Abstract
Severe aplastic anemia (SAA) is a life-threatening hematological disorder, and the optimal treatment for patients aged ≥40 years remains controversial. This retrospective study aims to conduct a comparative analysis of the efficacy of immunosuppressive therapy (IST) versus haploidentical hematopoietic stem cell transplantation (haplo-HSCT) specifically in elderly patients. To minimize selection bias, propensity score matching (PSM) was performed to pair samples based on age and Eastern Cooperative Oncology Group score, thereby addressing treatment responses and survival outcomes while ensuring balanced baseline characteristics between the 2 groups. The study reveals that while overall survival (OS) rates were similar (69.6% ± 13.3% vs. 68.6% ± 12.1%, P = 0.80), haplo-HSCT significantly outperformed IST in terms of failure-free survival (FFS) rates (68.7% ± 12.1% vs. 46.5% ± 20.2%, P = 0.025). Multivariate Cox regression analysis confirmed that treatment modality and disease severity were independent predictors of FFS. These findings support haplo-HSCT as a viable first-line treatment option for appropriately selected elderly SAA patients, providing evidence for treatment options in patients with SAA aged 40 and above.
Keywords: severe aplastic anemia, haploidentical hematopoietic stem cell transplantation, immunosuppressive therapy, propensity score matching, elderly patients
Graphical Abstract
Graphical Abstract.
Significance statement.
This retrospective study addresses a critical gap in severe aplastic anemia (SAA) care for patients aged ≥40, where immunosuppressive therapy (IST) remains standard despite limitations. Using propensity score matching, it demonstrates that haploidentical hematopoietic stem cell transplantation (haplo-HSCT) yields superior failure-free survival versus IST, with comparable overall survival. These findings challenge age-based treatment dogma, validate haplo-HSCT as a viable first-line option for appropriately selected older SAA patients, and inform updated clinical guidelines to improve long-term outcomes in this underserved population.
Introduction
Severe aplastic anemia (SAA) is a critical hematological disorder. Hematopoietic stem cell transplantation (HSCT) and immunosuppressive therapy (IST) are 2 effective options, decided based on patient age and disease status. According to the most recent British Society for Hematology (BSH) guidelines, for patients under 40 years with a matched related donor (MRD), HSCT is the first-line choice due to better failure free survival.1,2 For those patients over 40 years, IST remains the recommended first-line treatment. This recommendation is largely based on historical data showing increased transplant-related mortality (TRM) in older patients undergoing conventional myeloablative conditioning. Despite significant refinements in IST regimens over the past 2 decades, including the addition of eltrombopag and improved supportive care,3,4 substantial limitations persist. Long-term clonal evolution to myelodysplastic syndrome (MDS) or acute myeloid leukemia (AML), occurring in 10%-20% of patients; and a relapse rate of approximately 30%-40%. Consequently, long-term failure-free survival (FFS) with IST remains suboptimal, particularly in older adults with comorbidities.5
In parallel, the therapeutic landscape of allo-HSCT for SAA has evolved dramatically.6 Recent international studies have further optimized and validated the application of haploidentical HSCT in SAA treatment.7 The multicenter prospective BMT CTN 1502 trial from North America confirmed that haploidentical bone marrow transplantation with post-transplant cyclophosphamide (PTCy)-based prophylaxis achieves favorable 1-year overall survival (OS) of 81% and minimal graft-versus-host disease (GVHD) for patients with refractory or relapsed SAA.7 Consistently, the large-scale European Society for Blood and Marrow Transplantation Severe Aplastic Anemia Working Party (EBMT SAAWP) registry data demonstrated the safety and feasibility of PTCy prophylaxis in SAA patients receiving haploidentical, matched sibling, and unrelated donor transplantation, while also revealing relatively higher non-relapse mortality and inferior survival outcomes in haploidentical cohorts compared with other donor types.8 Beyond these advances in transplant modality optimization, emerging evidence has consistently challenged the conventional age barrier for transplantation. A landmark prospective study demonstrated that first-line matched sibling donor (MSD) allo-HSCT in SAA patients over 40 years yielded superior FFS compared to IST, establishing proof-of-concept that carefully selected older patients can benefit from upfront transplantation.9,10 Another registry data from the Chinese Blood and Marrow Transplant Registry Group (CBMTRG) have shown that in patients aged 40-50 years with SAA, outcomes following haploidentical HSCT were comparable to those after MSD transplantation, with OS rates exceeding 80%.9 In terms of patients age, retrospective data analysis of the CBMTRG cohort indicates good long-term results of HSCT in elderly SAA adults regardless of donor type.6,11,12 These collective findings strongly suggest that donor source may not be the primary determinant of survival in carefully selected older patients.
