Ann Hematol (2000) 79 : 299±303
Springer-Verlag 2000
ORIGINAL ARTICLE M. Carrasco ´ L. Muæoz ´ M. Bellido ´ S. Bernat E. Rubiol ´ J. Úbeda ´ J. Sierra ´ J.F. NomdedØu
CD66 expression in acute leukaemia
Received: 31 May 1999 / Accepted: 10 November 1999
Abstract Antibodies against CD66 identify antigens from the carcinoembryonic antigen (CEA) family of proteins, which belong to the immunoglobulin gene superfamily. Despite being usually restricted to cells of myeloid or monocytic origin, CD66 expression has also been reported in blasts from children with B-cell lineage acute lymphocytic leukaemia (ALL). An analysis of the CD66 expression was undertaken in a series of acute leukaemia patients. Antigenic expression was analysed using triple combinations of monoclonal antibodies (mAbs) in forty-five patients. The CD66 Kat4 fluorescein isothiocyanate clone was purchased from Dako (Glostrup, Denmark). CD66 was expressed in 2 of 29 patients with AML (acute myeloblastic leukemia) (6.8%) and in 8 of 12 patients with B-cell lineage ALL (66.7%; P < 0.001); in blast crisis (BC) of chronic myelocytic leukaemia (CML), CD66 was expressed in two patients with lymphoid BC but not in the two with myeloid BC. The co-expression of CD66 with other myeloid antigens was observed in all CD66+ ALL/ Ly-BC cases tested: CD 13 in six patients, CD33 in seven and CD117 in two patients. The CD66 expression is more frequent in ALL than in AML. Furthermore, we analysed minimal residual disease (MRD) in eight patients in complete remission. CD66 expression was associated with an abnormal B-cell differentiation pattern and with increases in CD34/CD19+ cells in all but one case. These findings suggest that an aberrant expression of CD66 could be used to investigate MRD in ALL. The association between CD66 reactivity and bcr-abl in adult ALL remains to be investigated.
M. Carrasco ´ L. Muæoz ´ M. Bellido ´ S. Bernat ´ E. Rubiol J. Úbeda ´ J. Sierra ´ J.F. NomdedØu ()) Department of Hematology, Hospital de la Santa Creu i Sant Pau, Avda Sant Antoni M. Claret 167, E-08025 Barcelona, Spain Tel.: +34-3-2919000 Fax: +34-3-2919192 e-mail:
[email protected]
Key words CD66 ´ Immunophenotype ´ Acute lymphoblastic leukaemia ´ Myeloid antigens ´ bcr-abl
Introduction CD66, which belongs to the family of carcinoembryonic-related antigens, can be separated into two groups: one group contains the carcinoembryonic antigen (CEA) and the classic non-specific cross-reacting antigens (NCSA), which are made up of seven genes; the other group includes the pregnancy-specific glycoproteins. These proteins have several splice variants and are involved in cellular interaction and signal transduction. The CEA subgroup has been divided into CD66a (NCA-160), -b (NCA-95), -c (NCA-50/90), -d (CGM1) and -e (CEA) [1, 2]. In normal haematopoiesis, CD66 is expressed on the surface of neutrophilic, eosinophilic and monocytic series. Its expression increases with neutrophilic maturation [3, 4]. Reverse-transcriptase polymerase chain reaction (RT-PCR) studies performed on peripheral blood revealed CD66a production by lymphocytes and CD4+ helper T cells. CD66c expression has been demonstrated in CD10+ early-B-cell malignancies [4, 5]. Aberrant CD66 expression has been reported in acute lymphocytic leukaemia (ALL) cell lines and in children with ALL [6, 7, 8, 9]. In this study, we analysed CD66 reactivity using a monoclonal antibody (mAb) against CD66a, -b, -c and -e in adult patients with acute leukaemia. Correlation with other immunophenotypic markers and molecular findings was established.
