Blood Res (2025) 60:7
Published online January 23, 2025
https://doi.org/10.1007/s44313-025-00057-7
© The Korean Society of Hematology
Correspondence to : Hyoung Jin Kang
kanghj@snu.ac.kr
Purpose This study compared the outcomes of haploidentical-related donor (HRD) and umbilical cord blood (UCB) hematopoietic stem cell transplantation (HSCT) in pediatric patients with hematologic malignancies.
Methods Data on patients who underwent HRD HSCT with post-transplant cyclophosphamide (n = 41) and UCB HSCT (n = 24) after targeted busulfan-based myeloablative conditioning with intensive pharmacokinetic monitoring between 2009 and 2018 were retrospectively analyzed.
Results The median follow-up durations in the HRD and UCB groups were 7.0 and 10.9 years, respectively. The cumulative incidence of acute graft-versus-host disease (GVHD) grades II–IV and moderate-to-severe chronic GVHD did not differ significantly between the groups. However, the HRD group demonstrated significantly lower rates of acute GVHD grades III–IV (4.9% vs. 29.2%, p = 0.009) and non-relapse mortality (2.6% vs. 34.2%, p < 0.001) but a higher relapse incidence (32.1% vs. 8.8%, p = 0.004) than the UCB group. The 5-year event-free and overall survival rates were 65.8% and 54.2% (p = 0.204) and 78.0% and 65.7% (p = 0.142) for the HRD and UCB groups, respectively. Multivariate analysis identified disease status as a significant risk factor for overall survival (hazard ratio, 3.24; p = 0.016). Additionally, UCB HSCT exhibited a trend toward worse event-free survival compared to HRD HSCT (hazard ratio, 2.63; p = 0.05).
Conclusions These findings indicate that HRD HSCT with post-transplant cyclophosphamide provides promising outcomes compared to UCB HSCT in pediatric patients, with a trend toward improved survival over a long-term follow-up period exceeding a median of 7 years. Thus, HRD HSCT may be a valuable option for pediatric patients without human leukocyte antigen-matched donors.
Keywords Haploidentical, Post-transplant cyclophosphamide, Cord blood, Children, Hematologic malignancy
Haploidentical-related donor (HRD) and umbilical cord blood (UCB) stem cell sources are alternative donor strategies for hematopoietic stem cell transplantation (HSCT) when a matched sibling or unrelated donor is unavailable These alternatives are particularly valuable given that the likelihood of finding a matched unrelated donor varies significantly with ethnicity, ranging from 25 to 80% [1]. Recently, the use of UCB has declined in cases where a matched donor is unavailable, while HRD has become more widely adopted [2]. The advantages of UCB include a lower risk of chronic graft-versus-host disease (GVHD) and rapid graft availability. However, slower immune recovery and higher rates of opportunistic infections remain significant challenges [3, 4, 5, 6–7]. In contrast, HRD HSCT approaches using either T-cell-depleted or T-cell-replete methods have shown promising outcomes, including high engraftment rates and low GVHD and non-relapse mortality (NRM) rates [8, 9, 10–11]. Among the T-cell-replete HRD HSCT methods, recent findings indicate that post-transplantation cyclophosphamide (PTCy) for GVHD prophylaxis improves leukemia- and GVHD-free survival compared with antithymocyte globulin (ATG) [12].
The choice between UCB and HRD in patients without suitable human leukocyte antigen (HLA)-matched donors remains unclear. While several studies have shown no significant differences in outcomes between UCB and HRD HSCT in adults with hematologic malignancies[13, 14], recent studies, including a prospective randomized trial, have shown better results for HRD in adults [15, 16, 17–18]. However, comparative studies in pediatric patients are limited. A retrospective study comparing T-cell-depleted, reduced-toxicity HRD HSCT with UCB HSCT in children with acute leukemia reported higher leukemia-free survival with HRD HSCT [19]. Similarly, Mo et al. observed improved clinical outcomes with HRD HSCT using ATG compared with UCB HSCT [20, 21]. In contrast, a Spanish multicenter study of pediatric acute myeloid leukemia reported comparable outcomes between HRD and UCB HSCT [22]. However, variations in treatment protocols and institutional differences complicate the ability to draw definitive conclusions.
Since 2009, our center has implemented a targeted busulfan-based myeloablative conditioning regimen with intensive pharmacokinetic (PK) monitoring and ATG-based GVHD prophylaxis for matched donors and UCB HSCT, which has demonstrated promising outcomes.[23] Additionally, we previously reported favorable results for HRD HSCT using PTCy-based GVHD prophylaxis in conjunction with this targeted busulfan-based conditioning approach [24]. Despite these advances, few studies have compared HRD HSCT with PTCy or UCB HSCT with ATG in pediatric and adolescent patients with hematological malignancies in Korea. Given the differing timelines for HRD and UCB HSCT at our institution, periodic comparisons of unrelated HSCT outcomes were performed to evaluate treatment differences. This study further investigated and compared the impact of graft source on outcomes in children and adolescents with hematologic malignancies following targeted busulfan-based myeloablative conditioning with PK monitoring.
The outcomes of 65 patients with high-risk hematologic malignancies who received their first allogeneic HSCT from HRD (n = 41) or UCB (double-unit, n = 21; single-unit, n = 3) at Seoul National University Children's Hospital between January 2009 and December 2018 were retrospectively analyzed. All patients underwent an intensive PK monitoring targeted busulfan-based myeloablative conditioning regimen.
UCB HSCT was conducted at our institution until 2016, whereas HRD HSCT with PTCy was initiated in 2014; therefore, the comparisons considered potential outcome differences due to timing. Consequently, the unrelated HSCT group (n = 87) served as the control group and was divided into two cohorts: the unrelated 09–13 (n = 51) and unrelated 14–18 (n = 36) groups, which received HSCT from 2009–2013 and 2014–2018, respectively. These groups were first compared as controls for temporal differences to facilitate comparisons between the HRD and UCB groups. During the overlap period from 2014 to 2016, UCB was prioritized when no suitable related or unrelated donors were available, with HRD selected if an appropriate UCB donor was not identified.
The Institutional Review Board of Seoul National University Hospital approved the procedure for reviewing the medical records and waived the requirement for obtaining consent (H-1107–024–368).
When selecting unrelated donors or HRDs, HLA-A, HLA-B, HLA-C, and HLA-DRB1 matching was confirmed in all patients and donors using high-resolution molecular methods, with a preference for HRDs with the killer cell immunoglobulin-like receptor B haplotype. For UCB donor selection, HLA-A and -B serological typing and HLA-DR allele typing were performed.
All patients received busulfan-based myeloablative conditioning regimens. An initial dose of busulfan (120 mg/m2 for patients aged ≥ 1 year and 80 mg/m2 for those aged < 1 year) was administered intravenously on day −8. Subsequent doses were adjusted daily from days −7 to −5, based on the therapeutic drug monitoring results from the previous day. The total target area under the curve (AUC) for busulfan was set at 74–76 mg × h/L [23].
The HRD group received a conditioning regimen consisting of targeted busulfan, fludarabine (40 mg/m2 intravenously once daily from days −8 to −4), and cyclophosphamide (14.5 mg/kg intravenously once daily from days −3 to −2). Most patients in the UCB and unrelated groups received a regimen of targeted busulfan, fludarabine (40 mg/m2 intravenously once daily from days −8 to −4) ± etoposide (20 mg/kg intravenously once daily from days −4 to −2), along with ATG (thymoglobulin, Genzyme Transplant, Cambridge, MA; 2.5 mg/kg intravenously once daily from days −8 to −6 for the UCB group or −4 to −2 for the unrelated group).