Despite these emerging data, a critical knowledge gap remains regarding the optimal approach in the era of modern haploidentical HSCT. Therefore, we conducted this retrospective study focusing on SAA patients older than 40 years old. To balance the bias introduced by baseline differences, particularly age as the key variable of interest in our study, we performed propensity score matching (PSM) based on age and ECOG score, directly comparing the outcomes of IST versus haploidentical HSCT in patients with SAA aged 40 years and older. By elucidating the differences in outcomes between these 2 groups, the study seeks to provide contemporary, real-world evidence to help redefine the age threshold and selection criteria for upfront transplantation, ultimately informing treatment algorithms for this underserved patient population.
Methods
This retrospective study was approved by the Ethics Committee of Peking University People’s Hospital. The requirement for written informed consent was waived by the institutional review board due to its retrospective design, and all clinical procedures were performed in accordance with the Declaration of Helsinki. After being counseled on the advantages and disadvantages of both treatment options, patient guardians selected either IST or haploidentical transplantation as first-line therapy.
IST and evaluation
Sixty-six patients with SAA aged over 40 years (range, 40-78 years) receiving IST at Peking University People’s Hospital between August 2016 and April 2025 were initially enrolled (Table 1). After PSM with the haploidentical HSCT group, 33 patients aged 40-65 years remained for final analysis.
Table 1.
Patient characteristics.
| All patients |
Patients matched by propensity score |
||||||
|---|---|---|---|---|---|---|---|
| IST (n = 66) | HSCT (n = 69) | P | IST (n = 33) | HSCT (n = 33) | P | ||
| Age, median (range) | 56 (40, 78) | 44 (40, 64) | <.001 | 49 (40, 64) | 47 (40, 64) | .396 | |
| Gender (%) | Male | 26 (39.4) | 41 (59.4) | .025 | 13 (39.4) | 20 (60.6) | .139 |
| Female | 40 (60.6) | 28 (40.6) | 20 (60.6) | 13 (39.4) | |||
| Severity (%) | SAA | 53 (80.3) | 52 (75.4) | .539 | 27 (81.8) | 28 (84.8) | 1.00 |
| vSAA | 13 (19.7) | 17 (24.6) | 6 (18.2) | 5 (15.2) | |||
| Disease course, months, (median [IQR]) | 2 [1, 2] | 24 [8, 168] | <.001 | 2 [1, 2] | 24 [11, 120] | <.001 | |
| ECOG (%) | 0 | 18 (27.3) | 19 (27.5) | <.001 | 14 (42.4) | 8 (24.2) | .17 |
| 1 | 22 (33.3) | 49 (71.0) | 16 (48.5) | 24 (72.7) | |||
| 2 | 26 (39.4) | 1 (1.4) | 3 (9.1) | 1 (3.0) | |||
| Treatment year (%) | 2021-2025 | 43 (65.2) | 42 (60.9) | .722 | 19 (57.6) | 25 (75.8) | .191 |
| 2016-2020 | 23 (34.8) | 27 (39.1) | 14 (42.4) | 8 (24.2) | |||
| Alive (%) | 43 (65.2) | 51 (73.9) | .349 | 22 (66.7) | 23 (69.7) | 1.00 | |
Abbreviations: ECOG, Eastern Cooperative Oncology Group Score; HSCT, hematopoietic stem cell transplantation; IQR, interquartile range; IST, immunosuppressive therapy; SSA, severe aplastic anemia; vSAA, very SAA.
The IST protocol comprised: rabbit ATG (Sangstat, Lyon, France) 3-4 mg/kg/day intravenously for 5 consecutive days; prednisone 1 mg/kg/day for 10 days (initiated before first ATG dose) with rapid taper over 1-2 weeks; cyclosporine (CsA) 3-5 mg/kg/day orally twice daily for 6 months or until plateau response, then continued for 12-24 months before discontinuation; dose-adjusted to maintain 150-250 μg/L levels.
Hematologic response was assessed at 3 and 6 months post-ATG + CsA. A complete response (CR) was defined as meeting all 3 peripheral blood count criteria: (1) neutrophil count >1.5 × 10⁹/L, (2) hemoglobin level >110 g/L, and (3) platelet count >100 × 10⁹/L. A partial response (PR) was defined as neutrophil count >0.5 × 10⁹/L, hemoglobin level >80 g/L, platelet count >20 × 10⁹/L, and transfusion independence.13,14
Haploidentical transplantation
Sixty-nine SAA/very SAA (vSAA) patients aged over 40 years (range, 40-65 years) receiving haploidentical transplantation at Peking University People’s Hospital between August 2016 and April 2025 were initially enrolled; all met the diagnostic criteria for severe or very SAA. After PSM with the IST cohort, 33 patients aged 40-65 years were retained for final analysis.
