Lower probability of relapse after myeloablative cord blood transplantation for patients with refractory AML

Matsuda et al.1 recently described the outcomes of patients with ‘non-remission’ acute myeloid leukaemia (AML) after haematopoietic stem cell transplantation (HSCT) using readily available alternative donors: cord blood (CB) and haplo-identical (Haplo) donors. This very well written manuscript, however, did not emphasize an important finding: that patients with active disease who received a myeloablative CB transplant had an approximately 1.5-fold higher probability of survival compared to those with haplo-donors. The improved survival was mainly due to lower incidence of relapse. Our letter to the Editor aims to highlight this finding, as it has implications for donor selection for these very high-risk patients. Allogeneic HSCT offers a potentially curative approach for AML patients, even when they are not in complete remission. Since these transplants are urgent, rapidly available donor sources are required, and these include CB and haplo-donors. The paper of Matsuda et al.1 describes a retrospective comparison of outcomes of recipients of unrelated CB grafts (n = 918) and those of haplo-HSCT (n = 459) from the Japanese Registry. These two groups were selected of a total of 2438 ‘non-remission’ AML patients with ages of 16–65 years transplanted during the period January 2008–December 2018, after 2:1 propensity matching. The disease risk was similar between the groups, and both included more than 50% patients who never achieved remission. There was no statistically significant difference in overall survival (OS), disease-free survival and relapse. However, 85% of the patients in both groups received reduced-intensity regimens. When the groups with myeloablative regimens (myeloablative cytoreduction [MAC], Total Body Irradiation- or busulfan-based) were analysed, the multivariate OS hazard ratio for haplo-HSCT was higher [hazard ratio (HR) 1.62, 95% confidence interval (CI) 1.09–2.40, p: 0.017] indicating a 1.5-fold lower survival compared to CB. While this information is presented in table 3 in Matsuda et al., the figure which shows superior survival for the CB grafts is only in their Supplement (figure S1). We believe that this clinically important finding deserves more attention, as it has implications for both graft selection and treatment. Therefore, it would serve better if it had been included in the main figure of the paper (figure 1A–I). Importantly, similar results were also reported in another recent analysis from the Japanese Registry of adult patients with refractory AML: CB recipients had superior five-year progression-free survival (p = 0.009) resulting from significantly lower relapse rates (p < 0.001) compared to those of matched related donors.2 Also, paediatric CB transplant outcomes from the UK reported at the Transplantation and Cellular Therapy meeting 2022 showed two-year disease-free, graft-versus-host disease (GvHD)-free survival of 60% for patients with AML, of whom 20/28 had refractory disease.3 Moreover, another recent analysis from Japan highlights the superior results after CB transplantation following intensified conditioning regimens.4 As described by Milano et al.5 and subsequently by others, CB transplantation leads to better disease control in AML patients with MRD-positive remission. In the Seattle study the relapse probability was twofold lower after CB compared to matched or mismatched bone marrow (BM) grafts. Matsuda et al.1 showed that CB transplants following myeloablative treatment have ‘preferable’ survival (quoting the authors) in comparison to haplo-HSCT, even in AML patients with refractory disease. The favourable outcomes were driven by the lower relapse rates in this group also. This potent graft-versus-leukaemia effect is thought to result from the unique immunological properties of CB T cells: CB-derived CD4+ cells have transcription profiles similar to fetal CD4+ cells,6 but very distinct from those of BM or peripheral blood (PB) T cells; further, naïve CB-derived CD8+ cells rapidly gain effector functions in the tumour microenvironment and mediate enhanced antitumor effects.7 The lower relapse rates after CB transplantation have important implications. The vast majority of transplant candidates, independently of their race/ethnicity,8 will have suitable CB grafts that can be obtained promptly for urgent transplantation. Furthermore, myeloablative cytoreduction needs to be considered early in the course of treatment, as additive toxicities from multiple courses of chemotherapy may affect the patients' HCT-CI9 and render them ‘unfit’ for MAC. Andromachi Scaradavou and Jaap Jan Boelens wrote and reviewed the paper. No conflicts of interest to disclose.