Female Donors Raise Graft-Versus-Host Disease Risk in Male Recipients, Meta-Analysis Confirms

For decades, transplant physicians have suspected that the sex of a bone marrow donor matters when the recipient is male. A new systematic review and meta-analysis published in Annals of Hematology has now pooled the available evidence and delivered one of the clearest quantitative answers yet: transplantation from a female donor into a male recipient is associated with a significantly elevated risk of moderate-to-severe acute graft-versus-host disease, the dangerous complication in which donor immune cells attack the patient’s own tissues. The finding, drawn from cohort studies of allogeneic hematopoietic stem cell transplantation, carries immediate practical weight for clinicians weighing donor options at the bedside, and it reopens a long-running debate about the immunological consequences of crossing the sex boundary in transplantation.

The research team, led by Xiao Han, Zhuang Liu, Yi Jin and colleagues at The Affiliated Huaian No. 1 People’s Hospital of Nanjing Medical University and collaborating institutions in Jiangsu, China, focused on grade II-IV acute graft-versus-host disease, the form of the condition severe enough to demand aggressive treatment. Acute GVHD arises because the immune cells in a transplanted graft, once engrafted in the recipient’s body, can recognize host tissues as foreign and mount an inflammatory assault, classically targeting the skin, gastrointestinal tract and liver. When the donor is female and the recipient male, the recipient carries male-specific antigens, collectively known as H-Y antigens, which are encoded on the Y chromosome and are absent from the donor’s immune repertoire. Donor T cells may therefore perceive these proteins as foreign, triggering an immune response that has been proposed as one mechanistic explanation for the higher GVHD risk observed in female-to-male transplants.

Previous individual studies have produced inconsistent results, and the question became especially pressing with the spread of post-transplant cyclophosphamide, a prophylactic regimen that has transformed the field by depleting the rapidly proliferating alloreactive T cells responsible for much of graft-versus-host pathology. To resolve the uncertainty, the investigators conducted a systematic review of cohort studies evaluating the risk of grade II-IV acute GVHD according to each donor-recipient sex combination, pooling hazard ratios with random-effects models, the standard statistical framework for combining estimates across studies that allows for between-study heterogeneity. The random-effects approach weights each study while acknowledging that the true effect may vary across populations, donor sources and prophylaxis protocols, producing a pooled estimate that reflects the broader landscape of transplantation practice rather than any single center’s experience.

The headline result was statistically robust. Across the pooled studies, female donor-to-male recipient transplantation was associated with a hazard ratio of 1.23 for grade II-IV acute GVHD compared with other donor-recipient sex combinations, with a 95 percent confidence interval of 1.08 to 1.39 and a P value of 0.001. In practical terms, male patients who received grafts from female donors faced roughly a 23 percent higher hazard of developing clinically significant acute GVHD than those in other sex-mismatched or sex-matched combinations. Because the confidence interval sits entirely above the null value of 1.0, the association is unlikely to be a statistical fluke, although, as with all observational pooling, it reflects correlation across cohorts rather than proof of causation in any individual patient.

The analysis then drilled into subgroups defined by donor type, and here the picture became more nuanced. Significant associations between female-to-male transplantation and acute GVHD risk emerged in both matched sibling donor and matched unrelated donor transplantation, the two settings in which human leukocyte antigen matching is most complete. This pattern is immunologically coherent: when HLA matching is near-perfect, the minor histocompatibility antigens such as the H-Y antigens become relatively more prominent as targets of donor-against-recipient reactivity, because the major barriers that normally dominate allorecognition have been removed. In contrast, the haploidentical donor subgroup, in which donors share only half of the HLA markers with their recipients, showed no significant association, suggesting that the sex-mismatch effect may be masked or overwhelmed by the far larger antigenic disparities inherent in haploidentical transplantation.

