REVIEW ARTICLE
BioDrugs 1999 Mar; 11 (3): 185-200 1173-8804/99/0003-0185/$08.00/0 © Adis International Limited. All rights reserved.
Treatment of Drug-Induced Agranulocytosis with Haematopoietic Growth Factors A Review of the Clinical Experience
Thierry Vial, Cédric Gallant, Geneviève Choquet-Kastylevsky and Jacques Descotes Centre Anti-Poison, Centre de Pharmacovigilance, Lyon, France
Contents Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Rationale for the Use of Granulocyte-Macrophage Colony-Stimulating Factor (GM-CSF) or Granulocyte Colony-Stimulating Factor (G-CSF) in Drug-Induced Agranulocytosis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2. Clinical Experience with GM-CSF or G-CSF in Drug-Induced Agranulocytosis . . . . . . 2.1 Studies in Unselected Patients . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2 Haematopoietic Growth Factors (HGF) in Vesnarinone-Induced Agranulocytosis 2.3 HGF in Clozapine-Induced Agranulocytosis . . . . . . . . . . . . . . . . . . . . . . . 2.4 HGF in Antithyroid Drug-Induced Agranulocytosis . . . . . . . . . . . . . . . . . . . 2.5 Tolerability of HGF in Patients with Drug-Induced Agranulocytosis . . . . . . . . . . 3. Perspectives and Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Abstract
. . . . . . 185
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
187 189 189 191 191 193 194 195
Although drug-induced agranulocytosis is infrequent, it is of concern as the mortality rate ranges from 6 to 10%. Since the approval of granulocyte colonystimulating factor (G-CSF) and granulocyte-macrophage colony-stimulating factor (GM-CSF), these drugs have been increasingly used in the management of drug-induced agranulocytosis. Unfortunately, most of the data regarding the use of these agents in patients with drug-induced agranulocytosis comes from case reports. In light of the low incidence of drug-induced agranulocytosis, the large variety of offending drugs with potentially different toxic mechanisms and the wide range of neutropenia duration among patients with agranulocytosis, randomised, double-blind studies are unlikely to be performed. Case reports provide promising results with a shortening in the duration of agranulocytosis, a possible reduction in the duration of hospitalisation and the fatality rate in patients treated with haematopoietic growth factors (HGF) compared with historical controls. A therapeutic effect is also suggested by reports of reductions in the neutrophil count after HGF discontinuation following an initial increase. The results of recent case series are less positive, with only a moderate, but usually not significant, reduction in the duration of neutropenia in patients treated with HGF, as compared with those receiving routine care. A Japanese study indi-
186
Vial et al.
cated that G-CSF was effective in patients with mild-to-moderate antithyroid drug-induced neutropenia, whereas no clear benefit was apparent in those with severe neutropenia. Several factors, for example, early recognition and improved management of individual cases with better supportive care, have contributed to a decrease in the overall mortality of drug-induced agranulocytosis. HGF are expected to further reduce mortality. Guidelines for the use of HGF in patients with febrile neutropenia, as established by the American Society of Clinical Oncology, are probably valuable for the management of drug-induced agranulocytosis. In accordance with these recommendations, the use of HGF may be recommended in patients with severe neutropenia and/or poor prognostic factors. Whether the absence of myeloid precursors or presence of promyelocytes or myelocytes in bone marrow examination represents optimal conditions for HGF treatment is still unknown. Most authors agree that treatment should be administered early in the course of the disease. An interesting approach, in which treatment decisions are based on the granulocyte count 4 hours after a single dose of G-CSF in patients with anthithyroid drug-associated neutropenia should be more extensively evaluated.
Agranulocytosis was originally described by Schultz[1] as an association between severe neutropenia and the sudden onset of infectious symptoms. In this review, agranulocytosis refers to severe neutropenia with or without infectious complications. Standardised definitions for agranulocytosis have been proposed: neutropenia is conventionally defined as a count of segmented polymorphonuclear leucocytes (PMNs) and band cells lower than 1.5 × 109 cells/L, and severe neutropenia (or agranulocytosis) as a count below 0.5 × 109 cells/L.[2] However, the definitions presented in table I will be used in this review. Agranulocytosis is one of the most frequently reported drug-related haematological complications and drugs are implicated in more than 90% of agranulocytosis cases.[3,4] Several criteria are used to assess causality in drug-induced agranulocytosis. A careful estimate of the incidence of druginduced agranulocytosis was provided by the International Agranulocytosis and Aplastic Anaemia Study (IAAAS) which was conducted from 1980 to 1986 in 6 European countries with a population of approximately 23 million.[5] The overall incidence of agranulocytosis was 4.7 cases per million per year, and the estimated incidence attributed to drugs was 3.1 cases per million per year. Therefore, © Adis International Limited. All rights reserved.
the excess risk of agranulocytosis due to drugs is seemingly very low. The number of drugs possibly involved in agranulocytosis has expanded in recent years to over 100 compounds.[6] From a combined analysis of 3 case-control studies, 20 drugs, or groups of drugs, were significantly associated with a risk of agranulocytosis.[7] In a multivariate model, the greatest relative risk (RR) estimates were found with sulfasalazine [RR: 207, 95% confidence interval (95% CI): 61 to 708], antithyroid drugs (e.g. carbimazole) (RR: 182, 95% CI: 74 to 449), procainamide (RR: 163, 95% CI: 45 to 594) and dipyrone (RR: 28, 95% CI: 10 to 74). Relative risks were estimated for any use in the relevant period (i.e. 7 or 14 days before the index day) compared to no use. Several drugs marketed later, such as ticlopidine,[8] clozapine[9] and vesnarinone[10,11] are relatively frequent causes of agranulocytosis with an incidence of approximately 1%. Although complete haematological recovery is Table I. Definitions of neutropenia used in this review Severity of neutropenia
Granulocyte count (×109 cells/L)
Mild
≥ 0.5 < 1
Moderate
≥ 0.1 < 0.5
Severe
< 0.1
BioDrugs 1999 Mar; 11 (3)
HGF for Drug-Induced Agranulocytosis
187
Table II. Summary of the main features of drug-induced agranulocytosis Immune type
Toxic type
Model drug
Aminopyrine
Phenothiazine
Drug dosage
Independent of the dose
Relatively dose-dependent
Clinical pattern
Often symptomatic with acute onset
Often asymptomatic or insidious onset
Time to onset
Usually short (7 to 21 days)
Usually more than 21 days
Drug readministration
Very rapid recurrence whatever the dose
Latent period, high-dose required
Laboratory exploration
Detection of drug-dependent antibodies
Evidence of direct or metabolite-mediated toxicity to cells
the rule, agranulocytosis remains a major and frequent cause of fatal reactions to drugs because of the development of severe infections, despite improved supportive measures and the availability of broad-spectrum antibiotics.[12,13] In the IAAAS, the overall mortality rate was 9% with a higher incidence in patients over 60 years and in patients who became septic or had prolonged (≥ 1 week) neutropenia.[5] Interestingly, this fatality rate was lower than previously reported[13,14] and closer to more recent estimates of 6 to 7%.[12,15,16] Before the availability of haematopoietic growth factors (HGF), several other factors had already contributed to decrease the mortality from this condition. For example, prescriber knowledge of the commonly implicated agents, patient education and regular haematological monitoring provide for earlier recognition of the condition and prompt discontinuation of the drug. Assessment of the severity of neutropenia, the identification of patient prognostic factors and better supportive care measures are also important. Several mechanisms that produce agranulocytosis have been described. An immunological mechanism, referred to as the aminopyrine type of mechanism, leads to the rapid destruction of circulating mature granulocytes and, probably, their progenitors in the bone marrow. Anti-PMN antibodies are sometimes detected in the serum of patients.[17] Another mechanism, which is referred to as the phenothiazine type of mechanism, involves a direct and toxic bone marrow suppressive effect through interference with protein synthesis or cell replication. In such cases, the time course of agranulocytosis is relatively dose-dependent with a © Adis International Limited. All rights reserved.
