Using T-Lymphocyte Subsets at Engraftment to Predict the Risk of Acute Graft Versus Host Disease in Patients with Thalassemia Major：Development of a New Predictive Nomogram

Background: Acute graft-versus-host disease (aGvHD) is the primary cause of mortality following allogeneic hematopoietic cell transplantation (HCT). Current prediction models for aGvHD based on clinical features often provide suboptimal results and require additional validation. Recent studies suggest that T cell count on the early days post-transplantation may a marker for aGvHD. This study aimed to predict the risk of aGvHD after HCT in patients with thalassemia major (TM) using a novel predictive nomogram based on T-lymphocyte subsets at engraftment.

Methods: We performed retrospective analyses on 402 consecutive thalassemia patients who underwent HCT from August 2019 to December 2023. The myeloablative conditioning regimen consisted of busulfan (BU), cyclophosphamide (Cy), fludarabine (FLU), and anti-thymocyte (ATG). The GVHD prophylaxis was cyclosporine/ tacrolimus, methotrexate, and mycophenolate mofetil. Clinical risk factors for aGvHD were analyzed using Cox proportional regression models. T lymphocyte subsets were collected from 240 patients at the time of neutrophil engraftment. LASSO regression was utilized to screen the indices, with cut-off values established through restricted cubic spline (RCS) regression. The predictive model was developed by integrating T lymphocyte subsets with clinical features. Receiver operating characteristic (ROC) curve, C-index, calibration curve and decision curve analysis (DCA) were used for the validation of the evaluation model.

Results: Among 402 thalassemia patients analyzed post-transplantation, 242 (60%) patients were male, the median age was 8 (range 2-20) years. One hundred and forty-six (36%) patients developed aGVHD, including 95 (24%) with grade 2-4 and 38 (9.5%) with grade 3-4 aGvHD. Significant independent risk factors for aGvHD included matched unrelated transplantation (HR3.63, 95%CI: 2.17-6.05, P<0.001), haploid transplantation (HR4.03, 95%CI: 2.40-6.77, P<0.001), peripheral blood stem cell infusion (HR2.11, 95%CI: 1.09-4.11, P=0.03) and donor age over 40 years (HR1.53, 95%CI: 1.00-2.33, P=0.049). These risk factors, CD4+ T cell count and CD8+ T cell count at the time of neutrophil engraftment, were screened as non-zero coefficients by LASSO regression. Our RCS analysis indicated a marked increased risk in aGvHD when CD4+T cell counts exceeded 36 cells/μl and CD8+T cell exceeded 43 cells/μl. In comparison with the Cox model using clinical risk factors, incorporating T lymphocyte subsets into the model demonstrated strong predictive performance for aGvHD, which showed a net reclassification improvement index (NRI) of 0.24 (95%CI 0.09-0.39, P=0.04) and an integrated discrimination improvement index (IDI) of 0.08 (95%CI 0.02-0.16, P<0.001). The total aGVHD, grade 2-4aGVHD and grade 3-4aGVHD area under ROC curve (AUC) at 100 days after transplantation were 0.802, 0.809 and 0.846 for the training set, and 0.777, 0.757 and 0.746 for the validation set, respectively. The calibration curve and DCA curve of training set and verification set are all good.

Conclusion: In the myeloablative conditioning regimen, combining T-lymphocyte subsets at engraftment and clinical characteristic, the aGVHD model provides better performance. This convenient model can rapidly identify the aGVHD risk profile of patients and facilitate clinical decision-making and reduce the occurrence of adversities.