential advantage of penetrating into the CSF, but as yet, this has not been analyzed systematically. AlloHCT for Ph+ ALL The EBMT reported on 127 ALL patients who underwent RIC and 449 who received myeloablative condi-tioning prior to alloHCT; all were >45 yrs.[13] In multivariable analyses, nonrelapse mortality was decreased in RIC recipients (p=.0001) whereas it was associated with a higher relapse rate (p=.03). Leukemia-free survival was not different (32% vs 38% at 2 yrs). Various myeloablative preparatory regimens have been studied in Ph+ ALL. Most are poorly tolerated in older patients due to high transplant related mortality (TRM). RIC regimens have not been well explored. Autologous HCT for Ph+ ALL Not all patients have a suitable donor for alloHCT. Thus, autologous (auto) HCT, using stem cells collected during the first molecular CR, offers an alternative approach. Using only conventional chemotherapy to eradicate Ph+ ALL prior to autoHCT resulted in poor outcomes with increased disease relapse. The combination of imatinib or dasatinib with chemotherapy may achieve a better success rate in cleansing the marrow from residual disease prior to stem cell collection. Several groups have reported on the feasibility and favorable outcomes following autoHCT for Ph+ ALL patients who lack suitable allogeneic donors.[4] In one study of 43 patients in CR1, the outcomes of patients undergoing autoHCT versus alloHCT were

S57 similar.[14] Even though the relapse rate was higher among patients who underwent autoHCT, the higher TRM associated with alloHCT led to similar final outcomes. TKI Maintenance for Ph+ ALL Patients with CML who have had a good clinical response to imatinib are recommended to continue the drug life-long. Discontinuation of imatinib after allo- or autoHCT continues to present a dilemma for Ph + ALL in CR1. Several studies continued imatinib until disease recurrence [15,16] while others had a set time (1–2 yrs) of imatinib maintenance.[17] It is difficult to discern a difference between the studies based on patients’ outcomes because of the relatively short follow-up. In the CALGB 10001 clinical trial of imatinib and chemotherapy followed by autoHCT [4], imatinib maintenance was continued for at least 1 yr and then until 2 consecutive quantitative PCR tests 3 months apart were negative. However, long-term data from this study are not yet available. Other studies have claimed that molecular relapse did not predict hematologic relapse [15,17,18]. Conflict of interest Consultant and research grant support from Bristol Myers Squibb and Novartis Pharmaceuticals. References 1. Stock W. Leuk & Lymph 2010; 51: 188–198 2. Ottmann OG & Pfeifer H. Hematology (Am Soc Hematol Education Program) 2009; 371–381 3. Fielding AK et al. Blood 2010; 116: abstr 169 4. Wetzler M et al. Ann Hematol 2008; 87(Suppl 1): S92–94 5. Pfeifer H et al. Blood 2010; 116: abstr 173 6. Pfeifer H et al. Blood 2007; 110: 727–734 7. Iacobucci I et al. Blood 2008; 112: 3847–3855 8. Lilly MB et al. Am J Hematol 2010; 85: 164–170 9. Mustjoki S et al. Leukemia 2009; 23: 1398–1405 10. Foa R et al. Blood 2008; 112: abstr 305 11. Rivandi F et al. Blood 2010; 116: 2070–2077 12. Rousselot P et al. Blood 2010; 116: abstr 172 13. Mohty M et al. Blood 2010; 116: 4439–4443 14. de Labarthe A et al. Blood 2007; 109: 1408–1413 15. Potenza L et al. Haematologica 2005; 90: 1275–1277 16. Vignetti M et al. Blood 2007; 109: 3676–3678 17. Lee KH et al. Leukemia 2005; 19: 1509–1516 18. Yanada M et al. Br J Haematol 2008; 143: 503–510 Acute Promyelocytic Leukemia: Results of the German AML Cooperative Group (AMLCG) E. LENGFELDER1, W.-K. HOFMANN1, and TH. BÜCHNER2 1 III. Medizinische Universitätsklink Mannheim, Universität Heidelberg, Germany 2 Medizinische Universitätsklinik A, Westfälische Wihelms-Universität Münster, Germany Background Acute promyelocytic leukemia (APL) is relatively uncommon in elderly patients. APL in the elderly appears to be as sensitive to therapy with all-trans-retinoic acid (ATRA) and chemotherapy as in

S58 younger adults. The main problem is a higher rate of complications in comparison to younger patients due to the worse tolerance of the toxicity of chemotherapy.1–3 The APL studies of the AMLCG were started in the year 1994, after the introduction of ATRA. Adult patients were uniformly treated with double induction therapy (TAD/HAM) incorporating high dose ara-C and ATRA, followed by consolidation and maintenance chemotherapy. In the group the of elderly patients (≥60 years), the intensity of the chemotherapy was reduced. The second induction cycle (HAM with reduced ara-C dose) was only recommended in the case of insufficient blast clearance after the first induction cycle. Results The results of the AMLCG with intensified double induction therapy including high dose ara-C and ATRA show that in patients <60 years (n=142) the initial WBC count above 10000/μl was no significant risk factor for relapse (p=0.46), and a low relapse rate was seen in high and low risk patients.4 In the elderly patients (n=68) included in studies of the AMLCG from 1994 to 2010, the outcome was worse in comparison to the younger patients. This includes the rate of induction death, the longterm overall, event free and disease free survival, as well as the cumulative incidence of relapse. Elderly patients with initial WBC count <10000/μl, who had received only one induction cycle (80% of the elderly patients), showed a significantly longer event free survival (p=0.01) and a lower the relapse rate (p=0.02) as compared to patients with WBC counts ≥10000/μ. The elderly patients who had received two induction cycles (20% of the elderly patients) seem to benefit from the intensification of the therapy with regard to relapse rate and survival. Separating the patients of the AMLCG studies according to age <60 years, 60–70 years and >70 years, the highest early death rate (60%) was seen in patients >70 years with pretreatment WBC counts ≥10000/μl. Multiorgan or cardiac failure were the most frequent causes of death in the elderly patients. Conclusions Our data indicate that the risk of early complications increases with older age and with a higher pretreatment WBC count. In elderly patients, a careful balance between toxicity and antileukemic effectiveness of the therapy as well as an optimization of the supportive measures is required to improve the cure rate. For APL patients unfit for chemotherapy at first diagnosis or at relapse, arsenic trioxide might be a reasonable treatment approach.5 In the future, a better characterization of the disease at the molecular level, might contribute to a better understanding of the biologic background of the disease including age-related differences and therefore may be utilized for further risk stratification.6,7 Conflict of interest None References 1) Mandelli F, Latagliata R, Avvisati G, Fazi P, Rodeghiero F, Leoni F, Gobbi M wt al (2003) Leukemia. 17:1085–1090. 2) Sanz MA, Vellenga E, Rayón C, Díaz-Mediavilla J, Rivas C, Amutio E et al. (2004) All-trans retinoic acid and anthracycline monochemotherapy for the treatment of elderly patients with acute promyelocytic leukemia. Blood 104:3490–3493. 3) Ades L, Chevret S, De Botton S, Thomas X, Dombret H, Beve B et al. (2005) Outcome of acute promyelocytic leukemia treated with

Ann Hematol (2011) 90 (Suppl 1):S25–S76 all trans retinoic acid and chemotherapy in elderly patients: the European group experience. Leukemia 19:230–233. 4) Lengfelder E, Haferlach C, Saussele S, Haferlach T, Schultheis B, Schnittger S et al. (2009) High dose ara-C in the treatment of newly diagnosed acute promyelocytic leukemia: long-term results of the German AMLCG. Leukemia. 23:2248–2258. 5) Lengfelder E, Lo-Coco F, Montesinos P, Grimwade D, Ades L, Bhuvan K et (2010) Treatment of Molecular and Clinical Relapse of Acute Promyelocytic Leukemia (APL) with Arsenic Trioxide: Results of the European Registry of relapsed APL, Blood, ASH Abstract 15. 6) Nolte F, Nowak D, Hecht A, Erben P, Hanfstein B, Benlasfer Q et al. (2010) High WT1 expression is associated with a shorter time to complete remission after ATRA based induction therapy in acute promyelocytic leukemia. Blood, ASH Abstract 1694. 7) Nowak D, Klaumuenzer M, Hanfstein B, Mossner M, Nolte F, Nowak V et al. (2010) High density SNP array analysis of acute promyelocytic leukemia detects new common genomic copy number alterations as possible cooperating lesions. Blood, ASH Abstract 2721.

What Has the t(11:17) Variant Forms of Acute Promyelocytic Leukemia Taught us About Disease Pathogenesis? J.D. LICHT Division of Hematology/Oncology, Robert H. Lurie Comprehensive Cancer Center, Northwestern University, Chicago, IL, USA Acute promyelocytic leukemia (APL) is always associated with rearrangement of the retinoic acid receptor gene with one of a number of partner proteins. In over 98% of cases the RARα gene is rearranged with the PML gene yielding the PML-RARα fusion protein. Studies from a number of laboratories have determined which characteristics of the RARα fusion protein are required for malignant transformation. Essential characteristics include, the ability of the fusions to form homodimers and multimers alone or in combination with RXR, as well as the ability to bind RXR. The PML-RARα and other fusion proteins have an enhanced affinity for co-repressor molecules and this leads to repression as well as activation of target genes in a DNA binding site dependent and DNA binding independent manner. The APL fusion proteins differ from the wild-type RARα in that these proteins bind to and expanded repertoire of sites in DNA, some of which resemble retinoic acid receptor half sites separated by more than 10 nucleotides, while the normal RAR binds direct repeats separated by 5 nucleotides. The t(11;17)(q23;q21) translocation, which occurs in <1% of cases of APL is associated with a retinoic acid-insensitive form of acute promyelocytic leukemia (APL) involving the production of reciprocal fusion proteins PLZF-RARα and RARα-PLZF. In order to investigate the molecular basis of PLZF-RARα induced leukemia, we performed a genome wide screen for PLZF-RARα direct target genes using a gain of function model in which PLZF-RARα was expressed in human U937 leukemia cells. Chromatin from U937PLZF-RARα cells was immunoprecipitated using PLZF antibodies, amplified by ligation-mediated PCR and biological triplicates were hybridized to NimbleGen 2.7 kB promoter arrays, which represent 24,659 human promoters. We identified 4916 genes directly bound by PLZF-RARα (2/3 biological replicates, FDR <0.2). These genes were highly enriched for ontological categories including immunity and defense (p<10−6), apoptosis (p<2×10−5), cell cycle (p<10−3) and oncogenesis (p<10−2). Gene expression profiling of U937:PLZF-RARα cells

Ann Hematol (2011) 90 (Suppl 1):S25–S76 revealed that 34% of direct targets were also transcriptionally regulated in response to PLZF-RARα induction. Despite the established role of PLZF-RARα as a transcriptional repressor, 56% of genes bound by PLZF-RAR αwere upregulated and 44% repressed. Bioinformatic analysis of PLZF-RARα bound sequences using the MATRIXReduce algorithm identified the ‘-AGGTCA-‘core sequence as the highest ranked position specific affinity matrix (PSAM). Comparison of this matrix with known transcription factor binding sites from the JASPAR core database revealed high similarity to the recognition sequence for the RAR-related orphan receptor A1 (RORA1) (E value: 5.2×10−3), RORA2 (3.5×10−2) and RXRAVDR (4.4×10−2). This suggests that the natural binding site of PLZFRAR difference modestly from that of wild-type RAR. The ‘GTCA’ core sequence is frequently observed in canonical retinoic acid receptor response elements and this motif was only associated with genes repressed with binding by PLZF-RAR. Together these results are consistent with the idea that PLZF-RARα acts at least in part as a dominant negative retinoic acid receptor. A comparison of genes bound directly by PLZF-RARα with gene expression profiles from 22 APL (4 PLZF-RARα, 18 PML-RARα) and 99 acute myeloid leukemias (AML) (randomly selected cytogenetic categories) using gene set enrichment analysis, revealed that direct targets of PLZFRARα were differentially repressed in APL when compared to other forms of AML. Overexpression of PLZF-RAR in murine hematopoietic progenitors and human CD34+ cord blood, blocked myeloid differentiation, an effect associated with the repression of C/EBP genes (α, β and ε), which were identified as direct targets of PLZF-RARα by ChIP-chip. Treatment of primary CD34+ cells with ATRA led to an increase in CEBPα and β, but repression of CEBPε was not relieved. Overexpression of PLZF-RARα in primary murine bone marrow led to an increase in the more primitive Sca1+ population, coincident with increased serial replating ability. Overexpression of PLZF-RARα in mouse and human progenitors led to increased proliferation with more cells in the S and G2/M phases of cell cycle. Correlating with this effect, genes with defined roles in hematopoietic stem cell selfrenewal including HOXA9 and MPL and c-Myc were bound and activated by the induction of PLZF-RARα in U937 cells. This is surprising given the ability of PLZF-RARα to bind co-repressors. However, PLZF-RARα binds to a region of the c-MYC promoter overlapping a functional PLZF site and antagonizes PLZF-mediated repression, suggesting that PLZF-RARα may act as a dominantnegative version of PLZF by affecting the regulation of shared targets. RA induced the differentiation of PLZF-RARα-transformed murine hematopoietic cells and reduced the frequency of clonogenic progenitors, concomitant with c-Myc down-regulation. Surviving RA-treated cells retained the ability to be replated and this was associated with sustained c-Myc expression and repression of Dusp6, suggesting a role for these genes in maintaining a self-renewal pathway triggered by PLZF-RARα. These data also suggest that ATRA treatment alone is not sufficient to eliminate disease reinitiating cells, a finding corroborated by the work of de Thé and colleagues. PLZF-RARα appears to transform cells through three interlinked modes of action, inhibition of differentiation by direct repression of key myeloid transcription factors, stimulation of proliferation by repression of a cyclin dependent kinase inhibitor and activation of genes critical for self-renewal. PLZF may affect gene expression in a qualitatively different manner than PML-RARα. We found that PLZF and PLZF-RARα both bound the H3K9 histone methyltransferase G9a. PLZF-RARα

S59 but not PML-RARα bound G9a, and this interaction was largely mediated through the BTB domain. We found that PLZF-RARα binding at the NFE2 promoter co-localized with increased H3K9me2. Furthermore, specific knockdown of G9a impaired the ability of PLZF-RARα to repress NFE2 transcription and impaired the differentiation block induced by PLZF-RARα in U937 cells. Our findings suggest that the specific recruitment of histone modifying enzymes to genes is required for PLZF-RARα to repress gene transcription and may be crucial for its transforming capabilities, and highlights that there are differences between the PLZF-RARα ands PML-RARα fusion proteins. APL associated with PLZF rearrangement may differ from typical APL in that PLZF itself is a regulator of normal hematopoiesis. Studies in collaboration with John Dick’s group showed that during homeostasis, PLZF inhibits proliferation and differentiation of myeloid progenitors and maintains a balance between progenitor and mature cell pools. PLZF repressed transcription factors involved in normal myeloid differentiation, including GFI-1, C/EBPalpha, and LEF-1, and induced DUSP6 and ID2, representing inhibitors of myelopoiesis. Induction of ERK1/2 by myeloid cytokines, reflective of a stress hematopoiesis, led to nuclear export and inactivation of PLZF, allowing mature myeloid cells to develop. In the case of APL, loss of normal PLZF activity might further enhance proliferation of the malignant cell. Conflict of interest None

