OBJECTIVES: It is widely believed that inflammation contributes to the pathogenesis of AD. TNF-α has been implicated in both AD and neurological inflammation. Therefore, anti-TNF-α therapies are surmised to be of potential benefit for preventing AD.

METHODS: Patients were identified as either receiving an anti-TNF-α drug (etanercept, adalimumab, certolizumab, infliximab, or golimumab) or patients who were anti-TNF-α naïve receiving methotrexate (MTX). Patient cohorts were generated in both an electronic medical records (EMR) research network of 52M patient lives and a 190M patient network of linked primary care data with medical and pharmacy claims. Neither cohort had AD (ICD-10 G30) prior to the ‘anytime’ outcome window. The risk of AD from 1 year after first exposure to either anti-TNF-α or MTX was analyzed using measures of association and survival analysis conducted via Kaplan-Meier estimator. Cohort comparisons were run with propensity score matching (PSM) to control for demographic, comorbidity, medication, and diagnostic procedure confounders.

RESULTS: There was significant AD prevention with anti-TNF-α treatment compared to MTX found in both EMR (RR 1.246, p=0.0334) and claims data networks (RR 1.788, p=0.0066). Time to AD outcome was measured up to 6,500 days in the EMR cohort with survival probability 98.677% for anti-TNF-α and 98.307% for MTX (p=0.0012). In the claims cohorts, survival probability was 97.807% for anti-TNF-α and 95.725% for MTX (p=0.0060) measured across 6,000 days post index.

CONCLUSIONS: This study provides RWD based evidence that points to anti-TNF-α therapies as a beneficial therapy in the prevention of AD. Further research is warranted to investigate these findings in larger cohorts and the AD outcome to be monitored in a longer observation window.