Materials and methods Patients Forty-five adult patients with acute leukaemia referred to our hospital from January 1998 to April 1999 took part in the study. The diagnosis of acute leukaemia was based on the morphological criteria established by the French±American±British (FAB)
300 group [10]. The distribution according to diagnosis was as follows: B-cell lineage ALL 12 cases, AML 29 cases (24 de novo, 5 secondary to myelodysplastic syndrome) and BC-CML (blast crisis-chronic myelocytic leukaemia) 4 cases (2 My-BC and 2 LyBC). Most of the ALL cases were referred for bone-marrow transplantation. CD66 reactivity was tested in the bone marrow of five control adults (healthy donors) and in two children with idiopathic thrombocytopenic purpura (ITP).
Immunological studies Sample preparation The number of total bone marrow cells was quantified by microscopy and adjusted to 2 106 in each tube. The immunophenotypic analysis was performed on lysed whole bone marrow samples with conjugated mAbs. Antigenic expression was analysed using triple combinations of mAbs conjugated with fluorescein isothiocyanate (FITC), phycoerytrin (PE), peridinin chlorophyll protein (PerCp) or phycoerytrin-cyanine 5 (PE/Cy 5). The mAbs used in the study were: CD 66 (Kat4c-FITC, a pan-CD66 able to recognise molecules CD66a, -b, -c and -e), CD22 (4KB128 FITC), glycophorin A (JC 159 PE), CD41 (5B 12 PE), IgM (rabbit anti-human, PE), CD79a (HM57 PE; Dako, Glostrup, Denmark); CD15 (MMA-FITC), CD34 (8G12-FITC, PE), HLA-Dr (L243 PetCp), CD10 (W8E7 FITC), CD20 (L27 PE), CD2 (S5.2 FITC), CD33 (67.6 PE), CD7 (4H9 FITC), CD45 (2D1 PerCp), CD13 (L138 PE), CD14 (M0P9 FITC), CD3 (SK7 PerCp), CD4 (Leu 3 FITC), CD5 (Leu 1 FITC), CD8 [Leu 2 PE; Becton Dickinson (BDIS), San Jose, Calif.], CD19 (SJ25-C1 PE/Cy 5), MPO (H-43-5 FITC; Caltag Laboratories, Burlingame, Calif.) and CD10 (HI10a, Cy-Chrome) (Phanningen, San Diego, Calif.); CD36 (FAG-52 FITC; Immunotech, Marseille, France) and TdT (VNF-830 and VTF-510 FITC; Harlam Sera Lab, Sussex, England). Triple combinations used in the study corresponding to FITC/PE/PerCP or Pe/Cy5 conjugated mAbs were CD15/CD34/HLA-DR, CD10/CD20/CD19, CD2/CD33/CD19, CD7/CD117/CD45, CD22/CD13/CD3, CD66/CD56/CD64, CD36/ glycophorinA/CD45, CD34/CD41/CD45, CD14/CD33/CD64, and CD33/CD34/CD45. The following triple combinations were used to investigate CD66 reactivity in normal controls: CD10/CD20/CD19 in order to assess the normal B-cell differentiation pathway, CD34/CD33/CD19 CD66/CD34/CD45 and CD66/CD34/CD10. Direct immunofluorescence was performed by incubating 2 106 cells with the specific mAb for 15 min in the dark at room temperature. An isotype-matched negative control (BDIS) was used in all cases to assess background fluorescence intensity. Cells were lysed (FACS lysis solution, BDIS) for 3±5 min and centrifuged at 250 g for 5 min. The cells were washed twice with phosphate buffered saline (PBS) before being re-suspended in PBS and examined.