For GVHD prophylaxis, the HRD group received PTCy (50 mg/kg intravenously once daily from days 3 to 4), along with tacrolimus and mycophenolate mofetil. The UCB group was treated with cyclosporine and mycophenolate mofetil, whereas the unrelated group received tacrolimus and methotrexate. Prophylactic treatments for veno-occlusive diseases and infections were administered according to institutional HSCT guidelines.[23]
Neutrophil engraftment was defined as the first of three consecutive days with a neutrophil count > 0.5 × 10⁹/L, while platelet engraftment was defined as the first of three consecutive days with a platelet count > 20 × 10⁹/L, without any transfusions for at least 7 days. Acute and chronic GVHD were diagnosed and graded according to standard criteria. [25, 26] Regimen-related toxicity, excluding GVHD, was graded up to 42 days post-transplantation using the National Cancer Institute Common Toxicity Criteria (v4.03).
Categorical variables were compared using the chi-square test, while continuous variables were analyzed using Student's t-test or one-way analysis of variance. The incidence rates of relapse, NRM, and GVHD were calculated using the cumulative incidence function. In this context, the competing risk for relapse was NRM; the competing risk for NRM was relapse; and the competing risks for GVHD included graft failure and NRM. Events were defined as death, relapse, or graft failure. Survival analyses were performed using the Kaplan–Meier method. Differences in the cumulative incidence curves were assessed using Gray’s test, while differences in survival rates were evaluated using the log-rank test. A Cox proportional hazards regression model was employed for the multivariate analysis of prognostic factors affecting survival, utilizing the backward elimination method (p < 0.05), and included independent variables with p < 0.2. Statistical significance was set at p < 0.05. Statistical analyses were performed using R version 3.2.2 (
www.r-project.org) and IBM SPSS Statistics for Windows, version 29.0.2.0 (IBM Corp., Armonk, NY, USA).
The clinical characteristics of the patients in the UCB (n = 24) and HRD (n = 41) groups are summarized in Table 1. The median age at HSCT was significantly lower in the UCB group than in the HRD group (2.4 vs. 11.3 years old, p = 0.004). The distribution of diagnoses and remission statuses at HSCT did not differ significantly between the groups. All patients underwent targeted busulfan- and fludarabine-based myeloablative conditioning. Etoposide was administered to patients in the UCB group with acute lymphoblastic or high-risk acute myeloid leukemia. All patients in the HRD group received a regimen consisting of busulfan, fludarabine, and cyclophosphamide with peripheral blood as the stem cell source. All patients in the UCB group underwent double-unit UCB HSCT, except for three who received a single unit. The median follow-up durations did not differ significantly between the UCB and HRD groups (10.9 years vs. 7.0 years). Detailed patient characteristics and comparisons, including those of the unrelated groups, are presented in Supplementary Table 1.
Patient characteristics
UCB (n = 24) | HRD (n = 41) | p-value | |
---|---|---|---|
Median age, years (IQR) | 2.4 (1.4–6.1) | 11.3 (5.8–14.2) | 0.004 |
Sex, No. (%) | 0.05 | ||
Male | 8 (33.3%) | 24 (58.5%) | |
Female | 16 (66.7%) | 17 (41.5%) | |
Median BSA, m2 (IQR) | 0.55 (0.48–0.80) | 1.11 (0.67–1.54) | < 0.001 |
Median body weight, kg (IQR) | 12.23 (9.8–21.1) | 31.2 (15.7–50.6) | < 0.001 |
Diagnosis, No. (%) | 0.271 | ||
Acute lymphoblastic leukemia | 10 (41.7%) | 16 (39.0%) | |
Acute myeloid leukemia | 10 (41.7%) | 14 (34.1%) | |
Myelodysplastic syndromea | 0 (0.0%) | 3 (7.3%) | |
Malignant lymphoma | 0 (0.0%) | 6 (14.6%) | |
Othersb | 4 (12.5%) | 2 (4.9%) | |
Conditioning regimen | < 0.001 | ||
Bu + Flu | 8 (33.3%) | 0 (0.0%) | |
Bu + Flu + VP | 14 (58.3%) | 0 (0.0%) | |
Bu + Mel + (Flu or Cy) | 2 (8.3%) | 0 (0.0%) | |
Bu + Flu + Cy | 0 (0.0%) | 41 (100.0%) | |
Status | 0.993 | ||
CR1 | 17 (70.8%) | 29 (70.7%) | |
≥ CR2 or persistence | 7 (29.2%) | 12 (29.3%) | |
Infused busulfan AUC, mg x h/L (IQR) | 73.7 (72.0–75.2) | 74.5 (74.0–76.0) | 0.074 |
Median follow-up years (IQR) | 10.9 (0.4–14.1) | 7.0 (4.4–8.7) | 0.318 |
UCB umbilical cord blood; HRD haploidentical related donor; IQR interquartile range; BSA body surface area; Bu busulfan; Flu fludarabine; VP etoposide; Mel melphalan; Cy cyclophosphamide; CR complete remission; AUC area under the curve
aTwo with therapy-related myelodysplastic syndrome, and one with myelodysplastic syndrome, with excess blasts in the HRD group
bTwo with juvenile myelomonocytic leukemia, one with mixed-phenotype acute leukemia, one with malignant histiocytosis in the UCB group, and two with mixed-phenotype acute leukemia in the HRD group
First, we compared the clinical outcomes of the unrelated 09–13 and 14–18 groups to investigate any differences in HSCT outcomes related to the timing at our institution. The 5-year relapse incidences in the unrelated 09–13 and 14–18 groups were 13.7% and 22.2%, respectively (p = 0.483). Furthermore, the 5-year cumulative incidence rate (CIR) of NRM did not differ significantly between the groups, at 9.8% for unrelated 09–13 versus 5.6% for unrelated 14–18 (p = 0.459). The 5-year event-free survival (EFS) and overall survival (OS) rates were 74.4% and 72.2% (p = 0.956) and 81.9% and 80.5% (p = 0.973), respectively (Supplementary Fig. 1). Given that the same conditioning regimen and similar supportive care were applied in our institution, the outcomes did not differ significantly between unrelated HSCTs performed from 2009–2013 and those performed from 2014–2018.
The median infused total nucleated and CD34 + cell counts were 14.8 × 10⁸/kg and 8.3 × 10⁶/kg in the HRD group, compared with 7.18 × 10⁷/kg/unit and 3.7 × 105/kg/unit (at cryopreservation) in the UCB group. All but three patients in the UCB group received double-unit UCB. The median total nucleated cell count and CD34 + cell count infused from double-unit UCB were 14.28 × 10⁷/kg (range, 6.92–40.5 × 10⁷/kg) and 6.95 × 105/kg (range, 3.25–21.97 × 105/kg), respectively. The median time to neutrophil engraftment was 15 days (range, 13–21 days) in the HRD group and 14 days (range, 12–40 days) in the UCB group. Despite similar medians, neutrophil engraftment was significantly lower in the HRD group than in the UCB group (p = 0.036) (Fig. 1A). Additionally, neutrophil engraftment pattern in the HRD group was more predictive and consistent than that in the UCB group, with all patients in the HRD group achieving neutrophil engraftment, compared with 95.8% in the UCB group (one primary engraftment failure). Platelet engraftment occurred significantly faster in the HRD group than in the UCB group (median 26 days [range, 13–71 days] vs. 46 days [range, 21–77 days], p < 0.001) (Fig. 1B).