The haploidentical SCT conditioning regimen consisted of intravenous busulfan (0.8 mg/kg 4 times daily, days −8 to −7), cyclophosphamide (25 mg/kg daily, days −5 to −2), fludarabine (Flu) 30 mg/m2 days −6 to −2, and rabbit-ATG (2.5 mg/kg daily, days −5 to −2). Stem cell source was G-CSF-primed bone marrow combined with peripheral blood. GVHD prophylaxis included CsA, mycophenolate mofetil, and short-course methotrexate. Additional procedural details have been previously reported.15–18
Definitions
vSAA or SAA was diagnosed according to guidelines.19 OS was defined as the interval from treatment initiation to death. Performance status prior to transplantation and at the last follow-up was assessed using the Eastern Cooperative Oncology Group (ECOG) scale.
In the HSCT group, neutrophil and platelet engraftment was defined as previous report.15 Failures after HSCT was defined as death, primary and secondary graft failure and relapse, whichever occurred first. Primary graft failure (GF) was defined as failure of myeloid engraftment until 28 days after transplantation. Secondary GF was defined as absence of graft function after full engraftment. Acute and chronic GVHD (aGVHD and cGvHD) were graded using standard criteria.19–21 Hematopoietic Cell Transplantation Comorbidity Index (HCT-CI) used to assess transplant risk.22,23 Epstein-Barr virus (EBV) DNAemia > 400 copies/mL, cytomegalovirus (CMV) > 1000 copies/mL were the thresholds for positive results.
In the IST group, IST treatment failure encompassed death, absence of response at or beyond 6 months, disease progression necessitating intervention, relapse, and clonal evolution.
Statistical analysis
Patient characteristics were compared using chi-square or Fisher’s exact tests for categorical variables, and the Mann–Whitney U test for continuous variables. OS and FFS were measured using the Kaplan–Meier estimator and compared using the log-rank test. The Univariate and multivariate analyses were performed using a Cox proportional hazard regression model. Univariate and multivariate analysis were performed using a Cox proportional hazard regression model. The statistical analyses were performed under R version 4.4.1. P < .05 was considered statistically significant; *P < .05, **P < .01, ***P < .001.
Results
Patients characteristics
The demographic characteristics of the IST and HSCT groups are presented in Table 1. The 2 groups were comparable in age, gender, disease severity, year of treatment, ECOG score and survival rate distribution. However, significant differences were observed in disease duration before treatment. The HSCT group had significantly longer disease duration than the IST group (a median 24 months vs. 2 months before treatment), 30 of 33 (90.9%) patients had disease duration longer than 4 months. Moreover, in the HSCT cohort, 6 of 33 (18.2%) patients had experienced IST failure (Table 2), and haplo-HSCT served as second-line therapy for these patients.
Table 2.
Patients and transplant characteristics in the HSCT group.
| Variable | Level | Statistic |
|---|---|---|
| ATG pre-HSCT, n (%) | No | 27 (81.8) |
| Yes | 6 (18.2) | |
| HCT-CI, n (%) | 0 | 12 (36.4) |
| 1 | 12 (36.4) | |
| 2 | 6 (18.2) | |
| 3 | 2 (6.1) | |
| 4 | 1 (3.0) | |
| MNCs, ×108, median [IQR] | 9.36 [8.38, 12.50] | |
| CD34+, ×106, median [IQR] | 3.38 [2.34, 4.77] | |
| Donor selection, n (%) | Parents | 1 (3.0) |
| Sibling | 10 (30.3) | |
| Child | 22 (66.7) | |
| Donor-recipient gender, n (%) | Male–male | 18 (54.5) |
| Male–female | 6 (18.2) | |
| Female–Female | 7 (21.2) | |
| Female–male | 2 (6.1) | |
| ABO match, (%) | Match | 15 (45.5) |
| Minor mismatch | 10 (30.3) | |
| Major mismatch | 4 (12.1) | |
| Bidirectional mismatch | 4 (12.1) | |
| Neutrophil engraftment, days, median [IQR] | 12 [11, 16] | |
| Platelet engraftment, days, median [IQR] | 16 [11, 23] | |
| aGVHD within 100 days, n (%) | No | 25 (75.8) |
| Yes | 7 (21.2) | |
| aGVHD grade (%) | 0 | 26 (78.8) |
| I | 3 (9.1) | |
| II | 3 (9.1) | |
| III-IV | 1 (3.0) | |
| cGVHD, n (%) | No | 21 (63.6) |
| Yes | 2 (6.1) | |
| cGVHD grade (%) | 0 | 31 (93.9) |
| I | 2 (6.1) | |
| Graft failure (%) | 0 | |
| CMV (%) | No | 15 (45.5) |
| Yes | 16 (48.5) | |
| CMV positive time point, days, median [IQR] | 37 [26, 40] | |
| EBV (%) | No | 23 (69.7) |
| Yes | 8 (24.2) | |
| EBV positive time point, days, median [IQR] | 41 [37, 49] |
Abbreviations: aGvHD, acute graft-versus-host disease; cGvHD, chronic graft-versus-host disease; CMV, cytomegalovirus; EBV, Epstein-Barr Virus; HCT-CI, Hematopoietic Cell Transplantation Comorbidity Index; HSCT, hematopoietic stem cell transplantation; IQR, interquartile range.
