Scholarlink

<p>Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a cornerstone of therapy for adult B-cell acute lymphoblastic leukemia (B-ALL), yet simultaneous comparison of all four major donor categories within a single transplant platform is scarce. We retrospectively analyzed 190 consecutive adults with B-ALL undergoing first allo-HSCT (2008–2025) at a single center on a predominantly rabbit antithymocyte globulin (rATG)–based myeloablative platform: matched sibling (MSD, n = 72), matched unrelated (MUD, n = 46), mismatched unrelated (MMUD, n = 40), and haploidentical (Haplo, n = 32) donors. Cumulative incidence of relapse (CIR) and non-relapse mortality (NRM) were compared in a competing-risk framework (Fine–Gray regression). Five-year overall survival did not differ significantly across donor groups (range 31.6–59.2%; <i>P</i> = 0.14). Competing-risk analysis revealed opposing gradients: five-year CIR decreased from MSD (38.5%) to Haplo (23.8%; <i>P</i> = 0.40), whereas NRM increased from MSD (13.3%) to Haplo (43.5%; <i>P</i> = 0.01). Relapse-free survival did not differ (<i>P</i> = 0.43). Haploidentical transplantation was independently associated with higher NRM (subdistribution hazard ratio [sHR] 4.44; <i>P</i> &lt; 0.001) but not relapse, compared with MSD. NRM was infection-related in 29/39 patients (74%; 83% in Haplo), whereas graft-versus-host disease (GVHD)–related deaths were rare. After inverse-probability-of-treatment weighting, NRM remained higher in haploidentical recipients (31.9% vs. 17.7%), while the apparent CIR advantage attenuated (38.2% vs. 34.8%). Within a predominantly rATG-based platform, donor source produced opposing gradients of relapse and NRM, with the apparent haploidentical relapse advantage largely explained by age and era confounding. The persistent NRM excess in haploidentical recipients was driven predominantly by infection rather than GVHD. These platform-specific findings support transitioning to post-transplant cyclophosphamide (PTCy)–based approaches.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Donor-source-specific competing risks of relapse and non-relapse mortality in adult B-ALL allogeneic hematopoietic stem cell transplantation on a rabbit antithymocyte globulin-based platform

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月22日

查看全文
  • Dae-Ho Choi,
  • Chul Won Jung,
  • Jun Ho Jang

摘要

Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a cornerstone of therapy for adult B-cell acute lymphoblastic leukemia (B-ALL), yet simultaneous comparison of all four major donor categories within a single transplant platform is scarce. We retrospectively analyzed 190 consecutive adults with B-ALL undergoing first allo-HSCT (2008–2025) at a single center on a predominantly rabbit antithymocyte globulin (rATG)–based myeloablative platform: matched sibling (MSD, n = 72), matched unrelated (MUD, n = 46), mismatched unrelated (MMUD, n = 40), and haploidentical (Haplo, n = 32) donors. Cumulative incidence of relapse (CIR) and non-relapse mortality (NRM) were compared in a competing-risk framework (Fine–Gray regression). Five-year overall survival did not differ significantly across donor groups (range 31.6–59.2%; P = 0.14). Competing-risk analysis revealed opposing gradients: five-year CIR decreased from MSD (38.5%) to Haplo (23.8%; P = 0.40), whereas NRM increased from MSD (13.3%) to Haplo (43.5%; P = 0.01). Relapse-free survival did not differ (P = 0.43). Haploidentical transplantation was independently associated with higher NRM (subdistribution hazard ratio [sHR] 4.44; P < 0.001) but not relapse, compared with MSD. NRM was infection-related in 29/39 patients (74%; 83% in Haplo), whereas graft-versus-host disease (GVHD)–related deaths were rare. After inverse-probability-of-treatment weighting, NRM remained higher in haploidentical recipients (31.9% vs. 17.7%), while the apparent CIR advantage attenuated (38.2% vs. 34.8%). Within a predominantly rATG-based platform, donor source produced opposing gradients of relapse and NRM, with the apparent haploidentical relapse advantage largely explained by age and era confounding. The persistent NRM excess in haploidentical recipients was driven predominantly by infection rather than GVHD. These platform-specific findings support transitioning to post-transplant cyclophosphamide (PTCy)–based approaches.