Upfront unrelated donor hematopoietic stem cell transplantation in patients with idiopathic aplastic anemia: A retrospective study of the Severe Aplastic Anemia Working Party of European Bone Marrow Transplantation

Abstract

Young patients with idiopathic aplastic anemia (AA) respond better to immunosuppressive therapy (IST) but long-term outcomes are suboptimal with 30% nonresponders, significant risk of relapse, cyclosporine (CSA) dependence, and clonal evolution.1 Excellent results of upfront unrelated donor (UD) hematopoietic stem cell transplantation (HSCT) have been reported in a cohort of 29 children with idiopathic AA using an alemtuzumab-based regimen, with low graft versus host disease (GVHD) rates and only one death. Outcomes of upfront UD HSCT were similar to matched sibling donor (MSD) HSCT and superior to IST and UD HSCT post-IST failure.1 Despite the small sample size in this study, the inclusion of exclusively pediatric patients, and the sole use of alemtuzumab in the conditioning regimen, many investigators are now offering upfront UD HSCT in patients as first-line therapy. Prospective trials are ongoing to confirm the feasibility of such procedure in comparison to IST, but no results are available yet. We used the European Society for Blood and Marrow Transplantation (EBMT) registry to analyze the outcome of 74 patients who received upfront UD HSCT in Europe during the last decade. All patients who received UD HSCT for AA between 2010 and 2018 registered in the EBMT registry for SAA were included. Patients who had received IST (single-agent CSA or anti-thymocyte globulin [ATG] plus CSA) before HSCT, cord blood, and haplo-identical transplant were excluded, as well as patients suffering from congenital bone marrow disorders. The primary endpoint was overall survival (OS) by 2 years. Secondary endpoints were GVHD-free/relapse-free survival (GRFS), defined as being alive, having engrafted without acute GVHD (aGVHD) grade III-IV, and extensive chronic GVHD (cGVHD) during follow-up. Cumulative incidences of day 100 aGvHD and 2-year cGvHD were estimated, including competing events relapse, graft failure, second transplant, and death. Seventy-four patients were included (characteristics are reported in Table S1). Patients were mainly adolescent and young adults (AYA) with a median age of 20 years (range 1–76). Sixty patients were transplanted from a matched UD and 12 patients had mismatched UD (HLA data were missing for two patients). Median time to neutrophil engraftment (> 0.5 × 109/L) was 18 (15–19) days and 19 (17–19) days for platelet recovery (> 20 × 109/L). Median follow-up was 49 months (95% CI: 39–57). The major complications of UD HSCT following failure of first-line IST are graft failure (> 10%)2 and severe GVHD. In this high-risk population, graft failure occurred only in 8% (95% CI: 2%–15%). Similarly, cumulative incidences of GVHD, both acute and chronic, with UD HSCT were expected to be higher compared to MSD HSCT (around 25%).3 Acute GVHD grade II-IV occurred in 13% (95% CI: 5%–20%) with grade III and IV diagnosed in only two patients (3%, 95% CI: 0%–7%). Chronic GVHD is the most feared and undesirable complication following HSCT especially in patients with AA where no graft versus leukemia effect is required. At 2 years, cumulative incidence of cGVHD was 17% (95% CI: 8%–26%), with no extensive form. In the previously reported upfront MUD study, 19% of the patients suffered from cGVHD (limited in all cases) using an alemtuzumab-based conditioning regimen.1 Overall, eight patients died, mainly due to infections (n = 5). One patient each died of secondary malignancy and GVHD, while cause of death was unknown in one patient. The 2-year survival rate was 89% (95% CI, 82%–96%) (Figure 1A) with an overall GRFS of 86% (95% CI, 77%–94%) at 2 years (Figure 1B). It is well established that both low-dose total body irradiation (TBI) and in vivo T-cell depletion using either ATG or alemtuzumab are efficient in preventing GVHD and graft failure in UD HSCT recipients.2, 3 In Europe, access to alemtuzumab is highly variable. In our cohort, 73 patients received T-cell depletion using ATG (n = 37) or alemtuzumab (n = 36), one patient received neither. Only nine patients (12%) received low-dose TBI, which is expected in young patients with nonmalignant disease to limit the risk of radiation-induced secondary malignancies.3 These data suggest that excellent outcomes can be achieved after upfront UD HSCT independently of access to alemtuzumab. The source