Outcomes of IST
Patient and treatment characteristics of the IST group are shown in Table 3. Among the 33 patients in the IST group, 100% patients underwent efficacy evaluation at 3 months post-treatment. Complete remission was achieved in 2 patients (6.1%), partial remission in 19 patients (57.6%), and no response in 12 patients (36.4%). At the 6-month time point, 1 patient (3%) had died within 6 months, 3 patients (9.1%) achieved complete remission, 21 patients (63.6%) achieved partial remission, 8 patients (24.2%) had no remission, and 11 patients (33.3%) relapsed after achieving remission with a median time to relapse of 2.25 years. A total of 3 patients in the IST group underwent HSCT after treatment failure.
Table 3.
Patient and treatment characteristics of the IST group.
| Variable | Level | Statistic |
|---|---|---|
| 3 months evaluation, n (%) | CR | 2 (6.1) |
| PR | 19 (57.6) | |
| NR | 12 (36.4) | |
| 6 months evaluation, n (%) | CR | 3 (9.1) |
| PR | 21 (63.6) | |
| NR | 8 (24.2) | |
| Dead | 1 (3.0) | |
| Adverse reactions, n (%) | No | 19 (57.6) |
| ATG-related | 2 (6.1) | |
| Hemorrhage-related | 2 (6.1) | |
| Infection-related | 10 (30.3) |
Abbreviations: CR, complete response; PR, partial response; NR, non-response.
During the initial 6 months of treatment, the incidence of ATG-related adverse reactions was 6.1%, hemorrhage-related adverse reactions was 6.1%, and infection-related adverse reactions was 30.3%.
Outcomes of HSCT
Descriptions of patients and transplant characteristics in the HSCT group are provided in Table 2. Among the 33 patients in the HSCT group, 1 patient died on day 4 post-infusion due to sudden cardiac death attributed to conditioning regimen-related cardiotoxicity, apart from this patient, all others achieved successful myeloid engraftment, with median neutrophil and platelet engraftment times of 12 and 16 days, respectively. Six patients (18.2%) had received prior IST before transplantation, and nine patients (27.3%) had pretransplant HCT-CI scores of 2 or higher.
aGVHD occurred in 7 of 33 patients, with 3 patients at grade I, 3 patients at grade II, and 1 patient at grade III-IV. As of the follow-up date, cGVHD developed in 2 surviving patients (6.1%), both of whom had limited involvement.
The 33 patients in the HSCT group achieved a 96.97% ± 3.03% cumulative engraftment rate. The cumulative incidence of CMV infection at day 100 was 48.48% ± 8.89%, with a median time to infection of 37 days post-transplantation. The cumulative incidence of EBV infection at day 100 was 24.24% ± 7.6%, with a median time to infection of 41 days post-transplantation. Among patients with CMV viremia, 2 progressed to clinical CMV disease. One patient developed CMV encephalitis and recovered completely after standardized antiviral therapy. The other patient presented with CMV pneumonia accompanied by mixed bacterial and fungal infections, and succumbed to infectious toxic shock. In terms of EBV viremia, a total of 3 patients developed PTLD, consisting of 2 cases of clinically suspected PTLD and 1 case of pathologically confirmed PTLD. Both patients with clinically suspected PTLD achieved clinical remission after targeted treatment; however, 1 patient died of severe gastrointestinal hemorrhage during follow-up. The patient with pathologically confirmed PTLD showed progressive disease despite intervention, and eventually died secondary to Gram-negative bacillary infectious toxic shock.
Survival outcomes
Median follow-up was 850 days for the IST cohort and 1122 days for the haploidentical HSCT cohort. After PSM, the 3-year OS from treatment start was estimated at 69.6% ± 13.3% in the IST group versus 68.6% ± 12.1% in the HSCT group (Figure 1, P = .80). The corresponding 3-year FFS was 46.5% ± 20.2% for IST and 68.7% ± 12.1% for HSCT (Figure 2, P = .025). Prior to PSM, the 3-year OS was 73.6% ± 8.6% in the IST group and 73.3% ± 7.4% in the HSCT group (Figure S1, P = .28). The pre-matching 3-year FFS was 49.6% ± 13.6% for IST versus 73.3% ± 7.4% for HSCT (Figure S2, P < .0001).
