Single‐unit unrelated cord blood transplantation versus HLA‐matched sibling transplantation in adults with advanced myelodysplastic syndrome: A registry‐based study from the adult MDS working group of the Japanese society for transplantation and cellular therapy

Allogeneic hematopoietic stem cell transplantation (HCT) remains the only potential curative therapeutic modality for advanced myelodysplastic syndrome (MDS). Within HCT, the advancement of cord blood transplantation (CBT) procedures has resulted in a drastic expansion of CBT as a donor source for MDS. However, data comparing matched sibling donors (MSDs) HCT with CBT for advanced MDS, which was defined as refractory anemia with an excess of blasts (RAEB)‐1 and RAEB‐2 according to the World Health Organization classification at the time of HCT, have not been explored. We retrospectively compared survival and other posttransplant outcomes in 999 adult patients with advanced MDS after receiving allogeneic HCT in Japan between 2011 and 2020, using either MSD (n = 331) or single‐unit unrelated cord blood (UCB) (n = 668). In the multivariate analysis, there were no significant differences in overall survival (hazard ratio [HR], 1.10; 95% confidence interval [CI], 0.90–1.34; P = 0.347), disease‐free survival (HR, 1.01; 95% CI, 0.84–1.23; P = 0.845), relapse (HR, 0.88; 95% CI, 0.68–1.15; P = 0.370), or non‐relapse mortality (HR, 1.15; 95% CI, 0.87–1.50; P = 0.310) between MSD recipients and UCB recipients. UCB was significantly associated with lower neutrophil (HR, 0.28; 95% CI, 0.24–0.33; P < 0.001) and lower platelet (HR, 0.29; 95% CI, 0.23–0.36; P < 0.001) recovery compared to MSD. UCB was significantly associated with a lower incidence of chronic graft‐versus‐host disease (GVHD) (HR, 0.57; 95% CI, 0.44–0.75; P < 0.001) and extensive chronic GVHD (HR, 0.46; 95% CI, 0.32–0.67; P < 0.001) compared to MSD. Similar results were observed after adjusting for differences between MSD and UCB recipients by propensity score matching analysis. Our study demonstrated that single CBT and MSD HCT had similar survival outcomes for adult patients with advanced MDS despite the lower hematopoietic recovery in CBT recipients and higher chronic GVHD in MSD recipients.


| INTRODUCTION
Myelodysplastic syndrome (MDS) is an acquired bone marrow failure disorder characterized by cytopenia and risk of transformation into acute myeloid leukemia.The only potential curative therapeutic modality for MDS is allogeneic hematopoietic cell transplantation (HCT).However, the donor search is a major challenge.This is because a matched sibling donor (MSD), which has been historically the primary donor for allogeneic HCT, is likely to be an elderly person because MDS is mainly diagnosed in the elderly population.2][3] Thus, MSD should not always be considered the primary choice for allogeneic HCT in patients with MDS, particularly elderly patients.
][5][6][7][8][9][10][11][12] The advancement of UCB transplantation (CBT) procedures has contributed to the recent improvement in engraftment and early mortality, 13 which were the main limitations of CBT in adults.These improvements have resulted in a drastic expansion of CBT for MDS in recent years.5][16][17][18] However, there is no comparative data for MSD HCT and CBT restricted to adult patients with advanced MDS.Here, we used data from a Japanese database to retrospectively compare the outcomes of allogeneic HCT in adult patients with advanced MDS who received either MSD HCT or single-unit CBT.

| Study design and objectives
This was a retrospective, nationwide registry-based analysis.Clinical data were provided by the dataset of the Transplant Registry Unified Management Program (TRUMP) of the Japanese Data Center for Hematopoietic Cell Transplantation (JDCHCT) and the Japanese Society for Transplantation and Cellular Therapy (JSTCT). 19,20The Adult MDS Working Group of the JSTCT and the Institutional Review Board of the Institute of Medical Science, The University of Tokyo (2023-18-0628) approved this study.
The primary objective was to compare overall survival (OS) between MSD and UCB recipients for advanced MDS.The secondary objectives were to compare disease-free survival (DFS), relapse rate, non-relapse mortality (NRM), neutrophil and platelet recovery rates, and the incidences of acute and chronic GVHD between the two groups.