Front-Line Therapy of Acute Promyelocytic Leukemia: Results of the Italian Cooperative Group GIMEMA F. LO COCO Department of Biopathology, Univerity Tor Vergata, Rome, Italy As reported in several large multicenter trials conducted in Europe, Asia and the US, front-line use of combined ATRA and anthracycline chemotherapy results in long-term remission and potential cure in >75% of newly diagnosed APL patients (reviewed in ref 1,2). Among these studies, the Italian multicenter Group GIMEMA reported in 1997 high rates of molecular remission in newly diagnosed and genetically confirmed APL using a simultaneous Atra plus IDArubicin (AIDA) combination for induction treatment, followed by 3 courses of intensive chemotherapy as consolidation (3). This protocol, with slight modifications, was adopted by other groups including the Spanish PETHEMA cooperative group who reported similar antileukemic efficacy by omitting Ara-C and other non-intercalating agents from the original AIDA, with the advantage of sparing toxicity and increasing compliance to treatment. Two independent risk-adapted studies were subsequently designed by the PETHEMA and GIMEMA in which treatment intensification was planned according to the relapse risk. The results of both studies showed improved outcomes by adding ATRA for consolidation to the original AIDA scheme (4,5). In line with these findings, most studies include nowadays risk-adapted approaches in which treatment intensification is based on initial WBC counts (1,2). Use of molecularly-driven protocols through detection of the PML/ RARA fusion has proved invaluable to improve treatment outcome in APL. In fact, in addition to its diagnostic relevance, detection of the PML/RARA hybrid by sensitive RT-PCR techniques is relevant to assess response to therapy and for the monitoring of minimal residual disease (MRD) during follow-up. As reported by several groups, the achievement of a PCR-negative status is associated with prolonged

S60 survival and higher probability of cure, whereas persistence of, or conversion to PCR-positivity in bone marrow after consolidation is invariably associated with subsequent hematological relapse. As a consequence, the achievement of molecular remission is nowadays universally considered as a therapeutic objective in this disease (6,7). Despite the dramatic progress achieved in front-line therapy with the ATRA/chemotherapy combination, relapses still occur in approximately 20% of patients. Moreover, these regimens are associated with significant toxicity due to severe myelosuppression frequently resulting in life-threatening infections, and with serious, though infrequent late complications such as cardiomyopathy and the occurrence of secondary myelodysplastic syndromes and/or acute myeloid leukemias (8,9). Several means are available to decrease toxicity in the treatment of newly diagnosed APL, including the availability of less toxic and highly effective agents such as arsenic trioxide (ATO) and the possibility of stringent MRD monitoring offered by RT-PCR. Following the demonstration of its striking activity in relapsed patients (10), arsenic trioxide (ATO) has been licensed in the USA and Europe for the treatment of relapsed and refractory APL. Arsenic derivatives had been used since ancient times in Chinese medicine for the treatment of malignant and inflammatory diseases. The mechanism of action of ATO in APL is complex and not yet known in detail. At a high concentration (0.5–2.0 μmol/L) ATO induces apoptosis in vitro, through induction of caspases 2 and 3, while at lower concentrations (0.1–0.5 μmol/L) it induces partial differentiation of leukemic promyelocytes through PML/RARa degradation; furthermore, ATO is known to inhibit angiogenesis via down—regulation of vascular endothelial growth factor (VEGF). As concerning its toxicity profile, ATO is usually well tolerated and its use is associated with a series of manageable adverse events including hyperleucocytosis, the APL differentiation syndrome, prolongation of the QT interval, peripheral neuropathy, mild myelosuppression, hyperglycemia and hypokalemia. Of these, QT prolongation and, particularly, the so called APL differentiation syndrome are the most serious ones as they can evolve into severe and potentially fatal ventricular arrhythmias (torsade de points) or respiratory failure, respectively (10). According to original clinical trials reported in China, ATO was able to induce hematological CR in >85% patients who relapsed after front-line ATRA. These results were subsequently reproduced in the USA first in a pilot, then in an expanded multicenter trial for patients relapsed after ATRA. A CR rate of 86% was reported in the US multicenter study. Significantly, unlike ATRA, ATO as a single agent was able to induce durable molecular remission after two cycles in the majority of patients treated for disease recurrence. Confirmation of the high efficacy of ATO for relapsed APL was provided successively by several trials conducted worldwide which reported CR rates >70% and 1–3 years survival rates in the range of 50–70% (10). In addition to trials in which ATO was used a single agent, some studies investigated its efficacy and toxicity profile in combination with other agents including ATRA. Synergism with ATRA and increased antileukemic efficacy in APL was demonstrated in a Chinese randomised study comparing ATO + ATRA vs. either ATO or ATRA used as single agents (11). No significant additional toxicity was reported in this or in other studies which analysed the effect of ATRA and ATO combination. Following the experience in relapsed patients and based on the favourable toxicity profile, several

Ann Hematol (2011) 90 (Suppl 1):S25–S76 investigators have more recently explored the effect of ATO in newly diagnosed APL patients and reported preliminary findings in front-line therapy. Results of studies from Shanghai, Houston, India and Iran conducted with ATO as single agent or combined to ATRA for newly diagnosed patients reported CR rates of 86–95%, molecular remission rates after two cycles of 76–100% and survival rates of 86–88%, with significantly better responses being obtained in patients with low and intermediate-risk disease as compared to high-risk patients (10–15). Although these data need to be strengthened by studies in larger series and with more prolonged observation, they strongly suggest that at least non-high risk APL patients may be cured without chemotherapy. This possibility is currently being tested in a randomised trial of the GIMEMA and MRC cooperative groups group which compares this approach with the current standard ATRA plus chemotherapy front-line therapy. Meanwhile, the antileukemic efficacy of arsenic trioxide in front-line therapy has also been confirmed in a large randomised study of the US Intergroup investigating the benefit of adding ATO for consolidation to the ATRA plus chemotherapy schedule (16). Conflict of interest None References 1. Sanz MA, Grimwade D, Tallman MS et al. Management of acute promyelocytic leukemia : recommendations from an expert panel on behalf of the European LeukemiaNet. Blood 2009; 113: 1875–91. 2. Fenaux P, Wang Z, Degos L. Treatment of acute promyelocytic leukaemia. Curr Top Microbiol Immunol. 2007;313:101–28 3. Mandelli F, Diverio D, Avvisati G, et al. Molecular remission in PML/RAR alpha positive acute promyelocytic leukemia by combined all-trans retinoic acid and idarubicin (AIDA) therapy. Blood 1997; 90: 1014–21. 4. Sanz MA, Martin G, Gonzalez M, Leon A, Rayon C, Rivas C, Colomer D, et al. Risk-adapted treatment of acute promyelocytic leukemia with all-trans retinoic acid and anthracycline monochemotherapy: a multicenter study by the PETHEMA Group. Blood 2004; 104: 3490–93. 5. Lo-Coco F, Avvisati G, Vignetti M et al. Front-line treatment of acute promyelocytic leukemia with AIDA induction followed by risk-adapted consolidation for adults patients younger than 61 years: results of the AIDA-2000 trial of the GIMEMA Group. Blood First Edition Paper, prepublished online July 19, 2010;DOI 10.1182/ blood-2010-03-276196 6. Cheson BD, Bennett JM, Kopecky KJ, et al. Revised recommendations of the International Working Group for Diagnosis, Standardization of Response Criteria, Treatment Outcomes, and Reporting Standards for Therapeutic Trials in Acute Myeloid Leukemia. J Clin Oncol. 2003;21:4642–49. 7. Grimwade D, Jovanovic JV, Hills RK, et al. Prospective minimal residual disease monitoring to predict relapse of acute promyelocytic leukemia and to direct pre-emptive arsenic trioxide therapy. J Clin Oncol. 2009 Aug 1;27(22):3650–58 8. Anderlini P, Benjamin, Wong F, et al. Idarubucin cardiotoxicity: a retrospective analysis of in acute myeloid myeloid leukemia and myelodysplasia. J Clin Oncol 1995; 13: 2827–34. 9. Latagliata R, Petti MC, Fenu S et al. Therapy related myelodysplastic syndrome and acute myelogenous leukemia in patients treated for acute promyelocytic leukemia: an emerging problem. Blood 2002; 99:822–24.

Ann Hematol (2011) 90 (Suppl 1):S25–S76 10. Sanz MA, Lo-Coco F. Arsenic trioxide: Its use in the treatment of acute promyelocytic leukemia. Am J Cancer 2006; 5:5183–91. 11. Hu J, Liu YF, Wu CF, et al. Long-term efficacy and safety of Alltrans retinoic acid/arsenic trioxide-based therapy in newly diagnosed acute promyelocytic leukemia. Proc Natl Acad Sci USA 2009;106:3342–47. 12. George B, Mathews V, Poonkuzhali B, et al. Treatment of children with newly diagnosed acute promyelocytic leukemia with arsenic trioxide: a single center experience. Leukemia 2004;18:1587–90 13. Ghavamzadeh A, Alimoghaddam K, Ghaffari SH, et al. Treatment of acute promyelocytic leukemia with arsenic trioxide without ATRA and/or chemotherapy. Ann Oncol 2006;17; 131–34. 14. Ravandi F, Estey E, Jones D et al. Effective treatment of acute promyelocytic leukemia with all-trans retinoic acid, arsenic trioxide and gemtuzumab ozogamicin. J Clin Oncol 2009; 27: 504–10. 15. Mathews V, George B, Lakshmi KM, et al. Single agent arsenic trioxide in the treatment of newly diagnosed acute promyelocytic leukemia: durable remissions with minimal toxicity. Blood 2006; 107: 2627–32. 16. Powell BL, Moser B, Stock W,et al. Arsenic trioxide improves event-free and overall survival for adults with acute promyelocytic leukemia: North American Leukemia Intergroup Study C9710. Blood 2010;116:3751–7 Treatment of Childhood Acute Lymphoblastic Leukemia without Cranial Irradiation C.H. PUI, D. CAMPANA, J.T. SANDLUND, DEEPA BHOJWANI, W.E. EVANS, M.V. RELLING, S. JEHA St. Jude Children’s Research Hospital and the University of Tennessee Health Science Center, Memphis, TN, USA Acknowledgment Supported in part by grant CA-21765 from the National Cancer Institute, by a Center of Excellence grant from the State of Tennessee, and by the American Lebanese Syrian Associated Charities (ALSAC). Key words: Acute lymphoblastic leukemia; cranial irradiation; central-nervous-system leukemia; triple intrathecal therapy Abstract The St. Jude Total Therapy XV Study yielded an outstanding treatment outcome for children and adolescents with acute lymphoblastic leukemia without the use of prophylactic cranial irradiation. However, approximately 2–3% of the patients required cranial irradiation for salvage therapy of central-nervous-system (CNS) relapse. In our current Total Therapy XVI study, we seek to determine if intensified intrathecal and systemic therapy for patients at high risk of CNS relapse can further improve outcome. Preliminary results are encouraging, with no CNS relapses observed among the first 180 patients treated in the first 3 years of the study. Introduction With the achievement of 5-year event-free survival rates of over 80% for childhood acute lymphoblastic leukemia (ALL) in contemporary

S61 clinical trials, recent efforts have focused on optimal risk-directed therapy to avoid over- or under-treatment. Because cranial irradiation can cause many serious late sequelae, including second cancers, neuro-cognitive deficits, and multiple endocrinopathies [1], this treatment modality has largely been replaced by intensive intrathecal and systemic chemotherapy, and is reserved for high-risk patients or those who develop central-nervous-system (CNS) relapse in current clinical trials. It is well recognized that high-risk genetic features, T-cell immunophenotype, large leukemic cell burden, and the presence of leukemic blast cells in cerebrospinal fluid (even from traumatic lumbar puncture) are associated with an increased risk of CNS relapse [2]. In the St. Jude Total Therapy XIIIB study, we intensified triple intrathecal therapy for these patients, used dexamethasone in postremission therapy, and limited prophylactic cranial irradiation to patients with T-cell ALL and a presenting leukocyte count of 100× 109/L, or those with a CNS-3 status (a total of 12%) [3]. This approach resulted in a 5-year event-free survival of 80.8% and a cumulative incidence of isolated CNS relapse of 1.7%. This encouraging result prompted us to test whether intensification of systemic drugs that affect control of ALL in the CNS, such as highdose methotrexate, dexamethasone, and asparaginase, together with optimal intrathecal treatment, would allow the complete omission of prophylactic cranial irradiation without compromising overall survival in our Total Therapy XIV and XV studies [4,5]. Although the Total Therapy XIV study was terminated early because of excessive toxicities during remisson induction, none of the 53 patients enrolled in the study developed an isolated CNS relapse [4]. In the Total therapy XV study, we achieved a 5-year event-free survival of 85.6% and a 5-year cumulative risk of isolated CNS relapse of 2.7%. All 11 patients with isolated CNS relapse remained in second remission for 0.4–5.5 years at the time of the report. Importantly, this outstanding result extended to older adolescent patients [6]. That prophylactic cranial irradiation can be safely omitted in all patients was confirmed by a study of Dutch Childhood Oncology Group, which yielded a 5-year event-free survival rate of 81% and a cumulative incidence of isolated CNS relapse of 2.6% [7]. Since prophylactic cranial irradiation was not used in the Total Therapy XV study, we could clearly identify risk factors for CNS relapse. These are the presence of the t(1;19) with TCF3-PBX1 fusion, T-cell immunophenotype, and any CNS involvement at diagnosis [5,8]. In our ongoing Total Therapy XVI study (October 2007 to present), one of the secondary therapeutic aims is to determine if further intensification of intrathecal and systemic therapy can improve the outcome of patients with the features indicative of a high risk of CNS relapse. Method Risk Classification In the Total Therapy XVI study, patients were classified into one of three risk categories (low, standard, and high) based on the presenting age, initial leukocyte count, the presence or absence of CNS-3 status or testicular leukemia, immunophenotype, cytogenetics and molecular genetics, DNA index, and the early response to therapy. The definitive risk assignment (for provisional low-risk or standard-risk cases according to presenting features) was based on the level of minimal residual disease after completion of remission induction therapy. Low-risk ALL was defined by B-cell precursor

S62 immunophenotype with DNA index≥1.16, ETV6-RUNX1 fusion, or age 1–9.9 years and presenting WBC <50×109/L; these patients must not have a CNS-3 status, overt testicular leukemia, the t(9;22) with BCR-ABL1 fusion, the t(1;19) with TCF3-PBX1 fusion, rearranged MLL, hypodiploidy <44 chromosomes, or poor early response (≥ 1% lymphoblasts on day 15 of remission induction or≥0.01% lymphoblasts upon completion of remission induction on day 42). High-risk cases included those with the t(9;22), infants with the t (4;11) or MLL rearrangement, induction failure or≥1% leukemic lymphoblasts in the bone marrow on remission date, ≥ 0.1% leukemic lymphoblasts in the bone marrow on week 7 of continuation treatment, or early T-cell precursor immunophenotype. All other cases were classified to have standard-risk ALL.