Figure 1.
Overall survival of IST versus haplo-HSCT cohort after propensity score matching (n = 33 vs. 33, P = 0.8).
Figure 2.
Failure free survival of IST versus haplo-HSCT cohort after propensity score matching (n = 33 vs. 33, P = 0.025). haplo-HSCT, haploidentical hematopoietic stem cell transplantation; IST, immunosuppressive therapy.
We subsequently performed univariate(Table 4) and multivariate Cox regression analyses (Table 5) for OS and FFS. The results demonstrated that in univariate analysis, variables significantly influencing OS included age (HR = 1.1, P = .029) and disease severity (HR = 4.1, P = .0017), while those significantly influencing FFS encompassed treatment modality (HR = .43, P = .029), age (HR = 1.1, P = .00051) and disease severity (HR = 4.1, P = .0017) (Table 4). To further adjust for the influence of confounding factors, multivariate analysis revealed that only disease severity (HR = 5.15, P = .001) remained significantly associated with OS, whereas both treatment modality and disease severity (HR = 5.39, P = .0003) were significantly associated with FFS. The hazard ratio for the effect of treatment modality on FFS was .39 (P = .03) (Table 5), indicating that HSCT was associated with significantly improved treatment outcomes.
Table 4.
Factors associated with outcomes in univariate analysis.
| Variable | Overall survival |
Failure-free survival |
||
|---|---|---|---|---|
| HR (95% CI) | P | HR (95% CI) | P | |
| Group | 0.89 (0.38-2.1) | .8 | 0.43 (0.2-0.92) | .029 |
| Age | 1.1 (1-1.2) | .029 | 1.1 (1.1-1.2) | .00051 |
| Gender | 0.87 (0.37-2.1) | .75 | 0.93 (0.45-1.9) | .85 |
| Severity | 4.1 (1.7-10) | .0017 | 3 (1.3-6.9) | .0076 |
| Disease course | 1 (0.99 -1) | .4 | 0.99 (0.99-1) | .12 |
| ECOG | 1.3 (0.82-2) | .27 | 1.3 (0.64-2.5) | .51 |
| Treatment year | 1.6 (0.64-4.2) | .31 | 1.9 (0.83-4.2) | .13 |
Abbreviation: ECOG, Eastern Cooperative Oncology Group Score.Values in bold denote statistically significant results with P‑value < 0.05.
Table 5.
Multivariate analysis of adverse factors associated with survival outcomes.
| Variable | Overall survival |
Failure-free survival |
||
|---|---|---|---|---|
| HR | P | HR | P | |
| Group | 1.20 | .73 | 0.39 | .03 |
| Age | 1.09 | .12 | 1.08 | .06 |
| Severity | 5.15 | .001 | 5.39 | .0003 |
| ECOG | 0.86 | .73 | 1.29 | .48 |
| Treatment year | 1.71 | .36 | 2.75 | .04 |
Abbreviation: ECOG, Eastern Cooperative Oncology Group Score.Values in bold denote statistically significant results with P‑value < 0.05.
Discussion
The management of SAA in patients aged 40 years and older represents a considerable therapeutic challenge, characterized by a delicate balance between competing risks. On one hand, IST aims at modulating the aberrant immune response that suppresses hematopoiesis but carries a substantial risk of relapse and clonal evolution, threatening long-term survival. On the other hand, HSCT has the potential to restore normal hematopoiesis but is limited by TRM due to diminished organ function reserve and increased comorbidities in the elderly patients.24 These challenges underscore the critical need for comparative evaluations of treatment modalities tailored to elderly patients to optimize clinical decision-making and improve outcomes. To our knowledge, this is the first retrospective study to directly compare haploidentical HSCT with IST in these specific elderly patient population.
In the treatment of SAA, rapid hematopoietic reconstitution is an advantage of HSCT, which is conent with our study and previous reports.9,25,26 In our analysis, the complete remission rate at 3 months in the IST group was only 6.1%, and at 6 months it was merely 9.1%. In the haploidentical transplantation group, our transplantation cohort demonstrated robust outcomes, 96.97% ± 3.03% patients achieved successful engraftment, with acceptable and manageable rates of both acute and chronic GVHD and viral reactivation, the median times to neutrophil and platelet engraftment were 12 days and 16 days, respectively, with significant differences in FFS. Therefore, haploidentical HSCT is associated with significantly faster hematopoietic recovery and superior FFS, suggesting that age alone should no longer be the sole determinant in treatment selection.
