R E S E A R C H A R T I C L E
Nationwide survey on the use of horse antithymocyte globulins
(ATGAM) in patients with acquired aplastic anemia: A report
on behalf of the French Reference Center for Aplastic Anemia
Regis Peffault de Latour
1|
Reza Tabrizi
2|
Ambroise Marcais
3|
Thierry Leblanc
4|
Thierry Lamy
5|
Mohamad Mohty
6|
Suzanne Tavitian
7|
Charlotte Jubert
8|
Marlène Pasquet
9|
Claire Galambrun
10|
Stephanie Nguyen
11|
Jean Yves Cahn
12|
Thorsten Braun
13|
Eric Deconinck
14|
Jacques Olivier Bay
15|
Flore Sicre de Fontbrune
16|
Fiorenza Barraco
17|
Gerard Socie
181Service Hematologie Greffe, Centre de Reference Aplasies Medullaires Acquises et Constitutionnelles, Universite Paris Diderot, Sorbonne Paris Cite, H^opital
Saint-Louis, Assistance Publique H^opitaux de Paris, Paris, France;2
Service d’Hematologie Clinique, Centre de Competence Aplasies Medullaires Acquises et Constitutionnelles, H^opital Haut Lev^eque, Bordeaux, France;3
Service d’hematologie Clinique, H^opital Necker, Assistance Publique H^opitaux de Paris, Paris, France;
4Service Hematologie Pediatrique, Centre de Reference Aplasies Medullaires Acquises et Constitutionnelles, H^opital Robert Debre, Assistance Publique H^opitaux de
Paris, Paris, France;5
Service d’hematologie Clinique, H^opital Pontchaillou, France;6
Service d’Hematologie Clinique, H^opital Saint-Antoine, Assistance Publique H^opitaux de Paris, Paris, France;7
Service d’Hematologie, Institut Universitaire du Cancer de Toulouse Oncopole, Centre hospitalier universitaire de Toulouse, France;
8Service Hematologie Pediatrique, Centre de Competence Aplasies Medullaires Acquises et Constitutionnelles, H^opital universitaire, Bordeaux, France;9
Service d’Hematologie Pediatrique, Oncop^ole, Toulouse, France;10Service Hematologie Pediatrique, Centre de Competence Aplasies Medullaires Acquises et
Constitutionnelles, H^opital La Timone, Marseille, France;11
Service d’Hematologie Clinique, H^opital Pitie-Salpetrière, Assistance Publique H^opitaux de Paris, Paris, France;12
Service d’Hematologie Clinique, H^opital universitaire, Grenoble, France;13
Service d’Hematologie Clinique, H^opital Avicennes, Bobigny, Assistance Publique H^opitaux de Paris, Paris, France;14
Service d’Hematologie Clinique, H^opital universitaire, Besançon, France;15
Service d’Hematologie Clinique, H^opital Estaing, Clermont Ferrand, France;16Service Hematologie Greffe, Centre de Reference Aplasies Medullaires Acquises et Constitutionnelles, H^opital Saint-Louis, Assistance
Publique H^opitaux de Paris, Paris, France;17
Service d’Hematologie Clinique, H^opital universitaire, Lyon, France;18Service Hematologie Greffe, Centre de Reference
Aplasies Medullaires Acquises et Constitutionnelles, Universite Paris Diderot, Sorbonne Paris Cite, Inserm UMR 1160, H^opital Saint-Louis, Assistance Publique H^opitaux de Paris, Paris, France
Correspondence
Regis Peffault de Latour, Hematologie Greffe, Centre de Reference Aplasies Medullaires Acquises et Constitutionnelles, Universite Paris Diderot, Sorbonne Paris Cite; 1 Av. Claude Vellefaux, 75010 Paris, France.
Email: regis.peffaultdelatour@aphp.fr and
Gerard Socie, Hematologie Greffe, Centre de Reference Aplasies Medullaires Acquises et Constitutionnelles, Universite Paris Diderot, Sorbonne Paris Cite, INSERM UMR 1160; 1 Av. Claude Vellefaux, 75010 Paris, France.
Email: gerard.socie@aphp.fr
Funding information
Pfizer pharmaceuticals, Unrestricted grant for data acquisition
Abstract
Antithymocyte globulins (ATG) plus cyclosporine (CSA) is the gold standard immunosuppressive treatment (IST) for patients with aplastic anemia. A prospective randomized trial showed in 2011 that hATG was superior to rabbit ATG for first-line treatment of severe AA. The French Health Agency (ANSM) permitted a patient-named authorization for temporary use (ATU) program of hATG (ATGAM, Pfizer) in patients with AA in 2011 since commercial access to hATG is not approved. We took advantage of this program to analyze the outcomes of 465 patients who received antithymocyte globulins (ATGAM) plus CSA as first line treatment (n5 379; 81.5%), or for refractory (n5 26) or relapsed disease (n 5 33), from September 2011 to March 2017. In the entire cohort one year, 72% of the patients had partial and 13% had complete response, with worse response for patients with severe AA and a longer interval between diagnosis and IST (more than 6 months). Severe adverse events were mainly linked to infections (24%), hemorrhages (6%), and elevated liver function tests (5%). Overall at 12 months, 9.7% of patients required second line IST and 15.6% received transplantation. Fifty-five patients died during the study mainly because of infections (53%). Factors predicting independently worse survival were age over 40 years, neutro-phils less than 0.53 109/L, male gender and longer delay between diagnosis and hATG (>6
months period). This study does illustrate the results of ATGAM with CSA in a true-life perspective and confirms ATGAM as standard of care IST to treat patients with AA not eligible for HSCT.