| Data collection and definitions
We retrospectively analyzed patients between 16 and 70 years of age with advanced MDS who received their first allogeneic HCT from either an MSD or single cord blood (CB) unit in Japan between 2011 and 2020.Advanced MDS was defined as refractory anemia with an excess of blasts (RAEB)-1 and RAEB-2 according to the World Health Organization (WHO) classification 21,22 at the time of HCT.The degree of HLA disparity was determined at the antigen level for HLA-A, HLA-B, and HLA-DR.Patients who received MSD HCT were matched at the HLA antigen levels for HLA-A, HLA-B, and HLA-DR.Patients who received CBT with single-unit CB were matched at ≥4 HLA antigen levels for HLA-A, HLA-B, and HLA-DR.OS was defined as the time from HCT to the time of death or final observation.The DFS was defined as the time from HCT to the time of relapse, death, or final observation.Hematological recurrence of MDS was used to define relapse.Death without relapse of MDS was referred to as NRM.The achievement of an absolute neutrophil count of ≥0.5 � 10 9 /L for three days was referred to as neutrophil recovery.An absolute platelet count of ≥20 � 10 9 /L for 7 days without transfusion was referred to as platelet recovery.According to previously described standards, the following variables were categorized: performance status (PS), 23 HCT-specific comorbidity index (HCT-CI), 24 and International Prognostic Scoring System (IPSS) at diagnosis. 25Poor-risk cytogenetics were characterized according to the IPSS, and included complex (≥3 abnormalities) or chromosome 7 abnormalities. 25The treating doctors at each location evaluated the diagnosis and severity of acute and chronic GVHD using consensus criteria. 26,27According to the criteria proposed by the Center for International Blood and Marrow Transplant Research (CIBMTR), the myeloablative conditioning (MAC) regimen was defined as a regimen containing either total body irradiation fractionated doses totaling >8 Gy, oral busulfan doses of ≥9 mg/kg, intravenous busulfan doses of ≥7.2 mg/kg, or melphalan doses of >140 mg/m 2 .Other regimens were classed as reduced-intensity conditioning (RIC). 28

| Statistical analysis
The chi-square test or Fisher exact test was used to evaluate patient characteristics for categorical data, whereas the Mann-Whitney U test was used for continuous data.Kaplan-Meier curves were used to calculate the probabilities of survival endpoints.Cumulative incidence curves were used to calculate the probabilities of other endpoints while accounting for competing risks.The competing event for neutrophil and platelet recovery was death prior to engraftment.The competing event for GVHD was death without GVHD.For NRM, relapse was a rival event, and vice versa.
The log-rank test and Gray's test were used in univariate analyses for survival curves and cumulative incidence curves, respectively.For multivariate analyses, either a Cox proportional hazard regression model for survival endpoints or a Fine and Gray proportional hazard model for other endpoints was used.The final model included adjustments for the following factors: donor type (MSD vs. UCB), age (<50 years vs. ≥ 50 years), recipients' gender (male vs. female), PS (0-1 vs. 2-4), cytogenetic risk (poor vs. other than poor), prior treatment with azacytidine (no vs. yes), prior treatment with chemotherapy (no vs. yes), conditioning regimen (MAC vs. RIC), GVHD prophylaxis (with methotrexate vs. without methotrexate), ABO incompatibility (major/bidirectional mismatch vs. match/minor mismatch), and year of HCT (2011-2015 vs. 2016-2020).
Additionally, propensity score matching was done to account for variations between MSD and UCB recipients.Age, sex, cytogenetic risk, the intensity of conditioning regimen, GVHD prophylaxis, ABO incompatibility, and year of HCT were all taken into account when calculating propensity scores using logistic regression.A 1:2 matching by propensity score with a fixed caliper width of 0.20 was carried out using the nearest neighbor matching approach (Supplementary Table S1).
EZR version 1.61 (Saitama Medical Center, Jichi Medical University, Saitama, Japan), 29 a graphical user interface for R 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria), was used to execute all statistical analyses.The two-tailed p-value of 0.05 was used to define the statistical significance level.
The cumulative incidences of acute GVHD of grades II to IV and grades III to IV were comparable between MSD recipients and UCB recipients (35.5% vs. 38.4% at 100 days, P = 0.359 for grades II to IV acute GVHD; 13.7% vs. 13.4% at 100 days, P = 0.688 for grades III to IV acute GVHD) (Figure 2C,D).In the multivariate analysis, there were no significant differences in the HRs of grades II to IV acute GVHD (HR, 1.05; 95% CI, 0.83-1.33;P = 0.640) or grades III to IV acute GVHD (HR, 0.80; 95% CI, 0.54-1.19;P = 0.290) between MSD recipients and UCB recipients (Table 3).

