Signal detection on a patient cohort: A disproportionality analysis of the ANRS CO22 HEPATHER cohort to identify associations between direct acting antivirals and adverse events in patients with hepatitis C virus chronic infection.

PURPOSE: Our aim was to explore a signal detection method for early identification of potential adverse drug reactions (ADRs) in a patient cohort.

METHODS: ANRS CO22 HEPATHER is a French multicentre prospective observational cohort started in 2012. The cohort includes patients with chronic hepatitis C virus (HCV) infection with reports of all adverse events (AEs) occurring in patients exposed to HCV drugs. We applied a disproportionality method, which calculated a measure of association, the Bayesian information component (IC), for each drug-AE pair. Information components were continuously updated and a positive drug-AE association was detected when the lower limit of an IC 95% credible interval (95% CI) exceeded 0. We illustrate how the method could result in timely detection of photosensitivity reaction with simeprevir use.

RESULTS: By August 28, 2016, 6600 patients with HCV infection had been treated or were undergoing current HCV treatment, and 3464 experienced at least one AE for a total of 12 720 reported AEs. We detected 52 positive drug-AE associations, including 44 that were known ADRs based on the summary of product characteristics. The association between simeprevir and photosensitivity reaction was detected on June 4, 2014. At this date, 68 patients had received simeprevir and 6 photosensitivity reaction (4 during simeprevir treatment) had been reported for an estimated IC of 1.90, 95% CI, 0.20-3.61.

CONCLUSIONS: The disproportionality method can help with early detection of potential ADRs in patient cohorts. Detected associations need to be confirmed by a review of clinical data.