Am J Hematol. 2018;93:635–642. wileyonlinelibrary.com/journal/ajh VC2018 Wiley Periodicals, Inc.
|
635A
J
H
1
|I N T R O D U C T I O N
Acquired aplastic anemia (AA), a rare disease (2 per million in Europe) associated with marrow failure, can be treated either by immunosup-pressive therapy (IST) or by allogeneic hematopoietic stem cell transplan-tation (HSCT). While HSCT is the preferred treatment in younger patients (<40 years) with available HLA-identical sibling donor HSCT, IST is the gold standard approach for the others.1–3Since the seminal German randomized trial, the reference IST includes a combination of antithymocyte globulin (ATG) with cyclosporine (CSA).4,5For patients with nonsevere AA (NSAA) who are transfusion dependent, the combi-nation of ATG and CSA is superior to CSA alone, with a higher response rate, higher blood counts, and improved disease-free survival.6 For severe AA (SAA), the combination of ATG and CSA results in a response rate of roughly 60% to 70% of patients, as first line therapy.1,3,7–9 Histor-ically, horse ATG (hATG) has been used as first line IST for AA, while rab-bit ATG (rATG) tended to be used for relapsed AA, at least in Europe.1,3
Lymphoglobuline (horse ATG, Genzyme) was withdrawn from most European markets in 2007. Since then, only 2 countries (Slovenia and Latvia) still have commercial access to hATG (ATGAM, Pfizer), and 19 are still awaiting regulatory approval. A prospective randomized clinical trial by NIH published in 2011 demonstrated that the response rates and sur-vival with hATG (ATGAM) was significantly better than rabbit ATG.10
Rabbit ATG (thymoglobulin, Genzyme) in combination with CSA, as first-line IST, was evaluated prospectively in a multicenter, European, phase 2 pilot study, in 35 patients with aplastic anemia and compared with histor-ical controls who received hATG. This later study also showed lower response and overall survival using rabbit ATG.8Other retrospective studies also indicate a significantly worse response rate and survival for AA patients treated with rATG in Europe and United States.9,11–18
In light of this evidence, European hematologists petitioned with their country health authorities to develop a procedure to make ATGAM available to certain patients. This resulted in the establishment of a program referred to as the patient-named authorization for tempo-rary use (ATU) program that allows certain patients to be treated with ATGAM in France, Italy and UK. In France the ATU program was estab-lished in September 2011 and allows for compassionate use of ATGAM for AA patients while the marketing authorization is pending with the regulatory authorities. We took advantage of this ATU program to per-form a Nation-wide survey of the use of ATGAM, and to analyze the outcome of a large cohort of patients with AA on behalf of the French national reference center for aplastic anemia.
2
|M E T H O D S
2.1
|Patient-named authorization for temporary use
program
The patient-named ATU program is aimed at making a medicinal prod-uct exceptionally available for a patient before marketing authorization, when the following conditions are met: (1) serious or rare disease with no appropriate treatment available on the market, (2) absence of ongoing biomedical research, (3) efficacy and safety are strongly
presumed considering the results of clinical data, and (4) the medicinal product is likely to constitute a real benefit for patients. In France, the French Health Agency (ANSM) approved a patient-named ATU pro-gram of hATG (ATGAM, Pfizer) in patients with AA in 2011.
2.2
|Cases identification
For the purpose of the study, a questionnaire was designed on behalf of the French reference center for aplastic anemia (national rare disease net-work) to collect informations from treating centers on patient-, and disease characteristics, and on outcomes. We identify 635 patients who prospec-tively registered in the hATG (ATGAM) ATU program from September 2011 to March 2017. We systematically screened files of 537 patients who were addressed to the French reference center for AA. The analysis was eventually done in 465 patients (86.6%) after checking indication for treatment with ATGAM for AA (i.e., excluding patients who received ATGAM for another indication). The study was conducted according to the Helsinki declaration. IRB of the AA French reference center approved the study and anonymous data collection was declared to the competent authorities. According to the French Law (MR-003), written informed con-sent was not required for this retrospective noninterventional study as patients provided a nonopposition statement at time of treatment.
2.3
|Procedures
Minimal diagnostic panel of aplastic anemia required to enter the study were a complete blood count, a bone marrow biopsy, and a bone mar-row karyotype analysis. The ATU program is associated with a protocol for therapeutic use. ATGAM was homogeneously administered in hos-pitalized patients at a dose of 40 mg/kg/day for 4 days. CSA was administered with the dose adjusted to maintain trough blood levels of 200 to 400 ng per milliliter for one year and then decrease to stop before end of year 2. Antimicrobial prophylaxis was heterogeneous among the different centers according to local policy. Blood products transfusions were provided according to national guidelines.
