Background. Gastric adenocarcinoma (GAC) is the third leading global cause of cancer mortality and the leading infection-associated cancer. In the U.S., GAC represents a major cancer disparity, with incidence rates 2-3X in non-white populations. Patients with gastric premalignant conditions (GPMC), intestinal metaplasia (IM) and atrophic gastritis (AG), represent a significant unmet need for chemopreventive approaches. Methods. We conducted an NCI-DCP-funded, Phase IIa placebo-controlled RCT of eflornithine in patients with GPMC (NCT02794428) between 09/2016 and 12/2022 in rural Honduras and Puerto Rico. This population is representative of Caribbean and Mesoamerican populations with a high prevalence of H. pylori (Hp) infection and GPMC. Hp positive and negative subjects ages 30-60 were randomized to eflornithine or placebo for 18 months, with endoscopy at baseline and 6, 18, and 24 months. Standard triple therapy was offered at 6 months to Hp positive patients. Outcomes were based upon tolerability, safety, and changes in Correa histopathology score and DNA damage (IHC, %pH2AX positive cells). A blinded, central pathologist assessed histology measures. Mixed effects linear regression models for %pH2AX were adjusted for Hp status and GPMC (AG, IM), including analyses vs baseline and vs adjacent time point. Results. 211 and 91 subjects were screening and randomized. The mean age was 45.2 (SD 8.8) and 67 (74%) were female. At baseline, 80% of subjects were Hp positive, and 46% and 54% had global histology of atrophy and IM, respectively. A total of 78, 69, 55 patients reached the 6-month primary outcome, the end-of-treatment 18-month (EoT), and the end-of-study (EoS) 24-month time points. The RCT was closed in 2022 due to the extended Pandemic challenges with 11/69 active EoT patients. Eflornithine treatment was safe and well-tolerated. Overall grade 1-2 adverse events were greater in the placebo group (108 vs 81). There were 4 unrelated grade 3 SAEs (COVID-19, hypertensive urgency, appendicitis, diverticulitis; 3 placebo; 1 eflornithine). There were no significant changes in laboratory parameters (CBC, creatinine, LFTs) at the study time points. There were no withdrawals due to ototoxicity/tinnitus (exception, one patient with a history of prior ear surgery). There were no significant differences in the Correa score between treatment arms at the 4 time points. In the regression model for DNA damage, the eflornithine group had a modest increase in %pH2AX at the 6 and 18-month time points. Importantly, a decrease in %pH2AX was observed at EoS, which was significant in the adjacent time point analysis (P = 0.012). (Figure). Conclusions Eflornithine treatment was safe and well tolerated in GPMC patients in Latin America. Our results suggest that eflornithine reduced long-term DNA damage in patients after completing treatment.

Figure 1. Regression Model, %pH2AX IHC change from adjacent time point

Marcia Cruz-Correa

Appearances