Previous studies in younger SAA patients have demonstrated that HSCT is superior to IST in terms of long-term FFS.2,25–27 Due to recent advances and optimization of transplantation techniques, we have reached the same conclusion in patients over 40 years of age. In recent years, remarkable advances in conditioning regimen optimization and GVHD prophylaxis strategies have substantially expanded the therapeutic ceiling of allo-HSCT for aplastic anemia. Notably, haploidentical transplantation has progressed at an unprecedented pace. A landmark prospective cohort study provided the first definitive evidence that first-line MSD HSCT yielded superior long-term FFS compared to IST in older patients, challenging decades of established practice.26 Subsequently, a propensity score-matched analysis in Cancer Letters demonstrated equivalent outcomes between MSD and haploidentical HSCT in patients over 40, with survival rates exceeding 80% in both groups.9 Further supporting this paradigm shift, a registry-based study illustrated the feasibility and efficacy of transplantation in patients over 50 years using both haploidentical and matched donors.10 These cumulative findings have prompted Chinese consensus guidelines to recommend HSCT as a viable first-line option for patients up to 50 years of age. Building upon this evidence base, our single-center retrospective analysis provides timely, real-world comparative data to further refine treatment algorithms.
In our IST cohort, the 6-month overall response rate was 72%, which is slightly higher than the data from the French Severe Aplastic Anemia Reference Center registry, attributable to the recent addition of eltrombopag (EPAG) in IST. However, the relapse rate remained high, with a failure event rate of 66.7%, significantly higher than the 33.3% observed in the HSCT cohort. The predominant causes of mortality in this group were hemorrhagic complications and infections, directly attributable to the slow and often incomplete hematologic recovery inherent to IST. In our haploidentical transplantation cohort, over 90% of patients received CsA-based treatment before transplantation, resulting in longer disease duration. The total disease duration of included patients was significantly longer than that in the IST cohort (median disease duration 24 vs 2 months), and 18.2% of patients had experienced failed IST prior to transplantation, both of which are adverse prognostic factors for SAA treatment. Nevertheless, the outcomes remained favorable, these factors did not affect its advantage. Successful engraftment was achieved among all alive patients, with acceptable and manageable rates of both acute and chronic GVHD and viral reactivation.
From baseline characteristics, the HSCT group had longer disease duration, these mismatched baseline features may introduce result bias. However, we employed multivariate analysis to eliminate the impact of mismatched baseline characteristics. After matching for age and ECOG score as PSM pairing factors and incorporating variables with significant risk in univariate analysis, multivariate analysis revealed that treatment modality and disease severity remained key variables affecting FFS. The long-term FFS was remarkably promising, supporting the viability of this approach in appropriately selected patients.
We acknowledge several limitations in our study. Patients in the HSCT group had significantly longer disease duration than those in the IST group, and transplantation was not first-line therapy, suggesting worse organ reserve function in the HSCT group. Encouragingly, the impact of prolonged disease duration was offset by the treatment modality. Nevertheless, while the retrospective cohort design allows for the analysis of existing patient data from 1 single center, which may introduce bias in treatment allocation and patient outcomes. The relatively small sample size limits the generalizability of our findings to the broader elderly population suffering from SAA. Prospective, multicenter validation studies are needed to confirm these findings and establish clear selection criteria.
In summary, our study represents the first retrospective analysis comparing haploidentical HSCT with IST specifically in patients with SAA aged 40 years and older. We demonstrated that haploidentical HSCT was associated with faster hematopoietic reconstitution and significantly improved FFS. These results provide compelling clinical evidence to support the consideration of haploidentical transplantation as a preferred therapeutic option for elderly patients with SAA, potentially redefining the age-based treatment paradigm in the modern era of transplantation. Nevertheless, several inherent limitations of this retrospective work should be acknowledged when interpreting our findings. As a single-center non-randomized investigation, residual confounding effects may persist even after PSM. The notable disparity in disease duration between cohorts may introduce selection bias. Moreover, the moderate sample size of 33 patients per group limits statistical power for subgroup analyses and constrains the broader generalizability of our observations.
Supplementary Material
Contributor Information
Sheng-nan Gao, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Jin-song Jia, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Xiao-jun Huang, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Yuan-yuan Zhang, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Ting-ting Han, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Hai-xia Fu, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Yun He, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Yao Chen, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Jing-zhi Wang, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Xiao-dong Mo, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Yu-qian Sun, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Wei Han, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Yu Wang, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Xiao-hui Zhang, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Lan-ping Xu, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Zheng-li Xu, Peking University People’s Hospital, Peking University Institute of Hematology, National Clinical Research Center for Hematologic Disease, Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China.