The incidence of hepatic veno-occlusive disease, the cytomegalovirus (CMV) antigenemia positivity rate, and antigenemia levels ≥ 10/200,000 cells did not differ significantly between the HRD and UCB groups. However, higher proportions of patients in the UCB group experienced ≥ grade 3 liver enzyme elevation (50.0% vs. 19.5%, p = 0.01) and hyperbilirubinemia (37.5% vs. 2.4%, p < 0.001). In contrast, hemorrhagic cystitis > grade 3 occurred more frequently in the HRD group (31.7% vs. 8.3%, p = 0.031).
Regarding acute GVHD grades II–IV, the CIR did not differ significantly between the HRD and UCB groups (41.5% vs. 54.2%; p = 0.589). However, the CIR of acute GVHD grades III–IV in the HRD group was significantly lower than that in the UCB group (4.9% vs. 29.2%, p = 0.009) (Figs. 2A and B). The CIRs of moderate-to-severe chronic GVHD were similar between the HRD and UCB groups (14.6% and 12.5%, respectively; p = 0.833).
The 5-year relapse rate was significantly higher in the HRD group than in the UCB group (32.1% vs. 8.8%; p = 0.004) (Fig. 2C). Conversely, the CIR of the 5-year NRM was markedly lower in the HRD group than in the UCB group (2.6% vs. 34.2%; p < 0.001) (Fig. 2D). Causes of NRM in the UCB group included five cases of pneumonia (with or without acute respiratory distress syndrome), two cases of hepatic veno-occlusive disease accompanied by pulmonary hemorrhage, and one case of hepatic failure. In contrast, the sole cause of NRM in the HRD group was pneumonia associated with acute respiratory distress syndrome.
The 5-year EFS and OS rates for the HRD and UCB groups were 65.8% (95% confidence interval [CI] 51.3–80.3%) versus 54.2% (95% CI 34.2–74.2%) (p = 0.204) and 78.0% (95% CI 65.3–90.7%) versus 65.7% (95% CI 46.3–85.1%) (p = 0.142), respectively (Fig. 3C–D). In multivariate analysis, UCB HSCT demonstrated a trend toward worse 5-year EFS compared with HRD HSCT, with a hazard ratio (HR) of 2.63 (95% CI 1.00–6.93; p = 0.05). Regarding the 5-year OS rates, disease status at HSCT, specifically second complete remission or higher, was a significant risk factor (HR 3.24, 95% CI 1.25–8.43; p = 0.016) (Table 2).
Multivariate analysis of EFS and OS rates in the HRD and UCB groups (N = 65)
5-year EFS | 5-year OS | |||||||
---|---|---|---|---|---|---|---|---|
Univariate | Multivariate | Univariate | Multivariate | |||||
Probability ± SE | p-value | HR (95% CI) | p-value | Probability ± SE | p-value | HR (95% CI) | p-value | |
HSCT type | 0.204 | 0.05 | 0.142 | 0.113 | ||||
HRD (n = 41) | 65.8 ± 7.4% | 1 | 78.0 ± 6.5% | 1 | ||||
CBT (n = 24) | 54.2 ± 10.2% | 2.63 (1.00–6.93) | 65.7 ± 9.9% | 2.17 (0.83–5.65) | ||||
Sex | 0.468 | 0.75 | ||||||
Male (n = 32) | 65.6 ± 8.4% | 74.4 ± 7.8% | ||||||
Female (n = 33) | 57.6 ± 8.6% | 72.4 ± 7.8% | ||||||
Age at HSCT (years) | 0.232 | 0.224 | 0.742 | |||||
2–10 (n = 27) | 51.9 ± 9.6% | 1 | 69.1 ± 9.1% | |||||
< 2 (n = 15) | 80.0 ± 10.3% | 0.30 (0.08–1.18) | 0.084 | 80.0 ± 10.3% | ||||
> 10 (n = 23) | 60.6 ± 10.3% | 0.82 (0.30–2.27) | 0.702 | 73.9 ± 9.2% | ||||
Disease status | 0.009 | 0.104 | 0.053 | 0.016 | ||||
CR1 (n = 46) | 71.7 ± 6.6% | 1 | 84.4 ± 5.4% | 1 | ||||
≥ CR2 (n = 19) | 36.8 ± 11.1% | 2.24 (0.85–5.94) | 63.2 ± 11.1% | 3.24 (1.25–8.43) | ||||
Diagnosis | 0.241 | 0.187 | 0.723 | |||||
ALL (n = 26) | 65.4 ± 9.3% | 1 | 76.9 ± 8.3% | |||||
AML (n = 24) | 54.2 ± 10.2% | 2.07 (0.77–5.60) | 0.151 | 69.5 ± 9.7% | ||||
Malignant lymphoma (n = 6) | 83.3 ± 15.2% | 0.49 (0.06–4.04) | 0.507 | 83.3 ± 15.2 | ||||
Myelodysplastic syndrome (n = 4) | 25.0 ± 21.7% | 0.73 (0.08–6.30) | 0.771 | 50.0 ± 25.0% | ||||
Others (n = 5) | 80.0 ± 17.9% | 3.81 (0.85–17.53) | 0.086 | 80.0 ± 17.9% |
ALL acute lymphoblastic leukemia; AML acute myeloid leukemia; AUC area under the curve; CI confidence interval; CR complete remission; EFS event-free survival; HR hazard ratio; HRD haploidentical related donor; HSCT hematopoietic stem cell transplantation; OS overall survival; SE standard error; UCB umbilical cord blood
The results of this retrospective study demonstrated that unmanipulated HRD HSCT using PTCy was associated with faster engraftment and a lower incidence of NRM in pediatric patients with hematologic malignancies. These benefits contributed to improved EFS and OS rates compared with UCB HSCT using ATG in the context of an intensively pharmacokinetically monitored busulfan-based myeloablative conditioning regimen. To our knowledge, this is the first study in Korea to compare HRD HSCT with PTCy and UCB HSCT in pediatric patients, despite previous studies reporting favorable outcomes for T-cell-depleted or ATG-based HRD HSCT compared with UCB HSCT in pediatric populations [19, 20–21].
Most studies comparing HRD and UCB HSCT, have focused on adults. While most reported better outcomes for HRD HSCT, including lower rates of GVHD and NRM[13, 15, 17, 18], limitations due to the retrospective study design and patient heterogeneity have made definitive conclusions challenging. A prospective randomized trial using a myeloablative regimen in adults with hematologic malignancies showed improved outcomes for HRD HSCT with PTCy compared with ATG-containing single-unit UCB HSCT [16]. Conversely, a recent randomized trial in the US (BMT CTN 1101) demonstrated similar 2-year progression-free survival rates between adults administered reduced-intensity, unmanipulated HRD HSCT or PTCy with double-unit UCB HSCT [27]. Given the challenges of patient accrual and variability in institutional HSCT practices, conducting a prospective randomized comparative study in pediatric patients may be difficult. Owing to the limited data in pediatric populations, this study holds potential clinical significance by offering insights into the selection of alternative donors for unmanipulated HRD HSCT with PTCy.