more insidious onset. Direct toxicity can be reproduced in vitro in bone marrow cultures, and was used to detect the causative drug in 59[18] and 71%[19] of patients in 2 reports. A more recently proposed mechanism involves the generation of unstable, chemically reactive drug metabolites by neutrophils or neutrophil precursor myelo-peroxidases which bind covalently to cell macromolecules and result in cytotoxic effects and/or neoantigen formation.[20] Unfortunately, regardless of the mechanism, laboratory tests to confirm the role of a given drug are not readily available or often yield equivocal results.[4] In addition, both toxic and immune mechanisms (see table II) have been described with the same drug in different individuals (e.g. beta-lactams).[6,21,22] 1. Rationale for the Use of Granulocyte-Macrophage Colony-Stimulating Factor (GM-CSF) or Granulocyte Colony-Stimulating Factor (G-CSF) in Drug-Induced Agranulocytosis After hospitalisation and discontinuation of all suspected drugs in a patient with suspected druginduced agranulocytosis, treatment is primarily preventive and supportive and includes the administration of antibiotics and antifungal agents. Therefore, treatment is very similar to the guidelines for the treatment of neutropenia associated with cancer chemotherapy.[23] The severity of drug-induced agranulocytosis depends on the absolute level of circulating neutrophils and bone marrow findings, with an expoBioDrugs 1999 Mar; 11 (3)
188
nential increase in the risk of severe infection as the neutrophil count drops.[13] The duration of granulocytopenia is also one of the most important factors correlated with a poor outcome. There is also a direct correlation between the granulocyte to erythrocyte count ratio in bone marrow and the peripheral leucocyte count,[24] and an inverse relationship between the extent of granulocyte compartment depletion and the duration of neutropenia.[13] When a bone marrow aspirate reveals a complete absence of cells in the myeloid series, the neutrophil recovery usually occurs within 2 weeks, but may require up to 2 months, whereas the presence of promyelocytes or myelocytes indicates that recovery will occur within 7 days.[16,25] Several additional factors, predictive of a poor outcome in drug-induced agranulocytosis, were carefully analysed in a retrospective study of 168 patients.[13] Renal insufficiency and bacteraemia were found to be significant prognostic factors. Theoretically, neutrophil precursor cells remaining within the bone marrow after drug-induced agranulocytosis should respond to HGF, which act on specific cells in the neutrophil lineage. A shortened duration of severe neutropenia, an accelerated increase in the neutrophil count during the recovery phase and, therefore, a reduced incidence of infection, duration of hospitalisation and decreased fatality rate are major end-points to be expected from the use of HGF in this indication. The beneficial effects of HGF in cytotoxic therapy-induced neutropenia, bone marrow transplantation and congenital, cyclic or idiopathic neutropenia, as well as in HIV-positive patients with ganciclovir- or zidovudine-induced neutropenia, generated interest in the use of these agents in the treatment of drug-induced agranulocytosis.[26-30] Both granulocyte colony-stimulating factor (GCSF) and granulocyte-macrophage colony stimulating factor (GM-CSF) stimulate myeloid granulocytic progenitors and induce their differentiation, proliferation and maturation.[26] In addition, they stimulate the functional activity of the few remaining mature cells. G-CSF enhances the bactericidal activity of mature neutrophils and increases pro© Adis International Limited. All rights reserved.
Vial et al.
duction of oxygen radicals. GM-CSF increases the functional capacity of monocytes and macrophages. These additional effects may enhance the control of infectious processes.[26] Interestingly, endogenous serum G-CSF levels have been reported in several cases of antithyroid drug-induced agranulocytosis. An increased concentration, close to that obtained after subcutaneous injection of the recommended dosage, of G-CSF was observed during the period of agranulocytosis in a patient with thiamizole (methimazole)-induced agranulocytosis.[31] This suggests that an inability to produce G-CSF is not the immediate cause of agranulocytosis and that G-CSF treatment is not always needed, provided that endogenous G-CSF production is sufficient to obtain a spontaneous recovery. Similar increases in endogenous serum GCSF levels, at the onset of severe neutropenia, have been reported in patients with antithyroid drug-induced agranulocytosis.[24,32] In contrast, 5 of 13 patients with mild or moderate neutropenia had normal G-CSF levels compared both with normal controls and pre-antithyroid therapy.[32] Thus, endogenous G-CSF production increases when bone marrow granulocyte production is markedly impaired. Based on these results, G-CSF was suggested to be effective only in patients with mild or moderate granulocytopenia, in whom endogenous G-CSF levels were low or undetectable.[32] Although no controlled clinical trials or population-based studies using a standard protocol are available to assess the efficacy of HGF in druginduced agranulocytosis, indirect evidence argues for their beneficial role. A decrease in the neutrophil count following an initial increase was observed after HGF withdrawal in some patients.[33] This, together with a subsequent, sustained recovery after a second or third course of HGF in these patients, suggests that HGF are useful in restoring and maintaining the neutrophil count.[16,28,32,34-36] Interestingly, 2 patients with antithyroid drug- or clozapine-induced granulocytopenia were successfully managed with G-CSF despite the continuation of the causal drug.[37,38] BioDrugs 1999 Mar; 11 (3)
HGF for Drug-Induced Agranulocytosis
In the absence of specific controlled trials, the use of HGF in the treatment of drug-induced agranulocytosis is based on retrospective case series or anecdotal cases reports in which granulocyte recovery time was compared with that in historical studies. Since the first report in 1989, the use of HGF in the treatment of drug-induced agranulocytosis has been reported with increasing frequency (table III), but a detailed analysis of these reports is beyond the scope of this article. Case reports are subject to publication bias because they generally describe the successful use of these agents. Several patients were also successfully treated for druginduced pancytopenia or aplastic anaemia,[67-70] but failure of treatment was also reported.[71-76] This review will focus on larger cases series, series with historical controls from the same centre, or reports in which agranulocytosis was induced by selected drugs and in which sufficient data regarding HGF treatment was provided.
189
2. Clinical Experience with GM-CSF or G-CSF in Drug-Induced Agranulocytosis 2.1 Studies in Unselected Patients
In a descriptive analysis of 22 symptomatic patients treated with G-CSF,[41] neutrophil recovery (i.e. absolute neutrophil count >1 × 109 cells/L) occurred after a median of 4 days (range 1 to 15 days), a duration shorter than expected in untreated patients, in whom a mean of about 10 days has been reported (see table IV). The death of 2 patients (9%) was reported without any other information. Sprikkelman et al.[43] extensively reviewed the use of HGF in drug-induced agranulocytosis in a retrospective analysis of 70 case reports. Twelve patients had mild granulocytopenia, 7 moderate granulocytopenia and 51 severe granulocytopenia. Aplasia of the myeloid lineage was found in 18 patients; 35 patients had features of myeloid hypo-
Table III. Reports of drug-induced agranulocytosis in which patients were treated with haematopoietic growth factors (HGF) Drug
Reference
Amodiaquine
39
Anticonvulsants (phenobarbital, phenytoin)
41
Antidepressants (clomipramine, mianserin, paroxetine)
41-43
Antimicrobials (cefuroxime axetil, ciprofloxacin, chloramphenicol; imipenem, pefloxacin, penicillin, cotrimoxazole, vancomycin)
41, 44, 45
Antipsychotics [chlorpromazine, clozapine (see section 2.3), promethazine]
46-48
Antirheumatics (gold, methotrexate, tiopronin)
39, 41, 49, 50
Antithyroid drugs [carbimazole, thiamazole (methimazole), propylthiouracil]
(See section 2.4) 24, 28, 32, 36, 95-102
Azathioprine
41
Captopril
45, 51
Chlorpropamide
52
Dapsone
44, 53
Immunoglobulins
54
Nonsteroidal antiinflammatory drugs [aminophenazone (amidopyrine), dipyrone, ketoprofen, ibuprofen, indomethacin, mefenamic acid, sulindac]
29, 33, 35, 39, 40, 41-55
Omeprazole
33
Procainamide
41, 56
Quinidine
56
Ranitidine
45, 56, 57
Sulfasalazine, mesalazine
39, 43, 49, 58-63a
Terbinafine
64, 65
Ticlopidine
39, 41, 66
Vesnarinone
(See section 2.2) 11, 78
a
The patient died from perforated duodenal ulcer after recovery from neutropenia.
© Adis International Limited. All rights reserved.
BioDrugs 1999 Mar; 11 (3)
190
Vial et al.