following myelomonocytic markers: cy MPO, CD13, CD33, and/or CD117 [12, 13]. Cytoplasmic CD3 was tested in every case. Data acquisition and analysis Measurements were performed on a FACScan flow cytometer (BDIS). For data acquisition, the LYSIS-II (BD) software program (BDIS) was used. At least 10,000 events/tube were measured. The PAINT-A-GATE PRO software program (BDIS) was employed for further data analysis. Thresholds for positivity were based on isotype negative controls. Analytical gates were set on desired viable cells based on forward light scatter and side light scatter. The positivity threshold was 20% for all markers except cytoplasmic or intranuclear antigens, for which a 10% threshold was used. Minimal residual disease (MRD) analysis MRD search was established in accordance with previously published criteria [14, 15]: (1) an abnormal B-cell differentiation pattern as revealed by abnormal percentages in the population of cells identified using the CD10/CD20/CD19 tube and (2) increases in the percentage of CD19/CD34 cells. Simultaneously, CD66/CD34 and CD22/CD34 co-expression was tested. Samples corresponding to patients in morphologic complete remission (CR) were analysed using a two-step acquisition procedure. In the first step, acquisition of 10,000 cells was performed and information stored for all these events. In the second step, a minimum of 105 cells were measured, information being stored only for the CD19+ or CD34+ cells, which were acquired employing a pre-established SSC/CD19 or CD34 live gate. The number of immature CD34+/CD19+, CD22+/CD34+ and CD66+ /CD34+ cells and the distribution of the different CD10/CD20/CD19 subpopulations were calculated as the percentage of events among the CD19+ or CD34+ cells in the second acquisition file. The number of CD34+/CD19+ cells in B-cell lineage ALL was considered high when its value exceeded 20% of CD19+ cells. The B-cell differentiation pathway was considered abnormal when B cells were located outside the dot plot regions occupied by the subsets of B cells detected in normal differentiation and/or when the percentages of the most immature subpopulations exceeded those of the most mature two B-cell subsets. Cytogenetic analysis Bone marrow was cultured according to standard methods. Twenty or more cells were analysed to exclude clonal abnormalities, which were defined in accordance with the International System for Human Cytogenetic Nomenclature (ISCN) guidelines [16].
Detection of intracellular antigens (cytoplasmic and nuclear) by flow cytometry
Molecular studies
The cells were incubated for fixation and permeabilisation with Fix and Perm solution (Caltag). They were then washed twice with PBS and incubated for 15 min with 20 ml antibody. After incubation, the cells were washed again with PBS before re-suspension and examination.
The presence of bcr/abl transcripts was investigated by means of RT-PCR. The primers and methods employed to amplify the b2a2, b3a2 types of bcr/abl hybrid mRNA and the e1a2 type in two different ªnestedº RT-PCR have been previously described [17].
Immunological criteria for lineage involvement
Statistical analysis
Lineage affiliation was done according to the EGIL recommendations [11]. The diagnosis of B-cell lineage acute leukaemia was established when the cytoplasmic CD79a (cyCD79a) was positive, regardless of the expression pattern of surface markers. AML was diagnosed by the expression of two or more of the
The chi-square test was used to compare the CD66 expression between ALL and AML.
301
