Monitoring Changes in Serum Albumin (SA) Concentrations As an Early and Objective Indicator of Potential CMX001-Associated Gastrointestinal (GI) Adverse Drug Effects

Background: CMX001 is an orally bioavailable, broad spectrum, lipid acyclic nucleoside phosphonate converted inside cells to the active antiviral, cidofovir diphosphate. In the preclinical toxicology program, GI AEs (diagnosed as gastropathy and enteropathy; dose-related changes included flattening or loss of epithelial cells lining the lumen of the small intestine on chronic dosing) were dose-limiting after daily administration; however, there were no GI AEs or gross/microscopic gut changes when animals were dosed twice-weekly (BIW) up to 9 months. Radiolabel studies in mice confirmed that CMX001 concentrates in the gut mucosa more than in other tissues. In a Phase 2 dose-escalation study (CMX001-201; ClinicalTrials.gov identifier: NCT00942305) evaluating CMX001 for CMV prophylaxis in allogeneic hematopoietic stem cell transplant (HSCT) recipients, an increased rate of profuse watery diarrhea was seen at 200 mg BIW and was considered dose-limiting in this population. A program-wide safety monitoring and management plan that included an option for dose interruption (≤ 2 weeks) in subjects with ≥ Grade 3 diarrhea was introduced. Subsequently, few subjects (≤ 10%) discontinued therapy because of GI AEs, indicating that dose interruption is an appropriate strategy to manage CMX001-associated GI AEs and to achieve effective CMV suppression in this population. Methods: Serum chemistry data were evaluated for changes in SA, a well-established marker of protein loss, to assess the potential relationship to diarrhea. Abnormally low SA concentrations were tabulated and the lowest value identified through +1 week post treatment. A clinically meaningful SA decrease was defined as value ≤ 30 g/L (lower limit of normal 33 g/L) and ≥ 4 g/L lower than baseline. Results: Increased grade and/or duration of diarrhea correlated with the decrease in SA concentrations over time as shown in the Kaplan-Meier plots with data grouped by “low” (≤ 100 mg/week) and “high” (≥ 200 mg/week) CMX001 dose vs. placebo. To rule out GI-GVHD (a common cause of diarrhea in HSCT recipients), the SA data from solid organ transplant (SOT) patients treated with CMX001 in an expanded access study (CMX001-350; ClinicalTrials.gov ID: NCT01143181) were also evaluated. A similar timing of decrease in SA concentrations was seen in these subjects who are unlikely to have GVHD; urinalysis data also excluded proteinuria as a cause. Conclusions: Our clinical experience in the HSCT population is consistent with preclinical findings. On chronic dosing, CMX001 likely accumulates in the gut mucosa in some patients and causes diarrhea that may be more pronounced in individuals with other causes of diarrhea (eg, GI-GVHD). Dose interruption gives the gut mucosa time to recover, allowing subjects the opportunity to resume therapy. Monitoring SA changes in patients may provide an early and objective indicator of potential drug-related GI AEs.