2.4
|Definitions
Patients were classified before ATGAM treatment as having very-severe, severe or nonsevere disease according to usual criteria.15,19 Paroxysmal nocturnal hemoglobinuria clone were considered positive in flow cytometry if more than 0.1% of neutrophils were deficient of glycosylphosphatidylinositol-anchored proteins on the cell surface.
Criteria for response to treatment follow the British Committee for Standards in Haematology Guidelines.15,20For severe AA; no response
means still severe; Partial response requires transfusion independence and that patient no longer meeting criteria for severe disease; Com-plete hemoglobin normal for age, neutrophil count>1.5 109/l, and pla-telet count>150 109/l. Response criteria for nonsevere AA aplastic
anemia; No response means; worse or not meeting criteria for response that include transfusion independence (if previously dependent), or doubling or normalization of at least one cell line or increase of baseline hemoglobin, or increase of baseline neutrophils or increase of baseline. Complete response uses same criteria as for severe disease.
Treatment groups were classified as follows. Patients were con-sidered first line treatment if (1) they never received horse or rabbit ATG (374/465; (80.4%); or (2) if no information about previous treat-ment with either horse or rabbit ATG were available but with a cur-rent treatment start with ATGAM within 45 days after date of diagnosis (5/465 (1.1%). Patients were considered refractory if they failed to respond to previous treatment with horse ATG, rabbit ATG or Eltrombopag within 12 months prior to treatment with ATGAM (regardless of AA severity) (26/465 (5.6%). Patients were considered in relapse if they failed to respond to previous treatment with either horse or rabbit ATG more than one year before current treatment with ATGAM (regardless of the severity of the AA) 33/465 (7.1%). We did not have enough data for some of the patients to allow classi-fication 27/465 (5.8%).
2.5
|Statistical analysis
Descriptive statistics were reported as mean, standard deviation, median, and range for continuous variables and frequencies and pro-portions for categorical variables.
The Kaplan-Meier method was used to estimate overall survival rates, and a 2-sided log-rank test was used to test the equality of
survival functions in different subgroups. Survival data were censored as of the date of last contact.
Probabilities of response and of new therapies were calculated by using the cumulative incidence estimator to accommodate competing risks. For response death, new IST treatment, or HSCT were competing risks. Comparisons between cumulative incidence curves were per-formed using the Gray test.
Multivariate analyses were performed to assess independent pre-dictors of response at 12 months. Backward stepwise logistic regres-sion methods included variables that demonstrated significant associations (P< .10) in the univariate models. Similarly, Cox propor-tional hazards regression model was used to assess predictors of the overall survival. Two-sided P< .05 was considered statistically signifi-cant. Statistical analyses were performed with the SAS 9.4 (SAS Insti-tute, Cary, North Carolina).
3
|R E S U L T S
3.1
|Patient and disease characteristics
From September 2011 to March 2017, 465 patients received ATGAM as first line treatment (n5 379; 81.5%), or for refractory
T A B L E 1 Patient and disease characteristics
First line treatment Moderate (n 5 140) Severe (n 5 133) Very severe (n 5 84) Refractory (n 5 26) Relapse (n 5 33) Follow-up (months) Median 20.2 15.7 11.9 9.2 12.9 Q1-Q3 8.9; 36.4 6.5; 35.6 5.5; 22.6 4.5; 15.3 11.1; 40.2 Age (year)a Mean6 SD 38.26 21.1 41.26 23.9 34.36 25.3 42.36 20.4 43.16 21.0 Age< 18 yearsa(n; %) 31 (22.5%) 36 (27.5%) 35 (41.7%) 2 (8.0%) 4 (12.1%) Genderb Male, n (%) 71 (51.1%) 70 (53.8%) 48 (57.1%) 13 (50.0%) 18 (54.5%) Female, n (%) 68 (48.9%) 60 (46.2%) 36 (42.9%) 13 (50.0%) 15 (45.5%) PNH clonec 45 (46.4%) 22 (31.0%) 7 (15.2%) 6 (40.0%) 8 (36.4%) Karyotyped Abnormal, n (%) 5 (4.5%) 4 (4.2%) 4 (7.3%) 3 (15.8%) 1 (3.4%) Etiology; n (%) Post hepatitis, 3 (2.2%) 10 (7.6%) 7 (8.4%) 3 (12.0%) Toxic 2 (2.4%) 1 (4.0%) PNH/AA 26 (19.0%) 15 (11.5%) 5 (6.0%) 3 (12.0%) 3 (9.1%) Idiopathic 108 (78.8%) 106 (80.9%) 69 (83.1%) 18 (72.0%) 30 (90.9%)
Abbreviations: AA: Aplastic Anemia; PNH: Paroxysmal Nocturnal Hemoglobinuria.
aMissing data, n5 5. bMissing data, n5 4. cMissing data, n5 180. dMissing data, n5 115.
