The persistence of graft-versus-host disease (GVHD) as the principal complication of allogeneic hematopoietic cell transplantation (HCT) demonstrates that HLA matching alone is insufficient to prevent alloreactivity. We performed molecular and functional characterization of 22 candidate cytokine genes for their potential to improve matching in 315 myeloablative, 10/10 HLA-matched donor–recipient pairs. Recipients of a graft carrying the -1082GG IL10 gene promoter region variant had a three-fold lower incidence of grade II–IV acute GVHD compared to IL10-1082AA graft recipients (SHR = 0.25, p = 0.005). This was most evident in matched unrelated donor (MUD) transplants, where the greatest alloreactivity is expected. IL10-1082GG transplants did not experience an increased incidence of relapse, and, consequently, overall survival was two-fold higher in IL10-1082GG MUD transplants (HR = 0.17, p = 0.023). Longitudinal post-transplant measurements demonstrated that -1082GG is a high-IL10-producing and -expressing genotype with attenuated CD8⁺ T-cell reconstitution. High post-transplant donor chimerism in T- and myeloid-cells (>95%) confirmed a predominant donor, rather than recipient, genotype effect on immune function and aGVHD. To date, this is the first study to report corroborating genome-to-cellular evidence for a non-HLA donor immunogenetic variant that appears to be protective against GVHD. The incorporation of IL10 variants in donor selection criteria and clinical-management decisions has the potential to improve patient outcomes.

Background: Cytomegalovirus (CMV) reactivation is a major complication after hematopoietic stem cell transplantation (HSCT). Natural killer (NK) cells help control CMV through killer-cell immunoglobulin-like receptors (KIRs) and their HLA ligands, but donor-derived CMV-specific T-cells may confound the interpretation of NK-mediated effects. Methods: We analyzed 276 HLA-matched (10/10) adults receiving ATG-based T-cell-depleted myeloablative HSCT with a known donor and recipient CMV serostatus. The donor and recipient KIR genotypes were scored by the Cooley B-content score (0–4; ≥2 = high). Clinically significant CMV reactivation (plasma viral load > 25,000 IU/mL, the institutional threshold for pre-emptive therapy) was analyzed with Fine–Gray competing-risks regression, stratified by the donor–recipient serostatus. Results: In seronegative-donor/seropositive-recipient (D−R+) pairs (n = 68), a high donor KIR B-content score was associated with a significantly lower reactivation risk (sub-hazard ratio, 0.46; 95% CI, 0.24–0.91; p = 0.024). No effect was seen in D+R+ pairs (n = 82; SHR, 0.65; p = 0.241); D+R− (n = 28) had too few events to model. A donor Tel-AA/recipient Tel-B+ mismatch was independently associated with a higher reactivation risk (adjusted HR, 2.41; 95% CI, 1.33–4.37; p = 0.004). The overall survival was unaffected in either stratum. Conclusions: A high donor KIR B-content score protects against CMV reactivation in D−R+, but not D+R+, HSCT recipients, consistent with NK dominance when CMV-specific donor T-cells are sparse. A specific donor–recipient telomeric mismatch independently modifies the risk. Donor KIR profiling warrants prospective evaluation in donor-selection algorithms.