The results of this study also demonstrated that a lower cumulative incidence of NRM and grade III–IV acute GVHD in the HRD group contributed to favorable EFS and OS rates compared with those in the UCB group. In the UCB group, the primary causes of NRM were infectious complications or hepatic toxicities, mostly occurring within seven months after HSCT. Notably, only one case of NRM occurred in the HRD group. Predictable neutrophil engraftment in the HRD group likely contributed to the reduction in early infectious complications after transplantation compared with the UCB group. Although the busulfan AUC in the UCB group was lower than that in the HRD group, this finding suggests that the higher NRM rates in the UCB group were not attributable to busulfan exposure. Additionally, while the reduced NRM rate in the HRD group could potentially be attributed to improved supportive care over time, the NRM rates did not differ significantly between the unrelated 09–13 and 14–18 cohorts. The introduction of PK-guided busulfan in our myeloablative conditioning regimen may have also contributed to the reduced HSCT-related toxicity. Although the relapse incidence was lower in the UCB group than in the HRD group, caution is warranted when interpreting this finding as high early NRM in the UCB group serves as a competing risk factor for relapse.
Given the recent advancements in HRD HSCT with PTCy, its relatively straightforward learning curve, and faster donor acquisition and engraftment times that facilitate more predictable clinical courses[11], our institution has prioritized HRD HSCT over UCB in cases without matched donors since 2016. This shift makes direct comparisons between HRD and UCB HSCT within the same timeframe challenging. During the transitional period from 2014 to 2016, UCB was the preferred alternative donor source when matched donors were not available. However, from 2016 onward, we adjusted our donor selection criteria to focus on reducing early NRM rates. Although outcomes did not differ significantly across the unrelated donor groups based on the period of treatment at our institution, caution should be exercised when comparing HRD and UCB HSCT outcomes in the pediatric population.
This study has several limitations. First, the retrospective comparison between the two groups was influenced by differing timelines. Notably, the UCB group received busulfan-based chemotherapy with ATG. Previous research has shown that total body irradiation offers better outcomes than chemoconditioning in pediatric patients with acute lymphoblastic leukemia undergoing UCB-HSCT[28], highlighting opportunities for further optimization. Additionally, while haploidentical donors with the KIR B haplotype were prioritized, a younger donor age was identified as a critical prognostic factor for HRD HSCT with PTCY [29, 30]. This factor is particularly relevant for pediatric patients who may have the option of receiving stem cells from a parent or sibling.
In summary, the results of this study suggest that HRD-HSCT with PTCy, combined with an intensive PK-monitored, targeted busulfan-based myeloablative conditioning regimen, is a safe and promising alternative for pediatric patients with hematological diseases who lack an HLA-matched donor. Nevertheless, prospective studies in children and adolescents are essential to enable more robust comparisons between these two stem cell sources.
HRD Haploidentical related donor
UCB Umbilical cord blood
HSCT Hematopoietic stem cell transplantation
GVHD Graft-versus-host disease
NRM Non-relapse mortality
PTCy Post-transplantation cyclophosphamide
ATG Anti-thymocyte globulin
PK Pharmacokinetics
AUC Area under the curve
CIR Cumulative incidence rate
EFS Event-free survival
OS Overall survival
HR Hazard ratio
CI Confidence interval
KTH participated in study design, data collection, statistical analysis/interpretation, manuscript drafting, and manuscript revisions. BKK, HYA, and JYC collected data. SHS, KSY, and IJJ participated in manuscript editing. HJK participated in study design and manuscript editing. All authors have read and approved the manuscript.
This study was supported by the Korean Society of Hematology (No. ICKSH-2022–08), and a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI) funded by the Ministry of Health & Welfare, Republic of Korea (No. HI14C1277).
The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.
Ethics approval and consent to participate
The Institutional Review Board of Seoul National University Hospital approved the procedure for reviewing the medical records and waived the requirement for obtaining consent (H-1107–024-368).
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
The online version contains supplementary material available at https://doi.org/10.1007/s44313-025-00057-7.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Blood Res 2025; 60():
Published online January 23, 2025 https://doi.org/10.1007/s44313-025-00057-7
Copyright © The Korean Society of Hematology.
Kyung Taek Hong1, Bo Kyung Kim1, Hong Yul An1, Jung Yoon Choi1, Sang Hoon Song2, Kyung‑Sang Yu3, In‑Jin Jang3 and Hyoung Jin Kang1,4*
1 Department of Pediatrics, Seoul National University College of Medicine, Seoul National University Cancer Research Institute, Seoul, Republic of Korea. 2 Department of Laboratory Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea. 3 Department of Clinical Pharmacology and Therapeutics, Seoul National University College of Medicine, Seoul, Republic of Korea. 4 Wide River Institute of Immunology, Seoul National University, Hongcheon, Republic of Korea.
Correspondence to:Hyoung Jin Kang
kanghj@snu.ac.kr
Purpose This study compared the outcomes of haploidentical-related donor (HRD) and umbilical cord blood (UCB) hematopoietic stem cell transplantation (HSCT) in pediatric patients with hematologic malignancies.
Methods Data on patients who underwent HRD HSCT with post-transplant cyclophosphamide (n = 41) and UCB HSCT (n = 24) after targeted busulfan-based myeloablative conditioning with intensive pharmacokinetic monitoring between 2009 and 2018 were retrospectively analyzed.
Results The median follow-up durations in the HRD and UCB groups were 7.0 and 10.9 years, respectively. The cumulative incidence of acute graft-versus-host disease (GVHD) grades II–IV and moderate-to-severe chronic GVHD did not differ significantly between the groups. However, the HRD group demonstrated significantly lower rates of acute GVHD grades III–IV (4.9% vs. 29.2%, p = 0.009) and non-relapse mortality (2.6% vs. 34.2%, p < 0.001) but a higher relapse incidence (32.1% vs. 8.8%, p = 0.004) than the UCB group. The 5-year event-free and overall survival rates were 65.8% and 54.2% (p = 0.204) and 78.0% and 65.7% (p = 0.142) for the HRD and UCB groups, respectively. Multivariate analysis identified disease status as a significant risk factor for overall survival (hazard ratio, 3.24; p = 0.016). Additionally, UCB HSCT exhibited a trend toward worse event-free survival compared to HRD HSCT (hazard ratio, 2.63; p = 0.05).
Conclusions These findings indicate that HRD HSCT with post-transplant cyclophosphamide provides promising outcomes compared to UCB HSCT in pediatric patients, with a trend toward improved survival over a long-term follow-up period exceeding a median of 7 years. Thus, HRD HSCT may be a valuable option for pediatric patients without human leukocyte antigen-matched donors.
Keywords: Haploidentical, Post-transplant cyclophosphamide, Cord blood, Children, Hematologic malignancy
Haploidentical-related donor (HRD) and umbilical cord blood (UCB) stem cell sources are alternative donor strategies for hematopoietic stem cell transplantation (HSCT) when a matched sibling or unrelated donor is unavailable These alternatives are particularly valuable given that the likelihood of finding a matched unrelated donor varies significantly with ethnicity, ranging from 25 to 80% [1]. Recently, the use of UCB has declined in cases where a matched donor is unavailable, while HRD has become more widely adopted [2]. The advantages of UCB include a lower risk of chronic graft-versus-host disease (GVHD) and rapid graft availability. However, slower immune recovery and higher rates of opportunistic infections remain significant challenges [3, 4, 5, 6–7]. In contrast, HRD HSCT approaches using either T-cell-depleted or T-cell-replete methods have shown promising outcomes, including high engraftment rates and low GVHD and non-relapse mortality (NRM) rates [8, 9, 10–11]. Among the T-cell-replete HRD HSCT methods, recent findings indicate that post-transplantation cyclophosphamide (PTCy) for GVHD prophylaxis improves leukemia- and GVHD-free survival compared with antithymocyte globulin (ATG) [12].