Author contributions
Xiao-jun Huang, Zheng-li Xu, and Lan-ping Xu conceived and designed the research. Jin-song Jia collected the immunosuppressive therapy cohort. Xiao-hui Zhang and Yu Wang supervised the data analysis project. Yuan-yuan Zhang, Ting-ting Han, Hai-xia Fu, Yun He, Yao Chen, Jing-zhi Wang, Xiao-dong Mo, Yu-qian Sun, and Wei Han collected transplant patient data. Zheng-li Xu and Sheng-nan Gao analyzed the data and wrote the manuscript. All authors critically reviewed and approved the final version of the manuscript.
Supplementary material
Supplementary material is available at Stem Cells Translational Medicine online.
Funding
This work was supported by the “High Innovation Plan” (no. G202531207), the National Key Research and Development Program of China (no. 2024YFC2510500), the National Key Research and Development Program of China (no. 2022YFA1103300), and the Major Program of the National Natural Science Foundation of China (no. 82293630).
Conflicts of interest
None declared.
Ethics statement
Given the retrospective nature of the study, the requirement for informed consent was waived by the ethics committee. All procedures were performed in accordance with the Declaration of Helsinki.
References
- 1. Zhang L, Li J, Liang W, et al. Comparison of hematopoietic stem cell transplantation and repeated intensified immunosuppressive therapy as second-line treatment for relapsed/refractory severe aplastic anemia. Front Immunol. 2024;15:1425076. 10.3389/fimmu.2024.1425076 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Zhang MX, Wang Q, Wang XQ. Hematopoietic stem-cell transplantation versus immunosuppressive therapy in patients with adult acquired severe aplastic anemia: a cost-effectiveness analysis. Int J Gen Med. 2021;14:3529-3537. 10.2147/IJGM.S310844 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Liu L, Han B, Zhang Y, et al. First-line treatment of severe aplastic anemia: immunosuppressive therapy plus eltrombopag versus haploidentical hematopoietic stem cell transplantation, a multicenter prospective study. Bone Marrow Transplant. 2024;59:1449-1457. 10.1038/s41409-024-02377-1 [DOI] [PubMed] [Google Scholar]
- 4. Li J, Shi Y, Zhang B, et al. The efficacy of immunosuppressive therapy with or without thrombopoietin receptor agonist in elderly patients with severe aplastic anemia. Ann Hematol. 2025;104:1515-1525. 10.1007/s00277-025-06335-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Fattizzo B, Gurnari C, Giammarco S, et al. Elderly patients with aplastic anemia: treatment patterns and outcomes in the real world. Am J Hematol. 2025;100:584-591. 10.1002/ajh.27611 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Xu ZL, Xu LP, Wang SQ, et al. Who is the best allogeneic hematopoietic stem cell transplant donor for severe aplastic anemia? Data from the Chinese bone marrow transplantation registry group (CBMTRG). Sci Bull (Beijing). 2026;71:1195-1203. 10.1016/j.scib.2026.01.057 [DOI] [PubMed] [Google Scholar]
- 7. Dezern AE, Eapen M, Wu J, et al. Haploidentical bone marrow transplantation in patients with relapsed or refractory severe aplastic anaemia in the USA (BMT CTN 1502): a multicentre, single-arm, phase 2 trial. Lancet Haematol. 2022;9:e660-e669. 10.1016/S2352-3026(22)00206-X [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Montoro J, Eikema DJ, Piepenbroek B, et al. Donor impact on allogeneic transplant outcomes with PTCy for severe aplastic anemia: a study of the SAAWP EBMT. Bone Marrow Transplant. 2025;60:1152-1159. 10.1038/s41409-025-02633-y [DOI] [PubMed] [Google Scholar]
- 9. Xu ZL, Zhang YP, Ye BD, et al. Comparable survival outcomes in HLA-matched and haploidentical hematopoietic stem cell transplantation for severe aplastic anemia patients aged 40-50: a CBMTR registry-based propensity score matching analysis over the last decade. Cancer Lett. 2025;621:217594. 10.1016/j.canlet.2025.217594 [DOI] [PubMed] [Google Scholar]
- 10. Xu ZL, Xu LP, Zhang YC, et al. The outcome of allogeneic hematopoietic stem cell transplantation among elderly patients with severe aplastic anemia and a predictive model from the Chinese Blood and Marrow Transplant Registry Group. Haematologica. 2024;109:2000-2004. 10.3324/haematol.2023.284581 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Zhao X, Zhang YP, Ye BD, et al. Clinical outcomes of offspring and matched unrelated donor transplants in relapsed or refractory severe aplastic anemia patients aged 40-50: a retrospective analysis from the CBMTR over the past decade. Bone Marrow Transplant. 2025;60:1282-1285. 10.1038/s41409-025-02617-y [DOI] [PubMed] [Google Scholar]
