CMV reactivation under foscarnet prophylaxis in cord blood transplantation: Increased NRM without impact on relapse or survival

Abstract

Abstract Background: Cytomegalovirus (CMV) infection remains an important post-transplant complication despite improved management, partly due to increased use of HLA-mismatched donors and expanded transplant indications to older patients. Human herpesvirus 6 (HHV-6) encephalitis is another severe infectious disease with high incidence in cord blood transplantation (CBT). Foscarnet (FOS) is an effective therapeutic agent against CMV infection and HHV-6 encephalitis; however, large-scale studies evaluating its prophylactic use in CBT are limited. Methods: We retrospectively analyzed patients who underwent first single-unit CBT at our institution between 2016 and 2022 and received prophylactic FOS (120mg/kg/day). FOS administration was planned from day 7 to day 50, but discontinuation or extension was permitted at the discretion of the attending physician. Prophylactic administration of letermovir was not performed. Tacrolimus and mycophenolate mofetil were used for GVHD prophylaxis. Eligible diseases included acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), and myelodysplastic syndrome (MDS). Patients with Eastern Cooperative Oncology Group performance status ≥3 or active infections before conditioning were excluded. CMV reactivation was defined by a positive C7-HRP antigenemia test, and CMV disease by clinical symptoms with supporting evidence from PCR (bronchoalveolar lavage fluid or cerebrospinal fluid) or biopsy. Primary endpoints were the cumulative incidence of CMV reactivation and its impact on relapse, non-relapse mortality (NRM), and overall survival (OS). Secondary endpoints included the incidence of CMV disease and HHV-6 encephalitis, as well as their impact on clinical outcomes. Multivariate analyses included age, underlying disease, year of transplant, conditioning intensity, remission status, performance status, HCT-CI, number of HLA mismatches, total body irradiation use, and development of grade II-IV acute graft-versus-host disease (aGVHD). Cumulative incidence functions accounting for competing risks were used for CMV reactivation, CMV disease, HHV-6 encephalitis, relapse, and NRM. Landmark analyses used events by day 150 for stratification to assess the impact of CMV reactivation, CMV disease, HHV-6 encephalitis and aGVHD. OS was estimated using the Kaplan-Meier method and compared via the log-rank test. Fine-Gray or Cox proportional hazards models were used for multivariate analysis. Reasons for early discontinuation of FOS (<30 days) were also assessed. Results: A total of 491 patients were included. The cumulative incidence of CMV reactivation by day 150 was 53.5%. CMV reactivation was more common in patients aged ≥60 years (51.0% vs. 37.1%, p<0.01) and those who received reduced-intensity conditioning (20.4% vs. 11.3%, p<0.01). In univariate analysis, CMV reactivation was not significantly associated with 3-year cumulative incidence of relapse (15.7% vs. 16.2%, p=0.77) or OS (60.5% vs. 66.1%, p=0.102), but was associated with higher NRM (23.9% vs. 15.5%, p=0.023). Subgroup analysis revealed that in the ALL cohort, reactivation was associated with significantly increased NRM (12.9% vs. 0.0%, p=0.017). However, multivariate analysis showed no independent association between CMV reactivation and relapse, NRM, or OS. The cumulative incidence of CMV disease by day 150 was 4.1%. Among the 20 cases that developed CMV disease within 150 days, the disease types were gastrointestinal disease (15 cases), pneumonia (3 cases), and encephalitis (2 cases). Although CMV disease did not significantly affect relapse, it was associated with markedly higher NRM (57.9% vs. 18.5%, p<0.01) and inferior OS (27.0% vs. 64.6%, p<0.01). The cumulative incidence of HHV-6 encephalitis by day 150 was 9.2%. HHV-6 encephalitis had no significant impact on NRM (23.4% vs. 20.4%, p=0.965) or OS (62.9% vs. 62.7%, p=0.607). FOS was discontinued within 30 days in 204 patients (41.5%), mainly due to renal dysfunction (155, 75.5%), which improved in 122 (79.2%) after cessation. Conclusions: In CBT with early post-transplant prophylactic administration of FOS, CMV reactivation was associated with increased NRM, although no significant impact on relapse or OS was observed. CMV disease had a substantial negative impact on NRM and OS, underscoring the importance of timely intervention.