The choice between UCB and HRD in patients without suitable human leukocyte antigen (HLA)-matched donors remains unclear. While several studies have shown no significant differences in outcomes between UCB and HRD HSCT in adults with hematologic malignancies[13, 14], recent studies, including a prospective randomized trial, have shown better results for HRD in adults [15, 16, 17–18]. However, comparative studies in pediatric patients are limited. A retrospective study comparing T-cell-depleted, reduced-toxicity HRD HSCT with UCB HSCT in children with acute leukemia reported higher leukemia-free survival with HRD HSCT [19]. Similarly, Mo et al. observed improved clinical outcomes with HRD HSCT using ATG compared with UCB HSCT [20, 21]. In contrast, a Spanish multicenter study of pediatric acute myeloid leukemia reported comparable outcomes between HRD and UCB HSCT [22]. However, variations in treatment protocols and institutional differences complicate the ability to draw definitive conclusions.
Since 2009, our center has implemented a targeted busulfan-based myeloablative conditioning regimen with intensive pharmacokinetic (PK) monitoring and ATG-based GVHD prophylaxis for matched donors and UCB HSCT, which has demonstrated promising outcomes.[23] Additionally, we previously reported favorable results for HRD HSCT using PTCy-based GVHD prophylaxis in conjunction with this targeted busulfan-based conditioning approach [24]. Despite these advances, few studies have compared HRD HSCT with PTCy or UCB HSCT with ATG in pediatric and adolescent patients with hematological malignancies in Korea. Given the differing timelines for HRD and UCB HSCT at our institution, periodic comparisons of unrelated HSCT outcomes were performed to evaluate treatment differences. This study further investigated and compared the impact of graft source on outcomes in children and adolescents with hematologic malignancies following targeted busulfan-based myeloablative conditioning with PK monitoring.
The outcomes of 65 patients with high-risk hematologic malignancies who received their first allogeneic HSCT from HRD (n = 41) or UCB (double-unit, n = 21; single-unit, n = 3) at Seoul National University Children's Hospital between January 2009 and December 2018 were retrospectively analyzed. All patients underwent an intensive PK monitoring targeted busulfan-based myeloablative conditioning regimen.
UCB HSCT was conducted at our institution until 2016, whereas HRD HSCT with PTCy was initiated in 2014; therefore, the comparisons considered potential outcome differences due to timing. Consequently, the unrelated HSCT group (n = 87) served as the control group and was divided into two cohorts: the unrelated 09–13 (n = 51) and unrelated 14–18 (n = 36) groups, which received HSCT from 2009–2013 and 2014–2018, respectively. These groups were first compared as controls for temporal differences to facilitate comparisons between the HRD and UCB groups. During the overlap period from 2014 to 2016, UCB was prioritized when no suitable related or unrelated donors were available, with HRD selected if an appropriate UCB donor was not identified.
The Institutional Review Board of Seoul National University Hospital approved the procedure for reviewing the medical records and waived the requirement for obtaining consent (H-1107–024–368).
When selecting unrelated donors or HRDs, HLA-A, HLA-B, HLA-C, and HLA-DRB1 matching was confirmed in all patients and donors using high-resolution molecular methods, with a preference for HRDs with the killer cell immunoglobulin-like receptor B haplotype. For UCB donor selection, HLA-A and -B serological typing and HLA-DR allele typing were performed.
All patients received busulfan-based myeloablative conditioning regimens. An initial dose of busulfan (120 mg/m2 for patients aged ≥ 1 year and 80 mg/m2 for those aged < 1 year) was administered intravenously on day −8. Subsequent doses were adjusted daily from days −7 to −5, based on the therapeutic drug monitoring results from the previous day. The total target area under the curve (AUC) for busulfan was set at 74–76 mg × h/L [23].
The HRD group received a conditioning regimen consisting of targeted busulfan, fludarabine (40 mg/m2 intravenously once daily from days −8 to −4), and cyclophosphamide (14.5 mg/kg intravenously once daily from days −3 to −2). Most patients in the UCB and unrelated groups received a regimen of targeted busulfan, fludarabine (40 mg/m2 intravenously once daily from days −8 to −4) ± etoposide (20 mg/kg intravenously once daily from days −4 to −2), along with ATG (thymoglobulin, Genzyme Transplant, Cambridge, MA; 2.5 mg/kg intravenously once daily from days −8 to −6 for the UCB group or −4 to −2 for the unrelated group).
For GVHD prophylaxis, the HRD group received PTCy (50 mg/kg intravenously once daily from days 3 to 4), along with tacrolimus and mycophenolate mofetil. The UCB group was treated with cyclosporine and mycophenolate mofetil, whereas the unrelated group received tacrolimus and methotrexate. Prophylactic treatments for veno-occlusive diseases and infections were administered according to institutional HSCT guidelines.[23]
Neutrophil engraftment was defined as the first of three consecutive days with a neutrophil count > 0.5 × 10⁹/L, while platelet engraftment was defined as the first of three consecutive days with a platelet count > 20 × 10⁹/L, without any transfusions for at least 7 days. Acute and chronic GVHD were diagnosed and graded according to standard criteria. [25, 26] Regimen-related toxicity, excluding GVHD, was graded up to 42 days post-transplantation using the National Cancer Institute Common Toxicity Criteria (v4.03).
Categorical variables were compared using the chi-square test, while continuous variables were analyzed using Student's t-test or one-way analysis of variance. The incidence rates of relapse, NRM, and GVHD were calculated using the cumulative incidence function. In this context, the competing risk for relapse was NRM; the competing risk for NRM was relapse; and the competing risks for GVHD included graft failure and NRM. Events were defined as death, relapse, or graft failure. Survival analyses were performed using the Kaplan–Meier method. Differences in the cumulative incidence curves were assessed using Gray’s test, while differences in survival rates were evaluated using the log-rank test. A Cox proportional hazards regression model was employed for the multivariate analysis of prognostic factors affecting survival, utilizing the backward elimination method (p < 0.05), and included independent variables with p < 0.2. Statistical significance was set at p < 0.05. Statistical analyses were performed using R version 3.2.2 (
www.r-project.org) and IBM SPSS Statistics for Windows, version 29.0.2.0 (IBM Corp., Armonk, NY, USA).
The clinical characteristics of the patients in the UCB (n = 24) and HRD (n = 41) groups are summarized in Table 1. The median age at HSCT was significantly lower in the UCB group than in the HRD group (2.4 vs. 11.3 years old, p = 0.004). The distribution of diagnoses and remission statuses at HSCT did not differ significantly between the groups. All patients underwent targeted busulfan- and fludarabine-based myeloablative conditioning. Etoposide was administered to patients in the UCB group with acute lymphoblastic or high-risk acute myeloid leukemia. All patients in the HRD group received a regimen consisting of busulfan, fludarabine, and cyclophosphamide with peripheral blood as the stem cell source. All patients in the UCB group underwent double-unit UCB HSCT, except for three who received a single unit. The median follow-up durations did not differ significantly between the UCB and HRD groups (10.9 years vs. 7.0 years). Detailed patient characteristics and comparisons, including those of the unrelated groups, are presented in Supplementary Table 1.
