Ganciclovir to Prevent Reactivation of Cytomegalovirus in Patients With Sepsis-Associated Acute Respiratory Failure: A Phase 3 Randomized Placebo-Controlled Trial

Abstract Introduction and Rationale: Cytomegalovirus (CMV) reactivation in critically ill adults is associated with worse respiratory and clinical outcomes, but a causal role remains uncertain. We sought to determine whether ganciclovir prophylaxis, by decreasing CMV reactivation, increases respiratory support-free days (RSFDs) in critically ill CMV-seropositive adults with sepsis-associated acute respiratory failure. Methods: This double-blind, placebo-controlled, 16-center randomized trial enrolled immunocompetent CMV-seropositive adults with sepsis-associated acute respiratory failure from June 2021-October 2024. Participants were randomized (1:1) to receive either intravenous ganciclovir (5 mg/kg twice daily for 5 days, then once daily) or placebo until the earlier of hospital discharge or 28 days. Participants were followed for 180 days. The primary outcome was RSFDs by day 28. Key secondary outcomes were also analyzed, including mortality through day 28 and day 180, hematologic and renal toxicities, CMV reactivation in plasma, and other reportable adverse events. Results: Among 213 randomized participants (mean age 59 years, 53% women, 76% on invasive mechanical ventilation at enrollment), 205 (96%) received ≥1 dose of study medication (N=106 ganciclovir, N=99 placebo). By day 180, 62 died, 17 were unable to be contacted but 8 of these were known to be alive, 8 withdrew/refused further contact, 24 continue in follow-up, and 94 have completed the study. The trial was terminated early, based on DSMB recommendation, with N=213 participants enrolled representing 43% of the N=500 target sample size. Study termination was recommended due to concerns of futility and increased mortality in the ganciclovir arm. Among participants receiving ≥1 dose of study drug, RSFDs were similar between study arms, and mortality was higher among ganciclovir recipients at both day 28 and day 180 (Table). There were no significant differences in neutropenia, thrombocytopenia, or renal dysfunction by day 28 between the arms. By day 28, CMV reactivation in plasma was significantly lower in the ganciclovir group (Table). Full follow-up data and additional secondary and exploratory endpoint data will be available at the ATS International Conference presentation. Conclusions: Among critically ill, CMV-seropositive adults with sepsis-associated respiratory failure, ganciclovir prophylaxis did not increase RSFDs and was associated with higher mortality, but without increased known ganciclovir-associated toxicities. The mechanism(s) underlying increased mortality with ganciclovir warrant further investigation. Ganciclovir should not be routinely used for prophylaxis of CMV reactivation in critically-ill adults without an underlying immunosuppressive condition.