| Cause of death
Supplementary Table S2 shows the causes of death according to graft type.The most common cause of death was relapse or progression of MDS, irrespective of graft type.Compared to MSD, infection, sinusoidal obstruction syndrome/transplantation-associated microangiopathy, and graft failure were generally more common causes of death after UCB.In contrast, GVHD and organ failure were more common causes of death after MSD compared to UCB.

F I G U R E 2
Unadjusted cumulative incidences of neutrophil recovery (A), platelet recovery (B), grades II to IV acute GVHD (C), grades III to IV acute GVHD (D), chronic GVHD (E), and extensive GVHD (F) in MSD hematopoietic cell transplantation (HCT) and single cord blood transplantation (CBT) for advanced MDS in the entire cohort.
T A B L E 3 Multivariate analysis of hematopoietic engraftment, and GVHD.conditioning intensity, GVHD prophylaxis, ABO incompatibility, and year of HCT (Figure 3).

| DISCUSSION
We compared the posttransplant outcomes in adults with advanced MDS who received MSD HCT or single-unit CBT.After adjusting the characteristics of several factors, MSD HCT provided higher hematopoietic recovery than CBT, whereas the incidences of overall and extensive chronic GVHD after CBT were lower than those after MSD HCT.Finally, there were no significant differences in survival or relapse between MSD HCT and CBT for advanced MDS.These data suggest that both MSD and UCB should be the primary donor options for allogeneic HCT in advanced MDS.
Although no studies comparing MSD HCT with CBT for MDS have been published, an older study by the CIBMTR showed that OS and DFS after MSD were similar to those after 8/8 allele-MUD for F I G U R E 3 Forest plots of the adjusted hazard ratios (HR) and 95% confidence intervals (CI) of mortality in the subgroup analysis.MDS. 30 However, several studies have shown that younger MUDs HCT has a better survival rate compared to older MSDs restricted to MDS patients older than 50 years. 1,2Indeed, our previous study showed that OS and DFS were comparable after MSD, 8/8 allele-MUD, 7/8 allele-MUD, and UCB transplantations for elderly MDS. 3 Importantly, our current study showed that CBT was not inferior to MSD HCT in terms of OS, DFS, and NRM not only for advanced MDS patients over 50 years of age but also for those under 50 years of age.Therefore, our data demonstrated that, for adult patients with advanced MDS, UCB did not show a disadvantage in survival compared to MSD HCT.
5][16][17][18] A lower and slower hematopoietic recovery is the major limitation of CBT for adults, which could contribute to higher risks of NRM and infectious complications early after CBT.In fact, infection and graft failure are more common causes of death after CBT compared with MSD HCT.
Furthermore, the lower and slower hematopoietic recovery might contribute to higher blood component costs and prolonged hospitalization after CBT compared to MSD HCT. 31,32Indeed, along with the use of UCB, the occurrence of acute GVHD, infections, and acute kidney injury (AKI) were significant predictors of longer hospital length of stay. 314][35] Moreover, extended hospitalization following CBT may result in higher room-and-board expenses, greater consumption of pharmaceutical and laboratory utilities, and higher blood component prices, resulting in larger absolute costs for UCB transplantation during the first 100 days compared to MSD transplantation. 32All these findings could contribute to the greater cost of CBT in the early phase of HCT.
Therefore, the choice between UCB and MSD may also take their cost-effectiveness into account.