of stem cell is also a known prognostic factor, the use of peripheral blood stem cells (PBSCs) being significantly associated with worse outcome in AA, except in the setting of alemtuzumab conditioning.2, 4, 5 In our cohort, 55% of the patients received PBSCs, 60% of them received alemtuzumab. Outcome seemed to be similar between PBSCs and BM HSCT recipient, but was not tested (Figure S1). Longer interval between AA diagnosis and transplantation is also correlated with poor outcome in refractory patients who received UD HSCT.4, 5 In our cohort, time to transplant was 7 months (IQR: 4–12), which is longer in comparison with previously reported UK study (3 months).1 In a previous study of HSCT after failed IST, a shorter time from diagnosis to transplant (< 6 months) has been shown significantly to improve outcome.2 However, due to the limited number of events post-HSCT, no specific factors correlating with outcome were identified. Age per se is a well-established prognostic factor in AA with better outcomes in younger patients.3 In our cohort, 48 patients (64%) were younger than 24 years old, while 9 were between 25 and 40 years old, and 17 (23%) were older than 40. Outcome was worse for patients over 25 but did not reach statistical significance, suggesting that age limit might be flexible (see Figure S2). In AA, secondary malignancies can be caused either by clonal evolution to acute leukemia or post-transplant lymphoproliferative disease (PTLD). Ten years cumulative incidence of secondary malignancies in AA is around 20% after IST and around 4% after HSCT.6 UD HSCT and in vivo T-cell depletion both increase the risk of PTLD.6 In our study, six patients acquired a secondary malignancy during follow-up, of which four patients presenting with non-Hodgkin lymphomas and one with an acute myeloid leukemia without previous graft failure. Overall, these data confirm that upfront UD HSCT in SAA is a valuable curative option in patients with AA. The choice between an upfront UD HSCT and a deferred HSCT may also take into account the improved results offered by nontransplant options, that is, triple therapy with ATG, cyclosporine A and eltrombopag. Though our study was retrospective, with lack of homogeneous treatment, the analysis involved a larger number of patients from multiple centers with overall satisfactory outcomes despite the disparity in stem cell source and conditioning regimen. This latter finding suggests excellent results are not exclusively related to use of low-dose TBI or alemtuzumab-based regimen. The large range of age in our cohort is also remarkable, with 23% being older than 40 years old. We did not assess quality of life and long-term complications, which may be significantly affected, especially by cGVHD. Because of obvious limitations related to the retrospective nature and the registry-based nature of the study, unreported events and information bias cannot be excluded. Upfront UD HSCT strategy in AA is highly dependent on rapidity of donor identification and donor availability, with the risk of infections/complications caused by unexpected donor delays. The data on intention-to-treat population, that is, patients who were planned for upfront UD but never did due to unexpected complications before HSCT or donor cancellation is missing in our study. Prospective trials ongoing in the United States and Europe will address this limitation to formally confirm upfront UD transplantation as standard of care for patients with idiopathic AA. The authors have no conflict of interest. A.P. and R.P.D.L designed the research and wrote the paper, D.-J.E. collected and analyzed the data, A.K. participated in the critical revision of the paper. All authors discussed the results and contributed to the final manuscript. The data that support the findings of this study are available from the corresponding author upon reasonable request. Table S1 Patient, disease, and graft characteristics (n = 74 patients) Figure S1 Outcome of patients after upfront UD HSCT according to stem cell source. Kaplan–Meier curves of (A) overall survival and (B) GvHD-free/relapse-free survival in the whole cohort. The shaded regions correspond with the 95% confidence intervals. Figure S2 Outcome of patients after upfront UD HSCT according to age groups. Kaplan–Meier curves of (A) overall survival and (B) GvHD-free/relapse-free survival in the whole cohort. The shaded regions correspond with the 95% confidence intervals. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Topics & Concepts