Patient characteristics.
UCB (n = 24). | HRD (n = 41). | p-value. | |
---|---|---|---|
Median age, years (IQR). | 2.4 (1.4–6.1). | 11.3 (5.8–14.2). | 0.004. |
Sex, No. (%). | 0.05. | ||
Male. | 8 (33.3%). | 24 (58.5%). | |
Female. | 16 (66.7%). | 17 (41.5%). | |
Median BSA, m2 (IQR). | 0.55 (0.48–0.80). | 1.11 (0.67–1.54). | < 0.001. |
Median body weight, kg (IQR). | 12.23 (9.8–21.1). | 31.2 (15.7–50.6). | < 0.001. |
Diagnosis, No. (%). | 0.271. | ||
Acute lymphoblastic leukemia. | 10 (41.7%). | 16 (39.0%). | |
Acute myeloid leukemia. | 10 (41.7%). | 14 (34.1%). | |
Myelodysplastic syndromea. | 0 (0.0%). | 3 (7.3%). | |
Malignant lymphoma. | 0 (0.0%). | 6 (14.6%). | |
Othersb. | 4 (12.5%). | 2 (4.9%). | |
Conditioning regimen. | < 0.001. | ||
Bu + Flu. | 8 (33.3%). | 0 (0.0%). | |
Bu + Flu + VP. | 14 (58.3%). | 0 (0.0%). | |
Bu + Mel + (Flu or Cy). | 2 (8.3%). | 0 (0.0%). | |
Bu + Flu + Cy. | 0 (0.0%). | 41 (100.0%). | |
Status. | 0.993. | ||
CR1. | 17 (70.8%). | 29 (70.7%). | |
≥ CR2 or persistence. | 7 (29.2%). | 12 (29.3%). | |
Infused busulfan AUC, mg x h/L (IQR). | 73.7 (72.0–75.2). | 74.5 (74.0–76.0). | 0.074. |
Median follow-up years (IQR). | 10.9 (0.4–14.1). | 7.0 (4.4–8.7). | 0.318. |
UCB umbilical cord blood; HRD haploidentical related donor; IQR interquartile range; BSA body surface area; Bu busulfan; Flu fludarabine; VP etoposide; Mel melphalan; Cy cyclophosphamide; CR complete remission; AUC area under the curve.
aTwo with therapy-related myelodysplastic syndrome, and one with myelodysplastic syndrome, with excess blasts in the HRD group.
bTwo with juvenile myelomonocytic leukemia, one with mixed-phenotype acute leukemia, one with malignant histiocytosis in the UCB group, and two with mixed-phenotype acute leukemia in the HRD group.
First, we compared the clinical outcomes of the unrelated 09–13 and 14–18 groups to investigate any differences in HSCT outcomes related to the timing at our institution. The 5-year relapse incidences in the unrelated 09–13 and 14–18 groups were 13.7% and 22.2%, respectively (p = 0.483). Furthermore, the 5-year cumulative incidence rate (CIR) of NRM did not differ significantly between the groups, at 9.8% for unrelated 09–13 versus 5.6% for unrelated 14–18 (p = 0.459). The 5-year event-free survival (EFS) and overall survival (OS) rates were 74.4% and 72.2% (p = 0.956) and 81.9% and 80.5% (p = 0.973), respectively (Supplementary Fig. 1). Given that the same conditioning regimen and similar supportive care were applied in our institution, the outcomes did not differ significantly between unrelated HSCTs performed from 2009–2013 and those performed from 2014–2018.
The median infused total nucleated and CD34 + cell counts were 14.8 × 10⁸/kg and 8.3 × 10⁶/kg in the HRD group, compared with 7.18 × 10⁷/kg/unit and 3.7 × 105/kg/unit (at cryopreservation) in the UCB group. All but three patients in the UCB group received double-unit UCB. The median total nucleated cell count and CD34 + cell count infused from double-unit UCB were 14.28 × 10⁷/kg (range, 6.92–40.5 × 10⁷/kg) and 6.95 × 105/kg (range, 3.25–21.97 × 105/kg), respectively. The median time to neutrophil engraftment was 15 days (range, 13–21 days) in the HRD group and 14 days (range, 12–40 days) in the UCB group. Despite similar medians, neutrophil engraftment was significantly lower in the HRD group than in the UCB group (p = 0.036) (Fig. 1A). Additionally, neutrophil engraftment pattern in the HRD group was more predictive and consistent than that in the UCB group, with all patients in the HRD group achieving neutrophil engraftment, compared with 95.8% in the UCB group (one primary engraftment failure). Platelet engraftment occurred significantly faster in the HRD group than in the UCB group (median 26 days [range, 13–71 days] vs. 46 days [range, 21–77 days], p < 0.001) (Fig. 1B).
The incidence of hepatic veno-occlusive disease, the cytomegalovirus (CMV) antigenemia positivity rate, and antigenemia levels ≥ 10/200,000 cells did not differ significantly between the HRD and UCB groups. However, higher proportions of patients in the UCB group experienced ≥ grade 3 liver enzyme elevation (50.0% vs. 19.5%, p = 0.01) and hyperbilirubinemia (37.5% vs. 2.4%, p < 0.001). In contrast, hemorrhagic cystitis > grade 3 occurred more frequently in the HRD group (31.7% vs. 8.3%, p = 0.031).
Regarding acute GVHD grades II–IV, the CIR did not differ significantly between the HRD and UCB groups (41.5% vs. 54.2%; p = 0.589). However, the CIR of acute GVHD grades III–IV in the HRD group was significantly lower than that in the UCB group (4.9% vs. 29.2%, p = 0.009) (Figs. 2A and B). The CIRs of moderate-to-severe chronic GVHD were similar between the HRD and UCB groups (14.6% and 12.5%, respectively; p = 0.833).
The 5-year relapse rate was significantly higher in the HRD group than in the UCB group (32.1% vs. 8.8%; p = 0.004) (Fig. 2C). Conversely, the CIR of the 5-year NRM was markedly lower in the HRD group than in the UCB group (2.6% vs. 34.2%; p < 0.001) (Fig. 2D). Causes of NRM in the UCB group included five cases of pneumonia (with or without acute respiratory distress syndrome), two cases of hepatic veno-occlusive disease accompanied by pulmonary hemorrhage, and one case of hepatic failure. In contrast, the sole cause of NRM in the HRD group was pneumonia associated with acute respiratory distress syndrome.
The 5-year EFS and OS rates for the HRD and UCB groups were 65.8% (95% confidence interval [CI] 51.3–80.3%) versus 54.2% (95% CI 34.2–74.2%) (p = 0.204) and 78.0% (95% CI 65.3–90.7%) versus 65.7% (95% CI 46.3–85.1%) (p = 0.142), respectively (Fig. 3C–D). In multivariate analysis, UCB HSCT demonstrated a trend toward worse 5-year EFS compared with HRD HSCT, with a hazard ratio (HR) of 2.63 (95% CI 1.00–6.93; p = 0.05). Regarding the 5-year OS rates, disease status at HSCT, specifically second complete remission or higher, was a significant risk factor (HR 3.24, 95% CI 1.25–8.43; p = 0.016) (Table 2).
Multivariate analysis of EFS and OS rates in the HRD and UCB groups (N = 65).
