Brincidofovir for Asymptomatic Adenovirus Viremia in Pediatric and Adult Allogeneic Hematopoietic Cell Transplant Recipients: A Randomized Placebo-Controlled Phase II Trial

Adenovirus infection in immunocompromised patients contributes to significant morbidity and mortality, especially after allogeneic hematopoietic cell transplantation (HCT). Brincidofovir (BCV, CMX001) is an orally bioavailable lipid conjugate of cidofovir that has in vitro activity against adenoviruses and other double-stranded DNA viruses. This randomized placebo-controlled phase II trial evaluated pre-emptive treatment with BCV for the prevention of adenovirus disease in pediatric and adult allogeneic HCT recipients with asymptomatic adenovirus viremia. Allogeneic HCT recipients with adenovirus viremia were randomized 1:1:1 to receive oral BCV 100 mg (2 mg/kg if <50 kg) twice weekly (BIW), BCV 200 mg (4 mg/kg if <50 kg) once weekly (QW), or placebo for 6 to 12 weeks, followed by 4 weeks of post-treatment follow-up. For randomization, subjects were stratified by screening absolute lymphocyte count (<300 cells/mm³ versus ≥300 cells/mm³). Assignment to BCV or placebo was double blinded; dose frequency was unblinded. The primary endpoint was the proportion of subjects experiencing treatment failure, defined as either progression to probable or definitive adenovirus disease or confirmed increasing adenovirus viremia (≥1 log₁₀ copies/mL) during randomized therapy. Between June 2011 and December 2012, 48 subjects were randomized to the BCV BIW (n = 14), BCV QW (n = 16), or placebo (n = 18) groups. The proportion of subjects with treatment failure in the BCV BIW group was 21% (odds ratio, .53; 95% confidence interval [CI], .11 to 2.71; P = .45), 38% (odds ratio, 1.23; 95% CI, .30 to 5.05, P = .779) in the BCV QW group, and 33% in the placebo group. All-cause mortality was lower in the BCV BIW (14%) and BCV QW groups (31%) relative to the placebo group (39%), but these differences were not statistically significant. After 1 week of therapy, 8 of 12 subjects (67%) randomized to BCV BIW had undetectable adenovirus viremia (<100 copies/mL), compared with 4 of 14 subjects (29%) randomized to BCV QW and 5 of 15 subjects (33%) randomized to placebo. In a post hoc analysis of subjects with viremia ≥1000 copies/mL at baseline, 6 of 7 BCV BIW subjects (86%) achieved undetectable viremia compared with 2 of 8 placebo subjects (25%; P = .04). Early treatment discontinuation because of adverse events was more common in subjects treated with BCV than with placebo. Diarrhea was the most common event in all groups (57% BCV BIW, 38% BCV QW, 28% placebo), but it led to treatment discontinuation in only 1 subject receiving BCV QW. Events diagnosed as acute graft-versus-host disease, primarily of the gastrointestinal tract, were more frequent in the BCV BIW group (50%) than in the BCV QW (25%) and placebo (17%) groups. There was no evidence of myelotoxicity or nephrotoxicity in BCV-treated subjects. The results of this trial confirm the antiviral activity of BCV against adenoviruses. Further investigation is ongoing to define the optimal treatment strategy for HCT recipients with serious adenovirus infection and disease.