The role of post-transplant cyclophosphamide was the question the authors approached with particular care, given that this prophylaxis has become a cornerstone of modern transplantation, especially in haploidentical settings. In the subgroup of patients receiving PTCy prophylaxis, the pooled hazard ratio for the female-to-male combination was 1.11, with a confidence interval of 0.90 to 1.37 and a P value of 0.33, meaning the association was not statistically significant. In the non-PTCy subgroup, the association remained significant, with a hazard ratio of 1.25 and a confidence interval of 1.09 to 1.44. On its face, this numerical pattern hints that cyclophosphamide might blunt the sex-mismatch penalty, a biologically plausible idea given that the drug selectively eliminates the proliferating alloreactive T-cell clones that would otherwise respond to H-Y antigens.

Yet the authors were careful not to overclaim. Formal testing for subgroup differences yielded a P value of 0.34, which provides no statistical evidence that post-transplant cyclophosphamide actually modifies the association between female-to-male mismatch and grade II-IV acute GVHD risk. The distinction between observing different point estimates in separate subgroups and demonstrating true effect modification is a subtle but crucial one in meta-analysis, and the researchers explicitly cautioned that the subgroup findings should be interpreted cautiously. The numerical differences across donor-type and prophylaxis strata are intriguing and hypothesis-generating, but they do not license the conclusion that PTCy erases the sex-mismatch risk; a larger, purpose-designed study would be needed to establish that.

What the analysis does support, according to the authors, is the consideration of donor-recipient sex in donor selection algorithms. Donor choice in allogeneic transplantation is a multi-dimensional optimization problem involving HLA matching, donor age, cytomegalovirus serostatus, kinship, donor availability and health, and graft source. Sex mismatch has often been treated as a secondary consideration, and the new pooled estimate gives it quantitative standing as a factor associated with a meaningful relative increase in moderate-to-severe GVHD risk. Where two otherwise comparable donors are available, the findings suggest that avoiding a female donor for a male recipient, when clinically feasible, may reduce the likelihood of a complication that remains a leading cause of non-relapse mortality after transplantation. The authors also highlight the immunomodulatory role of GVHD prophylaxis more broadly, noting that the patterns observed underscore how profoundly prophylactic strategies reshape the immune dynamics of transplantation.

The study arrives at a moment of rapid evolution in transplant immunology. Haploidentical transplantation with post-transplant cyclophosphamide has expanded access to transplantation for patients lacking fully matched donors, and newer prophylactic agents continue to shift the risk landscape. Against that backdrop, a meta-analysis that quantifies the contribution of a simple, easily ascertained variable such as donor sex is valuable both for clinical decision-making and for refining risk-stratification models used in trials and registries. The mechanistic thread of H-Y antigen mismatch, first proposed decades ago, now sits within a modern statistical framework that accounts for how contemporary prophylaxis changes the terrain on which those antigens are encountered.

Limitations inherent to the evidence base temper the conclusions. Cohort studies are vulnerable to confounding by indication, center effects and evolving transplantation practices, and the pooled hazard ratios describe populations rather than predict outcomes for individuals. The non-significant test for interaction with PTCy means the apparent attenuation of risk under that regimen remains unproven. Still, the core finding stands on firm statistical ground: female donor-to-male recipient transplantation is associated with a higher risk of grade II-IV acute graft-versus-host disease after allogeneic hematopoietic stem cell transplantation. For the thousands of patients each year who undergo this procedure, and for the physicians who select their donors, the message is that sex is not a neutral variable. It is a measurable, mechanistically grounded risk factor that deserves a deliberate place in the calculus of donor selection, alongside the growing recognition that the immunological dialogue between graft and host is shaped by factors far beyond the HLA genes that have long dominated the field.

Subject of Research: The effect of donor-recipient sex mismatch on acute graft-versus-host disease risk after allogeneic hematopoietic stem cell transplantation

Article Title: Donor gender selection and its impact on acute graft-versus-host disease: a systematic review and meta-analysis

Article References: Donor gender selection and its impact on acute graft-versus-host disease: a systematic review and meta-analysis. (n.d.). https://doi.org/10.1007/s00277-026-07304-6

Image Credits: AI Generated

DOI: 10.1007/s00277-026-07304-6

Keywords: graft-versus-host disease, allogeneic stem cell transplantation, donor sex mismatch, H-Y antigens, post-transplant cyclophosphamide, meta-analysis, hematology, transplant immunology, haploidentical donors, HLA matching, GVHD prophylaxis, risk factors