Hematopoietic stem cell transplantation from haploidentical offspring donors using post-transplant cyclophosphamide versus human leukocyte antigen-matched siblings in older patients with myelodysplastic syndrome

Allogeneic hematopoietic stem cell transplantation (HSCT) is the only curative treatment for myelodysplastic syndrome (MDS) and has been increasingly performed in older patients, the majority of whom have MDS.1 Donors are important factors in the success of allogeneic HSCT. Generally, a human leukocyte antigen (HLA)-matched sibling donor (MSD) is considered the most preferred donor.1, 2 However, several patients with MDS are diagnosed at an older age; therefore, the MSD is also likely to be of advanced age, which raises some concerns and could result in poor outcomes3 Hence, some reports have compared alternative donors and MSD and suggested that the alternative donor had comparable outcomes to MSD in older patients with MDS.3 Conversely, HSCT from an HLA haploidentical donor (haplo-HSCT) using post-transplant cyclophosphamide (PTCY) is effective and safe for patients with MDS. The offspring of older patients are generally selected as young haploidentical donors, making important donor candidates for older patients with MDS. Therefore, selecting whether haploidentical offspring or HLA-matched siblings for older patients with MDS was an important clinical question; however, there is limited available evidence. This study aimed to compare haplo-HSCT using PTCY with HSCT from MSD (MSD-HSCT) in older patients with MDS. Our study involved patients enrolled in the Transplant Registry Unification Management Program 2 (TRUMP 2) of the Japanese Data Center for Hematopoietic Cell Transplantation (JDCHCT) between 2013 and 2020. The inclusion criteria for our study were as follows: patients aged >50 years, who underwent their first allogeneic HSCT, and who had received transplants from haploidentical related donors, which were defined as HLA 2 locus or over mismatched related donors or HLA-MSDs. The exclusion criterion was patients who underwent haplo-HSCT from sibling donors. Patient consent was obtained before the TRUMP 2 registration. The study was approved by the Data Management Committee of the JDCHCT and the Ethics Committee of Kobe City Hospital Organization Kobe City Medical Center General Hospital (approval number: zn230909). Detailed definitions of variables and endpoints were described in supplemental materials. We set the primary endpoint as overall survival (OS). We also evaluated the progression-free survival (PFS), graft versus host disease (GVHD)-free relapse-free survival (GRFS), cumulative incidence of relapse, cumulative incidence of non-relapse mortality (NRM), neutrophil engraftment rate, platelet engraftment rate, grade II–IV acute GVHD, grade III–IV acute GVHD, chronic GVHD, and extensive chronic GVHD as secondary endpoints. Event rates were estimated using the Kaplan–Meier method for OS, PFS, and GRFS, and Gray's method for the cumulative incidence of the other endpoints. We used inverse probability treatment weight (IPTW) analysis to reduce the effect of confounding factors. In the IPTW analysis, we employed the multiple imputation by chained equation (MICE) approach to impute missing data in the variables of interest. The detailed methods of IPTW analysis and MICE were described in supplemental materials. The impact of haplo-HSCT using PTCY compared with MSD-HSCT was estimated using Cox proportional hazards models for OS, PFS, and GRFS, and the other endpoints using the Fine and Gray method. The endpoints were described using weighted hazard ratios (HRs) and 95% confidence intervals (CIs). Statistical significance was set at p < .05. All statistical analyses were performed using the R software (version 4.0.2; R Development Core Team). We analyzed 267 patients who underwent haplo-HSCT using PTCY (n = 90) or MSD-HSCT (n = 177). The patients' clinical characteristics are presented in Supplementary Table 1. The median age was 59 years (interquartile range, 56–63 years), and 200 patients (75%) were male. The haplo-HSCT with PTCY group included older male patients, patients who received reduced-intensity conditioning regimen, and patients who had recently received allogeneic HSCT (Supplementary Table 1). The 3-year OS was 42.5% (95% CI: 31.4–53.0) in the haplo-HSCT using PTCY group versus 48.3 (95% CI: 40.6–55.6) in the MSD-HSCT group (log-rank p = .287), and the HR of OS was 1.21 (95% CI: 0.85–1.72) in the haplo-HSCT using PTCY group compared to that in the MSD-HSCT group (p = .288) (Figure 1, Supplementary Table 2). IPTW analysis revealed no significant difference in OS between the two groups with a weighted HR of 0.75 (95% CI: 0.40–1.38, p = .346). Similarly, the two groups had no significant differences in the PFS, GRFS, relapse, and NRM (Figure 1). The weighted HR of PFS was 0.73 (95% CI: 0.40–1.34; p = .310), that of GRFS was 0.63 (95% CI: 0.35–1.14; p = .128), that of relapse was 0.89 (95% CI: 0.47–1.68; p = .719), and that of NRM was 0.55 (95% CI: 0.26–1.18, p = .123) (Supplementary Table 2). The haplo-HSCT using PTCY was associated with a low neutrophil and platelet engraftment rate compared to MSD-HSCT (Figure 1). IPTW analysis also revealed that haplo-HSCT using PTCY group had inferior engraftment with weighted HR of 0.49 (95% CI: 0.35–0.68; p < .001) for neutrophil engraftment and 0.55 (95% CI: 0.40–0.76; p < .001) for platelet engraftment (Supplementary Table 2). There were no significant differences between the two groups in the cumulative incidence of grade II–III acute GVHD and grade III–IV acute GVHD. The