A Co-Reactivation Model of Cytomegalovirus and Epstein-Barr Virus after Allogeneic Hematopoietic Stem Cell Transplantation Was Constructed to Predict Mortality after Transplantation

Abstract

Abstract Objective: The aim of this study was to create a predictive model for the occurrence of co-reactivation of cytomegalovirus (CMV) and Epstein-Barr virus (EBV) following allogeneic hematopoietic stem cell transplantation (allo-HSCT). Methods: The study included 373 patients who underwent allo-HSCT. The patient population was randomly divided into two groups: a training cohort comprising 70% of patients (n=261) and a validation cohort comprising 30% of patients (n=112). The predictive model was developed using several variables, including the time of reactivation of CMV and EBV, patient age and gender, presence of hepatitis B core antibody (HbcAb), type of transplant, use of anti-thymocyte globulin (ATG), and the occurrence of acute and chronic graft-versus-host disease (aGVHD and cGVHD, respectively). Results: The developed predictive model was expressed as y = -6.21338339 + 0.01923832 × (CMV reactivation time) + 0.00362628 × (EBV reactivation time) - 0.57604523 × (patient gender) + 0.01974136 × (patient age) + 0.85514424 × (presence of HbcAb) + 1.13782872 × (transplant type) + 0.95376138 × (use of ATG) - 0.62393602 × (aGVHD) - 1.09485262 × (cGVHD). The model demonstrated a g-mean value of 0.685 and a probability threshold of 0.297, allowing for the classification of patients into high-risk and low-risk groups. The incidence of co-reactivation of CMV and EBV over 100 days was significantly higher in the high-risk group compared to the low-risk group. In the high-risk group, the incidence was 49.3%, while in the low-risk group, it was 16.9% (P<0.0001, HR 95%CI 3.693(2.493,5.47)). For patients with aGVHD, the incidence was 41.7% in the high-risk group and 17.9% in the low-risk group (P=0.00026, HR 95%CI 2.7473(1.560,4.839)). In patients without aGVHD, the incidence was 53% in the high-risk group and 14.0% in the low-risk group (P<0.0001, HR 95%CI 4.9439(2.347,10.4)). Additionally, the model was able to predict post-transplant mortality and survival. Conclusion: Overall, the findings of this study suggest that the developed model is a valuable tool for predicting the risk of CMV and EBV co-reactivation following allo-HSCT. By identifying high-risk patients, clinicians can take proactive measures to prevent or manage viral infections, ultimately improving patient outcomes. Therefore, this model can potentially enhance clinical decision-making and optimize post-transplant care.