AIM: Pharmacovigilance data are primarily used to identify adverse drug reactions by screening for disproportionate reporting, i.e . more reports of certain combinations of adverse events and drugs than expected . However, scanning for associations of drugs and adverse events that occur less frequently than expected provides hypotheses for drug repurposing, i.e . a known drug could be therapeutically beneficial for a new indication like the coronavirus disease (COVID-19). As coronavirus related adverse events are scarce in pharmacovigilance data prior to 2020, we searched for drugs suitable against similar viral diseases .
METHODS: In this observational, retrospective, pharmacovigilance study, drugs associated with viral respiratory tract infections and/or diseases caused by RNA-viruses, which are phylogenetically similar to SARS-CoV-2, were extracted from the U.S. FAERS pharmacovigilance data 2004Q1 to 2020Q2 using OpenVigil 2.1-MedDRA17, filtered for significant inverse associations (ROR <1 and padj <0.05), checked for implausibility (e.g., only topically) or clinical infeasibility (e.g., strong cytotoxic effects), and categorised by their WHO Anatomical Therapeutic Chemical (ATC) classification code .
RESULTS: A total of 126 drugs were identified . ATC clustering of the manually curated list of 112 candidate drugs revealed female sex hormones, anti-diabetics, neuropharmacological sigma-receptor modulators, peptidase inhibitors, antiviral drugs, nicotinic acetylcholine receptor agonists, and tyrosine kinase inhibitors as putatively antiviral .
CONCLUSION: Scanning for inverse signals in pharmacovigilance data provides new hypotheses for drug repurposing, theoretically for all indications . Concerning the treatment of viral respiratory infections, there is affirmative data for some candidate drugs; the remaining proposed candidate drugs without already known antiviral mechanism of action should stimulate further exploration.