Table IV. Duration of neutropenia and fatality rate in patients with drug-induced agranulocytosis before and after the introduction of hematopoietic growth factors (HGF) Reference
Period
HGF available
Patients
Time to recoverya (days)
Mortality (%)
Julia et al.[13]
1970-1989
No
168
10 ± 8
27 (16)
Abt et al.[14]
1968-1976
No
30
10.2 ± 6.3
9 (30)
Sprikkelman et al.[43]b
1991-1994
Yes
58
5.4 ± 4.5c
3 (5)
a
Time to reach counts ≥ 0.5 × 109 cell/L.
b
Patients had moderate or severe neutropenia.
c
The mean time to neutrophil recovery was 2.6 days (range 1-5 days) in patients with moderate neutropenia and 5.8 days (range 1-31 days) in those with severe neutropenia.
plasia with maturation arrest and 6 a normal number of myeloid cells. The HGF dosage ranged from 2 to 10 μg/kg/day and 70% of patients were treated with G-CSF (duration of treatment not stated). G-CSF or GM-CSF enhanced the recovery of the myeloid lineage with a statistically significantly accelerated normalisation of peripheral blood granulocyte counts in patients with moderate or severe neutropenia, as compared with historical controls (table IV). In addition, a statistically significantly lower mortality rate was found in the moderate and severe granulocytopenia groups. No correlation was found between the dose of HGF and the duration of granulocyte recovery, and there was no difference in time to recovery between GCSF and GM-CSF treated groups. This review attracted criticism.[39,77] A major flaw was the biased selection (patients successfully treated with HGF were more likely to be described in published reports) which resulted in a comparison of isolated case reports with untreated patients in large case series. In addition, the time to recovery in patients treated with HGF was calculated from the first day of HGF treatment with an average delay of 3 days (range 0 to 17 days) after the diagnosis of drug-induced agranulocytosis, but not from the cessation of the suspected offending drug(s), as in the untreated cases. Some old cases selected for comparison occurred before the availability of modern antibiotics. A statistically significant difference in mortality rates was found between cases reported before or after 1975 (21.5% versus 9%) by pooling data from several studies published before HGF were available.[13] In addi© Adis International Limited. All rights reserved.
tion, it should be noted that neutrophil counts had returned to baseline within 8 days of drug withdrawal in approximately 50% of surviving patients in the IAAAS study. In a comparison of 7 historical controls and 6 patients treated with HGF for thiamazole- or nonsteroidal anti-inflammatory drug-induced agranulocytosis, a moderate, but significantly shorter period of neutropenia occurred in treated versus untreated patients (8.7 ± 1.9 days versus 11.0 ± 2.2 days), as well as a decreased duration of antibiotic treatment and hospitalisation.[55] Other case series have reported less positive results. Among 9 patients with drug-induced agranulocytosis treated with G-CSF or GM-CSF (5 μg/kg/day), a potential benefit from HGF was clear in only 3 patients.[39] Similarly, only 1 out of 5 patients identified by Delannoy and Ravoet[77] had an apparent reduction in the duration of neutropenia after treatment with GM-CSF. However, 3 of the 4 patients who derived no apparent benefit from GM-CSF had relapsing agranulocytosis, which may explain a slower recovery. Over a 7-year period (1988 to 1994) during which HGF were available to clinicians, 62 cases of druginduced agranulocytosis were reported to the regional drug monitoring centre in Lyon.[16] G-CSF and GM-CSF were used in 8 and 3 patients, respectively, who had an absolute neutrophil count lower than 0.1 × 109 cells/L. The mean duration of neutropenia in HGF-treated patients was 10.5 days, which was not different from the mean duration of 9.9 days in the 51 untreated patients. A comparison of patients treated with HGF and those with sponBioDrugs 1999 Mar; 11 (3)
HGF for Drug-Induced Agranulocytosis
taneous recoveries was restricted to patients with severe neutropenia and bone marrow aspirates which revealed an absence of cells of the myeloid series (6 patients in each group). The mean duration of neutropenia was 11.8 days (range 8 to 18 days) in patients treated with HGF and 13.0 days (range 8 to 24 days) in patients not receiving HGF. However, HGF was administered late in the course of neutropenia (i.e. a mean of 6 days after the diagnosis). 2.2 Haematopoietic Growth Factors (HGF) in Vesnarinone-Induced Agranulocytosis
Agranulocytosis has occurred in 1 to 3% of patients treated with vesnarinone, an oral inotropic agent.[10,11] In European and US studies in 22 patients with vesnarinone-induced neutropenia, the mean duration of neutropenia was not significantly lower in 8 patients treated with G-CSF (8 ± 3.6 days) compared with patients who recovered spontaneously (11.6 ± 11.6 days).[78] None of the patients receiving G-CSF died, whereas 4 of the 14 remaining patients died during hospitalisation. Although the deaths were considered to be from cardiovascular causes, 3 patients were neutropenic at the time of death. From the results of this study, it was suggested that G-CSF might improve survival by reducing in-hospital mortality.[78] Japanese investigators reached similar conclusions. An insignificant trend towards a reduction in the mean duration of neutropenia was observed in 27 patients with vesnarinone-induced neutropenia treated with G-CSF (10.3 ± 8.7 days) compared with 31 patients who did not receive G-CSF (13.7 ± 12.7 days).[11] 2.3 HGF in Clozapine-Induced Agranulocytosis
Clozapine, an atypical antipsychotic drug first marketed in 1972, was withdrawn from the market in many countries in 1975 following the repeated occurrence of agranulocytosis in clozapine-treated patients. However, because of its unique therapeutic advantages, the drug was reintroduced in the early 1990s.[9] A protocol of haematological mon© Adis International Limited. All rights reserved.
191
itoring is now mandatory, with weekly white blood cell counts during the first 18 weeks of treatment, and every 2 to 4 weeks thereafter (in the US, weekly haematological surveillance is recommended throughout treatment). Neutropenia typically develops between the 4th and 18th week of treatment with clozapine, with 95% of cases occurring within the first 24 weeks. The cumulative incidence of agranulocytosis is approximately 0.8% after 1 year and 0.9% after 18 months.[79,80] Bone marrow examination performed at the time of diagnosis of clozapine-induced agranulocytosis usually reveals an absence of myeloid precursors, or impaired maturation with increased numbers of cells in the promyelocytemyelocyte lineage. At least 17 case reports provide sufficient details to assess the benefit of HGF in patients with clozapine-induced agranulocytosis (table V). GCSF was used in 13 patients, GM-CSF in 2 and sequential treatment with G-CSF and GM-CSF or GM-CSF and interleukin-3 in the 2 remaining patients. HGF were initiated from 0 to 17 days after diagnosis and continued for 3 to 11 days. The duration of agranulocytosis after initiation of HGF treatment ranged from 1 to 24 days. Overall, the total duration of agranulocytosis ranged from 5 to 27 days, and time of hospitalisation ranged from 9 to 24 days. The place of HGF in the management of clozapine-induced agranulocytosis remains to be determined, as no controlled studies are available. Only 2 retrospective studies have compared the outcome of agranulocytosis in patients receiving HGF with untreated controls. In one study, 4 patients treated with G-CSF (300-600 μg/day), within 48 hours of being diagnosed with agranulocytosis, were compared with 7 matched historical controls.[92] All patients had an absolute neutrophil count (ANC) below 0.1 × 109 cells/L and no myeloid precursors in a bone marrow biopsy. The time required for the neutrophil count to increase to >0.5 × 109 cells/L was 8 days (range 6 to 11 days) in the G-CSF group and 15.7 days (range 10 to 21 days) in historical controls. However, in the G-CSF-treated group, BioDrugs 1999 Mar; 11 (3)
192
© Adis International Limited. All rights reserved.
