Risk factors and associations with clinical outcomes of cytomegalovirus reactivation after haploidentical versus matched-sibling unmanipulated PBSCT in patients with hematologic malignancies.

In allogeneic hematopoietic stem cell transplantation recipients, cytomegalovirus (CMV) infection can cause overt CMV-associated disease, which is a main cause of transplantation-associated mortality. CMV infection correlates closely with donor's type. We therefore examined whether risk factors of CMV reactivation and clinical endpoints in patients with hematologic malignancies after allogeneic peripheral blood stem cell transplantation (PBSCT) differed between using matched-sibling donors (MSD-SCT) and haploidentical donors (HID-SCT). In this retrospective cohort study, we enrolled in 200 consecutive patients received an unmanipulated G-CSF-mobilized allogeneic PBSCT. Ninety (45%) patients received MSD-SCT and 110 (55%) received HID-SCT. Quantitative PCR was used for monitoring of CMV reactivation after transplantation. One-year cumulative incidence of CMV DNAemia was 55.0%, ranging from 23.5% in MSD-SCT group to 81.0% in HID-SCT group (p < 0.001). Although univariate analyses showed that non-myeloid malignancies, disease in complete remission status at transplantation, pretreatment with antithymocyte globulin, HLA-haploidentical donors, male donors, previous Epstein-Barr virus DNAemia, and absolute lymphocyte count on day 30 < 0.6 × 109/L were respectively associated with CMV reactivation after transplantation in total cohort of recipients (all p < 0.05), haploidentical donors were found to be the only independent predictor in multivariate analyses (Hazard ratio = 6.4, p < 0.001). Furthermore, univariate analyses revealed that non-myeloid malignancies and previous Epstein-Barr virus DNAemia were respectively associated with CMV reactivation in MSD-SCT recipients, and female was associated with CMV reactivation in HID-SCT recipients (all p < 0.05). In HID-SCT recipients, but not MSD-SCT recipients, previous CMV DNAemia was associated with a lower cumulative incidence of acute graft-versus-host disease (49.2% vs. 72.6%, p < 0.001). CMV DNAemia did not play a role in the relapse rate, but it was strongly associated with an increased risk of non-relapse mortality either in total cohort of recipients (30.5% vs. 13.7%; p = 0.003) or in the HID-SCT subgroup (36.0% vs. 16.7%; p = 0.030). Relapse-free survival and overall survival in total cohort of recipients with CMV DNAemia were both inferior to those without CMV DNAemia (45.3% vs. 57.6% and 54.8% vs. 65.8%, respectively; both p < 0.05). However, in subgroup analysis according to donor's type, neither relapse-free survival nor overall survival was impacted by CMV status (both p > 0.05). This study addressed differences in incidence, risk factors, and associations with clinical outcomes of CMV reactivation after haploidentical versus matched-sibling PBSCT.