- 12. Zhao X, Zhou M, Ye BD, et al. The eastern cooperative oncology group score rather than donor type impacts clinical outcomes of allogeneic hematopoietic stem cell transplantation in severe aplastic anemia patients aged 51-60 years: a retrospective study from the Chinese blood and marrow transplant registry. Clin Transplant. 2025;39:e70359. 10.1111/ctr.70359 [DOI] [PubMed] [Google Scholar]
- 13. Choi YB, Yi ES, Lee JW, et al. Immunosuppressive therapy versus alternative donor hematopoietic stem cell transplantation for children with severe aplastic anemia who lack an HLA-matched familial donor. Bone Marrow Transplant. 2017;52:47-52. 10.1038/bmt.2016.223 [DOI] [PubMed] [Google Scholar]
- 14. Kosaka Y, Yagasaki H, Sano K, et al. J Prospective multicenter trial comparing repeated immunosuppressive therapy with stem-cell transplantation from an alternative donor as second-line treatment for children with severe and very severe aplastic anemia. Blood. 2008;111:1054-1059. 10.1182/blood-2007-08-099168 [DOI] [PubMed] [Google Scholar]
- 15. Xu LP, Wang SQ, Wu DP, et al. Haplo-identical transplantation for acquired severe aplastic anaemia in a multicentre prospective study. Br J Haematol. 2016;175:265-274. 10.1111/bjh.14225 [DOI] [PubMed] [Google Scholar]
- 16. Xu LP, Jin S, Wang SQ, et al. Upfront haploidentical transplant for acquired severe aplastic anemia: registry-based comparison with matched related transplant. J Hematol Oncol. 2017;10:25. 10.1186/s13045-017-0398-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Huang XJ, Liu DH, Liu KY, et al. Haploidentical hematopoietic stem cell transplantation without in vitro T-cell depletion for the treatment of hematological malignancies. Bone Marrow Transplant. 2006;38:291-297. 10.1038/sj.bmt.1705445 [DOI] [PubMed] [Google Scholar]
- 18. Xu LP, Liu KY, Liu DH, et al. A novel protocol for haploidentical hematopoietic SCT without in vitro T-cell depletion in the treatment of severe acquired aplastic anemia. Bone Marrow Transplant. 2012;47:1507-1512. 10.1038/bmt.2012.79 [DOI] [PubMed] [Google Scholar]
- 19. Xu L, Zhang X, Wu D, et al. Chinese society of hematology clinical practice guidelines for the comprehensive management of allogeneic hematopoietic stem cell transplantation in patients with severe aplastic anemia. Chin Med J (Engl). 2026;139:322-340. 10.1097/CM9.0000000000003917 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Przepiorka D, Weisdorf D, Martin P, et al. 1994 consensus conference on acute GVHD grading. Bone Marrow Transplant. 1995;15:825-828. [PubMed] [Google Scholar]
- 21. Shulman HM, Sullivan KM, Weiden PL, et al. Chronic graft-versus-host syndrome in man. A long-term clinicopathologic study of 20 Seattle patients. Am J Med. 1980;69:204-217. 10.1016/0002-9343(80)90380-0 [DOI] [PubMed] [Google Scholar]
- 22. Sorror ML, Maris MB, Storb R, et al. Hematopoietic cell transplantation (HCT)-specific comorbidity index: a new tool for risk assessment before allogeneic HCT. Blood. 2005;106:2912-2919. 10.1182/blood-2005-05-2004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Sorror ML, Storb RF, Sandmaier BM, et al. Comorbidity-age index: a clinical measure of biologic age before allogeneic hematopoietic cell transplantation. J Clin Oncol. 2014;32:3249-3256. 10.1200/JCO.2013.53.8157 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Yang Y, Ji J, Tang Z, et al. Comparisons between frontline therapy and a combination of eltrombopag plus immunosuppression therapy and human leukocyte antigen-haploidentical hematopoietic stem cell transplantation in patients with severe aplastic anemia: a systematic review. Front Oncol. 2021;11:614965. 10.3389/fonc.2021.614965 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Xu ZL, Zhou M, Jia JS, et al. Immunosuppressive therapy versus haploidentical transplantation in adults with acquired severe aplastic anemia. Bone Marrow Transplant. 2019;54:1319-1326. 10.1038/s41409-018-0410-3 [DOI] [PubMed] [Google Scholar]
- 26. Liu L, Lei M, Fu R, et al. Matched related transplantation versus immunosuppressive therapy plus eltrombopag for first-line treatment of severe aplastic anemia: a multicenter, prospective study. J Hematol Oncol. 2022;15:105. 10.1186/s13045-022-01324-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Andresen F, Yoshimi A, Bader P, et al. Comparable outcomes of HSCT up-front and after failure of IST in pediatric aplastic anemia in recent years. Blood Adv. 2026;10:1992-2005. 10.1182/bloodadvances.2025018285 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.