Impact of Atovaquone and Proguanil on Steroid Use, Hospitalization, and Advanced Therapeutic Use in Patients With IBD: A Propensity Matched Cohort Study

Introduction: The prevalence of inflammatory bowel disease (IBD) is rising globally, and many patients don’t respond to or are intolerant of current therapies. Atovaquone and proguanil (A/P) are antimalarial agents long-used in combination to prevent and treat malaria. Atovaquone has recently been shown to be effective in healing DSS-induced murine colitis, and anecdotally in ulcerative colitis (UC) (by JD). We therefore sought to evaluate the impact of A/P on surrogates for IBD activity utilizing a large population database. Methods: We performed a retrospective cohort study using TriNetX, a multi-institutional database to assess the effects of A/P on IBD patients. We identified all adult patients diagnosed with UC, Crohn’s disease (CD), and IBD (both UC & CD patients), between January 1, 2000 to December 31, 2020. Within each of these study populations, we stratified patients into 2 cohorts based on A/P use. For each of the 3 cohort comparisons, we performed 1:1 propensity score matching to balance cohorts on age, gender, race, comorbid conditions, and use of IBD advanced therapeutics to control for baseline disease severity. The primary outcomes were steroid use (prednisone, methylprednisolone, budesonide), all-cause hospitalization, and advanced therapeutic use (immunomodulators, biologics, and small molecules) within 1 year indexed on first A/P exposure for the A/P cohort, or the diagnosis of UC, CD or IBD for the no A/P cohort. Estimates of odds ratios (OR) and 95% confidence intervals (CI) were used to describe the outcome risks. P-values of α < 0.05 were considered statistically significant. Differences between groups were calculated using independent t-test or chi-square. Results: After matching, we identified 361 CD, 407 UC and 905 IBD patients in each of their respective A/P and no A/P cohorts. The A/P group had significantly lower odds of steroid use for CD (0.721) and IBD (0.654), and trended lower for UC (0.747) at 1 year compared to the no A/P group. The A/P group had significantly lower odds of hospitalization for CD (0.176), UC (0.285) and IBD (0.242) at 1 year compared to the no A/P group. The A/P group had significantly lower odds of advanced therapeutic use for CD (0.702) and IBD (0.700) and trended lower for UC (0.719) at 1 year compared to the no A/P group (Table 1). Conclusion: This study provides real world data that A/P could have a protective effect in IBD patients and justifies clinical trials looking at repurposing A/P for the treatment of IBD. Table 1. - Primary outcomes of steroid use, all-cause hospitalization and advanced therapeutic use within 1 year among CD, UC and All-IBD patients in the No A/P and A/P cohorts, after propensity score matching Crohn’s Disease (n=361) Outcome No A/P A/P OR (CI) P-Value Steroid Use 132 (36.6%) 106 (29.4%) 0.721 (0.528, 0.985) < 0.001 Hospitalization 79 (21.9%) 17 (4.7%) 0.176 (0.102, 0.305) < 0.001 Advanced Therapeutic Use 97 (26.9%) 74 (20.5%) 0.702 (0.497, 0.991) < 0.001 Ulcerative Colitis (n=407) Outcome No A/P A/P OR (CI) P-Value Steroid Use 136 (33.4%) 111 (27.3%) 0.747 (0.554, 1.009) 0.057 Hospitalization 68 (16.7%) 22 (5.4%) 0.285 (0.172, 0.471) < 0.001 Advanced Therapeutic Use 51 (12.5%) 38 (9.3%) 0.719 (0.461, 1.121) 0.144 IBD (n=905) Outcome No A/P A/P OR (CI) P-Value Steroid Use 351 (38.8%) 265 (29.3%) 0.654 (0.537, 0.795) < 0.001 Hospitalization 182 (20.1%) 52 (5.7%) 0.242 (0.175, 0.335) < 0.001 Advanced Therapeutic Use 210 (23.2%) 158 (17.5%) 0.700 (0.556, 0.882) 0.002