Treatment Remission induction Remission induction therapy begun with daily prednisone (40 mg/m2 for 28 days), weekly vincristine (1.5 mg/m2 for four doses), weekly daunorubicin (25 mg/m2 for two doses on days 1 and 8), and PEGasparaginase (3,000 units/m2 intravenously on day 3). Patients with≥ 1% residual leukemia in the bone marrow after 2 weeks of induction were given another dose of PEG-asparaginase (3,000 units/m2) on day 15. Subsequent induction therapy consisted of cyclophosphamide (1 g/m2) on day 22, thioguanine or mercaptopurine in patients with deficiency of thiopurine methyltransferase deficiency (60 mg/m2 per day) on days 22–35, and cytarabine (75 mg/m2) on days 23–26 and 30–33. Patients with≥5% residual leukemia in the bone marrow on day 15 were treated with fractionated cyclophosphamide (300 mg/m2 every 12 h for 4 doses) instead of a single dose of cyclophosphamide on day 22. Because of historically poor outcome of infants with MLL rearranged ALL, the second part of remission induction was intensified in them: intravenous clofarabine (40 mg/m2) plus etoposide (100 mg/m2) plus cyclophosphamide (300 mg/m2) daily from day 22 to day 26. For patients with the t(9;22) and BCR-ABL1 fusion, dasatinib was started on day 22 of remission induction treatment and continued until the end of therapy. Upon recovery of hematopoietic function around day 38 to day 42 of remission induction, bone marrow aspiration was performed to assess remission status and the level of minimal residual leukemia. Consolidation treatment Consolidation therapy consisted of high-dose methotrexate infused over 24 h and age-adjusted triple intrathecal therapy with methotrexate, hydrocortisone and cytarabine (every other week for 4 doses) and daily mercaptopurine (50 mg/m2 per day) for 8 weeks. The dosage of methotrexate was adjusted to achieve a steady-state concentration of 65 μM (corresponding to an average dose of approximately 5 g/m2) in standard- or high-risk cases, and 2.5 g/m2 in low-risk cases. Continuation therapy At the beginning of continuation treatment, patients were stratified and randomized to receive subsequent doses of PEG-asparaginase at 2,500 or 3,500 units/m2 per dose intravenously. In the first 19 weeks of continuation therapy, low-risk cases received daily mercaptopurine (75 mg/m2) and weekly methotrexate (40 mg/m2) with pulses of

Ann Hematol (2011) 90 (Suppl 1):S25–S76 mercaptopurine, dexamethasone (8 mg/m2 per day for 5 days) and vincristine (2 mg/m2; maximum 2 mg) given every 3–5 weeks. Two reinduction treatments consisting of dexamethasone, vincristine, and PEG-asparaginase were given between weeks 7 and 9 and weeks 17 and 20; one dose of doxorubicin (30 mg/m2) was given on week 7. Standard- and high-risk cases received daily mercaptopurine (50 mg/ m2) with PEG-asparaginase every other week for 15 doses and pulses of doxorubicin (30 mg/m2 per dose) plus vincristine plus dexamethasone (12 mg/m2 per day for 5 days) every 3 weeks. They also received two reinduction treatments consisting of dexamethasone, vincristine, and PEG-asparaginase together with doxorubicin in the first reinduction (weeks 7–9) and high-dose cytarabine (2 g/m2 every 12 h for 4 doses) in the second reinduction (weeks 17–20). Patients who developed allergic reactions to PEG-asparaginase, were given Erwinia asparaginase subsequently. The remaining continuation therapy for patients with low-risk ALL consisted of daily mercaptopurine (75 mg/m2) and weekly methotrexate (40 mg/m2), interrupted by pulse therapy every 4 weeks (up to week 101) with dexamethasone (8 mg/m2 per day for 5 days), vincristine (2 mg/m2) and mercaptopurine (75 mg/m2 per day for 7 days). In patients with standard-risk ALL, the remaining continuation therapy consisted of three drug pairs given in 4-week blocks: mercaptopurine (75 mg/m2 daily for 7 days) plus methotrexate (40 mg/m2) in the first and second weeks, cyclophosphamide (300 mg/m2) plus cytarabine (300 mg/m2) in the third week (replaced by mercaptopurine and methotrexate after week 68) and dexamethasone (12 mg/m2 per day for 5 days) plus vincristine (2 mg/m2) in the fourth week (replaced by mercaptopurine and methotrexate after week 101). Dexamethasone dose was reduced to 6 mg/m2 per day for 5 days for all patients between week 69 and week 101. The total duration of continuation treatment was 120 weeks. The total scheduled doses of anthracyclines were 80 mg/m2 and 230 mg/m2, and cyclophosphamide 1 g/m2 and 4.6– 4.8 g/m2 for low-risk and standard- or high-risk patients, respectively. The dosages of mercaptopurine and methotrexate were tailored to the limits of tolerance (white cell count between 1.5 and 3.0×109/L and absolute neutrophil count above 0.5×109/L); higher count was desirable a week following dexamethasone treatment and caution was taken to avoid overzealous dose escalation leading to interruption of chemotherapy. Thiopurine methyltransferase phenotype and genotypes are determined prospectively in all patients, and the starting dose of mercaptopurine was reduced to 60 mg/m2 in those with heterozygous enzyme deficiency and to 5–10 mg/m2 in the rare patients with homozygous enzyme deficiency. A trimethoprim and sulfamethoxazole combination was given to all patients twice daily for three consecutive days per week from day 15 of remission induction to 6 weeks after completion of all chemotherapy as prophylaxis against Pneumocytis carinii pneumonia. CNS-directed therapy Triple intrathecal therapy, at age-appropriate dose, was instilled immediately after collection of cerebrospinal fluid with diagnostic lumbar puncture and was repeated on day 15 and at the end of remission induction (coinciding with the start of consolidation therapy with high-dose methotrexate). Additional triple intrathecal therapy was given to patients with the t(9;22), MLL rearrangement, hypodiploidy <44 chromosomes or WBC >100×109/L on days 8 and 22, and to those with T-cell ALL, the t(1;19) or any CNS involvement (CNS2, CNS-3, traumatic lumbar puncture with blasts) on days 4, 8, 11 and 22 of remission induction. Triple intrathecal chemotherapy was given

Ann Hematol (2011) 90 (Suppl 1):S25–S76 every 2 weeks during consolidation treatment, and then every 8 weeks in low-risk cases and every 4 weeks in standard- or high-risk cases during the first year of continuation treatment. Patients at high risk of CNS relapse continued to receive intrathecal therapy every 8 weeks beyond the first year until week 97 of continuation therapy. Depending on the CNS status, low-risk patients received 13–21, and standardor high-risk cases 16–27 intrathecal treatments. Allogeneic hematopoietic stem-cell transplantation This procedure was offered as a treatment option for patients with high-risk leukemia and high level of minimal residual leukemia (i.e., ≥ 1% leukemic cells in bone marrow at the end of remission induction or≥0.1% in week 7 of continuation treatment). To maximize leukemic cell kill before transplantation, reintensification therapy with highdose dexamethasone (20 mg/m2 per day for 5 days), high-dose cytarabine (2 g/m2 every 12 h for 4 doses) followed by etoposide (100 mg/m2 every 12 h for 5 doses), and PEG-asparaginase (3,000 units/m2) on day 6 was given to these patients. This course was repeated if the patient had responded to the first course but still had measurable residual leukemia. Results and Discussion Treatment on Total XVI study is similar to that used in the Total XV study but includes several modifications. Intramuscular native E Coli asparaginase is replaced by intravenous PEG-asparaginase. Mercaptopurine during days 22–25 of remission induction is replaced by thioguanine at 60 mg/m2 per day for patients with wild-type thiopurine methyltransferase activity. Cyclophosphamide is intensified to 4 fractionated doses (300 mg/m2 every 12 h), instead of 1 g/m2 for one dose, for standard- or high-risk patients with poor early treatment response (i.e., 5% or more leukemic blast cells on day 15 of remission induction). For patients with high risk of CNS relapse, triple intrathecal treatment is further intensified during remission induction: twice weekly in the first 2 weeks, and weekly for the subsequent 2 weeks. All intrathecal treatments in induction are followed by leucovorin rescue (5 mg/m2 per dose, maximum 5 mg) at 24 and 30 h after instillation to reduce the risk of neurotoxicity and myelosuppression. Because of its dual targeting of ABL and SRC, more potent suppression of BCR-ABL1 signaling, efficacy to most imatinib-resistant BCR-ABL1 mutants, as well as good CNS penetration and tolerability in adult clinical trials, dasatinib was used instead of imatinib for patients with the t(9;22) [9]. To improve quality of life, dexamethasone dose is decreased to 6 mg/m2 per day (from 8 to 12 mg/m2 per day) for all patients after week 69 of continuation treatment, and the duration of continuation treatment is shortened from 146 weeks to 120 weeks for boys. Preliminary results are encouraging with no CNS relapse observed among the first 180 patients enrolled in the study. There has been no excessive neurotoxicity noted thus far. These early results suggest that intensification of triple intrathecal therapy with leucovorin rescue is feasible and safe. Longer follow-up is necessary to determine if our current treatment approach can totally eliminate CNS relapse. Conflict of interest None References 1. Pui CH, Cheng C, Leung W, et al. (2003) Extended follow-up of long-term survivors of childhood acute lymphoblastic leukemia. N Engl J Med 349:640–649

S63 2. Pui CH, Howard SC (2008) Current management and challenges of malignant disease in the CNS in paediatric leukaemia. Lancet Oncol 9:257–268 3. Pui CH, Sandlund JT, Pei D, et al. (2004) Improved outcome for children with acute lymphoblastic leukemia: results of Total Therapy Study XIIIB at St Jude Children’s Research Hospital. Blood 104:2690–2696 4. Pui CH, Pei D, Sandlund JT, et al. (2010) Long-term results of St Jude Total Therapy Studies 11, 12, 13A, 13B, and 14 for childhood acute lymphoblastic leukemia. Leukemia 24: 371–382 5. Pui CH, Campana D, Pei D, et al. (2009) Treating childhood acute lymphoblastic leukemia without cranial irradiation. N Engl J Med 360:2730–2741 6. Pui CH, Pei D, Campana D, et al. (2010) Improved prognosis for older adolescents with acute lymphoblastic leukemia. J Clin Oncol (in press) 7. Veerman AJ, Kamps WA, van den Berg H, et al. (2009): Dexamethasone-based therapy for childhood acute lymphoblastic leukaemia: results of the prospective Dutch Childhood Oncology Group (DCOG) protocol ALL-9 (1997–2004). Lancet Oncol. 10:957–966 8. Jeha S, Pei D, Raimondi SC, et al. (2009) Increased risk for CNS relapse in pre-B cell leukemia with the t(1;19)/TCF3-PBX1. Leukemia. 2009 Aug;23(8):1406–1409 9. Pui CH, Jeha S (2007) New therapeutic strategies for the treatment of acute lymphoblastic leukaemia. Nat Rev Drug Discov 6:149–165

Sequential Therapy for High Risk AML CH. SCHMID Stem Cell Transplantation Unit, Department of Medicine II, Klinikum Augsburg, Ludwig-Maximilians-Universität, Munich, Germany Introduction Acute myeloid leukemia (AML) represents a heterogenous group of diseases, ranging from rather sensitive diseases with a reasonable chance for cure by conventional chemotherapy, to high risk forms with a median survival of only few months. Increased risk can defined (1) at time of diagnosis and (2) during the course of the disease. At diagnosis, the karyotype of the malignant clone is the best established risk factor1-3. Although some difference exist among the definitions of prognostic cytogenetic subgroups, all research teams have identified t(15;17), inv16 and t(8;21) as favorable aberrations, whereas monosomy 5 and 7, del 7q, del 5q (in AML, not in MDS), and a komplex aberrant karyotype (i.e. 3 or more aberrations) are associated with a dismal outcome. Recently, the description of the monosomal karyotype has identified a subgroup with a particularly bad prognosis.4 In addition to cytogenetic aberrations, molecular research has identified certain mutations that allow a risk estimation in particular among patients with a normal karyotype.5 Although the rapidly ongoing detection of new molecular markers as the IDH gene mutations6, as well as the growing insight into the role of the combination of certain mutations7 may change our opinion on the prognostic relevance of molecular markers, there is little debate at present on the favorable prognosis of AML bearing a NPM1 or CEPBα mutation, whereas AML with an internal tandem duplication (ITD) in the FLT3 gene is considered to have a much worse prognosis. During the course of the disease, increased risk can be defined by failure or delayed response to induction chemotherapy8;9, early relapse after a

S64 remission duration of less then 6–12 months, second or higher relapse and relapse following high-dose therapy and autologous stem cell transplantation.10 In the light of unsatisfying results with conventional therapy, allogeneic stem cell transplantation (alloSCT) is the recommended treatment for high risk AML and MDS.11–13 However, in spite of the undisputable curative potential of alloSCT, which is mainly based on the allogeneic Graft-versus-Leukemia (GvL) effect, results are hampered both by treatment-related mortality and high incidence of relapse. Particularly in elderly patients, patients with MDS and secondary AML, and patients with advanced disease, non-relapse mortality (NRM) may reach >70%.14–17 Therefore, reduced-intensity conditioning (RIC) regimen have been introduced, which, as a consequence of low acute toxicity, allow to offer an alloSCT also to patients beyond the age of 60, and those with relevant co-morbidities. However, so far it is uncertain, to what extend the conditioning regimen can be reduced without jeopardizing the efficacy of the whole procedure. The intensity of the preparative regimen has been shown to directly influence relapse incidence and leukemia-free survival after allografting for AML and MDS,18 and disappointing results have been reported from RIC transplants in advanced disease.19,20 This might be due on one side to the missing direct cytotoxic effect of high-dose radio-chemotherapy against the underlying disease. On the other side, the GvL reaction might also be less powerful after RIC, if the limited cytoreduction applied does not allow enough time for an allogeneic immune effect against highly proliferative AML. Studies on the kinetics of disappearance of the bcr-abl transcript after donor lymphocyte infusion (DLI) in CML have shown, that it may take up to 1 year until a complete molecular remission is achieved. Hence, early progression of acute leukemia after RIC transplantation may simply overrun an ongoing GvL reaction. The principle of sequential therapy Based on these considerations, new concepts had to be developed, in order to find a balance between preserved antileukemic activity and lowered treatment-associated toxicity of RIC. The chosen strategy attempted to separate the two effects of conditioning— direct antileukemic cytotoxicity for disease control, and immunosuppression to allow for engraftment—by a sequence of a course of intensive chemotherapy, followed by a mainly immunosuppressive RIC. A similar approach had been developed earlier in the treatment of multiple myeloma: Reduction of tumor burden is achieved by high-dose chemotherapy with autologous stem cell rescue, followed by a RIC alloSCT within 2 months to implement an allogeneic immune response. Due to the higher proliferative activity of high-risk and advanced AML, the interval between the two blocks was reduced to 3 days in the sequential protocol presented here. In addition, the GvL effect should be augmented by early tapering of the immunosuppressive medication, and prophylactic transfusion of donor lymphocytes. For cytoreduction, a chemotherapy course similar to common induction or salvage protocols was used, including fludarabine (30 mg/m2), high-dose AraC (2 g/m2) and amsacrine (100 mg/m2) for 4 days each (FLAMSA regimen). It was followed by 3 days of rest, allowing time to recover from acute toxicity, concerning in particular the intestinal mucosa and liver. Then, RIC consisted of 4 Gy total body irradiation (TBI) on day –5, cyclophosphamide (40 mg/kg with HLA-id sibling, 60 mg/kg for unrelated or mismatched donors) on days –4 and –3, and rabbit antithymococyte globulin (ATG; 10 mg/

Ann Hematol (2011) 90 (Suppl 1):S25–S76 kg for HLA-id sibling, 20 m/kg for unrelated or mismatched donors) from day –4 until day –2. For transplantation, G-CSF mobilized peripheral blood stem cells (PBSC) were preferred, bone marrow (BM) was accepted at the donor’s preference. GvHD prophylaxis consisted of cyclosporine A (CyA) from day –1, and mycophenolat mofetil (MMF, 2× 15 mg/kg), starting from day 0 In the absence of GvHD, MMF was discontinued by day +50, CyA was tapered from day +60 to +90. To increase the allogeneic GvL effect, patients received prophylactic donor lymphocyte transfusions (pDLT), if they were in CR without evidence of GvHD at day +120 or 30 days after discontinuation of immunosuppression. The initial dose was 1× 106 CD3+ cells/kg; it could be increased to 5×106 CD3+ cells/kg in patients without a history of aGvHD. In the absence of GvHD, pDLT was repeated up to 3x, using escalating cell doses (5-10 fold increase/transfusion) at 4–6 weeks’ intervals. A summary of the protocol is shown in figure 1. PBSCT Fludara AraC Amsacrine