(n5 26) or relapsed disease (n 5 33) and were included in this sur-vey. Median age was 39.4 years and 235 (53.8%) patients were male. The median follow-up was 13.5 months with 60% of the patients being followed for a minimum of one year. Most patients were adult but 25.9% were children (less than 18 years). Baseline patient, disease- and treatment- characteristics of the 438 patients who have enough data to be classified are summarized in Table 1. Nearly 60% of patients had severe or very severe AA at the time of ATGAM treatment. CSA was given in addition to ATGAM in 98% of the cases. G-CSF was delivered in addition to ATGAM1 CSA in 5 patients (all but one as part of first line treatment) and Eltrombo-pag in 9 other patients (all refractory). Median neutrophil count was 0.48 109/L, and that of reticulocytes was 20 109/L; 95% of the patients required red blood cell and/or platelets transfusions at time of treatment.
3.2
|Response to hATG
Overall, at one year 72% (144/200 evaluable patients) of the patients had partial and 13% (26/200 evaluable patients) had com-plete response as defined by Camitta et al. in 2000.20 Response
rates at 12 months were similar in adults (71.6% partial and 12.1% complete) and in children (73.2% partial and 16.1% complete). Response rate at 12 months in children in patients receiving
ATGAM1 CSA as first line treatment (n 5 106) was also in the same range (73.1% partial and 17.3% complete). The detailed response to IST is summarized in Table 2. Using cumulative Incidence estimates (CumI), we were able to take into account competing risk of death, or new treatment including HSCT. The CumI of achieving a response at 12 month was 65.3% (95% CI; 60.2–70.0%) as illustrated in Sup-porting Information Figure S1A. The CumI of response was lower in patients with refractory disease [42.4%; 95% Confidence Interval (20.6–62.7%)], but not statistically significant (Gray’s test 5 0.089) (Supporting Information Figure S1B). However, using Fine and Gray’s model, the hazard ratio for refractory vs. first line treated patients was almost significant [Hazard ratio 0.5; 95% Confidence Interval (0.25-1.00); P5 .051]. The CumI of response for children treated with ATGAM first line was similar to the overall population [59.3%; 95% Confidence Interval (47.9–69.0%)]. The CumI of achieving a response at 12 month for first line treatment is also shown in Sup-porting Information Figure S1C.
Finally, factors predictive for response at 12 months were ana-lyzed. Disease severity (severe versus non severe) and time interval between diagnosis to IST with ATGAM plus CSA (delay between diag-nosis and IST superior to 180 months) emerged as significant risk fac-tor, by multivariate analysis (Supporting Information Table S1). When we restricted this analysis to the patients who were treated in first line (relapse and refractory patients were excluded), the only factor
T A B L E 2 Response to horse ATG
First line treatment Moderate (n 5 140) Severe (n 5 133) Very severe (n 5 84) Refractory (n 5 26) Relapse (n 5 33) Month 3a None, n (%) 18 (16.7%) 75 (74.3%) 47 (83.9%) 10 (62.5%) 6 (27.3%) Partial, n (%) 87 (80.6%) 26 (25.7%) 9 (16.1%) 4 (25.0%) 16 (72.7%) Complete, n (%) 3 (2.8%) 2 (12.5%) Month 6b None, n (%) 17 (15.7%) 42 (47.7%) 33 (61.1%) 6 (42.9%) 7 (31.8%) Partial, n (%) 82 (75.9%) 43 (48.9%) 20 (37.0%) 7 (50.0%) 15 (68.2%) Complete, n (%) 9 (8.3%) 3 (3.4%) 1 (1.9%) 1 (7.1%) Month 12c None, n (%) 5 (6.9%) 10 (15.2%) 5 (15.2%) 5 (62.5%) 4 (21.1%) Partial, n (%) 56 (77.8%) 50 (75.8%) 21 (63.6%) 3 (37.5%) 13 (68.4%) Complete, n (%) 11 (15.3%) 6 (9.1%) 7 (21.2%) 2 (10.5%) Month 12—LOCFd None, n (%) 22 (16.7%) 47 (41.6%) 35 (51.5%) 10 (55.6%) 9 (34.6%) Partial, n (%) 96 (72.7%) 60 (53.1%) 26 (38.2%) 7 (38.9%) 15 (57.7%) Complete, n (%) 14 (10.6%) 6 (5.3%) 7 (10.3%) 1 (5.6%) 2 (7.7%)
Abbreviation: LOCF5 last observation carried forward.
aMissing data, n5 100. bMissing data, n5 84. cMissing data, n5 79. dMissing data, n5 46.
associated with better response was age less than 40 years [HR 0.35; 95% CI (0.13-0.96); P5 .042, n 5 169].