Table V. Reports of clozapine-induced agranulocytosis in which parents were treated with haematopoietic growth factors (HGF) Reference
Duration of clozapine treatment before diagnosis
ANC nadir (× 109 cells/L) after clozapine withrawal
Interval (days) from diagnosis until initiation of HGF treatment
HGF type, daily dosage (duration of treatment)
Time (days) to recovery (ANC >1 × 109 cells/L) after HGF
Time (days) to recovery (ANC > 1 × 109 cells/L) after clozapine withdrawal
Duration of Hospitalisation in days
Barnas et al.[81]
6 weeks
0,162 at day 3
0
GM-CSF 150-300g (8 days)
5
5
9
Bradford et al.[82]
12 months
0 at day 0
17
G-CSF 5 g/kg (6 days) then GM-CSF 5 g/kg (3 days)
7
24
NS
Geibig & Marks[83]
21 months
<0.1 at day 0
3
G-CSF 300g (7 days)
6
9
13
Gruner et al.[84]
11 weeks
0 at day 0
0
G-CSF 300g (10 days)
12
12
22
Yeh[85]
6 weeks
0 at day 6
4
G-CSF 300-900g (10 days)
8
12
13
Guillion & Yeh[85]
4 weeks
0 at day 6
3
G-CSF 300-900g (6 days) 6
9
NS
Guillion & Yeh[85]
8 weeks
0 at day 3
2
G-CSF 300-900g (11 days)
10
12
NS
Guillion & Yeh[85]
8 weeks
0 at day 2
1
G-CSF 300-900g (8 days) 10
11
NS
Guillion & Yeh[85]
3.5 months
0 at day 2
1
G-CSF 300-900g (6 days) 6
7
NS
Lamberti et al.[86]
11 weeks
0 at day 3
2
G-CSF 300g (8 days)
6
8
NS
Lamberti et al.[86]
7 weeks
0 at day 4
3
G-CSF 300g (6 days)
4
7
NS
Lamberti et al.[86]
10 weeks
0 at day 5
6
G-CSF 300g (5 days)
4
10
NS
Nielsen[87]
10 weeks
0 at day 1
0
G-CSF 5-10 g/kg (11 days)
10
10
11
Oren et al.[88]
7 weeks
0 at day 0
4
GM-CSF 3 g/kg (4 days)
1
5
NS
7 weeks
0 at day 3
0
G-CSF 300-480g (10 days)
10
10
NS
Valevski et al.[90]
6 weeks
0 at day 4
0
GM-CSF 300g (3 days) then IL-3 300g (16 days)
24
27
NS
Weide et al.[91]
3.8 weeks
0 at day 0
9
G-CSF 10 g/kg (6 days)
5
14
14
Mean ± SD
15.4 ± 22.3 weeks
3.2 ± 4.3
(7.5 ± 2.4 days)
7.9 ± 5.0
11.3 5.9
13.7 ± 4.5
Guillion &
Raison et
al.[89]
Vial et al.
BioDrugs 1999 Mar; 11 (3)
ANC = absolute neutrophil count; G-CSF = granulocyte colony-stimulating factor; GM-CSF = granulocyte-macrophage colony-stimulating factor; IL-3 = interleukin-3; NS = not stated.
HGF for Drug-Induced Agranulocytosis
clozapine treatment was discontinued immediately when the white blood cell count was <3 × 109 cells/L or when the ANC was <1.5 × 109 cells/L, whereas no information as to the timing of clozapine discontinuation or the ANC at the time of clozapine withdrawal was provided for the historical control group. In addition, the duration of hospitalisation was not evaluated. In another retrospective study, the medical records of 11 patients with clozapine-induced agranulocytosis were evaluated.[93] 6 patients received G-CSF (usually 300 μg/day), which was initiated from 0 to 7 days after the ANC was discovered to be <0.5 × 109 cells/L and continued for a mean of 5.8 days (range 3 to 8 days). The ANC in patients receiving G-CSF recovered to >0.5 × 109 cells/L after a mean of 3.5 days (range 1 to 6 days) of treatment. However, the total duration of agranulocytosis was 6.5 days (range 1 to 11 days) in the G-CSF-treated group and 6.6 days (range 3 to 11 days) in 5 the untreated patients. No fatalities occurred. A significant difference was found in the mean duration of hospitalisation, with 8.2 days (range 4 to 13 days) in the treated patients and 13.5 days (range 10 to 18 days) in 4 of the untreated group. 1 patient in the latter group was treated as an outpatient and was not included in the calculations. No clear explanation was provided for the longer duration of hospitalisation in the untreated group, despite the similar duration of neutropenia in both groups. Of note, at the nadir in the ANC only moderate neutropenia was present in 2 patients in the G-CSF group, whereas neutrophils were absent in all 5 patients in the untreated group. A difference in the severity of neutropenia between both groups was shown by the presence of fever or worsening thrombophlebitis in 3 patients in the untreated group, whereas only 1 patient in the G-CSF group developed fever, which is suggestive of selection bias. 2.4 HGF in Antithyroid Drug-Induced Agranulocytosis
Agranulocytosis is one of the major adverse effects of antithyroid thionamide drugs, namely thia© Adis International Limited. All rights reserved.
193
mazole, carbimazole and propylthiouracil. Mild leucopenia is observed in 12 to 25% of patients and agranulocytosis occurs in 0.1 to 0.5% of patients.[94] A substantially shortened recovery time after thionamide withdrawal and treatment with GCSF or GM-CSF treatment was suggested in several case reports[28,36,95-102] and 2 case series in which the management of antithyroid drug-induced agranulocytosis was described.[24,32] An interesting, but retrospective, comparison was done in 34 patients with moderate or severe thiamazole-induced neutropenia who were treated in a Japanese hospital between 1979 and 1991.[24] Patients who did not receive G-CSF were divided into 2 groups: group A received antibiotics alone (n = 11), and group B received antibiotics and dexamethasone (n = 11). When G-CSF became available, 12 further patients (group C) were divided into 2 groups according to the ratio of granulocyte to erythrocyte (G : E) counts in bone marrow. Six patients had a G : E ratio of <0.5 (group C1), and 6 patients had a ratio of ≥0.5 (group C2), all of whom received antibiotics plus G-CSF (75 μg/day, duration not reported). The time to neutrophil recovery (>1 × 109 cells/L) was significantly shorter in patients treated with G-CSF compared with those in group A or B (6.8 ± 1.2 days versus 10.1 ± 2.2 and 12.3 ± 1.9 days, respectively). As expected, recovery was significantly faster in patients in group C2 compared with group C1 (2.2 ± 0.6 versus 9.8 ± 1.3 days). However, the time to neutrophil recovery was not statistically different in patients in groups A or B compared with those in group C1, but the validity of comparison is limited since bone marrow examinations were not performed in groups A or B. Similar results were reported in 16 patients with thiamazole or propylthiouracil-induced neutropenia who were observed for 1-year.[32] Patients were stratified as to the severity of neutropenia: 10 patients had mild granulocytopenia and 3 patients each had moderate or severe granulocytopenia. All patients received G-CSF 75 μg/day until the granulocyte count was ≥7 × 109 cells/L. Neutrophil recovery was defined as a granulocyte count >1 × 109 BioDrugs 1999 Mar; 11 (3)
194
cells/L. Twelve patients in the mild or moderate groups had neutrophil recoveries (range 1.5 to 12.3 × 109 cells/L) within 24 hours of the first injection. The other patient in this group had a neutrophil count of 0.6 × 109 cells/L after 1 day and >1 × 109 cells/L within 4 days. In contrast, the time to neytrophil recovery ranged from 6 to 14 days in patients with severe granulocytopenia. The results of these studies suggest that G-CSF is effective only in mild to moderate granulocytopenia. However, since Japanese investigators used G-CSF doses at least 4 times lower than those used in Western countries, a beneficial effect of higher doses in patients with severe granulocytopenia cannot be excluded. 2.5 Tolerability of HGF in Patients with Drug-Induced Agranulocytosis
Data on the tolerability of HGF in patients with drug-induced agranulocytosis are limited. Based on the known tolerability profile of G-CSF and GM-CSF and owing to the short duration of treatment in this setting, expected adverse effects include musculoskeletal and transient bone pain, headache, increased lactate dehydrogenase levels and mild rashes.[103] In contrast to G-CSF, GMCSF produced fever in as many as 50% of patients receiving more than 3 μg/kg/day;[103] an adverse effect which must be correctly identified to avoid misdiagnosing infection. This is illustrated by the persistent fever in a patient receiving GM-CSF treatment for sulfasalazine-induced agranulocytosis.[59] In this patient, GM-CSF was the assumed cause because high fever (temperature range 38.7 to 40.2°C) persisted during the whole GM-CSF treatment period (8 days) in spite of intensive culturing of blood, a wide range of viral serology testing, antibiotic treatment, and declining C-reactive protein, and disappeared within a day of discontinuing GM-CSF. Several criteria have been proposed to recognise GM-CSF–induced fever: peak fever at a constant time after GM-CSF injection, lack of clinical and biological signs of infection, and prompt response to paracetamol administration.[103] However, GM-CSF–induced fever may © Adis International Limited. All rights reserved.