Day -12 -11

-10 -9

TBI ATG ATG ATG 4Gy CY CY

-8

-7

-6

-5

-4

-3

-2

Off immunosuppression

-1

0

+90

Prophylactic DLT

+ 120 + 150 + 180

CSA MMF

Figure 1. The FLASMA-RIC regimen for allogeneic stem cell transplantation Results The protocol was first evaluated in a prospective pilot trial of 75 consecutive patients at the Ludwig-Maximilians-University Hospital of Munich.21 In contrast to other trials investigating modern RIC regimen,22–24 inclusion criteria were not based on medical contraindications against standard conditioning, but on an increased risk disease, as defined by one or more of the following criteria: (1) primary or secondary refractory leukemia (as defined by persisting disease following≥1 course of high-dose AraC), (2) delayed response to induction chemotherapy (3) relapse within 3 months from induction or consolidation therapy, (4)≥2nd relapse, or relapse following autologous HSCT (5) unfavorable cytogenetics2, (6) AML secondary to MDS or other malignancies, and (7) progressive MDS/RAEB. Median patient age was 52.3 years, unrelated donors were used half of the patients. Twenty-seven patients had PIF, 22 had untreated relapse, 10 had progressive MDS, and 16 were in remission. The karyotype of 71 informative patients was favorable in 3, intermediate in 30, unfavorable in 35 (including 19 with complex abnormalities), and of unknown significance in 3. The regimen proved to be myeloablative in all cases. Neutrophil engraftment occurred at a median of 14 days, and donor chimerism of> 90% was achieved at day +30 in the vast majority of patients. The overall non-relapse mortality (NRM), including deaths related to concomitant disease, was 20% at day +100 and 33% at 1 year. After a median follow up of 31.5 months, overall survival (OS) and disease free survival (DFS) were 42 and 40%. In a multivariate analysis, a higher number of CD34+ cells in PBSC recipients was the only pre-transplant variable being associated with better survival. After transplantation, acute and chronic GvHD had a significant influence on outcome: Severe forms were

Ann Hematol (2011) 90 (Suppl 1):S25–S76 associated with high NRM and were deleterious for outcome, whereas patients with mild GvHD did significantly better. Later on, the rapid spreading of the protocol allowed for the analysis of more precisely defined subsets of patients. In a study on patients transplanted in first complete remission (CR1), high-risk was defined by AML secondary to MDS or radio/chemotherapy (52%), unfavorable cytogenetics (65%) or delayed response to induction chemotherapy (52%). 43% had one, 42% had two, and 7% had all three risk factors. Following FLAMSA-RIC for alloSCT, OS at 4 years was 72%, thereby equalizing the results of a control group of patients with standard risk disease, who received a conventional conditioning before sibling alloSCT during the same time period.25 Another study analyzed the outcome of 103 patients transplanted with refractory disease. Refractoriness was defined by primary induction failure (PIF, 36%), early (52%), refractory (7%), or second (5%) relapse. OS at 2 and 4 years from transplantation was 54% and 32%; the respective leukemia free survival (LFS) was 47%, and 30%. Patients transplanted in PIF had a 2y OS of 62.5% In a multivariate analysis, a history of more than 2 courses of prior chemotherapy were the strongest predictor for poor outcome (P = .007; HR=3.01 [OS]; P = .002; HR=3.25 [LFS]).26 More recently, the final analysis of a prospective study designed specifically for patients with complex aberrant karyotype was presented.27 Considering the low sensitivity of these diseases to conventional induction chemotherapy, an urgent search for a suitable stem cell donor, and transition from the induction protocol to alloSCT according to the FLAMSA-RIC regimen were part of the studied strategy. 18 patients from 4 institutions were included. They received alloSCT in CR/CR1 (n=8) or persistant disease (n=10). Median time from diagnosis to transplantation was 91 days. After a median follow up among survivors of 23 (range: 3–33) months, 8 patients have died from leukemia (n=3) or treatment-related causes (n=4). Overall survival at 1 and 2 years from transplantation is 74% and 58%, the corresponding leukemia free survival is 69% and 58%. Even with low numbers, that result from the relatively low incidence of complex karyotype AML in patients <60 years, these results suggest a promising activity of the approach of early alloSCT in those patients who otherwise have a dismal outcome both after conventional treatment and standard conditioning alloSCT. Finally, the FLAMSA-RIC registry, that has been established, allows to address relevant questions of AML in the context of alloSCT against the background of a uniform conditioning registry. Among other studies, the role of molecular subgroups in cytogenetically normal AML (CNAML) has recently been studied.28 As observed earlier, the surprisingly good outcome of patients transplanted in primary induction failure was confirmed in CN-AML. 2y OS was 66% among these patients, with no difference between genotypes based of the presence or absence of NMP1- and FLT3-ITD mutations. In contrast, the genotype was of major importance when the results of patients transplanted in CR1 vs. beyond CR1 were compared: In patients with an isolated NPM1 mutation, nearly identical results were obtained after alloSCT in CR1 or more advanced stage. However, in other genotypes, namely FLT3-ITD mutated or double negative diseases, results of alloSCT in advanced disease were significantly worse as compared to transplantation in CR1, arguing for early alloSCT also in these subtype of AML. Outlook The strategy of sequential therapy has been accepted as a successful approach to alloSCT in high-risk AML5 Beside the FLAMSA-RIC protocol, similar concepts have been presented by several other groups.

S65 The group from Marseille has shown promising results by RIC-alloSCT in CR1, following intensified consolidation.29 An even more consequent approach was chosen by the Dresden group, who initiated an urgent donor search at time of diagnosis of a high-risk disease patient, and performed alloSCT in aplasia after induction treatment. An impressively short interval of 40 days between diagnosis and SCT was reached, and results were promising, showing a 3y OS of 61%.30 More recently, several modifications of the FLAMSA-RIC protocol have been developed. Among others, the feasibility of a combination of FLAMSA chemotherapy and a classical standard conditioning has been shown. Other groups have included modern drugs such as clofarabin into the cytoreductive course preceding RIC.31 Within our own group, substitution of i.v busulfane for total body irradiation was successful in reducing treatment-related toxicity with at least maintained antileukeimic efficacy in elderly patients. As shown by the promising preliminary results coming out of these studies, the concept of sequential strategy might serve as a backbone for an individualized and highly effective treatment for high-risk AML Conflict of interest None References 1. Grimwade D, Walker H, Oliver F et al. The importance of diagnostic cytogenetics on outcome in AML: analysis of 1,612 patients entered into the MRC AML 10 trial. The Medical Research Council Adult and Children's Leukaemia Working Parties. Blood 1998;92:2322–2333. 2. Slovak ML, Kopecky KJ, Cassileth PA et al. Karyotypic analysis predicts outcome of preremission and postremission therapy in adult acute myeloid leukemia: a Southwest Oncology Group/Eastern Cooperative Oncology Group Study. Blood 2000;96:4075–4083. 3. Schoch C, Kern W, Schnittger S, Hiddemann W, Haferlach T. Karyotype is an independant prognostic parameter in therapyrelated acute myeloid leukemia (t-AML): an analysis of 93 patients with t-AML in comparision to 1092 patients with de novo AML. Leukemia 2004;18:120–125. 4. Breems D, Van Putten W, de Greef G et al. Monosomal Karyotype in Acute Myeloid Leukemia: A Better Indicator of Poor Prognosis Than a Complex Karyotype. Journal of Clinical Oncology 2008;26:4791–4797. 5. Dohner H, Estey E, Amadori S et al. Diagnosis and management of acute myeloid leukemia in adults: recommendations from an international expert panel, on behalf of the European LeukemiaNet. Blood 2010;115:453–474. 6. Paschka P, Schlenk RF, Gaidzik VI et al. IDH1 and IDH2 mutations are frequent genetic alterations in acute myeloid leukemia and confer adverse prognosis in cytogenetically normal acute myeloid leukemia with NPM1 mutation without FLT3 internal tandem duplication. J.Clin.Oncol. 2010;28:3636–3643. 7. Bacher U, Haferlach C, Kern W, Haferlach T, Schnittger S. Prognostic relevance of FLT3-TKD mutations in AML: the combination matters— an analysis of 3082 patients. Blood 2008;111:2527–2537. 8. Estey EH, Shen Y, Thall PF. Effect of time to complete remission on subsequent survival and disease-free survival time in AML, RAEB-t, and RAEB. Blood 2000;95:72–77. 9. Kern W, Haferlach T, Schoch C et al. Early blast clearance by remission induction chemotherapy is a major independant prognostic factor for both achievement of complete remission and longterm outcome in acute myeloid leukemia: data from the German

S66 AML cooperative group (AMLCG) 1992 trial. Blood 2003;101: 64–70. 10. Estey E. Treatment of refractory AML. Leukemia 1996;10:932– 936. 11. Edenfield WJ, Gore SD. Stage-specific application of allogeneic and autologous marrow transplantation in the management of acute myeloid leukemia. Semin.Oncol. 1999;26:21–34. 12. Appelbaum FR. Who should be transplanted for AML. Leukemia 2001;15:680–682. 13. Robak T, Wrzesien-Kus A. The search for optimal treatment in relapsed and refractory acute myeloid leukemia. Leuk.Lymphoma 2002;43:281–291. 14. Ringden O, Horowitz M, Gale RP. Outcome after allogeneic bone marrowe transplantfor leukemia in older adults. JAMA 1993;270:57–60. 15. Arnold R, de Witte T, van Biezen A et al. Unrelated bone marrow transplantation in patients with myelodysplastic syndromes and secondary acute myeloid leukemia: an EBMT survey. European Blood and Marrow Transplantation Group. Bone Marrow Transplant. 1998;21:1213–1216. 16. Singhal S, Powles R, Henslee-Downey PJ et al. Allogeneic transplantation from HLA-matched sibling or partially HLAmismatched related donors for primary refractory acute leukemia. Bone Marrow Transplant. 2002;29:291–295. 17. Witherspoon RP, Deeg HJ, Storer B et al. Hematopoietic stem-cell transplantation for treatment-related leukemia or myelodysplasia. J.Clin.Oncol. 2001; 19:2134–2141. 18. de Lima M, Anagnostopoulos A, Munsell M et al. Nonablative versus reduced-intensity conditioning regimens in the treatment of acute myeloid leukemia and high-risk myelodysplastic syndrome: dose is relevant for long-term disease control after allogeneic hematopoietic stem cell transplantation. Blood 2004;104:865– 872. 19. de Lima M, Couriel D, Thall PF et al. Once-daily intravenous busulfan and fludarabine: clinical and pharmacokinetic results of a myeloablative, reduced-toxicity conditioning regimen for allogeneic stem cell transplantation in AML and MDS. Blood 2004;104:857–864. 20. Stelljes M, Bornhauser M, Kroger M et al. Conditioning with 8 Gy total body irradiation and fludarabine for allogeneic hematopoietic stem cell transplantation in acute myeloid leukemia. Blood 2005;106:3314–3321. 21. Schmid C, Schleuning M, Ledderose G, Tischer J, Kolb HJ. Sequential regimen of chemotherapy, reduced-intensity conditioning for allogeneic stem-cell transplantation, and prophylactic donor lymphocyte transfusion in high-risk acute myeloid leukemia and myelodysplastic syndrome. J.Clin.Oncol. 2005;23:5675– 5687. 22. Bertz H, Potthoff K, Finke J. Allogeneic stem-cell transplantation from related and unrelated donors in older patients with myeloid leukemia. J.Clin.Oncol. 2003;21:1480–1484. 23. Taussig DC, Davies AJ, Cavenagh JD et al. Durable remissions of myelodysplastic syndrome and acute myeloid leukemia after reduced-intensity allografting. J.Clin.Oncol. 2003;21:3060–3065. 24. Wong R, Giralt SA, Martin T et al. Reduced-intensity conditioning for unrelated donor hematopoietic stem cell transplantation as treatment for myeloid malignancies in patients older than 55 years. Blood 2003;102:3052–3059.

Ann Hematol (2011) 90 (Suppl 1):S25–S76 25. Schmid C, Schleuning M, Hentrich M et al. High antileukemic efficacy of an intermediate intensity conditioning regimen for allogeneic stem cell transplantation in patients with high-risk acute myeloid leukemia in first complete remission. Bone Marrow Transplant. 2008;41:721–727. 26. Schmid C, Schleuning M, Schwerdtfeger R et al. Long term survival in refractory acute myeloid leukemia after sequential treatment with chemotherapy and reduced intensity conditioning for allogeneic stem cell transplantation. Blood 2006;108:1092– 1099. 27. Schmid C, Schleuning M., Tischer J et al. Long term survival in patients suffering from AML with a complex karyotype after early allogeneic stem cell transplantation using the FLAMSA-RIC regimen—Results from a prospective phase II trial [abstract]. Bone Marrow Transplant. 2008;41:S75. 28. Pfeiffer T, Schleuning M., Eder M et al. Allogeneic Stem Cell Transplantation in Acute Myeloid Leukemia with Normal Karyotype: A Risk Factor Analysis in 247 Patients, Based on Molecular Markers and Stage at Transplantation [abstract]. Blood (ASH Annual Meeting Abstracts), Nov 2009; 114: 1209. 2009; 29. Blaise D, Boiron JM, Faucher C et al. Reduced Intensity Conditioning Prior to Allogeneic Stem Cell Transplantation for Patients with AML as First-Line Treatment. Cancer 2005;104: 1931–1938. 30. Platzbecker U, Thiede C, Fussel M et al. Reduced intensity conditioning allows for up-front allogeneic hematopoietic stem cell transplantation after cytoteductive induction therapy in newly-diagnosed acute myeloid leukemia. Leukemia 2006;20: 707–714. 31. Buchholz S, Krauter J, Dammann E et al. Cytoreduction with clofarabine/AraC combined with RIC and alogeneic stem cell transplantation in patients with high-risk, relapsed or refractory acute leukaemia [abstract]. Bone Marrow Transplant 2009;43: S280. Peptide Vaccination in Patients with Acute Myeloid Leukemia M. SCHMITT, A. SCHMITT and R. CASALEGNO-GARDUNO University of Rostock, Germany Patients with acute myeloid leukemia (AML) are currently treated with polychemotherapy and hematopoietic stem cell transplantation (HSCT). However a high percentage of the patients relapse due to the persistence of malignant cells. The success of allogeneic HSCTs and particularly of donor lymphocytes infusions (DLIs) underlines the crucial role of T lymphocytes in the anti-leukemic immune response. In order to improve the survival of AML patients in general, novel therapeutic approaches have to be defined. One of these approaches consists in taking benefit of the adaptive arm of the immune system. In a synergetic manner targeted immunotherapies might help to eliminate minimal residual disease (MRD) after chemotherapy and also after HSCT. Different antigens have been identified to be overexpressed in leukemia blasts, so called leukemia associated antigens (LAAs). Such antigens have been related to proliferation, survival, inhibition of apoptosis, angiogenesis and metastasis of the malignant cells. The broad cluster of LAAs comprises the receptor for hyaluronic acid mediated motility (RHAMM), m-phase phosphoprotein 11 (MPP11), Wilms tumor 1 (WT1), proteinase-3 (PR-3), survivin, ovalbumin fetal

Ann Hematol (2011) 90 (Suppl 1):S25–S76 antigen incomplete laminar receptor protein (OFA-iLRP), FMS-like tyrosine kinase 3 internal tandem duplication (FLT-ITD), carbonic anhydrase 9 (CA IX/G250), breakpoint cluster region Abelson tyrosine kinase (BCR-ABL), human telomerase reverse transcriptase (hTERT), B-cell chronic lymphocytic leukemia/lymphoma 2, and the preferentially expressed antigen of melanoma (PRAME). These LAAs can be recognized by CD8+ cytotoxic T lymphocytes (CTL). The ideal LAA should be restricted to leukemia blasts, associated to conserved the leukemic phenotype and able to induce strong T cell responses. RHAMM, WT1, and PR-3 are LAAs that fulfill these requirements. The focus of our work is on RHAMM which was first described as a soluble binding protein release by subconfluent migrating cells that interacts with hyaluronan acid (HA) to promote cell motility and invasion. RHAMM has been related to motility, proliferation and transformation. Adhesion, wound healing, angiogenesis, migration, metastasis, invasion and growth are funtions related to RHAMM as well. RHAMM mRNA and protein are poorly expressed in most normal human tissue and in the immune privileged sites testis, placenta and adnexes. In contrast high levels of mRNA and protein RHAMM were detectable in all leukemia cell lines. Recently, peptides derived from several leukemia-specific antigens WT-1, RHAMM and PR3 have been tested in clinical peptide vaccination trials. More than 100 patients with AML have been vaccinated with CD8+ T-cell epitope peptides derived from these leukemia-specific antigens. After peptide vaccination the patients showed no severe side effects according to the Common Toxicity Criteria (CTC), except from grade I toxicity of the skin and grade I elevation of liver function tests.