3.3
|Second line treatment after hATG
We also estimated the CumI of second line treatment with the associa-tion hATG plus CSA (i.e., first treatment failure), which was 9.7% at 12 months [95% CI (7.0–12.9%)] and 12.7% at 24 months [95% CI (9.4– 16.4%)]. The CumI according to treatment group is illustrated in Sup-porting Information Figure S2. The chance to have a new IST after a first hATG was higher for patients treated for relapse (Supporting Infor-mation Figure S2A). The CumI of proceeding to an HSCT was 15.6% at 12 months [95% CI (12.1–19.4%)] and 17.5% at 24 months [95% CI (13.7–21.6%)]. The CumI of HSCT according to treatment group is illus-trated in Supporting Information Figure S2B. As expected it was higher in patients with refractory disease. Using Fine and Gray’s model, none of the hazard of HSCT between groups reached significance (not shown). The CumI of second line treatment for patients who were treated in first line only is also shown in Supporting Information Figure S2C (time to new IST) and in Supporting Information Figure S2D (time to transplantation).
3.4
|Overall survival
Overall survival (OS) and survival by treatment group were also esti-mated (Figure 1A,B, respectively). The 12- and 24-months Kaplan-Meier estimates of OS were 90.2% [95% CI (86.8–92.7%)] and 86.4% [95% CI (82.3–89.7%)]. The estimates by treatment group did not differ significantly. OS by age group is illustrated in Figure 1C. Overall sur-vival is significantly better for patients younger than 15 years when compared with patients aged older than 40. Worst overall survival was observed for patients older than 40. Moreover, in the group of 379 patients who received ATGAM plus CSA as first line treatment, OS was impacted by the severity of the disease (significant worst outcome for patients with very severe disease, Figure 1D). When we analyzed fac-tors predicting overall survival, age over 40 years, neutrophils less than 0.5 3 109/L, male gender and longer delay between diagnosis and
treatment predicted independently for worse outcome (Supporting Information Table S2). In the group of patients who were treated first line (moderate, severe, and very severe aplastic anemia), factors associ-ated with worse overall survival were age of 40 years or more [HR 3.4; 95% CI (1.6–7); P 5 .001, n 5 353] and disease severity (severe aplastic anemia, [HR 5; 95% CI (2.2–11.6); P < .001, n 5 353].
F I G U R E 1 Overall survival after hATG plus CSA. (A) Time to death—Kaplan–Meier estimates; population: all treated patients (n 5 465). (B) Time to death according to treatment group—Kaplan–Meier estimates; P value 5 . 224 (n 5 465). (C) Time to death according to age groups —Kaplan–Meier estimates; P value 5 .003. (D) Time to death according to baseline disease severity—Kaplan–Meier Estimates; Population: first line treatment; P value< .001 [Color figure can be viewed at wileyonlinelibrary.com]
3.5
|Side effects and cause of death
As expected in this patient population with AA, severe adverse events (SAE) were mainly linked to infections (24%), hemorrhages (6%) and elevated liver function tests (5%). Moreover, SAE also included 6 cases of acute myeloid leukemia (4 in patients treated as first line and 2 in refractory patient), and 8 cases of myelodysplastic syndrome (7 in patients treated as first line and 1 in refractory patient); of note 3 patients already died of clonal evolution (2 acute leukemia and 1 mye-lodysplastic syndrome). During the study, 55 patients died (11.8%) with for more than half of the patients infections identified as the main rea-son. Causes of death are detailed in Table 3.
4
|D I S C U S S I O N
IST using the combination of ATG plus CSA is associated with response rates between 60% and 80% with current 5-year survival rates of around 75–85%.1,15,21,22 With possibly the exception of Eltrombo-pag,23any attempt to add or substitute any other IST agent (like myco-phenolate or sirolimus) to this gold standard drug combination has been a failure in phase II/III during the past 15 years.3While horse ATG (ATGAM) has been demonstrated to be significantly better to rab-bit ATG in the setting of a randomized clinical trial24with similar results seen in a prospective case-control European study (historical controls patients treated with hATG),8 this finding were subject to controversies.9,11–18 In a recent meta-analysis, Hayakawa et al.10
searched online databases for studies that compared the two ATGs brand for AA, including both randomized and nonrandomized con-trolled trials. Thirteen studies were included in this analysis. The risk ratio (RR) of early mortality for rabbit vs. horse ATG was 1.33, with sig-nificant heterogeneity. A sensitivity analysis demonstrated higher early mortality rate in patients who received rATG. The overall response rate
was significantly higher in patients who received hATG with a signifi-cant risk ratio of 1.27. Thus, 6 year after the publication of the random-ized NIH trial, one can safely state that hATG must be the reference first line IST for patients with AA (in association with CSA). However, as described in the introduction, the access to ATGAM for this orphan disease remains a major problem in the majority of European countries. Moreover, real life experience on large cohort of patients treated homogeneously is missing. We took advantage of the patient-named ATU program of hATG (ATGAM, Pfizer) in patients with AA in France to assess the results of ATGAM1 CSA in a real life perspective in a large unbiased cohort of 465 patients with AA.