Results CD66 expression in normal bone marrow Cells co-expressing CD34/CD66 were not found in the five healthy adult donors. A rare population with a CD66dim CD10+ CD19+ CD34+ CD45+ phenotype was detected in the two children with ITP (Fig. 1). These cells accounted for less than 1 10 ± 3 of all the bone marrow nucleated cells. CD66 expression in AML CD66 was expressed in 2 of 29 patients with AML (6.8%). One patient was diagnosed with AML-M2 and a normal karyotype, and the other suffered from an AML-M0 with abnormalities at chromosome 7. CD66 expression in BC-CML CD66 was negative in two of two My-BC patients, whereas it was positive in the two patients with LyBC-CML. Fig. 1 a±c Dot plots showing CD66 reactivity in a normal B-cell precursor in a child with idiopathic thrombocytopenia purpura. Cells with the CD66dim CD34+ CDIG+ CD19+ phenotype represent 1 104 of all bone-marrow nucleated cells. d±f Dot plots showing blast cells in an adult acute lymphocytic leukaemia (ALL) patient at diagnosis with CD66 expression. Abnormal cells (2 10 ± 3) co-expressing CD66bright CD34+ CD45+ were detected in an ALL patient in complete remission (G±I). d±f Blast cells were identified by forward scatter (FSC)/side scatter (SSC) gating; a±c, g±i B-cell gate was established on the basis of the CD19 expression, and low right angle light scatter (SSC) and additional live gates were defined using the CD34 expression
CD66 expression in ALL CD66 was positive in 8 of 12 patients with ALL (66.7%), P < 0.001. The characteristics of the patients with CD66 + ALL and Ly-BC are shown in Table 1. All patients had adverse prognostic factors (old age, high leukocyte count or the presence of bcr±abl transcripts). CD66 was negative in four ALL cases: two ALL-L3, one L2 with a pre-B phenotype and one L2 with a CD 10+ phenotype. None of these patients showed the bcr/abl rearrangement. CD66 expression in samples with CR Bone marrow from eight patients (23 samples) with ALL in morphologic CR were analysed for MRD. Of 23 samples, 19 were positive. In all the samples, MRD was analysed using three triple-antigen combinations as described above. All the MRD cases showed an abnormal B-cell differentiation pathway. The MRD+ cases were associated with greater than 1 10 ± 3 cells co-expressing CD22/CD34 and CD66/CD34 in all but two samples. The CD66/CD34 combination allowed us to detect residual leukaemic cells in one of these sam-
t(9;22) t(9;22) t(9;22) t(9;22) t(9;22) ± + + ± ± + ± ± +dim +dim + + + + + + + + ± + + + + + + + + + + +
± ± ± ± ±
46,XX t(9;22) ± + + +dim ± + + ± ± + + +
± ±
t(9;22) Hyperdiploid t(9;22) ± ± ± + + + ± ± + ± + ± + + + + + +
+ ± ±
CD10
+ + + + ± + + + + + + + + + + + + + + + ± ± ± + ± ± ± ± ± ± b
a
44 66 47 61 28 Female Female Male Male Male 6 7 8c 9 10c
64 42 Female Male 4 5
( 109/1) Cytoplasmic c Immunophenotype studied at relapse
+ + + + + 69 10.3 7.5 41 84
± + + + + + + + ± + ± ± + + 13.9 277
± + + + + + + + + + + + ± ± + ± ± ± + + + 15 5.6 NA
L2 L2 Ly-BC CML L1 Ly-BC CML L2 L2 L2 L1 L1 43 50 70 Female Female Female 1 2 3
CD22 HLA-Dr CD34 MPOb mb CD79ab
ples, and CD22/CD34 co-expression was the only evidence of MRD in the other. In five patients (numbers 2, 5, 6, 7 and 9) who had blast cells with CD66 reactivity at diagnosis, flow cytometric criteria for MRD were noted. Cells co-expressing CD66/CD34 and CD22/CD34 were identified, ranging from 5 10 ± 4 to 6 10 ± 3. These five patients relapsed: three died and two reached a second CR. In these two cases, MRD was detected at the end of the second induction therapy. One of these patients died of disease progression and the other is currently in CR after an allogeneic peripheral stem cell transplantation (alloPSCT) with a positive MRD. CD66 was not tested at diagnosis in three patients (numbers 1, 8 and 10) because these patients were referred in CR for stem cell transplantation (SCT). Residual leukaemic cells (3 10 ± 3) were detected pre- and post-alloPSCT in the patient relapsing 3 months after the transplantation. MRD was detected 6 months after alloPSCT in the patient relapsing after 2 months. Finally, residual leukaemic cells (2 10 ± 3) were detected before a mobilisation procedure on two occasions before morphological relapse in patient 8.