Figure 1: Recognition of peptides derived from leukemia-antigens by the immune system Vaccines consisting of CD8+ T cell restricted epitopes can be presented by HLA-class I molecules on the surface of antigenpresenting cells (APCs) such as dendritic cells towards T cell receptor (TCR) molecules on the surface of CD8+ T cells (left hand panel). This presentation might be enhanced through longer peptides which comprise both class I and II epitopes of the given leukemia antigen (middle panel). By that means helper epitopes are presented to CD4+ helper T cells which can provide the approriate cytokine milieu for CD8+ cytotoxic T cells. Truncated proteins can comprise a plethora of class I and class II epitopes presented on different HLA molecules. Thus (truncated) proteins circumvent HLA restriction (right hand panel). Immunological responses have been detected by flow cytometry employing both intracellular cytokine staining and tetramer staining as well as by enzyme-linked immunosorbent spot (ELISPOT) assays, proliferation assays and cytotoxicity assays. Hematological responses like normalization of the peripheral blood count, reduction in blast

S67 cells in the bone marrow, reduction in expression of the respective LAA in the bone marrow, cessation of red blood cell transfusions could be detected. Peptide vaccinations have been administrated in patients with hematological malignancies in both complete and partial remissions as well as after autologous and allogeneic stem cell transplantation. To prevent tolerance of T cells versus a given LAA, two approaches have been undertaken: 1) addition of a peptide derived from a second LAA in the same syringe for vaccination, 2) generation of so-called analogue peptides with a mutation of the amino acid residue in defined positions of the peptide. Moreover, several categories of adjuvants have been used: 1) cytokines like granulocyte-macrophage colony stimulating factor (GM-CSF), 2) emulsifiers like incomplete Freund’s adjuvant (IFA; ISA 51) or quillaja saponaria (QS-21), 3) toll-like receptor ligands like oligodinucleotides, 4) HLA class II stimulating substances like keyhole limpet hemocyanine (KLH) or the pan-HLA-DR epitope (PADRE). All of these studies are phase I and early phase II clinical trials. Phase III clinical studies are fervently needed. However recent regulatory restrictions have led to a tremendous increase in the costs of peptide vaccines. Therefore the performance of such phase III clinical trials in the academic setting is extremely challenging. Several questions need to be addressed: the route of administration, the proper adjuvant and formulation of the vaccine, the frequency and duration of vaccination as well as the format of vaccination. The later might be further developed into truncated protein (Figure 1), mRNA or DNA formats to cover CD8+ and even CD4+ T cell epitopes and to circumvent human leukocyte antigen (HLA) restriction. This approach, maybe even realized for two or three LAAs would be highly rational and desirable, as it is imitating the procedure for vaccines against infectious diseases. However, the production of such truncated proteins under conditions of good manufacturing practice (GMP) costs roughly two million Euro per protein which can be only realized on the industrial level, but rather not in the academic setting. At present we combine peptide vaccination with allogeneic stem cell transplantation in the context of DLIs as so-called vaccine enhanced DLIs (veDLIs). Taken together peptide vaccination for leukemia patients has been demonstrated to be feasible and save. Immunological and even hematological responses have been observed. Future clinical trials have to be performed to define the clinical angle of vaccination therapy as a synergistic option in the therapy of leukemia patients. Conflict of interest None References 1. Bocchia M, Gentili S, Abruzzese E, Fanelli A, Iuliano F, Tabilio A, Amabile M, Forconi F, Gozzetti A, Raspadori D, Amadori S, Lauria F. Effect of a p210 multipeptide vaccine associated with imatinib or interferon in patients with chronic myeloid leukaemia and persistent residual disease: a multicentre observational trial. Lancet. 365,657– 62 (2005). 2. Rezvani K, Yong AS, Mielke S, Savani BN, Musse L, Superata J, Jafarpour B, Boss C, Barrett AJ. Leukemia-associated antigen-specific T-cell responses following combined PR1 and WT1 peptide vaccination in patients with myeloid malignancies. Blood. 111,236–42 (2007). 3. Schmitt M, Schmitt A, Rojewski M, Chen J, Götz M, Heyduk M, Giannopoulos K, Ritter G, Gnjatic S, Guillaume P, Ringhoffer M, Schlenk R, Liebisch P, Bunjes D, Shiku H, Döhner H, Greiner J. RHAMM-R3 peptide vaccination in patients with acute myeloid

S68 leukemia, myelodysplastic syndrome and multiple myeloma elicits immunological and clinical responses. Blood 111,1357–1365 (2008). 4. Keilholz U, Letsch A, Busse A, Asemissen AM, Bauer S, Blau IW, Hofmann WK, Uharek L, Thiel E, Scheibenbogen C. A clinical and immunologic phase 2 trial of Wilms tumor gene product 1 (WT-1) peptide vaccination in patients with AML and MDS. Blood. 113,6541–8 (2009). 5. Greiner J*, Schmitt A*, Giannopoulos K, Rojewski MT, Götz M, Funk I, Ringhoffer M, Bunjes D, Hofmann S, Ritter G, Döhner H, Schmitt M. High dose RHAMM-R3 peptide vaccination for patients with acute myeloid leukemia (AML), myelodysplastic syndrome (MDS) and multiple myeloma (MM). Haematologica 95,1191–7, (2010) (*both authors contributed equally). 6. Giannopoulos K, Dmoszynska A, Kowal M, Rolinski J, Gostick E, Price DA, Greiner J, Rojewski MT, Stilgenbauer S, Döhner H, Schmitt M. Peptide vaccination elicits leukemia-associated antigenspecific cytotoxic CD8+ T-cell responses with potential clinical relevance in patients with chronic lymphocytic leukemia. Leukemia 24,798–805 (2010). Results from the AIEOP-BFM ALL 2000 Trials M. SCHRAPPE1, G. CARIO1, A. SCHRAUDER1, C. R. BARTRAM2, A. MÖRICKE1, M. STANULLA1, G. MANN3, M. ZIMMERMANN4, M.G. VALSECCHI5, V. CONTER6 1 Department of Pediatrics, University Medical Center SchleswigHolstein, Campus Kiel, Germany; 2Institute of Human Genetics, University of Heidelberg; 3St. Anna Children’s Hospital and CCRI, Vienna, Austria; 4Department of Pediatric Hematology and Oncology, Hannover Medical School, Germany; 5Medical Statistics Unit, Department of Clinical Medicine and Prevention, University of Milano-Bicocca, Italy; 6San Gerardo Hospital, Monza, Italy

The AIEOP-BFM ALL 2000 study has for the first time introduced standardized quantitative assessment of minimal residual disease (MRD) based on immunoglobulin and T-cell receptor gene rearrangements as PCR targets (PCR-MRD), at two time points (TP), to stratify patients in a large prospective study (1.7.2000–31.7.2006). Patients with precursor B (pB) ALL (n=3184) and T-ALL (n=464) were considered MRD standard risk (MRD-SR) if MRD was negative at day 33 (analysed by two markers, with a sensitivity of at least 10−4); MRD intermediate risk (MRD-IR) if positive at day 33 or 78, and <10−3 at day 78; MRD high risk (MRD-HR) if ≥10−3 at day 78. While the distribution of pB ALL into the MRD risk groups was largely different compared to that of T-ALL, the outcome per risk group was very similar: 5y-EFS (SE) for pB in MRD-SR, -IR, and -HR was 92.3% (0.9), 77.6% (1.3), and 50.1% (4.1), as compared to 93.0% (3.0), 80.6% (2.3), and 49.8% (5.1), respectively, in T-ALL. If pB ALL patients were analyzed by conventional risk groups (as in ALL-BFM 95), MRD remained to be the best discriminating factor in all subgroups. Even in genetic subgroups, the same separation into prognostic categories by MRD response applied. In MRD-non-HR pB ALL, patients could further be separated by levels of MRD at TP1: If MRD was≥10−3 at TP1 and still positive (<10−3) at TP2, the cumulative incidence of relapse was 40.7% at 5 years. This formed the rationale to add this subset to the HR group in the new trial AIEOP-BFM ALL 2009. In T-ALL, relapse incidence was strongly depending on levels of MRD at TP2. Thus, MRD-based response assessment proved feasible, and highly informative of risk to relapse in all subgroups of ALL.

Ann Hematol (2011) 90 (Suppl 1):S25–S76 Despite the power of MRD to predict the risk of relapse, the majority of relapses is still being observed in the large intermediate risk group. Thus, new tools for a more precise risk assessment are needed. High-level expression of the cytokine receptor-like factor 2 gene, CRLF2, in pB-ALL was shown to be caused by a translocation involving the IGH@ locus or a deletion juxtaposing CRLF2 with the P2RY8 promoter. The incidence of these abnormalities in pB-ALL was estimated at approximately 7–10%. To assess its possible prognostic value, leukemic CRLF2 gene expression was measured by RQ-PCR in an unselected population of 555 children with ALL uniformly treated according to the ALL-BFM 2000 protocol. To define the best cut-off to distinguish a CRLF2-high from a CRLF2– low-expression group, samples were screened for known CRLF2 involving genomic aberrations beginning with those having the highest CRLF2 expression. Besides CRLF2-rearrangements high-level CRLF2 expression was seen in cases with supernumerary copies of the CRLF2 locus. Based on the detection of CRLF2-rearrangements, a CRLF2-high-expression group (n=49) was defined. This group had a 6-year relapse incidence of 31±8% compared to 11±1% in the CRLF2-low-expression group (P=0.006). This difference was mainly attributable to an extremely high relapse incidence (71±19%) in non-high-risk patients with P2RY8-CRLF2 rearrangement. Of importance, these patients were not identified by current stratification tools as they were treatmentsensitive measured by prednisone response, bone marrow clearance on treatment day 33 and/or measurement of minimal residual disease. The assessment of CRLF2-aberrations may therefore serve as new stratification tool in BFM-based protocols to further refine true nonhigh-risk patients with a low risk of relapse by identifying additional patients at high relapse risk who may benefit from an intensified treatment. Moreover, high-level CRLF2 expression and/or aberrant CRLF2/JAK signalling may serve as therapeutic targets for this important subgroup of patients. Conflict of interest None

Principles of Cellular Immunotherapy of Cancer H. SCHREIBER1 and K.SCHREIBER1 1 Department of Pathology, The University of Chicago, 5841 South Maryland Ave., Chicago, IL 60615, USA Abstract Many different treatments arrest malignant disease or cause cancer regression in patients and mice but are not curative. Critical issues are toxicity of the treatment and duration of the beneficial effects. Relapse from therapy is tightly linked to the development of heritably resistant variants. Because of the genetic instability of cancer cells, specific resistance to a particular therapy occurs at high frequency. We show that cancer cells release antigenic materials into their immediate stromal microenvironment. Targeting this material with T cells leads to IFN-gamma- and TNF-dependent stromal destruction and bystander death of cancer variants that would otherwise cause relapse. Introduction One of most stunning, yet sometimes forgotten, advances in cancer medicine is our ability to cure most childhood leukemias. Major credit for this goes to Howard Skipper. He laid down the principles for effective

Ann Hematol (2011) 90 (Suppl 1):S25–S76 combination chemotherapy in patients by his research in the L1210 murine leukemia model (1). Without question, his findings transformed our thinking of cancer medicine in how to go about eradicating cancer. One of the reasons for his success has been his realistic assessment of what it would take to obtain useful data from animal studies. He realized that virtually any patient newly diagnosed with cancer harbors at least 109 cancer cells, be it a liquid or solid malignancy. In fact, the smallest cancer clinically detected is usually 1 cm in average diameter and contains about 109 cancer cells (2). This is true for cancer in mice and men since the sizes of the cells in both species are similar. Skipper postulated that eradicating 109 cancer cells would require sequential use of at least three independently acting agents. Each of them must kill at least three logs of cancer cells in order to have the very last cancer cell eliminated. Thus, resistant variants will have to be less frequent than 1 in 1000 cancer cells in the malignancy treated for any of the three drugs used. Also, the second and third drug must follow the first drug at correct intervals before the surviving resistant cancer cells have expanded in the population to be eradicated. The key problem of resistant cancer cells causing relapse was also realized and analyzed by much earlier studies of three leading immunologists in the cancer field decades ago (3–5). The hope was that the immune system might kill a smaller antigen negative population as bystander when the targeted population was sensitive to destruction. Two of the laboratories failed to find any bystander killing of antigen negative cells, while the third found a dramatic bystander effect. All three used different models not realizing at this time that bystander killing depended on stromal sensitization and destruction as shown further below. The fallacy of targeting small cancer cell populations in mice Cancer cells often have an unlimited proliferation potential. Thus, a single cancer cell surviving treatment potentially leads to relapse of the malignancy eventually killing the patient. Why then would for example eradicating a tumor consisting of 106 cancer cells have little clinical significance? The answer is that at a given frequency of variant development, there may only be 1 or no resistant variant in 106 cancer cells treated while, at the same frequency, a clinically relevant inoculum of 109 cancer cells would have 1000 cancer variants. Thus, escape from therapy of the latter inoculum would be by three orders of magnitude more likely (6). The fallacy of targeting cancer cells in mice early after inoculation Tumor inoculation in mice is associated with a strong M1/N1 inflammatory response due to experimental manipulation and many of the cancer cells inoculated may die before they can adjust to the new environment (7). This early inflammatory response elicits innate immune cells followed by adaptive immune responses. This may easily lead to the eradication of the cancer cells that survived the inoculation. By contrast, clinical cancer when first diagnosed will likely only have chronic rather but not acute inflammatory reactions; the latter suppress rather than help immune destruction. Two week-old tumors escape T cells targeting cancer cells only Two week-old tumors lack the early inflammatory response and contain 109 cancer cells or more (8). Adoptive transfer of T cells with high affinity to an antigen on the surface of the cancer cells is highly effective in destroying the bulk of the cancer cells. However, resistant

S69 variants quickly overgrow and fill the surviving stromal network of the tumor followed by rapid re-growth of the tumors (9, 10). Large tumors are eradicated when T cells target cancer cells as well as stroma Cancer cells impregnate neighboring tumor stroma by releasing exosomes, blebs or microvesicles. One of the consequences of this process is the cross-presentation of tumor antigens present in these materials and picked up by stromal cells surrounding the cancer cells. The cross-presentation can be demonstrated directly by flow cytometry using TCR-tetramers that bind the tumor peptide loaded MHC Class I molecules (pMHC) on the stromal cells. Adoptive transfer of T cells specific for these pMHC complexes cause destruction of the stromal cells isolated from the treated tumors. Presence or absence of these effects correlates with cure or relapse after treatment (10). Experiments with chimeric mice demonstrate that the bone marrow(BM-)derived, hematopoietic component and the non BM-derived, sessile or fibroblastic component of the tumor stroma must both express the appropriate peptide-presenting MHC molecules to allow tumor eradication by T cell transfer.