Over 80% of the patients responded to IST with ATGAM1 CSA in our series, with most responses being partial. These results are similar to those reported in various phase II/III trials published by the NIH and with the case control study reported by the EBMT, for first line treat-ment (reviewed in Refs. 3,15). Only few patients in our cohort received the association of ATGAM plus CSA because of relapse/refractory dis-ease. In 2014, Scheinberg et al.25,26investigated standard ATGAM plus CSA in patients who were refractory to initial rATG/CSA. Of the 19 patients who received rATG as initial therapy, 4 (21%) achieved a hematologic response by 3 months. The overall survival for the cohort at 3 years was 68%. The authors suggested that only a minority of patients can be successfully salvaged after receiving as first therapy with rATG. Our own results also found poor response rate for salvage hATG in patients with refractory disease but quite good results in patients with relapse.
Somewhat surprisingly, overall response rates are almost univer-sally reported in the literature as crude estimates. However, this does not take into account competing risk of death, or new treatment including HSCT. For the best of our knowledge, we reported here for the first time on a large cohort of patients the CumI of response at 12 months (65.3%) overall and according to treatment group, and using
T A B L E 3 Main causes of death after hATG
Causes of death (n 5 55) Time of death (days after hATG) Age (years) Infections, n5 29
Bacterial sepsis, n5 23 111/4/91/5/39/39/30/20/575 36/79/73/22/57/3/71/51/62 380/50/79/12/57/11/320/240 59/59/59/73/60/61/38/39
23/207/15/90/17/U 58/30/71/64/81/62 Fungal infection, n5 5a U/136/30/U/U 46/65/57/56/67
Virus infection, n5 1 U 55
Cardiac failure, n5 4 14/6/551/180 64/73/24/68
Hemorrhage, n5 4 22/140/31/4 15/0.8/64/68
Hematological malignancies, n5 3
Acute leukemia U/306 66/75
Myelodysplastic syndrome 219 41
Solid cancer, n5 3 62/1058/U 69/54/61
Other causes, n5 12b 7/U/1/353/U/124/547 69/67/70/67/57/18/66
Abbreviations: hATG: horse ATG; U: unknown.
aFungal causes: 1 candida infection; 1 aspergillosis; 1 zygomycosis; 2 Pneumocystis jirovecii. bNo information is available on etiology, time of death after ATG and age at death for 5 patients.
Fine and Gray’s model we report, as expected, a decrease hazard of response for refractory vs. first line treated patients. Factors predictive for response at 12 months were then analyzed. In multivariate analysis, disease severity and longer interval between diagnostic to IST (>6 months) were associated with worse outcomes. The latter factor most likely reflects the fact that our analysis mixed patients treated at differ-ent time post diagnosis (i.e., first line vs. refractory vs. relapse). Disease severity per see has been associated with survival but few studies [reviews JM and NSY] have specifically looked to factors associated with response after ATGAM1 CSA combination. Marsh et al. in a randomized trial demonstrated that patient treated with combination therapy responded better than those who received CSA alone.6The
NIH group did not find neutrophil counts to be predictive for response in patients with severe disease while response was associated with higher reticulocytes and lymphocytes counts as well as younger age.27 Lymphocyte counts were not available in our cohort in a sufficient number of patients to include this factor in the analysis. Reticulocytes count although significantly associated with response in univariate analysis but were not independently predictive of response by multi-variate analysis, most probably because they are integrated in the severity score. Moreover, PNH clone were also not associated with response in our cohort possibly related to the number of patients lack-ing this data (n5 180).
The estimates of subsequent IST treatment or of HSCT have been usually calculated by Kaplan-Meier estimates where HSCT, second IST and death are recorded as events in the survival curves [reviewed in Refs. 1,3,15]. However, using current standard of statistical methodol-ogy Kaplan Meier estimates in this situation is inadequate because death, relapse/second IST and HSCT are competing events. We used here, for the first time Fine and Gray methodology. These analyses demonstrated that nearly 40% of the patients in the relapsed group and 4% in the refractory group proceeded to new IST or HSCT, respectively.
Finally, we assessed factors influencing survival. Overall causes of death were as expected mainly infections (53% of the patients). One should not forget clonal evolution in this particular disease with 8 patients overall and 3 deaths justifying regular marrow aspirations dur-ing follow-up. Age over 40 years was independently associated with worse outcome. Older age has repeatedly been associated with worse outcome.1,3,7,15,21,22,28 Although expected from previous
stud-ies,1,3,7,15,21,22,28low neutrophil counts were also associated with bad prognosis. We also found a detrimental effect of the delay between diagnosis and IST, best results being obtained for patients treated in the first 6 months after diagnosis. Male gender was also associated with overall survival in our study, which is difficult to explain. The Basel group did find in the early 1990s that female gender was associated with worse response to ATG alone29 but no with overall survival.
Androgens were the first drug associated with response in idiopathic AA30,31as well as more recently in Dyskeratosis Congenita32and might
thus play a role.