5-year EFS. | 5-year OS. | |||||||
---|---|---|---|---|---|---|---|---|
Univariate. | Multivariate. | Univariate. | Multivariate. | |||||
Probability ± SE. | p-value. | HR (95% CI). | p-value. | Probability ± SE. | p-value. | HR (95% CI). | p-value. | |
HSCT type. | 0.204. | 0.05. | 0.142. | 0.113. | ||||
HRD (n = 41). | 65.8 ± 7.4%. | 1. | 78.0 ± 6.5%. | 1. | ||||
CBT (n = 24). | 54.2 ± 10.2%. | 2.63 (1.00–6.93). | 65.7 ± 9.9%. | 2.17 (0.83–5.65). | ||||
Sex. | 0.468. | 0.75. | ||||||
Male (n = 32). | 65.6 ± 8.4%. | 74.4 ± 7.8%. | ||||||
Female (n = 33). | 57.6 ± 8.6%. | 72.4 ± 7.8%. | ||||||
Age at HSCT (years). | 0.232. | 0.224. | 0.742. | |||||
2–10 (n = 27). | 51.9 ± 9.6%. | 1. | 69.1 ± 9.1%. | |||||
< 2 (n = 15). | 80.0 ± 10.3%. | 0.30 (0.08–1.18). | 0.084. | 80.0 ± 10.3%. | ||||
> 10 (n = 23). | 60.6 ± 10.3%. | 0.82 (0.30–2.27). | 0.702. | 73.9 ± 9.2%. | ||||
Disease status. | 0.009. | 0.104. | 0.053. | 0.016. | ||||
CR1 (n = 46). | 71.7 ± 6.6%. | 1. | 84.4 ± 5.4%. | 1. | ||||
≥ CR2 (n = 19). | 36.8 ± 11.1%. | 2.24 (0.85–5.94). | 63.2 ± 11.1%. | 3.24 (1.25–8.43). | ||||
Diagnosis. | 0.241. | 0.187. | 0.723. | |||||
ALL (n = 26). | 65.4 ± 9.3%. | 1. | 76.9 ± 8.3%. | |||||
AML (n = 24). | 54.2 ± 10.2%. | 2.07 (0.77–5.60). | 0.151. | 69.5 ± 9.7%. | ||||
Malignant lymphoma (n = 6). | 83.3 ± 15.2%. | 0.49 (0.06–4.04). | 0.507. | 83.3 ± 15.2. | ||||
Myelodysplastic syndrome (n = 4). | 25.0 ± 21.7%. | 0.73 (0.08–6.30). | 0.771. | 50.0 ± 25.0%. | ||||
Others (n = 5). | 80.0 ± 17.9%. | 3.81 (0.85–17.53). | 0.086. | 80.0 ± 17.9%. |
ALL acute lymphoblastic leukemia; AML acute myeloid leukemia; AUC area under the curve; CI confidence interval; CR complete remission; EFS event-free survival; HR hazard ratio; HRD haploidentical related donor; HSCT hematopoietic stem cell transplantation; OS overall survival; SE standard error; UCB umbilical cord blood.
The results of this retrospective study demonstrated that unmanipulated HRD HSCT using PTCy was associated with faster engraftment and a lower incidence of NRM in pediatric patients with hematologic malignancies. These benefits contributed to improved EFS and OS rates compared with UCB HSCT using ATG in the context of an intensively pharmacokinetically monitored busulfan-based myeloablative conditioning regimen. To our knowledge, this is the first study in Korea to compare HRD HSCT with PTCy and UCB HSCT in pediatric patients, despite previous studies reporting favorable outcomes for T-cell-depleted or ATG-based HRD HSCT compared with UCB HSCT in pediatric populations [19, 20–21].
Most studies comparing HRD and UCB HSCT, have focused on adults. While most reported better outcomes for HRD HSCT, including lower rates of GVHD and NRM[13, 15, 17, 18], limitations due to the retrospective study design and patient heterogeneity have made definitive conclusions challenging. A prospective randomized trial using a myeloablative regimen in adults with hematologic malignancies showed improved outcomes for HRD HSCT with PTCy compared with ATG-containing single-unit UCB HSCT [16]. Conversely, a recent randomized trial in the US (BMT CTN 1101) demonstrated similar 2-year progression-free survival rates between adults administered reduced-intensity, unmanipulated HRD HSCT or PTCy with double-unit UCB HSCT [27]. Given the challenges of patient accrual and variability in institutional HSCT practices, conducting a prospective randomized comparative study in pediatric patients may be difficult. Owing to the limited data in pediatric populations, this study holds potential clinical significance by offering insights into the selection of alternative donors for unmanipulated HRD HSCT with PTCy.
The results of this study also demonstrated that a lower cumulative incidence of NRM and grade III–IV acute GVHD in the HRD group contributed to favorable EFS and OS rates compared with those in the UCB group. In the UCB group, the primary causes of NRM were infectious complications or hepatic toxicities, mostly occurring within seven months after HSCT. Notably, only one case of NRM occurred in the HRD group. Predictable neutrophil engraftment in the HRD group likely contributed to the reduction in early infectious complications after transplantation compared with the UCB group. Although the busulfan AUC in the UCB group was lower than that in the HRD group, this finding suggests that the higher NRM rates in the UCB group were not attributable to busulfan exposure. Additionally, while the reduced NRM rate in the HRD group could potentially be attributed to improved supportive care over time, the NRM rates did not differ significantly between the unrelated 09–13 and 14–18 cohorts. The introduction of PK-guided busulfan in our myeloablative conditioning regimen may have also contributed to the reduced HSCT-related toxicity. Although the relapse incidence was lower in the UCB group than in the HRD group, caution is warranted when interpreting this finding as high early NRM in the UCB group serves as a competing risk factor for relapse.
Given the recent advancements in HRD HSCT with PTCy, its relatively straightforward learning curve, and faster donor acquisition and engraftment times that facilitate more predictable clinical courses[11], our institution has prioritized HRD HSCT over UCB in cases without matched donors since 2016. This shift makes direct comparisons between HRD and UCB HSCT within the same timeframe challenging. During the transitional period from 2014 to 2016, UCB was the preferred alternative donor source when matched donors were not available. However, from 2016 onward, we adjusted our donor selection criteria to focus on reducing early NRM rates. Although outcomes did not differ significantly across the unrelated donor groups based on the period of treatment at our institution, caution should be exercised when comparing HRD and UCB HSCT outcomes in the pediatric population.
This study has several limitations. First, the retrospective comparison between the two groups was influenced by differing timelines. Notably, the UCB group received busulfan-based chemotherapy with ATG. Previous research has shown that total body irradiation offers better outcomes than chemoconditioning in pediatric patients with acute lymphoblastic leukemia undergoing UCB-HSCT[28], highlighting opportunities for further optimization. Additionally, while haploidentical donors with the KIR B haplotype were prioritized, a younger donor age was identified as a critical prognostic factor for HRD HSCT with PTCY [29, 30]. This factor is particularly relevant for pediatric patients who may have the option of receiving stem cells from a parent or sibling.
In summary, the results of this study suggest that HRD-HSCT with PTCy, combined with an intensive PK-monitored, targeted busulfan-based myeloablative conditioning regimen, is a safe and promising alternative for pediatric patients with hematological diseases who lack an HLA-matched donor. Nevertheless, prospective studies in children and adolescents are essential to enable more robust comparisons between these two stem cell sources.