Radiation or chemotherapy overcomes failure of stromal destruction Local radiation or systemic chemotherapy may not have any measurable effect on tumor growth but nevertheless allow cure when combined with subsequent T cell treatment (11). Apparently radiation or chemotherapy can cause increased release of antigenic materials from the cancer cells. This material is then picked up by the stromal cells and leads to stromal sensitization. Maximal sensitization occurs 48 h after radio- or chemotherapy at which time T cells must be transferred to achieve eradication. This mechanism appears to be critically important for eradicating malignancies that constitutively express antigen levels insufficient to sensitize their stroma.

TNF and IFN-gamma are critical for stromal destruction and cure Further experiments demonstrated that prevention of relapse after T cell treatment depended on the T cells being producing TNF and interferon-gamma (12). Furthermore, T cell treatment of tumors growing in chimeric mice showed that both BM- and non-BMderived components of the tumor stroma must express the receptors for TNF as well as the receptors for IFN-gamma.

Targeting stroma only may lead to long-lasting growth arrest of large tumors Cancer cells cannot be targeted directly when they lack the MHC Class I necessary for presenting the antigen they harbor (13). Cancers and viruses alike use this strategy for immune evasion. Nevertheless, these cancer cells still release their antigens into the surrounding stroma where T cells may detect them as targets and destroy the stromal cells. Stromal destruction by itself can lead to almost complete destruction of large solid tumors by killing, as bystanders, the cancer cells caught in the destroyed stroma. A thin rim of proliferating and dying cancer cells, however, persists at the tumor margin that is oxygenated by pre-existent non-tumor vasculature in the surrounding healthy tissue. Nevertheless, this mode of stromal destruction leads to a dramatic growth arrest keeping the tumor size in equilibrium, sometimes for months.

S70 CD4+ and CD8+ T cooperate in the effector phase tumor stroma Recent experiments explored how the bystander killing of cancer cells can be enhanced when T cells cannot kill directly but can only recognize the cross-presented antigens on stromal cells in the tumor microenvironment (14). We have found that CD4+ T cells can synergize with CD8+ T cells very effectively in bystander killing of such cancer cells that cannot be targeted directly. However, this synergy occurred only when the each cancer cells expressed both types of antigens. Thus, inoculation of mixtures of cancer cells escaped when 50% expressed only the CD4+ T cell-recognized antigen and the other 50% expressed only the CD8+ T cell-recognized antigen. This finding emphasizes the importance of proximity of the CD4+ T cell, the CD8+ T cell, the stromal cell picking up the antigens and the cancer cell releasing both antigens. Possibly there is a need for a fourcell cluster. Conclusion Realistic animal models must take into account that cancer cells are very prone to escape treatment by generating heritably resistant variants at high frequencies. Heritable escape from virtually any treatment modality has been reported. Tumor stroma consists of nonmalignant cells, is absolutely essential for tumor growth and cannot be replaced by cancer cells, even when cancer cells undergo epithelialmesenchymal transformation (EMT). Stromal destruction alone only arrests tumor growth but usually fails to eradicate because of a small rim of cancer cell surviving at the tumor margin. This margin is oxygenated by pre-existing non-tumor vessels. However, when the levels of antigen expression by the tumors are high and the cancer cells as well as the stroma are targeted, tumor eradication occurs. Presumably when more antigen being released, stromal sensitization and destruction extends into the healthy tumor–free margins thereby allowing T cells to “excise” the cancer completely like a surgeon removing a cancer leaving only tumor-free margins behind. However, levels of antigen could be low in many human cancers. In this case, a single course of radiation or chemotherapy can force cancer cells to transiently release more antigens. This may then sensitize the tumor stroma sufficiently to allow T cells to destroy stroma directly and cancer variants as bystanders, thereby achieving cure. Ongoing efforts concentrate on extending our findings to micro-disseminated cancer cells. Finally, the new principles discovered could well be applicable to the treatment of leukemias and other hematopoietic malignancies. These cancers also depend on stromal cells for their growth and survival just like cancer cells from solid malignancies.

Ann Hematol (2011) 90 (Suppl 1):S25–S76 Prehn, R. T. 1973. Destruction of tumor as an “innocent bystander” in an immune response specifically directed against nontumor antigens. Isr J Med Sci 9:375–379. 6. Skipper, H. E. 1983. The forty-year-old mutation theory of Luria and Delbruck and its pertinence to cancer chemotherapy. Adv Cancer Res 40:331–363. 7. Schreiber, K., D. A. Rowley, G. Riethmuller, and H. Schreiber. 2006. Cancer immunotherapy and preclinical studies: why we are not wasting our time with animal experiments. Hematol Oncol Clin North Am 20:567–584. 8. Singh, S., S. R. Ross, M. Acena, D. A. Rowley, and H. Schreiber. 1992. Stroma is critical for preventing or permitting immunological destruction of antigenic cancer cells. J. Exp. Med. 175:139– 146. 9. Wick, M., P. Dubey, H. Koeppen, C. T. Siegel, P. E. Fields, F. W. Fitch, L. Chen, J. A. Bluestone, and H. Schreiber. 1997. Antigenic cancer cells can grow progressively in immune hosts without evidence for T cell exhaustion or systemic anergy. J. Exp. Med. 186:229–237. 10. Spiotto, M. T., D. A. Rowley, and H. Schreiber. 2004. Bystander elimination of antigen loss variants in established tumors. Nat Med 10:294–298. 11. Zhang, B., N. A. Bowerman, J. K. Salama, H. Schmidt, M. T. Spiotto, A. Schietinger, P. Yu, Y. X. Fu, R. R. Weichselbaum, D. A. Rowley, D. M. Kranz, and H. Schreiber. 2007. Induced sensitization of tumor stroma leads to eradication of established cancer by T cells. J Exp Med 204:49–55. 12. Zhang, B., T. Karrison, D. A. Rowley, and H. Schreiber. 2008. IFN-g- and TNF-dependent bystander eradication of antigen-loss variants in established mouse cancers. J Clin Invest 118:in press. 13. Zhang, B., Y. Zhang, N. A. Bowerman, A. Schietinger, Y. X. Fu, D. M. Kranz, D. A. Rowley, and H. Schreiber. 2008. Equilibrium between host and cancer caused by effector T cells killing tumor stroma. Cancer Res. 68:in press. 14. Schietinger, A., M. Philip, R. B. Liu, K. Schreiber, and H. Schreiber. 2010. Bystander killing of cancer requires the cooperation of CD4(+) and CD8(+) T cells during the effector phase. J Exp Med 207:2469–2477. 5.

FLT3 Inhibitors in Acute Myeloid Leukemia: An Update R.M. STONE Dana-Farber Cancer Institute, Boston, MA 02115, USA

Conflicts of Interest None References 1. Skipper, H. E. 1971. Cancer chemotherapy is many things: G.H.A. Clowes Memorial Lecture. Cancer Res 31:1173–1180. 2. Kumar, V., Abbas, A.K., Fausto, N. 2004. Neoplasia. In Robbins & Cotran Pathologic Basis of Disease. A. A. K. Kumar. V., Fausto, N., ed. 3. Klein, E., and G. Klein. 1972. Specificity of homograft rejection in vivo, assessed by inoculation of artificially mixed compatible and incompatible tumor cells. Cell Immunol 5:201–208. 4. Weissman, I. L. 1973. Tumor immunity in vivo: evidence that immune destruction of tumor leaves “bystander” cells intact. J Natl Cancer Inst 51:443–448.

A gain-of-function FLT3 mutation which occurs in blasts from about 30% of patients with AML, was identified as a potential therapeutic target over 10 years ago.(1) The path to proving that inhibition of this target would be clinically beneficial has been difficult and circuitous. Activation of the FLT3 transmembrane tyrosine kinase can occur via one of two types of mutations: a length or internal tandem duplication (ITD) mutation in the juxtamembrane region (repeat of between 3 to >100 amino acids) or a point mutation in the tyrosine kinase domain which accounts for 10–20% of all FLT3 mutations.(1) Depending on such variables as length and location of the ITD mutation(2), and allelic burden(3), this type of genetic abnormality portends an adverse prognosis, particularly in patients with normal chromosome AML, compared to patients with wild-type disease. The prognostic significance of the tyrosine kinase domain mutation is less clear.(4,5)

Ann Hematol (2011) 90 (Suppl 1):S25–S76 Murine stem cells transduced with a mutant FLT3 construct produce a fatal myeloproliferative disease.(6) However, a potential intrinsic problem with FLT3 as a therapeutic target is its acquisition late in leukemogenesis, based on evidence that FLT3 mutant patients can relapse with FLT3 wild type disease (and vise versa) and from genomic sequencing studies which suggest that FLT3 might be a ‘late hit’ in the leukemogenic process.(7) Nonetheless, FLT3 inhibitors can specifically inhibit leukemic cell lines transformed by a FLT3 mutation (8) and prolong the survival of mice with model FLT3 induced disease. (9) Initial trials with pharmacological inhibitors of FLT3, even those restricted to patients with mutant FLT3 AML, were disappointing. While agents like CEP701(10)(lestauranib), MLN518 (tendutinib) (11), and PKC412(midostaurin) (12) decreased peripheral blasts in mutant FLT3 patients, clinical complete remissions were rare. These disappointing results could have been due to lack of prolonged pharmacological inhibition of the target, inability to overcome survival pathways in the leukemic stem cell marrow niche, or activation of leukemogenic survival pathways not dependent on FLT3. Recent studies documenting complete remission with a more potent and specific FLT3 inhibitor (AC220) suggest that single agent FLT3 inhibition may still have a therapeutic role in AML.(13). Sorafenib, an agent approved as a vascular endothial growth factor receptor inhibitor in kidney cancer, also a reasonably potent and specific FLT3 inhibitor, has activity in AML patients(14), even in post-transplant relapse. Because of the lack of clinical activity with single agent FLT3 inhibition despite biological activity blast the developmental therapeutics have shifted to combining FLT3 inhibitors with the most active agents in AML, namely chemotherapy. For example, a phase I/II study in which sorafenib was combined with standard induction chemotherapy in AML showed a surprising high complete remission rate (but no obvious improvement in disease-free survival)(15). A phase II study of chemotherapy plus sorafenib or placebo in older AML patients of any FLT3 status was disappointing, especially in terms of added toxicity(16). Nonetheless, the CALGB is planning a phase II trial of standard induction chemotherapy plus sorafenib in older adults with AML who have mutant FLT3 disease. A phase IB study in which midostaurin was combined with chemotherapy in newly diagnosed adults with wild type and mutant FLT3 (17) demonstrated how difficult it can be to combine a so-called targeted agent with standard chemotherapy. In initial iterations of this combination, a high degree of nausea and vomiting suggested doubtful feasibility. However, when chemotherapy was combined with a 14 day course of midostaurin at a somewhat lower dose (50 mg twice daily, a dose in which a prior randomized phase IB trial showed activity)(18) the drug was tolerated when the FLT3 inhibitor was given concomitantly with chemotherapy or sequentially with chemotherapy.(17) A recent retrospective analysis of this trial(17) demonstrated that patients with FLT3 mutant AML survived as long as those with wild type FLT3 AML. Although there were very small numbers in this analysis (and four of the 13 patients in this trial with mutant FLT3 AML had a tyrosine kinase domain mutation), some hope was offered that the addition of FLT3 inhibitor to chemotherapy in the upfront mutant FLT3 AML setting in younger patients might be of benefit. As such, standard daunorubicin (60 mg/day2 for 3 days) and cytarabine (200 mg/day2 for 7 days) plus placebo or midostaurin at 50 mg/twice daily d8-21 of induction and high dose ara-C 3 g/m2 over 3 h q12h d1,3,and 5, plus placebo or midostaurin 50 mg twice daily on d8-21 of

S71 four consolidation courses followed by maintenance with placebo or midostaurin (50 mg twice daily continuous x 12 months) was chosen for further analysis in a large prospective randomized double blind controlled trial in patients between 18 and 60 with mutant FLT3 AML. This trial, CALGB 10603 requires rapid identification of a mutant FLT3 state so that the patients can be enrolled onto trial before they receive any chemotherapy. So far, almost 600 patients have been enrolled in this trial worldwide which is expected to meet its new accrual goal of 714 by mid 2011. A phase III trial conducted in relapsed patients with mutant FLT3 AML showed that the addition of lestaurtinib (CEP701) to salvage chemotherapy did not yield a higher complete remission rate or overall survival rate than noted in patients who received chemotherapy alone. (19) While these results were disappointing, a retrospective analysis of the reasons for treatment to failure indicated that patients who were able to achieve a sufficient level of FLT3 inhibiting activity in their plasma had a higher CR rate, implying that a higher dose of CEP701 or a more potent agent that could inhibit the target for a longer period of time might have been optimal. Of course, AML cells in relapsed patients probably have more survival pathways activated than in upfront patients. Because of the success of AC220 as a single agent, plans are underway to perform a U.S. Intergroup randomized trial in mutant FLT3 younger adults of chemotherapy plus placebo or AC220. Summary In summary, while single agent pharmacologic inhibition of FLT3 might not be sufficient to provide a significant degree of therapeutic benefit to allow approval, it is still possible that more potent agents, different mechanisms of FLT3 inhibition, or adding FLT3 inhibitors, or adding other pathway inhibitors, (e.g. CXCR4/CXCL12 inhibition or mTOR inhibition) might be fruitful. The results of current and future studies in upfront mutant FLT3 patients with chemotherapy plus FLT3 inhibitors ‘vs’ placebo are eagerly awaited. Conflict of interest Novartis clin res support; Consultant. Genzyme, Consultant. References 1. Gilliland DG, Griffin JD: Role of FLT3 in Leukemia. Curr Opin Hematol 2002;9:274–281. 2. Breitenbuecher F, Schnittger S, Grundler R, Markova B, Cariu8s B, Brecht A, Duyster J, Haferlach T, Huber C, Fischer T: Identification of a novel type of ITD mutations located in nonjuxtamembrane domains of the FLT3 tyrosine kinase 3. Whitman SP, Archer KJ, Feng L, Baldus C, Becknell B, Carlson BD, Carroll AJ, Mrozek K, Vardiman JW, George SL, Kolitz JE, Larson RA, Bloomfield CD, Caligiuri MA. Absence of the wildtype allele predicts poor prognosis in adult de novo acute myeloid leukemia with normal cytogenetics and the internal tandem duplication of FLT3: a Cancer and Leukemia Group B study. Cancer Res 2001;61:7233–7239. 4. Whitman SP, Ruppert AS, Radmacher MD, Mrozek K, Paschka P, Langer C, Baldus CD, Wen J, Racke F, Powell BL, Kolitz JE, Larson RA, Caligiuri MA, Marcucci G, Bloomfield CD. FLT3 D8351836 mutations are associated with poor disease-free survival and a distinct gene-expression signature among younger adults with de novo cytogenetically normal acute myeloid leukemia lacking FLT3 internal tandem duplications. Blood 2008;11:1552–1559.