Although representing a large unbiased cohort of consecutive patients treated in French centers in a recent period of time in a homo-geneous way, the cohort analyzed here does not fit the classical clinical
trial requirements. Due to voluntary basis of providing clinical details, 15% of the centers did not report their cases, and thus we cannot rule out that results presented here might have been slightly different if all patients treated within the ATU program have been included. Never-theless, all patients within centers who agree to provide data have been included and we believe this study does illustrate the results of ATGAM1 CSA in a true-life perspective. This large national real-life cohort confirms the findings from randomized clinical trials that the ATGAM1 CSA is a gold standard IST for patients with AA. Whether Eltrombopag might revolutionized this plateform is at the moment under investigation through a large, randomized, controlled trial (RACE; ClinicalTrials.gov number, NCT02099747) in Europe.
A C K N O W L E D G M E N T S
The authors thank all French hematological team, physicians and patients who take part of that study. This study was supported by Pfizer (unrestricted grant for data acquisition).
D I S C L O S U R E S
RPL has received research funding from, consulted for, and received honoraria from Alexion, Pfizer, and Novartis, and received research funding from Amgen.
A U T H O R C O N T R I B U T I O N S Conception and design: RPL and GS.
Provision of study materials and patients: RPL, RT, FS, TLe, TLa, MM, ST, CJ, MP, CG, SN, JYC, TB, ED, JOB, FS, FB, and GS.
Collection and assembly of the data: RPL and GS. Data analysis and interpretation: RPL and GS. Manuscript writing: RPL and GS.
Final approval of manuscript: RPL, RT, FS, TLe, TLa, MM, ST, CJ, MP, CG, SN, JYC, TB, ED, JOB, FS, FB, and GS.
O R C I D
Regis Peffault de Latour http://orcid.org/0000-0001-6222-4753
R E F E R E N C E S
[1] Bacigalupo A, Brand R, Oneto R, et al. Treatment of acquired severe aplastic anemia: bone marrow transplantation compared with immu-nosuppressive therapy–The European Group for blood and marrow transplantation experience. Semin Hematol. 2000;37(1):69–80. [2] Young NS, Calado RT, Scheinberg P. Current concepts in the
patho-physiology and treatment of aplastic anemia. Blood. 2006;108(8): 2509–2519.
[3] Scheinberg P, Young NS. How I treat acquired aplastic anemia. Blood. 2012;120(6):1185–1196.
[4] Frickhofen N, Heimpel H, Kaltwasser JP, Schrezenmeier H, German Aplastic Anemia Study G. Antithymocyte globulin with or without cyclosporin A: 11-year follow-up of a randomized trial comparing treatments of aplastic anemia. Blood. 2003;101(4):1236–1242. [5] Frickhofen N, Kaltwasser JP, Schrezenmeier H, et al. Treatment
methylprednisolone with or without cyclosporine. The German Aplastic Anemia Study Group. N Engl J Med. 1991;324(19):1297– 1304.
[6] Marsh J, Schrezenmeier H, Marin P, et al. Prospective randomized multicenter study comparing cyclosporin alone versus the combina-tion of antithymocyte globulin and cyclosporin for treatment of patients with nonsevere aplastic anemia: a report from the Euro-pean Blood and Marrow Transplant (EBMT) Severe Aplastic Anae-mia Working Party. Blood. 1999;93(7):2191–2195.
[7] Locasciulli A, Oneto R, Bacigalupo A, Severe Aplastic Anemia Work-ing Party of the European B, Marrow Transplant G, et al. Outcome of patients with acquired aplastic anemia given first line bone mar-row transplantation or immunosuppressive treatment in the last decade: a report from the European Group for Blood and Marrow Transplantation (EBMT). Haematologica. 2007;92(1):11–18. [8] Marsh JC, Bacigalupo A, Schrezenmeier H, Marrow Transplant
Group Severe Aplastic Anaemia Working P, et al. Prospective study of rabbit antithymocyte globulin and cyclosporine for aplastic ane-mia from the EBMT Severe Aplastic Anaeane-mia Working Party. Blood. 2012;119(23):5391–5396.
[9] Rosenfeld S, Follmann D, Nunez O, Young NS. Antithymocyte glob-ulin and cyclosporine for severe aplastic anemia: association between hematologic response and long-term outcome. JAMA. 2003;289(9):1130–1135.
[10] Hayakawa J, Kanda J, Akahoshi Y, et al. Meta-analysis of treatment with rabbit and horse antithymocyte globulin for aplastic anemia. Int J Hematol. 2017;105(5):578–586.
[11] Afable MG, 2nd, Shaik M, Sugimoto Y, et al. Efficacy of rabbit anti-thymocyte globulin in severe aplastic anemia. Haematologica. 2011; 96(9):1269–1275.
[12] Atta EH, Dias DS, Marra VL, de Azevedo AM. Comparison between horse and rabbit antithymocyte globulin as first-line treatment for patients with severe aplastic anemia: a single-center retrospective study. Ann Hematol. 2010;89(9):851–859.
[13] Chang MH, Kim KH, Kim HS, et al. Predictors of response to immu-nosuppressive therapy with antithymocyte globulin and cyclospo-rine and prognostic factors for survival in patients with severe aplastic anemia. Eur J Haematol. 2010;84(2):154–159.