Vial et al.
also be considered as an indication of improved host defences rather than a true adverse effect. In several case series, no or only minor adverse effects developed after HGF treatment for druginduced agranulocytosis.[16,24,43,93] However, acute respiratory distress syndrome (ARDS) was recently reported to be a possible complication of G-CSF treatment for drug-induced agranulocytosis. Diffuse pulmonary infiltrates and severe hypoxaemia occurred in 3 of 9 patients with severe neutropenia and infectious complications who were treated with G-CSF 5 μg/kg/day for a mean of 8.8 (range 5 to 23) days.[33] Two patients had severe sepsis upon presentation and the third had preexisting pulmonary failure. ARDS was also reported in 1 of 4 patients receiving HGF (GM-CSF for 5 days, dosage not reported) for clozapine-induced agranulocytosis.[80] Interestingly, a liver transplant patient with antibiotic-induced neutropenia and an infection experienced a worsening of ARDS and apparent arterial desaturation following each injection of G-CSF (5 μg/kg/day for 2 days).[104] In these patients, ARDS occurred during neutrophil recovery and a rapid rise in the neutrophil count might have exacerbated an underlying predisposition. For example, white blood cell sequestration within the lung may have contributed to this effect.[105] Very few data on the comparative clinical tolerability of G-CSF and GM-CSF are available. In one sequential but non-randomised trial in 50 patients with cancer, patients receiving intravenous molgramostim (GM-CSF) required dose reduction or treatment discontinuation more frequently than patients receiving intravenous figrastim (G-CSF; 41% versus 17%, p = 0.09).[106] Symptoms found more frequently in the GM-CSF group included fever, cutaneous rash, weight gain and hypotension, whereas myalgia and bone pain were more frequently reported in the G-CSF group. However, no significant differences in the adverse effect profile of filgrastim and mogramostim were found in a recent prospective randomised trial in 42 patients with breast cancer who underwent autologous peripheral stem cell transplantation.[107] BioDrugs 1999 Mar; 11 (3)
HGF for Drug-Induced Agranulocytosis
3. Perspectives and Conclusions The overall mortality from drug-induced agranulocytosis has decreased with the early recognition of this condition by prescribers, patient education, and by prompt withdrawal of the causal agent and improved supportive care measures. HGF are increasingly used in the treatment of drug-induced agranulocytosis, but their efficacy has not been rigorously investigated. Whether a subtype of neutropenia responds better to HGF and whether HGF reduces the duration of neutropenia and the fatality rate is still uncertain. In most patients, a spontaneous and sometimes rapid neutrophil recovery cannot be excluded, especially in patients who appear to have maturation arrest according to a bone marrow biopsy. Unfortunately, the time-course of recovery is unpredictable from 1 patient to another and for a given drug. Evaluation of HGF in drug-induced agranulocytosis is hindered by the limited number of evaluable patients. The very low incidence of drug-induced agranulocytosis and the variety of offending drugs with potentially different toxic mechanisms severely limits the ability to design a randomised, placebo-controlled study. As HGF are now increasingly used for the treatment of drug-induced agranulocytosis in some centres, several authors have emphasised the ethical problems related to such studies.[57,108] An approach in which data are collected in a sufficient number of patients who have been treated with a uniform protocol would allow for a more accurate comparison between patients treated with HGF and those receiving routine care, and would be preferred to the use of historical controls. Increasing the quality of adverse drug reaction reports contributes to a continuing analysis of drug-induced agranulocytosis.[16] The minimal data included in a report of drug-induced agranulocytosis should include: the likely offending drug(s), an assessment of the severity of agranulocytosis by complete peripheral blood count and bone marrow biopsy, patient prognostic factors and other underlying diseases or conditions (e.g. age, renal function, presence of bacteraemia, support© Adis International Limited. All rights reserved.
195
ive care measures) and whether and how HGF were used together with the outcome of the disorder.[2] Patients with drug-induced agranulocytosis who may benefit from HGF remain to be more clearly identified. According to the manufacturers of HGF, prerequisites for the administration of HGF in patients with drug-induced agranulocytosis include an ANC <1 × 109 cells/L (in the US) or <0.5 × 109 cells/L (in Europe), withdrawal of the suspected drug, the absence of exposure to a cytotoxic drug and evidence of bone marrow aplasia or hypoplasia.[41] It is still not known whether the absence of myeloid precursors or the presence of features of maturation arrest in bone marrow represent the optimal conditions for HGF treatment. Several authors conclude that HGF accelerate neutrophil recovery in patients with mild or moderate neutropenia, or with normal marrow cellularity, by reducing the transit time of cells in the maturing myeloid cell compartment, whereas HGF are not effective in patients with severe neutropenia.[24,32,39] Although patients experiencing mild to moderate neutropenia seem to benefit from HGF, they usually recover rapidly in the absence of HGF treatment and it is unclear whether they require such treatment. Based on a study in 37 patients with antithyroid drug-induced neutropenia, it was proposed that disease severity could be predicted by the granulocyte count obtained 4 hours after a 75μg injection of G-CSF.[109] 25 of 28 patients (89%) with mild granulocytopenia, and 4 of 6 patients (67%) with moderate granulocytopenia, had granulocyte counts >1 × 109 cells/L at 4 and 24 hours after the G-CSF injection and had sustained neutrophil recoveries thereafter. In contrast, 5 further patients with mild or moderate granulocytopenia had granulocyte counts <1 × 109 cells/L 4 hours after the G-CSF injection and had further decreases in granulocyte count, despite daily G-CSF administration. These patients, together with the 3 patients with severe granulocytopenia who also failed to respond to a single injection of G-CSF, required hospital treatment including isolation and received GCSF until recovery. Guidelines for the treatment of BioDrugs 1999 Mar; 11 (3)
196
Vial et al.
higher dosages.[108] Although there is uncertainty as to when therapy should begin, the beneficial effects of treatment seem more pronounced when HGF are administered early, i.e. within 48 hours of the diagnosis of agranulocytosis.[39,108] The duration of treatment should probably be limited to 4 to 5 days or until a neutrophil count above 0.5 to 1 × 109 cells/L is reached. The potential risk of ARDS should be considered in patients who have an excessive rebound in the neutrophil count. Whether G-CSF offers any advantages over GM-CSF in the treatment of drug-induced agranulocytosis is not clear. G-CSF produced 6-fold greater stimulation of granulocyte production in patients with cancer than equivalent doses of GMCSF and predominantly stimulated the proliferation and differentiation of myeloid progenitors cells, with a shorter transit time in the maturing myeloid cell compartment.[26] In addition, G-CSF seems to induce a shift in granulocytes from the marginal to the circulating pool and to squeeze out mature granulocytes from the bone marrow.[32] These differences may offer some potential advantages; however, no clinical data in patients with agranulocytopenia support these contentions. Because of the limited available data, and the absence of appropriate comparisons between G-SCF and GM-CSF in patients with drug-induced agranulocytosis, no specific recommendations on which
antithyroid drug-induced neutropenia are presented in figure 1. Extrapolation of these guidelines to patients with neutropenia induced by other drugs remains to be investigated. This approach may be helpful to identify patients with a good prognosis and may preclude the need for a bone marrow examination and inpatient management. If a trial of a single dose of G-CSF is considered, the assessment period could be extended to 24 hours and/or the dosage of G-CSF increased to 300μg. Treatment decisions regarding the use of HGF in drug-induced agranulocytosis are based on a careful analysis of prognostic factors, and a caseby-case approach is used because of limited clinical data. Elderly patients, patients with bone marrow aplasia or severe peripheral neutropenia, and febrile neutropenic patients with poor prognostic factors (i.e. renal failure, hypotension, sepsis syndrome),[110] are expected to benefit from HGF treatment. The underlying clinical condition should also be considered. For example, in a psychiatric patient who might experience decompensation and severe psychosis after clozapine withdrawal, a significant reduction in the length of severe neutropenia and hospitalisation may allow a more rapid psychiatric management.[92,93] A recommended dosage of G-CSF or GM-CSF is 5 μg/kg/day in patients with drug-induced granulocytopenia,[41] but other authors recommend
Antithyroid-induced granulocytopenia
Granulocyte count < 0.1 G/I
Granulocyte count > 0.1 G/I
G-CSF 75μg single dose injectiona
Granulocyte count 4h aftera
< 1 G/I
Hospitalisation and therapy with G-CSF
> 1 G/I
No hospitalisation and no additional treatment
Fig. 1. Proposed guidelines for the management of antithyroid drug-induced neutropenia.[109] a The assessment period could be extended to 24 hours and/or the dose of G-CSF increased to 300μg.
