Risk factors for transplant-associated thrombotic microangiopathy in pediatric patients undergoing haploidentical hematopoietic stem cell transplantation

Abstract

Abstract Background: Transplant-associated thrombotic microangiopathy (TA-TMA) is a severe life-threatening endothelium-mediated complication of allogeneic hematopoietic stem cell transplantation (allo-HSCT). In children, haploidentical HSCT (haplo-HSCT) is associated with a significantly higher incidence of TA-TMA compared to other donor types, with one study reporting a rate of 17% (PMID: 32492158). While HLA mismatch is an established risk factor for TA-TMA in pediatric allo-HSCT (PMID: 33877298), and specific alleles like HLA-DRB1*11 have been implicated in adult cohorts of HLA matched HSCT(PMID: 26146809), the specific HLA determinants and other factors that drive TA-TMA risk in pediatric haplo-HSCT are poorly defined. This study aimed to identify risk factors for TA-TMA in a large, pediatric cohort undergoing haplo-HSCT. Methods: We retrospectively analyzed a multicenter cohort of 186 pediatric patients who underwent a first haplo-HSCT in China between 2019 and 2023, including 115 with malignant diseases and 71 with non-malignant conditions. All patients and donors had high-resolution HLA typing results for the HLA-A, -B, -C, -DRB1, and -DQB1 loci. The primary endpoint was the 180-day cumulative incidence (CI) of TA-TMA. Secondary endpoints included acute graft-versus-host disease (aGVHD) and overall survival (OS). The CI of TA-TMA was estimated using cumulative incidence function method with Gray-test. For multivariate analysis, Cox proportional hazards model was used to identify independent risk factors for TA-TMA. Results: Of the 186 enrolled patients, 23 developed TA-TMA, with a 180-day CI of TA-TMA reaching 12.7%. The median time to onset of TA-TMA was 67 days (interquartile range [IQR], 42-110 days). A statistically significant difference was observed in the 1-year post-transplant survival rate between patients who developed TA-TMA and those who did not (53% vs. 86%, p < 0.001). A total of 25 patients carried the HLA-DRB1*11 genotype, and compared with patients without this genotype, no statistically significant difference was observed in the 180-day CI of TA-TMA (12.5% vs 12.7%, p=0.885). Recipients with HLA-B-matched donors exhibited a lower 180-day CI of TA-TMA compared to recipients with HLA-B-mismatched donors (2.9% vs 15.0%, p=0.058). Additionally, patients who developed grade II–IV aGVHD had a significantly higher incidence of TA-TMA compared with those who did not (23.9% vs 3.9%, p < 0.001). Univariate analysis showed no significant association with TA-TMA for disease type, pre-transplant disease status, gender mismatch, or HLA mismatches at loci HLA-A/C/DRB1/DQB1. Multivariate analysis incorporating grade II–IV aGVHD demonstrated that HLA-B mismatch failed to achieve statistical significance (p=0.14). Conclusion: In this multicenter pediatric haplo-HSCT cohort, TA-TMA was linked to significantly lower 1-year survival. Grade II–IV aGVHD emerged as a key risk factor. HLA-B matching showed a trend toward reduced TA-TMA risk, while HLA-DRB1*11 was not associated. These findings enhance our understanding of TA-TMA's risks and determinants in pediatric haplo-HSCT, aiding in clinical management.