Discussion
c
CD20
CD19
CD15
CD13
CD33
CD45
CD117
Cytogenetics Immunophenotype
WBC counta FAB Age Patient Gender
Table 1 Biological characteristics of patients with acute lymphocytic leukaemia (ALL) and blast crisis-chronic myelocytic leukaemia (BC-CML) expressing CD66. NA not available; WBC white blood cell; FAB French±American±British group; MPO myeloperoxidase; HLA-Dr human leukocyte antigen-Dr
302
The frequency of the myeloid antigen expression in ALL ranges between 5% and 54% [8, 9, 18, 21]. Its significance in adults with ALL is not clear. Studies with children have shown that the myeloid antigen expression has no influence on the prognosis of these patients [21]. However, studies with adults have yielded controversial results and the adverse prognostic factor of myeloid antigen expression could be linked with particular entities (bcr/abl rearrangements) [8, 18, 19, 20]. The most common myeloid antigen expressed in ALL are CD33 and CD13, less frequently CD15 and CD14 and only rarely CD117 [12, 13, 20]. We found a high frequency of CD66 expression in ALL patients. In contrast to other myeloid antigens (i.e. CD13 and CD33), CD66 was more commonly expressed by ALL than by AML. We were not able to detect the CD10+ restriction using an antibody that recognises different epitopes clustered under CD66. CD66 reactivity could be attributed to CD66c reactivity and/or CD66a upregulation in B-lymphoblasts. In our experience, CD66 is a very satisfactory associated ALL marker. Studies performed with children have indicated that the CD66 expression could identify different prognostic groups. In our adult series, the majority of CD66 + ALL corresponded to Ph + leukaemia, which is the most common molecular lesion in this age group. The high frequency of bcr/abl could be attributed to the fact that most patients at our centre are referred for SCT. It remains to be ascertained whether CD66 positivity is restricted to a particular stage of B-cell lympho-
303
poiesis, in the light of our findings regarding a rare precursor in children with ITP, or whether it is preferentially associated with the leukaemia phenotype. The detection of leukaemia-associated immunophenotypes by means of flow cytometry is an interesting method for MRD analysis [22, 23]. Recently, alterations in normal B-lymphopoiesis have also been employed to study MRD in B-cell lineage ALL [14, 15]. Our parallel results testing CD66 and CD22 reactivity in cases with abnormal B-cell differentiation pattern suggest that leukaemic cells express CD66. These data are supported by our findings in normal volunteers. Thus, CD66 combinations could be used to detect residual leukaemic cells in a large number of cases of B-cell lineage ALL. The prognostic significance of the CD66 expression in adult ALL harbouring bcr/abl transcripts warrants further study.
References 1. Thomson J, Grunert F, Zimmermann W (1991) Carcinoembryonic antigen gene family: molecular biology and clinical perspectives. Clin Lab Anal 5 : 344±366 2. Madjic O (1989) CD66 cluster report. In: Knapp W, Dorken B, Gilks WR, Rieber EP, Schmidt R, Stein H, von der Bome (eds) AEGKr, leukocyte typing IV: white cell differentiation antigens. Oxford University Press, Oxford, pp 838±839 3. Watt SM, Sala±Newby G, Hoang T, Gilmore DJ, Grunert J, Nagel G, Murdoch SJ, Tchilian E, Lennox ES, Waldmann H (1991) CD66 identifies a neutrophil±specific epitope within the hematopoietic system that is expressed by members of the carcinoembryonic antigen family of adhesion molecules. Blood 78 : 63±74 4. Hansen L, Meyer K, Hokland P (1998) Flow cytometric identification of myeloid disorders by asynchronous expression of the CD14 and CD66 antigens. Eur J Haematol 61 : 339±346 5. Boccum P, Di Noto F, Lo Pardo C, Villa MR, Ferrara F, Rotoli B, Del Vecchio L (1998) CD66c antigen expression is myeloid restricted in normal bone marrow but is a common feature of CD10+ early-B-cell malignancies. Tissue Antigens 52 : 1±8 6. Drach J, Drach D, Glassl H, Gattringer C, Huber H (1992) Flow cytometric determination of atypical antigen expression in acute lymphoblastic leukemia for the study of minimal residual disease. Cytometry 13 : 893±901 7. Hanenberg H, Baumann M, Quentin I, Nagel G, GrosseWilde H, von Kleist S, Göbel U, Burdach