© Adis International Limited. All rights reserved.
BioDrugs 1999 Mar; 11 (3)
HGF for Drug-Induced Agranulocytosis
HGF should be used in this setting can be proposed at the moment. As the clinical utility of HGF in the treatment of drug-induced agranulocytosis is as yet unproven, and because the cost of these agents is high, restriction of their use to selected cases (elderly patients, patients with severe peripheral neutropenia, and febrile neutropenic patients with poor prognostic factors such as renal failure, hypotension, sepsis syndrome) has been proposed.[108,110] The cost-effectiveness of HGF was estimated by comparing 2 patients with druginduced agranulocytosis who were treated with routine care and 3 who were treated with GCSF.[44] G-CSF was less costly than supportive care because it reduced the duration of neutropenia and the length of hospital stay. According to these results and by simple decision analysis, the authors concluded that HGF treatment is justified on the basis of cost if 4 days of treatment shortens the hospital stay by 1 day. In addition, with regard to the low number of annual cases of drug-induced agranulocytosis and the usually short duration of HGF-treatment for this condition, the financial consequences are probably minimal and do not represent a real limitation to treatment. The validity of this viewpoint may, however, vary from country to country depending on fiscal constraints. In conclusion, HGF are frequently used in patients with drug-induced agranulocytosis, but their efficacy has not yet been studied appropriately in this setting. Data from case reports or short series of patients suggest that the early use of HGF in patients with severe peripheral neutropenia or patients with poor prognostic factors (i.e. renal failure, hypotension, sepsis syndrome) might be beneficial. However, the potential for HGF-induced adverse effects, particularly ARDS or GM-CSF– induced fever, should also be taken into consideration. References 1. Schultz W. Über eigenartige Halserkrankungen. Dtsch Med Wochenschr 1922; 48: 1965-50 2. Standardization of definitions and criteria of causality assessment of adverse drug reactions: drug-induced cytopenia [editorial]. Int J Clin Pharmacol Ther Toxicol 1991; 29 (2): 75-81
© Adis International Limited. All rights reserved.
197
3. Wiholm BE, Emanuelsson S. Drug-related blood dyscrasias in a Swedish reporting system, 1985-1994. Eur J Haematol 1996; 57 Suppl.: 42-6 4. Heimpel H. When should the clinician suspect a drug-induced blood dyscrasia, and how should he proceed. Eur J Haematol 1996; 57 Suppl.: 11-5 5. Kaufman DW, Kelly JP, Levy M, et al. The drug etiology of agranulocytosis and aplastic anaemia: monographs in epidemiology and biostatistics. No 18. New York: Oxford University Press, 1991 6. Heimpel H. Drug-induced agranulocytosis. Med Toxicol 1988; 3: 449-62 7. Kaufman DW, Kelly JP, Jurgelon JM, et al. Drugs in the aetiology of agranulocytosis and aplastic anaemia. Eur J Haematol 1996; 57 Suppl.: 23-30 8. Haas WK, Easton JD, Adams HP, et al. A randomized trial comparing ticlopidine hydrochloride with aspirin for the prevention of stroke in high-risk patients. N Engl J Med 1989; 321: 501-7 9. Mendelowitz AJ, Gerson SL, Alvir JMJ, et al. Clozapine-induced agranulocytosis; risk factors, monitoring and management. CNS Drugs 1995; 4 (6): 412-21 10. Bertolet BD, White BG, Pepine CJ. Neutropenia occurring during treatment with vesnarinone (OPC-8212). Am J Cardiol 1994; 74: 968-70 11. Furusawa S, Ohashi Y, Asanoi H. Vesnarinone-induced granulocytopenia: incidence in Japan and recommendations for safety. J Clin Pharmacol 1996; 36 (5): 477-81 12. D’Antonio D, Iacone A, Fioritoni G, et al. Patterns of infections in 41 patients with idiosyncratic drug-induced agranulocytosis. Ann Hematol 1991; 63: 84-8 13. Julia A, Olona M, Bueno J, et al. Drug-induced agranulocytosis: prognostic factors in a series of 168 episodes. Br J Haematol 1991; 79: 366-71 14. Abt C, Dietrich M, Niethammer D, et al. Akute Arzneimittel induzierte Agranulocytosis. Dtsch Med Wochenschr 1978; 103: 108-13 15. Jick H, Myers MW, Dean AD. The risk of sulfasalazine- and mesalazine- associated blood disorders. Pharmacotherapy 1995; 15 (2): 176-81 16. Vial T, Pofilet C, Pham E, et al. Agranulocytoses aïgues médicamenteuses: expérience du Centre Régional de Pharmacovigilance de Lyon sur 7 ans. Thérapie 1996; 51 (5): 508-15 17. Salama A, Schutz B, Kiefel V, et al. Immune mediated agranulocytosis related to drugs and their metabolites: mode of sensitization and heterogeneity of antibodies. Br J Haematol 1989; 72: 127-32 18. Irvine AE, French A, Daly A, et al. Drug-induced neutropenia due to direct effects on CFU-C: ten years of culture experience. Eur J Haematol 1994; 52: 21-7 19. Parent-Massin DM, Sensébé L, Léglise MC, et al. Relevance of in vitro studies of drug-induced agranulocytosis: report of 14 cases. Drug Saf 1993; 9 (6): 463-9 20. Uetrecht JP, Spielberg SP. Mechanisms of neutropenia: the role of neutrophil-specific pathways of drug metabolism. Pharmacoepidemiol Drug Saf 1993; 2 Suppl.: 37-42 21. Neftel KA, Hauser SP, Müller MR. Inhibition of granulopoiesis in vivo and in vitro by betalactam antibiotics. J Infect Dis 1985; 1: 90-8
BioDrugs 1999 Mar; 11 (3)
198
22. Rouveix B, Lassoued K, Vittecoq D, et al. Neutropenia due to β-lactamine antibodies. Lancet 1983; 287: 1832-4 23. Bodensteiner DC, Doolittle GC. Adverse haematological complications of anticancer drugs: clinical presentation, management and avoidance. Drug Saf 1993; 8 (3): 213-24 24. Tamai H, Mukuta T, Matsubayashi S, et al. Treatment of methimazole-induced agranulocytosis using recombinant human granulocyte colony-stimulating factor (rhG-CSF). J Clin Endocrinol Metab 1993; 77 (5): 1356-60 25. Patton WN, Duffull SB. Idiosyncratic drug-induced haematological abnormalities: incidence, pathogenesis, management and avoidance. Drug Saf 1994; 11 (6): 445-62 26. Lieschke GJ, Burgess AW. Granulocyte colony-stimulating factor and granulocyte-macrophage colony-stimulating factor (first of two parts). N Engl J Med 1992; 327: 28-35 27. Delannoy A, Géhenot M. Colony-stimulating factor and druginduced agranulocytosis. Ann Intern Med 1989; 110 (11): 942-3 28. Heinrich B, Gross M, Goebel FD. Methimazole-induced agranulocytosis and granulocyte-colony stimulating factor. Ann Intern Med 1989; 111 (7): 621-2 29. Muroi K, Ito M, Sasaki R, et al. Treatment of drug-induced agranulocytosis with granulocyte-colony stimulating factor [letter]. Lancet 1989; II: 55 30. Groopman JE, Feder D. Hematopoietic growth factors in AIDS. Semin Oncol 1992; 19: 408-14 31. Fukuta S, Murakami Y, Kuma K, et al. G-CSF levels during spontaneous recovery from drug-induced agranulocytosis [letter]. Lancet 1993; 342: 1495 32. Tajiri J, Noguchi S, Okamura S, et al. Granulocyte colony-stimulating factor treatment of antithyroid drug-induced granulocytopenia. Arch Intern Med 1993; 153: 509-14 33. Demuynck H, Zachée P, Verhoef GEG, et al. Risks of rhG-CSF treatment in drug-induced agranulocytosis. Ann Hematol 1995; 70: 143-7 34. Delannoy A. GM-CSF therapy for drug-induced agranulocytosis. J Intern Med 1992; 231: 269-71 35. Pasquale D, Newton M, Goss JB, et al. Granulocyte colonystimulating factor treatment of clozapine-induced agranulocytosis. Am J Psychiatry 1996; 153 (11): 1503-4 36. Yokoyama K, Sato T, Nakajima S, et al. Successful treatment of methimazole-induced agranulocytosis by granulocyte colony-stimulating factor. Am J Hematol 1992; 40: 76-7 37. Miyasaka Y, Yoshimura M, Tabata S, et al. Successful treatment of a patient with Graves’disease on hemodialysis complicated by antithyroid drug-induced granulocytopenia and angina pectoris. Thyroid 1997; 7 (4): 621-4 38. Sperner-Unterweger B, Czeipek I, Gaggl S, et al. Treatment of severe clozapine-induced neutropenia with granulocyte colony-stimulating factor (G-CSF): remission despite continuous treatment with clozapine. Br J Psychiatry 1998, 172: 82-4 39. Hunault M, Dombret H, Gardin C, et al. Hematopoietic growth factors in drug-induced agranulocytosis. Leukemia 1995; 9: 1286-7 40. Hurtado MR, Candelaria MH, Majluf-Cruz A, et al. Drug-induced agranulocytosis treated with granulocyte-macrophage colony-stimulating factor. Rev Invest Clin 1994; 46 (1): 59-61 41. Teitelbaum AH, Bell AJ, Brown SL. Filgrastim (r-metHuGCSF) reversal of drug-induced agranulocytosis. Am J Med 1993; 95: 245-6
© Adis International Limited. All rights reserved.