S72 5. Mead AJ, Linch DC, Hills RK, Wheatley K, Burnett AK, Gale RE. FLT3 tyrosine kinase domain mutations are biologically distinct from and have a significantly more favorable prognosis than FLT3 internal tandem duplications in patients with acute myeloid leukemia. Blood 2007;110:1262–1270. 6. Kelly LM, Liu Q, Kutok JL, Williams IR, Boulton CL, Gilliland DG. FLT3 internal tandem duplication mutations associated with human acute myeloid leukemias induce myeloproliferative disease in murine bone marrow transplant model. Blood 2002;99:310– 318. 7. Mardis ER, Ding L, Dooling DJ, Larsen DE, McLellan MD, Chen K, Koboldt DC, Fulton RS, Delehaunty KD, McGrath SD, Fulton LA, Locke DP, Magrini VJ, Abbott RM, Vickery TL, Reed JS, Robinson JS, Wylie T, Smith SM, Carmichael L, Eldred JM, Harris CC, Walker J, Peck JB, Du F, Dukes AF, Sanderson GE, Brummett AM, Clark E, McMichael JF, Meyer RJ, Schindler JK, Pohl CS, Walis JW, Shi X, Lin L, Schmidt H, Tang Y, Haipek C, Wiechert ME, Ivy JV, Kalicki J, Elliott G, Ries RE, Payton JE, Westervelt P, Tomasson MH, Watson MA, Baty J, Heath S, Shannon WD, Nagarajan R, Link DC, Walter MJ, Graubert TA, DiPersio JF, Wilson RK, Ley TJ. Recurring mutations found by sequencing an acute myeloid leukemia genome. N Engl J Med 2009;361(11):1058–66. 8. Weisberg E, Boulton C, Kelly LM, Manley P, Fabbro D, Meyer T, Gilliland DG, Griffin JD. Inhibition of mutant FLT3 receptors in leukemia cells by the small molecule tyrosine kinase inhibitor PKC412. Cancer Cell 2002;1:433–443. 9. Kelly LM, Yu JC, Boulton CL, Apatira M, Li J, Sullivan CM, Williams I, Amaral SM, Curley DP, Duclos N, Neuberg D, Scarborough RM, Pandey A, Hollenbach S, Abe K, Lokker NA, Gilliland DG. CT53518, a novel selective FLT3 antagonist for the treatment of acute myelogenous leukemia (AML). Cancer Cell 2002;1:421–432. 10. Smith BD, Levis M, Beran M, Giles F, Kantarjian H, Berg K, Murphy KM, Dauses T, Allebach J, Small D. Single-agent CEP701, a novel FLT3 inhibitor, shows biologic and clinical activity in patients with relapsed or refractory acute myeloid leukemia. Blood 2004;103:3669–3676. 11. DeAngelo DJ, Stone RM, Heaney ML, Nimer SD, Paquette RL, Klisovic RB, Caligiuri MA, Cooper MR, Lecerf J-M, Karol MD < Sheng S, Holford N, Curtin PT. Druker BJ, Heinrich MC. Phase I clinical results with tandutinib (MLN518), a novel FLT3 antagonist, in patients with acute myelogeneous leukemia or high-risky myelodysplastic syndrome: safety, pharmacokinetics, and harmacodyn-amics. Blood 2006;108: 3674–3681. 12. Stone RM, DeAngelo DJ, Klimek V, Galinsky I, Estey E, Nimer SD, Grandin W, Lebwohl D, Wang Y, Cohen P, Fox EA, Neuberg D, Clark J, Gilliland DG, Griffin JD. Patients with acute myeloid leukemia and an activating mutation in FLT3 respond to a smallmolecule FLT3 tyrosine kinase inhibitor, PKC412. Blood 2005;105:54–60. 13. Cortes J, Foran J, Ghirdaladze D, DeVetten MP, Zodelava M, Holman P, Levis MJ, Kantarjian HM, Gorthakur G, James J, Zarringkar PP, Gunawardane RN, Armstrong RC, Padre NM, Wierenga W, Corringham R, Trikha M. AC220, a potent, selective, second generation FLT3 receptor tyrosine kinase (RTK) inhibitor, in a first-in-human (FIH) phase I AML study. Blood 2009;114:[636], 264.

Ann Hematol (2011) 90 (Suppl 1):S25–S76 14. Zhang W, Konopleva M, Shi YX, McQueen T, Harris D, Ling X, Estrov Z, Quintas-Cardamas A, Small D, Cortes J, Andreeff M. Mutant FLT3: a direct target of sorafenib in acute myelogenous leukemia. J Natl Cancer Inst 2008;100:184–198. 15. Ravandi F, Cortes JE, Jones D, Jones D, Faderl S, Garcia-ManeroG, Konopleva MY, O’Brien S, Estrov Z, Borthakur G, Thomas D, Pierce SR, Brandt M, Byrd A, Bekele BN, Pratz K, Luthra R, Levis M, Andreeff M, Kantarjian HM. Phase I/II study of combination therapy with sorafenib, idarubicin, and cytarabine in younger patients with acute myeloid leukemia. J Clin Oncol 2010;28:1856–1862. 16. Serve H, Wagner R, Sauerland C, Brunnberg U, Krug U, Schaich M, Ottmann OG, Duyster J, Wandt H, Herr W, Giagounidis A, Neubauer A, Reichle A, Aulitzky WE, Noppeney R, Blau IG, Kunzmann V, Schmitz N, Kreuzer K-A, Kramer A, Brandts C, Steffen B, Heinecke A, Theide C, Muller-Tidow C, Ehninger G, Berdel WE. Sorafenib in combination with standard induction and consolidation therapy in elderly AML patients: results from a randomized placebo-controlled phase II trials. Blood 2010;115:151[333]. 17. Stone RM, Fischer T, Paquette R, Schiller G, Schiffer CA, Ehninger G, Cortes J, Kantarjian HM, DeAngelo DJ, HuntsmanLabed A, Dutreix C, Rai S, Giles F. A phase IB study of midostaurin (PKC412) in combination with daunorubicin and cytarabine induction and high-dose cytarabine consolidation in patients under age 61 with newly diagnosed de novo acute myeloid leukemia: Overall survival of patients whose blasts have FLT3 mutations is similiar to those with wild-type FLT3. ASH Annual Meeting Abstracts 2009;11:{22}634. 18. Fischer T, Stone RM, DeAngelo DJ, Galinsky I, Estey E, Lanza C, Fox E, Ehninger G, Feldman EJ, Schiller GJ, Klimek VM, Nimer SD, Gilliland DJ, Dutreix C, Huntsman-Labed A, Virkus J, Giles FJ. A phase IIIB trial of oral midostaurin (PKC412), the FLT3 and multi-targeted kinase inhibitor, in patients with acute myeloid leukemia (AML) and high-risk myelodysplastic syndrome (DS) with either wild-type or mutated-FLT3. J Clin Oncol 2010;28:4339–4345. 19. Levis M, Ravandi F, Wang ES, Baer MR, Perl A, Coutre S, Erba H, Stuart RK, Baccarani M, Cripe LD, Tallman MS, Meloni G, Godley LA, Langston A, Amadoir S, Lewis ID, Nagler A, Stone R, Yee K, Advani A, Douer D, Jedrzejczak WW, Juliusson G, Litzow MR, Petersdorf S, Sanz M, Kantarjian HM, Sato T, Tremmel L, Bensen-Kennedy DM, Small D, Smith BD. Results from a randomized trial of salvage chemotherapy followed by lestaurtinib for FLT3 mutant AML patients in first relapse. Blood 2009;114: 325[788]. Vaccination with Dendritic Cells in AML M. SUBKLEWE 1 , F. LICHTENEGGER 1 , I. BIGALKE 2 , CH. GEIGER2, B. BECK1, M. JAVOROVIC2, D. DÖRFEL1, D. SCHENDEL, 2W. HIDDEMANN1 1 Department of Hematology and Oncology, LMU – Universität Munich, Klinikum Großhadern, Munic, Germany2; Institute of Molecular Immunology, Helmholtz Center Munich, German Research Center for Environmental Health Munich, Germany Nowadays, 60–70% of all patients with acute myeloid leukemia (AML) achieve a complete remission with long-term disease-free

Ann Hematol (2011) 90 (Suppl 1):S25–S76 survival and potential cure in 25–40% of cases. Cytogenetic and molecular markers assign patients to a low, intermediate and high prognostic risk group and serve as guidance for the choice of postremission therapy. Allogeneic stem cell transplantation (SCT) is employed to reduce the rate of AML relapse. This is mediated by a cellular immune response of the graft against residual leukemic cells. However, this apporach is restricted to patients with good performance status and younger age. Several autologous immunotherapeutic strategies have been employed with variable success in the induction of immune responses. We want to use dendritic cells (DCs) as the most powerful antigen presenting cells for active immunotherapy in patients with AML in complete hematological remission and a non-favorable genetic risk profile. In our previous work we developed a 3 day protocol for generation of DCs from autologous monocytes using a maturation cocktail containing six components including a synthetic TLR7/8 and TLR3 agonist. The protocol generated DCs with excellent profile with respect to recovery, phenotype, cytokine secretion, migration as well as activation of NK- and T-cells. Similar results were obtained using monocytes from AML patients after intense induction or consolidation therapy. We choose RNA as the source of antigen for peptide loading upon MHC class I and II of DCs. Based on extensive literature research and T cell stimulation assys ex vivo we decided to use the following three, welldefined antigens for our clincial phase I/II study: WT1, one of the most prominent LAA and commonly used in immunotherapeutic trials; PRAME, a cancer-testis antigen that has been shown to be immunogeneic in vivo; and CMVpp65, a dominant target antigen of the cellular immune response in latently infected people. A protocol for generation of RNA-transfected DCs under GMPcompatible conditions using leukapheresis as our starting product was developed. We aim for 10 vaccinations with 1×107 DCs per vaccine injected i.d. over 26 weeks. The primary end point of our prospective, non-randomized, single center study is feasibility and safety. Secondary end points are induction of immune responses, control of minimal residual disase (MRD) and clincial response. We believe that our TLR ligand matured DCs with the capacity to secrete cytokines for Th1polarizing and NK cell activation require clinical evaluation and are promising tools in the eradication of MRD in A Conflict of interest None Curing Acute Promyelocytic Leukemia without Chemotherapy M. S. TALLMAN1 and J. H. PARK2 1 Memorial Sloan-Kettering Cancer Center, Weill Cornell Medical College, New York, NY, USA; 2Medical Oncology and Hematology, Department of Medicine, Memorial Sloan-Kettering Cancer Center, New York, NY, USA Introduction Acute Promyelocytic Leukemia (APL) is a subtype of acute myeloid leukemia (AML) characterized by distinctive morphology of blast cells1,2, a life-threatening bleeding diathesis3, and a unique balanced reciprocal translocation t(15;17)4, which fuses the PML (promyelocyte) gene on chromosome 15 to the RARα (retinoic acid receptor-α) gene on chromosome 175. The disease is relatively rare in adults, accounting for only 10–15% of the approximately 12,330 adults diagnosed with AML in the United States each year.6 Historically, the disease was among the most fatal of the acute myeloid leukemias

S73 primarily because of the associated complex and potentially clinically catastrophic bleeding disorder6. However, since the introduction of all-trans retinoic acid (ATRA) in the late 1980s and arsenic trioxide (ATO) in the late 1990s, progress in the treatment of APL has resulted in the cure of the majority of patients. Despite the remarkable improvement in outcome of APL, treatment failure still occurs, primarily among high-risk patients, due most often to early death. Relapse has become increasingly less frequent, and is now only observed for the most part in patients with high-risk disease. Furthermore, now that patients are surviving longer, long-term complications of ATRA plus anthracycline-based chemotherapy have emerged including therapy-related myelodysplastic syndromes, extramedullary disease, and cardiomyopathy. A major focus of research for the past decade has been to develop risk-adapted treatment strategies to reduce treatment-related morbidity and mortality in low- and intermediate-risk and older patients while targeting more intensive or alternative therapy to those patients at highest risk of relapse. Arsenic trioxide has emerged as the most potent antileukemic agent in APL and has been combined with ATRA in newly diagnosed patients. This combination may replace standard ATRA plus anthracycline-based regimens for the majority of patients.

INDUCTION The evolution of induction therapy in APL Induction in APL has historically consisted of an anthracycline and cytarabine7-9. Among patients who achieve CR with initial chemotherapy, 50–65% subsequently relapse and only 30–40% of such patients remained alive at 2 years7–10. The introduction of ATRA prompted cooperative groups to exploit the role of ATRA in combination with chemotherapy. Although treatment with single-agent ATRA during induction results in CR rates of 72–90%11–15, early mortality due to hemorrhage and the differentiation syndrome remain problematic11,16–21. Importantly, patients who achieve CR with ATRA alone generally relapse without additional chemotherapy12,13,16. Therefore, subsequent clinical trials combined ATRA with chemotherapy, leading to CR in 90–95% of patients with primary resistance reported in very few anecdotal cases6. The European APL (EAPL) group demonstrated that concurrent ATRA plus chemotherapy (daunorubicin and cytarabine) resulted in a better outcome than sequential ATRA followed by chemotherapy with a reduction in relapse rate at 2 years (6 vs 16%)22. This approach has the benefit of reducing the incidence of the differentiation syndrome from approximately 25% with ATRA alone19,20 to approximately 10% with concurrent chemotherapy and ATRA23,24.

The role of cytarabine in induction therapy The benefits of cytarabine given in addition to ATRA and an anthracycline during induction has been clarified in several recent studies. The Italian Cooperative group GIMEMA and the Spanish Cooperative group PETHEMA have omitted cytarabine from induction, and demonstrated that ATRA combined with idarubicin alone (AIDA) is as effective in inducing CR as cytarabine-containing regimens with CR rates of 89–95%23,24. The Medical Research Council (MRC, now NCRI) in the United Kingdom randomized newly diagnosed patients under age 60 to AIDA or ATRA with daunorubicin and cytarabine. No differences in response, relapse or overall survival rates were observed, and less myelosuppression was

S74 observed with the regimen omitting cytarabine26. In contrast, a randomized study by the EAPL group demonstrated an increased risk of relapse when cytarabine was omitted from both induction and consolidation therapy27. This discrepancy may be explained by differences in the consolidation regimens (ATRA vs. no ATRA in EAPL trial), the specific choice of anthracycline (idarubicin vs. daunorubicin), the number of consolidation courses, and the cumulative doses of anthracyclines. However, it is clear that most patients with APL can be cured without cytarabine and with potentially less toxicity. The non-chemotherapy approach of ATO+ ATRA Arsenic trioxide is extremely effective in patients with relapsed and refractory APL.28–30 In these patients, ATO induces complete molecular remission in 50% of patients after a single 5-week course and in 86% of patients after 2 cycles.29 As a single agent, ATO induces CR in approximately 85% of patients with untreated APL, comparable to that achieved with ATRAbased therapy31,32. Single agent ATO during induction and consolidation was most effective in patients with WBCs of <5,000/ul and platelet counts >20,000/ul at diagnosis with an event-free survival (EFS) of 100% at 3 years31. However, the outcome of patients with WBC >5,000/ul at diagnosis was inferior to a similar subset treated with ATRA plus chemotherapy, with an EFS of only 67% and a higher early death rate of 14.4% attributable generally to bleeding31. Therefore, single agent ATO alone is sufficient for many patients. Combined use of ATRA and ATO are synergistic in inducing differentiation and apoptosis in vitro33,34. Investigators at the Shanghai Institute of Hematology (SIH) conducted a randomized clinical trial in which patients were randomized to receive ATRA, ATO, or the combination of ATRA plus ATO as induction therapy35. Although all patients in CR after induction therapy subsequently received consolidation and maintenance chemotherapy, the combination treatment of ATRA and ATO resulted in the lowest rate of relapse, a more rapid achievement of CR, and a greater reduction in the number of PML-RARα transcripts with no apparent greater toxicity than each agent alone. The updated report confirms the high efficacy and minimal toxicity of the combination treatment for newly diagnosed APL in long-term follow up of 7 years36. Investigators at the MD Anderson Cancer Center (MDA) similarly demonstrated that the combination treatment is an effective treatment in untreated APL37. In this patient population, a CR rate of 96% was achieved, with few late relapses even when chemotherapy was eliminated during consolidation37. However, high-risk patients achieved a lower CR rate of 79% despite the addition of gemtuzumab ozogamicin (GO) or idarubicin during induction to prevent hyperleukocytosis and the APL differentiation syndrome. Recently, the Australasian Leukaemia and Lymphoma Group (ALLG) treated 124 patients with newly diagnosed APL with induction consisting of ATRA, ATO, and idarubicin. This was followed by two courses of consolidation with ATRA and ATO and two years of maintenance with ATRA, methotrexate and 6mercaptopurine. Remarkably, the early death rate (survival≤36 days) was only 3%. At a median follow-up of 20 months, the 3-year overall survival and event-free survival rates were 93% and 87%, respectively. This approach resulted in a high CR rate with significant reduction in exposure to standard chemotherapy38.