[14] Jeong DC, Chung NG, Cho B, et al. Long-term outcome after immu-nosuppressive therapy with horse or rabbit antithymocyte globulin and cyclosporine for severe aplastic anemia in children. Haematolog-ica. 2014;99(4):664–671.
[15] Marsh JC, Ball SE, Cavenagh J, British Committee for Standards in H, et al. Guidelines for the diagnosis and management of aplastic anaemia. Br J Haematol. 2009;147(1):43–70.
[16] Rosenfeld SJ, Kimball J, Vining D, Young NS. Intensive immunosup-pression with antithymocyte globulin and cyclosporine as treatment for severe acquired aplastic anemia. Blood. 1995;85(11):3058–3065. [17] Shin SH, Yoon JH, Yahng SA, et al. The efficacy of rabbit mocyte globulin with cyclosporine in comparison to horse antithy-mocyte globulin as a first-line treatment in adult patients with severe aplastic anemia: a single-center retrospective study. Ann Hematol. 2013;92(6):817–824.
[18] Yoshimi A, Niemeyer CM, Fuhrer MM, Strahm B. Comparison of the efficacy of rabbit and horse antithymocyte globulin for the treatment of severe aplastic anemia in children. Blood. 2013;121(5): 860–861.
[19] Camitta BM, Rappeport JM, Parkman R, Nathan DG. Selection of patients for bone marrow transplantation in severe aplastic anemia. Blood. 1975;45(3):355–363.
[20] Camitta BM. What is the definition of cure for aplastic anemia? Acta Haematol. 2000;103(1):16–18.
[21] Bacigalupo A, Bruno B, Saracco P, et al. Antilymphocyte globulin, cyclosporine, prednisolone, and granulocyte colony-stimulating fac-tor for severe aplastic anemia: an update of the GITMO/EBMT study on 100 patients. European Group for Blood and Marrow Transplantation (EBMT) Working Party on Severe Aplastic Anemia and the Gruppo Italiano Trapianti di Midolio Osseo (GITMO). Blood. 2000;95:1931–1934.
[22] Fuhrer M, Rampf U, Baumann I, et al. Immunosuppressive therapy for aplastic anemia in children: a more severe disease predicts bet-ter survival. Blood. 2005;106(6):2102–2104.
[23] Townsley DM, Scheinberg P, Winkler T, et al. Eltrombopag added to standard immunosuppression for aplastic anemia. N Engl J Med. 2017;376(16):1540–1550.
[24] Scheinberg P, Nunez O, Weinstein B, et al. Horse versus rabbit antithymocyte globulin in acquired aplastic anemia. N Engl J Med. 2011;365(5):430–438.
[25] Scheinberg P, Rios O, Scheinberg P, Weinstein B, Wu CO, Young NS. Prolonged cyclosporine administration after antithymocyte glob-ulin delays but does not prevent relapse in severe aplastic anemia. Am J Hematol. 2014;89(6):571–574.
[26] Scheinberg P, Townsley D, Dumitriu B, et al. Horse antithymocyte globulin as salvage therapy after rabbit antithymocyte globulin for severe aplastic anemia. Am J Hematol. 2014;89(5):467–469. [27] Scheinberg P, Wu CO, Nunez O, Young NS. Predicting response to
immunosuppressive therapy and survival in severe aplastic anaemia. Br J Haematol. 2009;144(2):206–216.
[28] Tichelli A, Socie G, Henry AM, et al. Effectiveness of immunosup-pressive therapy in older patients with aplastic anemia. European Group for Blood and Marrow Transplantation Severe Aplastic Anae-mia Working Party. Ann Intern Med. 1999;130:193–201.
[29] Nissen C, Gratwohl A, Tichelli A, et al. Gender and response to anti-lymphocyte globulin (ALG) for severe aplastic anaemia. Br J Haema-tol. 1993;83(2):319–325.
[30] Najean Y. Long-term follow-up in patients with aplastic anemia. A study of 137 androgen-treated patients surviving more than two years. Joint Group for the Study of Aplastic and Refractory Anemias. Am J Med. 1981;71(4):543–551.
[31] Najean Y, Haguenauer O. Long-term (5 to 20 years) Evolution of nongrafted aplastic anemias. The Cooperative Group for the Study of Aplastic and Refractory Anemias. Blood. 1990;76:2222–2228. [32] Townsley DM, Dumitriu B, Liu D, et al. Danazol Treatment for
Telo-mere Diseases. N Engl J Med. 2016;374(20):1922–1931.
S U P P O R T I N G I N F O R M A T I O N
Additional Supporting Information may be found online in the sup-porting information tab for this article.
How to cite this article: Peffault de Latour R, Tabrizi R, Marcais A, et al. Nationwide survey on the use of horse antithymocyte globulins (ATGAM) in patients with acquired aplastic anemia: A report on behalf of the French Reference Center for Aplastic Anemia. Am J Hematol. 2018;93:635–642.https://doi.org/10. 1002/ajh.25050