Vial et al.
42. Hunt KA, Resnick MP. Clomipramine-induced agranulocytosis and its treatment with G-CSF. Am J Psychiatry 1993; 150 (3): 522-3 43. Sprikkelman A, De Wolf JTM, Vellenga E. The application of hematopoietic growth factors in drug-induced agranulocytosis: a review of 70 cases. Leukemia 1994; 8 (12): 2031-6 44. Mani S, Barry M, Concato J. Granulocyte-colony stimulating factor therapy in drug-induced agranulocytosis. Arch Intern Med 1993; 153: 2500-1 45. Patton WN, Holyoake TL, Yates JM, et al. Accelerated recovery from drug-induced agranulocytosis following G-CSF therapy. Br J Haematol 1992; 80: 564-5 46. Kendra JR, Rugman FP, Flaherty TA, et al. First use of G-CSF in chlopromazine-induced agranulocytosis: a report of two cases. Postgrad Med J 1993; 69: 885-7 47. Means RT, Sandidge DR, Rankin KM, et al. Treatment of phenothiazine-induced agranulocytosis with recombinant granulocyte colony-stimulating factor [letter]. Am J Hematol 1992; 41: 296 48. Weiss M, Gross-Weege W, Wernet P. Rapid neutrophil recovery from acquired agranulocytosis by recombinant human granulocyte-macrophage colony-stimulating factor in an intensive care patient. Crit Care Med 1992; 20: 1490-1 49. Bauduer F, Mignard D, Ducout L, et al. GM-CSF et agranulocytoses médicamenteuses. Sem Hop 1996; 72 (11-12): 370-2 50. Ellman MH, Telfer MC, Turner AF. Benefit of G-CSF for methotrexate-induced neutropenia in rheumatoid arthritis. Am J Med 1992; 92: 337-8 51. Chia HM, Kalra L, Lakhani AK, et al. Filgrastim for lowdose, captopril-induced agranulocytosis [letter]. Lancet 1993; 342: 304 52. Wallach FR, Zirn J, Murray HW, et al. Use of granulocyte-macrophage colony-stimulating factor for treatment of drug-induced agranulocytosis [letter]. Rev Infect Dis 1991; 13: 523 53. Miyagawa S, Shiomi Y, Fukumoto T, et al. Recombinant granulocyte colony-stimulating factor for dapsone-induced agranulocytosis in leukocytoclastic vasculitis. J Am Acad Dermatol 1993; 28 (4): 659-61 54. Martino R, Martinez E, Sureda A, et al. A favorable response to the granulocyte colony-stimulating factor in a patient with agranulocytosis associated with the use of intravenous immunoglobulin. Med Clin 1993; 100: 358-9 55. Gora Tybor J, Krykowski E, Robak T. Treatment of drug-induced agranulocytosis with colony-stimulating factors (GCSF or GM-CSF). Arch Immunol Ther Exp 1996; 44 (4): 255-8 56. Nand S, Bayer R, Prinz RA, et al. Granulocyte-macrophage colony stimulating factor for the treatment of drug induced agranulocytosis. Am J Hematol 1991; 37: 267-9 57. Cappellari G, Gaio A, Cavallaro A, et al. Use of colony stimulating factors for the treatment of drug-induced agranulocytosis. Br J Haematol 1993; 84: 183-4 58. Gales BJ, Gales MA. Granulocyte colony-stimulating factor for sulfasalazine-induced agranulocytosis. Ann Pharmacother 1993; 27 (9): 1052-4 59. Kuipers EJ, Vellenga E, de Wolf JTM, et al. Sulfasalazine induced agranulocytosis treated with granulocyte-macrophage colony stimulating factor. J Rheumatol 1992; 19 (4): 621-2
BioDrugs 1999 Mar; 11 (3)
HGF for Drug-Induced Agranulocytosis
60. Palmblad J, Jonson B, Kanerud L. Treatment of drug-induced agranulocytosis with recombinant GM-CSF. J Intern Med 1990; 228: 537-9 61. Roddie P, Dorrance H, Cook MK, et al. Treatment of sulphasalazine-induced agranulocytosis with granulocyte macrophage-colony stimulating factor. Aliment Pharmacol Ther 1995; 9: 711-2 62. Rospond RM, Glowacki RC, Mailliard JA. Sargramostim for sulfasalazine-induced agranulocytosis [letter]. Clin Pharm 1993; 12 (3): 179 63. Wyatt S, Joyner MV, Daneshmend TK. Filgrastim for mesalazine-associated neutropenia [letter]. Lancet 1993; 341: 1476 64. Barete S, Bissuel F, Longuet P, et al. Agranulocytosis caused by treatment with terbinafine. Rev Méd Interne 1997; 18 (3): 258-9 65. Kovacs MJ, Alshammari S, Guenther L, et al. Neutropenia and pancytopenia with oral terbinafine [letter]. J Am Acad Dermatol 1994; 31: 806 66. Ruiz-Irastorza G, Alonso JJ, Iglesias JJ, et al. Granulocyte colony-stimulating factor for neutropenia secondary to ticlopidine. Acta Haematol 1994; 91 (2): 106-7 67. Katz R, Chuang LC, Sutton JD. Use of granulocyte colonystimulating factor in the treatment of pancytopenia secondary to colchicine overdose. Ann Pharmacother 1992; 26 (9): 1087-8 68. Lopez-Karpovitch X, Ulloa-Aguirre A, von Eiff C, et al. Treatment of methimazole-induced severe aplastic anemia with recombinant human granulocyte-monocyte colony-stimulating factor and glucocorticosteroids. Acta Haematol 1992; 27: 148-50 69. Rawanduzy A, Sarkis A, Rovit RL. Severe phenytoin-induced bone marrow depression and agranulocytosis treated with human recombinant granulocyte-macrophage colony-stimulating factor. J Neurosurg 1993; 79: 121-4 70. Tsatalas C, Chalkia P, Garyfallos A, et al. Ticlopidine-induced aplastic anaemia: case report and review of the literature. Clin Drug Invest 1995; 9 (2): 127-30 71. De Jaureguiberry JP, Galzin M, Talard P, et al. Aplasie médullaire associée à la ticlopidine. Rev Méd Interne 1996; 17: 1032-6 72. Escobar-Morreale HF, Bravo P, Garcia-Robles R, et al. Methimazole-induced severe aplastic anemia: unsuccessful treatment with recombinant human granulocyte-monocyte macrophage colony stimulating factor. Thyroid 1997; 7 (1): 67-70 73. Kao TW, Hung CC, Chen YC, et al. Ticlopidine-induced aplastic anemia: report of three Chinese patients and review of the literature. Acta Haematol 1997; 98: 211-3 74. Khelif A, Assouline D, Ffrench M, et al. Ticlopidine and aplastic anemia. Br J Haematol 1993; 83: 678-9 75. Mallet L, Mallet J. Ticlopidine and fatal aplastic anemia in an elderly woman. Ann Pharmacother 1994; 28 (10): 1169-71 76. Shapiro CM, Walk D. Aplastic anemia associated with ticopidine. Neurology 1996; 47: 300 77. Delannoy A, Ravoet C. Hematopoietic growth factors in druginduced agranulocytosis. Leukemia 1995; 9: 1289-90 78. Feldman AM, Bristow MR, Parmley WW, et al. Effects of vesnarinone on morbidity and mortality in patients with heart failure. N Engl J Med 1993; 329: 149-55
© Adis International Limited. All rights reserved.