haplo-HSCT using PTCY was associated with a low cumulative incidence of chronic GVHD and similar extensive chronic GVHD compared to MSD-HSCT (Figure 1). The IPTW analysis revealed that haplo-HSCT in the PTCY group was not significantly associated with acute or chronic GVHD (Supplementary Table 2). Older donor age is problematic, including deviations from donor eligibility owing to donor illness, concerns regarding poor graft collection, and the possibility of clonal hematopoiesis.3 In addition, some studies have shown an association between an older donor age and poor outcomes in patients who underwent unrelated donor transplantation and haplo-HSCT using PTCY.4 Although, the validation in patients who receive MSD-HSCT is difficult because donor age is usually associated with recipient age in patients transplanted from MSD; several studies have indirectly verified the utility of alternative donor comparing MSD in older patients with MDS.3 Our study compared PTCY-Haplo and MSD transplantation in patients with MDS aged >50. The PTCY group had comparable OS with a weighted HR of 0.75 (95% CI: 0.40–1.38). Consistent with this study, there were no significant differences in the other endpoints, except for neutrophil and platelet engraftment. Thus, our results suggest the utility of young haploidentical donors who use PTCY in older patients with MDS. Haplo-HSCT using PTCY has been widely used worldwide because its safety and efficacy have been demonstrated in clinical studies. Therefore, some studies compared MSD-HSCT, which was the standard donor source, in patients with MDS. For example, the largest study from the European Group for Blood and Marrow Transplantation group compared 415 patients with haplo-HSCT using PTCY and 1414 patients with MSD-HSCT.5 In the study, haplo-HSCT using PTCY had inferior OS compared to MSD-HSCT with 2-year OS of 50% and 58%, respectively. The inferior OS of haplo-HSCT using PTCY was remarkable in the first 6 months after allogeneic HSCT, with an HR of 1.93 due to a lower engraftment rate and higher NRM. Although inconsistent with previous studies, our study revealed that haplo-HSCT using PTCY had outcomes comparable to those of 0.75 compared to MSD-HSCT in older patients with MDS. This inconsistency was due to selection bias. Previous studies on MDS included the entire age range of patients with MDS and their siblings and parent donors in haplo-HSCT using PTCY. In addition, the relatively low engraftment rate of patients with haplo-HSCT using PTCY, which was probably because previous reports including patients who received bone marrow transplantation, also influenced the inconsistent outcomes. Conversely, a retrospective study using the Center for International Blood and Marrow Transplant Research and the European Group for Blood and Marrow Transplantation registry data compared haplo-HSCT using PTCY from offspring and MSD-HSCT in older patients with acute leukemia.6 The Haplo-HSCT using PTCY group had inferior OS with an adjusted HR of 1.32. The point estimation was slightly lower than that in previous studies comparing haplo-HSCT using PTCY and MDS-HSCT, but was inconsistent with our results.5 In this study, 70% of patients in the haplo-HSCT using PTCY group underwent bone marrow transplantation, leading to a lower engraftment rate and higher NRM. In contrast, most patients in our study received peripheral blood transplantation based on earlier nationwide clinical studies. This difference in the donor source of haplo-HSCT using PTCY might be the reason for the inconsistent results. Nonetheless, the present study had several limitations. First, we were unable to assess genetic data, which has been shown to be a strong prognostic factor for patients with MDS. Secondly, unmeasured confounding factors may have influenced the selection of the conditioning regimen due to the nature of the registry-based study, although we used IPTW analysis. However, our data raises the possibility of haplo-HSCT using PTCY in older patients with MDS and provides hematologists with valuable information for donor selection. In conclusion, in this Japanese cohort, haplo-HSCT using PTCY had a comparable OS to MSD-HSCT with a weighted HR of 0.75 in patients with MDS aged over 50 years. Yoshimitsu Shimomura designed the study, developed the models, performed statistical analyses, and wrote the first draft of the manuscript. Tetsuhisa Kitamura contributed to the data analysis. Takaaki Konuma, Yosuke Nakaya, Toshiro Kawakita, and Hidehiro Itonaga critically reviewed the data analysis and the manuscript. All other authors contributed to data collection. All authors approved the final version of the manuscript. The authors thank all the physicians and data managers at the centers that contributed to data collection on transplantation for the Japanese Data Center for Hematopoietic Cell Transplantation and Transplant Registry Unified Management Program 2. HN received honoraria from Astellas, Otsuka, Sumitomo Dainippon, Takeda, Novartis, Bristol-Myers Squibb, Nippon Shinyaku, Astellas-Amgen, Shire, Celgene, Daiichi Sankyo, and Pfizer; consulting/advisory roles for Ono, Simon-Kucher and Partners, Daiichi Sankyo, and Novartis; and research funding (to institution) from Novartis, Astellas, Otsuka, Alexion, PPD-SNBL, Cmic, Meiji Seika, and Bristol-Myers Squibb outside of the submitted work. The data of this study are not publicly available due to ethical restrictions that it exceeds the scope of the recipient/donor's consent for research use in the registry. Data S1. Supporting Information. 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.