S, Grunert F (1994) Expression of the CEA gene family members NCA-50/90 and NCA-160 (CD66) in childhood acute lymphoblastic leukemias (ALLs) and in cell lines of B-cell origin. Leukemia 8 : 2127±2133 8. Hann WL, Richards SM, Eden OB, Hill FGH (1998) Analysis of the immunophenotype of children treated on the Medical Research Council United Kingdom Acute Lymphoblastic Leukemia Trial XI (MCR UKALLXI). Leukemia 12 : 1249±1255 9. Kammerer R, Hahn S, Singer BB, Luo JS, Von Kleist S (1998) Biliary glycoprotein (CD66a), a cell adhesion molecule of the immunoglobulin superfamily, on human lymphocytes: structure, expression and involvement in T cell activation. Eur J Immunol 28 : 3664±3674
10. Bennett JM, Catovsky D, Daniel MT, Flandrin G, Galto DA, Gralnick HR, Sultan C (1985) Proposed revised criteria for the classification of acute myeloid leukemia: a report of the French±American±British Cooperative Group. Ann Intern Med 103 : 620±624 11. Bene MC, Castoldi G, Knapp W, Ludwig WD, Matutes E, Orfao A, van't Veer MB (1995) Proposals for the immunological classification of acute leukemias. Leukemia 9 : 1783±1786 12. Bene MC, Bernier M, Casanovas RO, Castoldi G, Knapp W, Lanza F, Ludwing WD, Matutes E, Orfito A, Sperling C, van't Veer MB for EGIL (1998) The reliability and specificity of c-kit for the diagnosis of acute myeloid leukemias and undifferentiated leukemias. Blood 92 : 596±599 13. Nomdedeu JF, Mateu R, AltØs A, Llorente A, Rio C, Estivill C, López O, Ubeda J, Rubiol E (1999) Enhanced myeloid specificity of CD117 compared with CD 13 and CD33. Leuk Res 23 : 341±347 14. Ciudad J, San Miguel JF, López-Berges MC, García Marcos MA, Gonzalez M, Vazquez L, Del Caæizo MC, López A, Van Dongen JJM, Orfao A (1999) Detection of abnormalities in B-cell differentiation pattern is a useful tool to predict relapse in precursor-B-ALL. Br J Haematol 104 : 695±705 15. Lucio P, Parreira, A, van den Beemd MWL, van Lochem EG, van Wering E, Baarrs E, Porwit-MacDonald A, Bjorklund E, Gaipa G, Orfao A, Janossy G, van Dongen JJM, San Miguel JM (1999) Flow cytometric analysis of normal B cell differentiation: a frame of reference for the detection of minimal residual disease in precursor B-ALL. Leukemia 13 : 419±427 16. Mitelman F (ed) (1995) ISCN an international system for human cytogenetic nomenclature. Karger, Basel 17. Saglio G, Pane F, Gottardi E, Frigeri F, Buonaiuto MR, Guerrasio A, De Micheli D, Parziale A, Forcani MN, Martinelli G, Salvatore F (1996) Consistent amounts of acute leukemia-associated p190 bcr/ab1 transcripts are expressed by chronic myelogenous leukemia patients at diagnosis. Blood 87 : 1075 18. Boldt DK, Kopecky KJ, Head D, Gehly G, Radich JP, Appelbaum FR (1994) Expression of myeloid antigens by blast cells in acute lymphoblastic leukemia of adults. The Southwest Oncology Group Experience. Leukemia 12 : 2118±2126 19. Cuneo A, Demuynck H, Ferrant A, Louwagie A, Doyen C, Stul M, Cassiman JJ, Dal Cin P, Negrini M, Carli MG, Boogaerts M, Michaux JL, Castoldi G, van den Berghe H (1994) Minor myeloid component in Ph chromosome-positive acute lymphoblastic leukaemia: correlation with cytogenetic pattern and implication for poor response therapy. Br J Haematol 87 : 515±522 20. Lauria F, Raspadori D, Martinelli G, Rondelli D, Ventura MA, Farabegoli P, Tosi P, Testoni N, Visani G, Zaccaria A, Gamberi B, Cenachi A, Tura S (1994) Increased expression of myeloid antigen markers in adult acute lymphoblastic leukaemia patients: diagnostic and prognostic implications. Br J Haematol 87 : 286 21. Putti MC, Rondelli R, Cocito MG, Aricó M, Sainati L, Conter V, Gugliemi C, Cantfa-Rajnoldi A, Consolini R, Pession A, Zanesco L, Masera G, Biondi A, Basso G (1998) Expression of myeloid markers lacks prognostic impact in children treated for acute lymphoblastic leukemia: Italian experience in AIEOP-ALL 88±91 studies. Blood 92 : 795±801 22. Campana D, Coustan±Smith E, Janossy G (1990) The immunologic detection of minimal residual disease in acute leukemia. Blood 76 : 163±171 23. Coustan-Smith E, Behm FG, Sanchez J, Boyett J, Hancock M, Raimondi S, Rubnitz J, Rivera G, Sandlund J, Pui C, Campana D (1998) Immunological detection of minimal residual disease in children with acute lymphoblastic leukaemia. Lancet 35 : 550±554