199
79. Alvir JMJ, Lieberman JA, Safferman AZ, et al. Clozapine-induced agranulocytosis. N Engl J Med 1993; 329: 162-7 80. Miller PR, Cutten AEC. Haematological side effects of clozapine: patient characteristics. NZ Med J 1997; 110: 125-7 81. Barnas C, Zwierzina H, Hummer M, et al. Granulocyte-macrophage colony-stimulating factor (GM-CSF) treatment of clozapine-induced agranulocytosis: a case report. J Clin Psychiatry 1992; 53 (7): 245-7 82. Bradford CR, Ong EL, Hendrick DJ, et al. Use of colony stimulating factors for the treatment of drug-induced agranulocytosis. Br J Haematol 1993; 84: 182-3 83. Geibig CB, Marks LW. Treatment of clozapine- and molindoneinduced agranulocytosis with granulocyte colony-stimulating factor. Ann Pharmacother 1993; 27 (10): 1190-2 84. Gruner U, Pesch S, Spittler S, et al. Treatment of clozapine-induced agranulocytosis with granulocyte colony-stimulating factor. Dtsch Med Wochenschr 1994; 119: 1467-70 85. Gullion G, Yeh HS. Treatment of clozapine-induced agranulocytosis with recombinant granulocyte colony-stimulating factor. J Clin Psychiatry 1994; 55 (9): 401-5 86. Lamberti JS, Bellnier TJ, Schwarzkopf SB, et al. Filgrastim treatment of three patients with clozapine-induced agranulocytosis. J Clin Psychiatry 1995; 56 (5): 256-9 87. Nielsen-H. Recombinant human granulocyte colony-stimulating factor (rhG-CSF; filgrastim) treatment of clozapine-induced agranulocytosis. J Intern Med 1993; 234: 529-31 88. Oren R, Granat E, Shtussberg S, et al. Clozapine-induced agranulocytosis treated with granulocyte macrophage colony stimulating factor. Br J Psychiatry 1993; 162: 686-7 89. Raison CL, Guze BH, Kissell-RL. Successful treatment of clozapine-induced agranulocytosis with granulocyte colony-stimulating factor. J Clin Psychopharmacol 1994; 14 (4): 285-6 90. Valevski A, Modai I, Lahav M, et al. Clozapine-lithium combined treatment and agranulocytosis. Int Clin Psychopharmacol 1993; 8: 63-5 91. Weide R, Köppler H, Heymanns J, et al. Successful treatment of clozapine induced agranulocytosis with granulocyte colony-stimulating factor (G-CSF). Br J Haematol 1992; 80: 557-9 92. Gerson SL, Gullion G, Yeh HS, et al. Granulocyte colony-stimulating factor for clozapine-induced agranulocytosis [letter]. Lancet 1992; 340: 1097 93. Chengappa KNR, Gopalani A, Haught MK, et al. The treatment of clozapine-associated agranulocytosis with granulocyte colony-stimulating factor (G-CSF). Psychopharmacol Bull 1996; 32 (1): 111-21 94. Bartalena L, Bogazzi F, Martino E. Adverse effects of thyroid hormone preparations and antithyroid drugs. Drug Saf 1996; 15 (1): 53-63 95. Balkin MS, Buchholtz M, Ortiz J, et al. Propylthiouracil (PTU)induced agranulocytosis treated with recombinant human granulocyte colony-stimulating factor (G-CSF). Thyroid 1993; 3 (4): 305-9 96. Björkholm M, Pisa P, Arver S, et al. Haematologic effects of granulocyte-macrophage colony stimulating factor in a patient with thiamazole-induced agranulocytosis. J Intern Med 1992; 232: 443-5 97. Freebairn RC, Young RJ, Gomersall CD, et al. Successful treatment of carbimazole-induced agranulocytosis and severe sep-
BioDrugs 1999 Mar; 11 (3)
200
98.
99. 100.
101.
102.
103. 104.
105.
Vial et al.
sis with granulocyte macrophage colony stimulating factor. Anaesth Intensive Care 1995; 23 (4): 510-2 Genet P, Pulik M, Lionnet F, et al. Use of colony-stimulating factors for the treatment of carbimazole-induced agranulocytosis. Am J Hematol 1994; 47: 334-5 Gerl A, Gerhartz H, Wilmanns W. GM-CSF for agranulocytosis. J Intern Med 1991; 230: 90-1 Magner JA, Snyder DK. Methimazole-induced agranulocytosis treated with recombinant human granulocyte colony-stimulating factor (G-CSF). Thyroid 1994; 4 (3): 295-6 Somogyi A, Rosta A, Lang I, et al. Treatment of drug-induced bone marrow suppression with recombinant human granulocyte/monocyte colony stimulating factor. Adverse Drug React Toxicol Rev 1996; 15 (2): 119-24 Willfort A, Lorber C, Kapiotis S, et al. Treatment of drug-induced agranulocytosis with recombinant granulocyte colonystimulating factor (rh G-CSF). Ann Hematol 1993; 66: 241-4 Vial T, Descotes J. Clinical toxicity of cytokines used as haemopoietic growth factors. Drug Saf 1995; 13 (6): 371-406 Schilero GJ, Oropello J, Benjamin E. Impairment in gas exchange after granulocyte colony-stimulating factor (G-CSF) in a patient with the adult respiratory distress syndrom. Chest 1995; 107: 276-8 Devereux S, Linch DC, Campos Costa D, et al. Transient leucopenia induced by granulocyte-macrophage colony-stimulating factor. Lancet 1987; II: 1523-4
© Adis International Limited. All rights reserved.
106. Bregni M, Siena S, Di Nicola M, et al. Comparative effects of granulocyte-macrophage colony-stimulating factor and granulocyte colony-stimulating factor after high dose cyclophosphamide cancer therapy. J Clin Oncol 1996; 14 (2): 628-35 107. Cabalerro MD, Vazques L, Barragnan JM, et al. Randomised study of filgrastim versus molgramostim after peripheral stem cell transplant in breast cancer. Haematologica 1998; 83: 514-8 108. Gerson SL. G-CSF and the management of clozapine-induced agranulocytosis. J Clin Psychiatry 1994; 55 (9 Suppl. B): 139-42 109. Tajiri J, Noguchi S, Murakami N. Usefulness of granulocyte count measurement four hours after injection of granulocyte colony-stimulating factor for detecting recovery from antithyroid drug-induced granulocytopenia. Thyroid 1997; 7 (4): 575-8 110. American Society of Clinical Oncology. Recommendations for the use of hematopoietic colony-stimulating factors: evidenced-based, clinical practice guidelines. J Clin Oncol 1994; 12 (11): 2471-508
Correspondence and reprints: Dr Thierry Vial, Centre AntiPoison, Centre de Pharmacovigilance, Pavillon N, Hôpital Edouard Herriot, 69437 Lyon Cedex 03, France.
BioDrugs 1999 Mar; 11 (3)