Ann Hematol (2011) 90 (Suppl 1):S25–S76 Although combination therapy with ATRA and ATO yielded a remarkable improvement in the survival rates, early death rates of 6– 11.4% were reported in these trials due to intracranial hemorrhage35–38. The incidence of CNS relapse remains a problem, possibly attributable to the relatively poor penetrance of the drugs through the blood-brain barrier into sanctuary sites. These studies suggest that the combination of ATRA and ATO can potentially cure most patients with low- and intermediate-risk disease, and has emerged as the most exciting new treatment in APL. A series of studies demonstrating evolution of induction therapy in APL is summarized in Table 1. Table 1: Evolution in Induction Strategies, Less Chemotherapy and Early Arsenic (ATO) Trial

N

APL 200027 95

Induction

CR% CIR% D(E)FS% OS%

ATRA/DA

94

5

93

95

16

77

90

3

94

98

97

NR

87

93

90

7

90

85

86

10

87/75

85

APL 200027 101 ATRA/D SIH36

60

ALLG

124 ATRA/ATO/ Chemo 82 ATRA/ATO+/ −CT 72 ATO

MDA41 India42

ATRA/ATO/CT 93

98

*NR=Not Reported CONSOLIDATION Risk-adapted treatment strategy Non-chemotherapy consolidation regimens for low- and intermediate-risk patients In order to reduce toxicity and chemotherapy exposure, several groups have investigated the role of ATO in consolidation as part of the firstline treatment. In a phase II study at least one cycle of cytotoxic chemotherapy was substituted with ATO, and a comparable outcome to previous trials using a risk-adapted approach was reported39. The North American Intergroup addressed the role of ATO during consolidation in a randomized trial, in which patients received 2 courses of 25 days of ATO (5 days/week x 5 weeks) as a first consolidation followed by further consolidation with 2 courses of ATRA plus daunorubicin40. This trial showed that the addition of ATO as initial consolidation therapy significantly improves DFS and OS in patients of all risk groups. Investigators at the MDA completely eliminated cytotoxic chemotherapy from consolidation and reported a comparable clinical outcome (3-year survival 85%) using ATRA and ATO for 28 weeks as the only post-remission therapy, but with a somewhat less favorable outcome in high-risk patients41. Investigators from India treated patients in all risk groups with a single agent ATO during induction and consolidation, and reported the clinical outcome comparable to that achieved with conventional combination chemotherapy in lowand intermediate-risk patients. However, for the high-risk patients, the relapse rate was higher (17% 5-year cumulative incidence of relapse) than with conventional therapy42. The majority of studies suggest a potential benefit in terms of reduction of relapse risk for the addition of cytarabine during consolidation for high-risk patients, but not for low- and intermediaterisk patients. The synergism of ATRA plus ATO combination regimens

Ann Hematol (2011) 90 (Suppl 1):S25–S76 have shown that minimizing or even eliminating chemotherapy, particularly in patients with low- and intermediate-risk patient, is effective. Further results from ongoing randomized trials comparing ATRA + ATO to ATRA + chemotherapy will be forthcoming. MAINTENANCE Despite two randomized trials demonstrating a substantial benefit from ATRA-based maintenance therapy, the routine use of maintenance therapy is controversial, particularly for low- and intermediate-risk patients. Maintenance therapy in the low- and intermediate-risk elderly patients may be harmful as shown by the EAPL group study that reported an 11% death rate in CR mainly from sepsis related to myelosuppression 22. Furthermore, the introduction of ATO as initial treatment for APL may eliminate the need for maintenance therapy. The first US Intergroup study first reported a clinical benefit with the ATRA-based maintenance therapy for a year in terms of a reduction in relapse rate (22% vs. 39%) and a higher 5-year DFS (61% vs. 36%)11,43. Similarly, the EAPL group confirmed the beneficial effect of adding ATRA to maintenance therapy in a randomized study with the regimen consisting of continuous low-dose 6-mercaptopurine and methotrexate in addition to intermittent ATRA22. The EAPL group recently updated the results of their previous study reporting for the first time that the clinical benefit of maintenance therapy was mainly observed in high-risk patients while only a marginal benefit was noted in patients with low- and intermediate-risk disease44. In contrast, two other randomized trials by the GIMEMA and Japanese Adult Leukemia Study Group (JALSG) have recently reported no benefit from maintenance therapy45,46. Importantly, the JALSG group trial did not include ATRA as maintenance. The discrepancy in these studies suggests that the benefit of maintenance treatment may depend on prior induction and consolidation therapy, and PML/RARα status after consolidation therapy. For example, both GIMEMA45 and JALSG46 trials included three cycles of consolidation therapy and used idarubicin as anthracycline for induction and consolidation chemotherapy, whereas the EAPL22 and the US Intergroup11 study gave only two consolidation courses and used daunorubicin. Furthermore, while studies by the GIMEMA and JALSG groups were carried out in patients testing negative for PML/RARα at the end of consolidation, the US Intergroup and EAPL studies did not examine the molecular remission status at the end of consolidation, raising a question whether the benefit provided by the maintenance therapy is largely from patients with residual disease following consolidation. These studies suggest that patients with low- and intermediate-risk APL and who achieve complete molecular remission after consolidation therapy may not benefit from the combination maintenance therapy. For high-risk patients, maintenance therapy for 1–2 years with intermittent ATRA and low-dose chemotherapy with 6mercaptopurine and methotrexate is currently recommended. Whether the inclusion of ATO as part of initial induction and/or consolidation will obviate the need for ATRA plus low-dose chemotherapy as maintenance remains to be determined. Conflict of interest None References 1. Bennett, J.M., et al. A variant form of hypergranular promyelocytic leukaemia (M3). Br J Haematol 44, 169–170 (1980).

S75 2. Bennett, J.M., et al. Proposals for the classification of the acute leukaemias. French-American-British (FAB) co-operative group. Br J Haematol 33, 451–458 (1976). 3. Tallman, M.S. & Kwaan, H.C. Reassessing the hemostatic disorder associated with acute promyelocytic leukemia. Blood 79, 543–553 (1992). 4. Rowley, J.D., Golomb, H.M. & Dougherty, C. 15/17 translocation, a consistent chromosomal change in acute promyelocytic leukaemia. Lancet 1, 549–550 (1977). 5. Grignani, F., et al. The acute promyelocytic leukemia-specific PML-RAR alpha fusion protein inhibits differentiation and promotes survival of myeloid precursor cells. Cell 74, 423–431 (1993). 6. Tallman, M.S. & Altman, J.K. Curative strategies in acute promyelocytic leukemia. Hematology Am Soc Hematol Educ Program, 391–399 (2008). 7. Cunningham, I., et al. Acute promyelocytic leukemia: treatment results during a decade at Memorial Hospital. Blood 73, 1116– 1122 (1989). 8. Fenaux, P. & Degos, L. Treatment of acute promyelocytic leukemia with all-trans retinoic acid. Leuk Res 15, 655–657 (1991). 9. Head, D.R., Kopecky, K.J., Willman, C. & Appelbaum, F.R. Treatment outcome with chemotherapy in acute promyelocytic leukemia: the Southwest Oncology Group (SWOG) experience. Leukemia 8 Suppl 2, S38-41 (1994). 10. Rodeghiero, F., Avvisati, G., Castaman, G., Barbui, T. & Mandelli, F. Early deaths and anti-hemorrhagic treatments in acute promyelocytic leukemia. A GIMEMA retrospective study in 268 consecutive patients. Blood 75, 2112–2117 (1990). 11. Tallman, M.S., et al. All-trans-retinoic acid in acute promyelocytic leukemia. N Engl J Med 337, 1021–1028 (1997). 12. Chen, Z.X., et al. A clinical and experimental study on all-trans retinoic acid-treated acute promyelocytic leukemia patients. Blood 78, 1413–1419 (1991). 13. Castaigne, S., et al. All-trans retinoic acid as a differentiation therapy for acute promyelocytic leukemia. I. Clinical results. Blood 76, 1704–1709 (1990). 14. Huang, M.E., et al. Use of all-trans retinoic acid in the treatment of acute promyelocytic leukemia. Blood 72, 567–572 (1988). 15. Castaigne, S., et al. Retinoic acids in the treatment of acute promyelocytic leukemia. Nouv Rev Fr Hematol 32, 36–38 (1990). 16. Warrell, R.P., Jr., et al. Differentiation therapy of acute promyelocytic leukemia with tretinoin (all-trans-retinoic acid). N Engl J Med 324, 1385–1393 (1991). 17. Fenaux, P., et al. Effect of all transretinoic acid in newly diagnosed acute promyelocytic leukemia. Results of a multicenter randomized trial. European APL 91 Group. Blood 82, 3241–3249 (1993). 18. Fenaux, P., et al. All-transretinoic acid followed by intensive chemotherapy gives a high complete remission rate and may prolong remissions in newly diagnosed acute promyelocytic leukemia: a pilot study on 26 cases. Blood 80, 2176–2181 (1992). 19. Vahdat, L., et al. Early mortality and the retinoic acid syndrome in acute promyelocytic leukemia: impact of leukocytosis, low-dose chemotherapy, PMN/RAR-alpha isoform, and CD13 expression in patients treated with all-trans retinoic acid. Blood 84, 3843–3849 (1994).

S76 20. Tallman, M.S., et al. Clinical description of 44 patients with acute promyelocytic leukemia who developed the retinoic acid syndrome. Blood 95, 90–95 (2000). 21. De Botton, S., et al. Incidence, clinical features, and outcome of all trans-retinoic acid syndrome in 413 cases of newly diagnosed acute promyelocytic leukemia. The European APL Group. Blood 92, 2712–2718 (1998). 22. Fenaux, P., et al. A randomized comparison of all transretinoic acid (ATRA) followed by chemotherapy and ATRA plus chemotherapy and the role of maintenance therapy in newly diagnosed acute promyelocytic leukemia. The European APL Group. Blood 94, 1192–1200 (1999). 23. Avvisati, G., et al. AIDA (all-trans retinoic acid + idarubicin) in newly diagnosed acute promyelocytic leukemia: a Gruppo Italiano Malattie Ematologiche Maligne dell’Adulto (GIMEMA) pilot study. Blood 88, 1390–1398 (1996). 24. Sanz, M.A., et al. A modified AIDA protocol with anthracyclinebased consolidation results in high antileukemic efficacy and reduced toxicity in newly diagnosed PML/RARalpha-positive acute promyelocytic leukemia. PETHEMA group. Blood 94, 3015–3021 (1999). 25. Sanz, M.A., et al. Management of acute promyelocytic leukemia: recommendations from an expert panel on behalf of the European LeukemiaNet. Blood 113, 1875–1891 (2009). 26. Burnett AK, H.R., Grimwade D, et al. Idarubicin and ATRA is as effective as MRC chemotherapy in patients with acute promylocytic leukemia with lower toxicity and resource usage: preliminary results of the MRC AML 15 trial. Blood 110, 181a (2007). 27. Ades, L., et al. Is cytarabine useful in the treatment of acute promyelocytic leukemia? Results of a randomized trial from the European Acute Promyelocytic Leukemia Group. J Clin Oncol 24, 5703–5710 (2006). 28. Niu, C., et al. Studies on treatment of acute promyelocytic leukemia with arsenic trioxide: remission induction, follow-up, and molecular monitoring in 11 newly diagnosed and 47 relapsed acute promyelocytic leukemia patients. Blood 94, 3315–3324 (1999). 29. Soignet, S.L., et al. United States multicenter study of arsenic trioxide in relapsed acute promyelocytic leukemia. J Clin Oncol 19, 3852–3860 (2001). 30. Soignet, S.L., et al. Complete remission after treatment of acute promyelocytic leukemia with arsenic trioxide. N Engl J Med 339, 1341–1348 (1998). 31. Mathews, V., et al. Single-agent arsenic trioxide in the treatment of newly diagnosed acute promyelocytic leukemia: durable remissions with minimal toxicity. Blood 107, 2627–2632 (2006). 32. Ghavamzadeh, A., et al. Treatment of acute promyelocytic leukemia with arsenic trioxide without ATRA and/or chemotherapy. Ann Oncol 17, 131–134 (2006). 33. Zheng, P.Z., et al. Systems analysis of transcriptome and proteome in retinoic acid/arsenic trioxide-induced cell differentiation/apoptosis of promyelocytic leukemia. Proc Natl Acad Sci U S A 102, 7653–7658 (2005).

Ann Hematol (2011) 90 (Suppl 1):S25–S76 34. Gianni, M., et al. Combined arsenic and retinoic acid treatment enhances differentiation and apoptosis in arsenic-resistant NB4 cells. Blood 91, 4300–4310 (1998). 35. Shen, Z.X., et al. All-trans retinoic acid/As2O3 combination yields a high quality remission and survival in newly diagnosed acute promyelocytic leukemia. Proc Natl Acad Sci U S A 101, 5328– 5335 (2004). 36. Hu, J., et al. Long-term efficacy and safety of all-trans retinoic acid/arsenic trioxide-based therapy in newly diagnosed acute promyelocytic leukemia. Proc Natl Acad Sci U S A 106, 3342– 3347 (2009). 37. Estey, E., et al. Use of all-trans retinoic acid plus arsenic trioxide as an alternative to chemotherapy in untreated acute promyelocytic leukemia. Blood 107, 3469–3473 (2006). 38. [Iland H, Firkin F, Supple S, et al. Interim analysis of the APML4 trial incorporating all-trans retinoic acid (ATRA), idarubicin, and intravenous arsenic trioxide (ATO) as initial therapy in acute promyelocytic leukaemia (APL): an Australasian Leukaemia and Lymphoma Group study (ALLG) International ORAL Arsenic Union, Hong Kong. March 2010 [Abstract]. 39. Gore, S.D., et al. Single cycle of arsenic trioxide-based consolidation chemotherapy spares anthracycline exposure in the primary management of acute promyelocytic leukemia. J Clin Oncol 28, 1047–1053 (2010). 40. Powell, B.L., et al. Arsenic trioxide improves event-free and over-all survival for adults with acute promyelocytic leukemia: North American Leukemia Intergroup Study C9710. Blood (2010). 41. Ravandi, F., et al. Effective treatment of acute promyelocytic leukemia with all-trans-retinoic acid, arsenic trioxide, and gemtuzumab ozogamicin. J Clin Oncol 27, 504–510 (2009). 42. Mathews, V., et al. Single-agent arsenic trioxide in the treatment of newly diagnosed acute promyelocytic leukemia: long-term followup data. J Clin Oncol 28, 3866–3871 (2010). 43. Tallman, M.S., et al. All-trans retinoic acid in acute promyelocytic leukemia: long-term outcome and prognostic factor analysis from the North American Intergroup protocol. Blood 100, 4298–4302 (2002). 44. Ades, L., et al. Very long-term outcome of acute promyelocytic leukemia after treatment with all-trans retinoic acid and chemotherapy: the European APL Group experience. Blood 115, 1690– 1696 (2010). 45. Lo-Coco, F., et al. Front-line treatment of acute promyelocytic leukemia with AIDA induction followed by risk-adapted consolidation for adults patients younger than 61 years: results of the AIDA-2000 trial of the GIMEMA Group. Blood (2010). 46. Asou, N., et al. A randomized study with or without intensified maintenance chemotherapy in patients with acute promyelocytic leukemia who have become negative for PML-RARalpha transcript after consolidation therapy: the Japan Adult Leukemia Study Group (JALSG) APL97 study. Blood 110, 59–66 (2007).