HRD Haploidentical related donor
UCB Umbilical cord blood
HSCT Hematopoietic stem cell transplantation
GVHD Graft-versus-host disease
NRM Non-relapse mortality
PTCy Post-transplantation cyclophosphamide
ATG Anti-thymocyte globulin
PK Pharmacokinetics
AUC Area under the curve
CIR Cumulative incidence rate
EFS Event-free survival
OS Overall survival
HR Hazard ratio
CI Confidence interval
KTH participated in study design, data collection, statistical analysis/interpretation, manuscript drafting, and manuscript revisions. BKK, HYA, and JYC collected data. SHS, KSY, and IJJ participated in manuscript editing. HJK participated in study design and manuscript editing. All authors have read and approved the manuscript.
This study was supported by the Korean Society of Hematology (No. ICKSH-2022–08), and a grant from the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI) funded by the Ministry of Health & Welfare, Republic of Korea (No. HI14C1277).
The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.
Ethics approval and consent to participate
The Institutional Review Board of Seoul National University Hospital approved the procedure for reviewing the medical records and waived the requirement for obtaining consent (H-1107–024-368).
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
The online version contains supplementary material available at https://doi.org/10.1007/s44313-025-00057-7.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Patient characteristics.
UCB (n = 24). | HRD (n = 41). | p-value. | |
---|---|---|---|
Median age, years (IQR). | 2.4 (1.4–6.1). | 11.3 (5.8–14.2). | 0.004. |
Sex, No. (%). | 0.05. | ||
Male. | 8 (33.3%). | 24 (58.5%). | |
Female. | 16 (66.7%). | 17 (41.5%). | |
Median BSA, m2 (IQR). | 0.55 (0.48–0.80). | 1.11 (0.67–1.54). | < 0.001. |
Median body weight, kg (IQR). | 12.23 (9.8–21.1). | 31.2 (15.7–50.6). | < 0.001. |
Diagnosis, No. (%). | 0.271. | ||
Acute lymphoblastic leukemia. | 10 (41.7%). | 16 (39.0%). | |
Acute myeloid leukemia. | 10 (41.7%). | 14 (34.1%). | |
Myelodysplastic syndromea. | 0 (0.0%). | 3 (7.3%). | |
Malignant lymphoma. | 0 (0.0%). | 6 (14.6%). | |
Othersb. | 4 (12.5%). | 2 (4.9%). | |
Conditioning regimen. | < 0.001. | ||
Bu + Flu. | 8 (33.3%). | 0 (0.0%). | |
Bu + Flu + VP. | 14 (58.3%). | 0 (0.0%). | |
Bu + Mel + (Flu or Cy). | 2 (8.3%). | 0 (0.0%). | |
Bu + Flu + Cy. | 0 (0.0%). | 41 (100.0%). | |
Status. | 0.993. | ||
CR1. | 17 (70.8%). | 29 (70.7%). | |
≥ CR2 or persistence. | 7 (29.2%). | 12 (29.3%). | |
Infused busulfan AUC, mg x h/L (IQR). | 73.7 (72.0–75.2). | 74.5 (74.0–76.0). | 0.074. |
Median follow-up years (IQR). | 10.9 (0.4–14.1). | 7.0 (4.4–8.7). | 0.318. |
UCB umbilical cord blood; HRD haploidentical related donor; IQR interquartile range; BSA body surface area; Bu busulfan; Flu fludarabine; VP etoposide; Mel melphalan; Cy cyclophosphamide; CR complete remission; AUC area under the curve.
aTwo with therapy-related myelodysplastic syndrome, and one with myelodysplastic syndrome, with excess blasts in the HRD group.
bTwo with juvenile myelomonocytic leukemia, one with mixed-phenotype acute leukemia, one with malignant histiocytosis in the UCB group, and two with mixed-phenotype acute leukemia in the HRD group.
Multivariate analysis of EFS and OS rates in the HRD and UCB groups (N = 65).
5-year EFS. | 5-year OS. | |||||||
---|---|---|---|---|---|---|---|---|
Univariate. | Multivariate. | Univariate. | Multivariate. | |||||
Probability ± SE. | p-value. | HR (95% CI). | p-value. | Probability ± SE. | p-value. | HR (95% CI). | p-value. | |
HSCT type. | 0.204. | 0.05. | 0.142. | 0.113. | ||||
HRD (n = 41). | 65.8 ± 7.4%. | 1. | 78.0 ± 6.5%. | 1. | ||||
CBT (n = 24). | 54.2 ± 10.2%. | 2.63 (1.00–6.93). | 65.7 ± 9.9%. | 2.17 (0.83–5.65). | ||||
Sex. | 0.468. | 0.75. | ||||||
Male (n = 32). | 65.6 ± 8.4%. | 74.4 ± 7.8%. | ||||||
Female (n = 33). | 57.6 ± 8.6%. | 72.4 ± 7.8%. | ||||||
Age at HSCT (years). | 0.232. | 0.224. | 0.742. | |||||
2–10 (n = 27). | 51.9 ± 9.6%. | 1. | 69.1 ± 9.1%. | |||||
< 2 (n = 15). | 80.0 ± 10.3%. | 0.30 (0.08–1.18). | 0.084. | 80.0 ± 10.3%. | ||||
> 10 (n = 23). | 60.6 ± 10.3%. | 0.82 (0.30–2.27). | 0.702. | 73.9 ± 9.2%. | ||||
Disease status. | 0.009. | 0.104. | 0.053. | 0.016. | ||||
CR1 (n = 46). | 71.7 ± 6.6%. | 1. | 84.4 ± 5.4%. | 1. | ||||
≥ CR2 (n = 19). | 36.8 ± 11.1%. | 2.24 (0.85–5.94). | 63.2 ± 11.1%. | 3.24 (1.25–8.43). | ||||
Diagnosis. | 0.241. | 0.187. | 0.723. | |||||
ALL (n = 26). | 65.4 ± 9.3%. | 1. | 76.9 ± 8.3%. | |||||
AML (n = 24). | 54.2 ± 10.2%. | 2.07 (0.77–5.60). | 0.151. | 69.5 ± 9.7%. | ||||
Malignant lymphoma (n = 6). | 83.3 ± 15.2%. | 0.49 (0.06–4.04). | 0.507. | 83.3 ± 15.2. | ||||
Myelodysplastic syndrome (n = 4). | 25.0 ± 21.7%. | 0.73 (0.08–6.30). | 0.771. | 50.0 ± 25.0%. | ||||
Others (n = 5). | 80.0 ± 17.9%. | 3.81 (0.85–17.53). | 0.086. | 80.0 ± 17.9%. |
ALL acute lymphoblastic leukemia; AML acute myeloid leukemia; AUC area under the curve; CI confidence interval; CR complete remission; EFS event-free survival; HR hazard ratio; HRD haploidentical related donor; HSCT hematopoietic stem cell transplantation; OS overall survival; SE standard error; UCB umbilical cord blood.
Wonjin Jang, Suejung Jo, Jae Won Yoo, Seongkoo Kim, Jae Wook Lee, Pil-Sang Jang, Nack-Gyun Chung, Bin Cho
Blood Res 2022; 57(4): 256-263Woojung Jeon, Young Kwon Koh, Sunghan Kang, Hyery Kim, Kyung-Nam Koh, Ho Joon Im
Blood Res 2022; 57(1): 41-50Ezzatollah Fathi, Zohreh Sanaat, Raheleh Farahzadi
Blood Res 2019; 54(3): 165-174