7][38] The development of chronic GVHD adversely affects not only physical and mental health but the social quality of life in allogeneic HCT survivors. 39,40deed, CBT resulted in a higher rate of immunosuppressive therapy discontinuation, a higher Karnofsky performance scale score, and a higher rate of return to school or full-time employment, compared with MSD and MUD transplantation, primarily due to the reduction of chronic GVHD. 37,41Moreover, UCB recipients have a lower healthcare burden from the time of HCT through to year five compared with other graft sources. 42In addition, previous data by the JSTCT adult MDS working group showed a survival benefit in MDS patients with chronic GVHD but not acute GVHD after singleunit CBT. 9 In contrast, our previous study demonstrated that only MSD recipients with limited chronic GVHD and UCB recipients with grade I-II acute GVHD showed a survival benefit from the graftversus-MDS effect in advanced MDS. 43 Therefore, the presence of mild acute and chronic GVHD could have contributed to the survival benefit for advanced MDS in both MSD and UCB recipients.
This study had several limitations.First, it was a retrospective study using a registry.The treating doctors' preferences and those of the center were not taken into consideration when choosing the donor.Thus, selection bias and the center effect could have affected the results of our study.Second, the use of post-transplant cyclophosphamide as GVHD prophylaxis has led to a significant increase in haploidentical-related donor transplants for various hematological diseases, including MDS. 11,[44][45][46] Our previous study demonstrated comparable survival rates between haploidentical-related donor transplants and CBT in adult patients with MDS. 11Donor selection might therefore be dependent not only on donor availability, patient and disease circumstances but also on physician and center preference.Third, our findings should be interpreted cautiously when extended to other racial cohorts, because CBT in Japan has several distinctive traits, including the use of single-unit UCB for adults, more flexible selection criteria for UCB units, and a broader indication for advanced disease. 13,47Moreover, GVHD has a varied effect on outcomes following adult single-unit CBT in European and Japanese populations. 48In the United States and Europe, the number of transplants from haploidentical-related donors is growing while CBT is dropping. 49Based on these situations, our findings may be less directly relevant to cohorts of different races.Despite such limitations, the strength of our study was the relatively larger patient population and the focus on advanced MDS in comparing the efficacy and safety of MSD HCT and single CBT.
In summary, higher rates of hematopoietic recovery in MSD HCT were counterbalanced by lower incidences of chronic GVHD in CBT, leading to comparable survival outcomes for advanced MDS.
Therefore, both MSD and UCB should be considered as the primary donor options for advanced MDS patients who required urgent allogeneic HCT.

1
Unadjusted probabilities of overall survival (OS) (A) and disease-free survival (DFS) (B) and the unadjusted cumulative incidences of relapse (C) and non-relapse mortality (NRM) (D) in MSD hematopoietic cell transplantation (HCT) and single cord blood transplantation (CBT) for advanced MDS in the entire cohort.

Table 1
Characteristics of patient, disease, and transplantation.
Among the entire cohort, subgroup analyses showed that the effects of donor type on OS were not significant between MSD and UCB recipients based on age, recipients' gender, PS, cytogenetics, prior azacytidine treatment, prior chemotherapeutic treatment,T A B L E 2Note: The p values in bold are statistically significant (<0.05).Abbreviations: Aza, azacytidine; CI, confidence interval; DFS, disease-free survival; GVHD, graft-versus-host disease; HCT, hematopoietic cell transplantation; HR, hazard ratio; MAC, myeloablative conditioning; MSD, matched sibling donor; MTX, methotrexate; NRM, non-relapse mortality; OS, overall survival; PS, performance status; RIC, reduced-intensity conditioning; UCB, unrelated cord blood.