Drug-Induced ANCA-Associated Vasculitis: An Update

Drug-induced antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) has been associated with the use of specific drugs, with 3 additional drugs now being implicated in causing AAV, according to study data published in Arthritis & Rheumatology.

The current analysis is the first worldwide systemic pharmacovigilance study to describe the largest series of drug-induced cases of AAV, some of which have been linked to AAV (eg, hydralazine and antithyroid drugs), as well as other recently developed drugs (eg, sofosbuvir, mirabegron, and nintedanib).

The researchers sought to update the list of drugs associated with AAV using a pharmacovigilance-based data mining approach. They collected information on adverse drug reactions reported using the preferred term of the Medical Dictionary for Regulatory Activities —“anti-neutrophil cytoplasmic antibody positive vasculitis”— through November 2020 from the World Health Organization pharmcovigilance datatbase (VigiBase). For each of the retrieved drugs, a case-noncase analysis was conducted, with disproportionate reporting calculated using the information component (IC). In the current analysis, a positive IC₀₂₅ value — ie, the lower end of the 95% credibility interval — was considered significant.

A total of 483 deduplicated individual case safety reports of drug-induced AAV that included 15 drugs with an IC₀₂₅ of greater than 0 were located. Among individuals with drug-induced AAV for whom data regarding sex were available, 71.2% (n=264/371) were women. The median patient age at onset of drug-induced AAV was 62 years (IQR, 45-72 years). Further, the median time from introduction of the suspected drug to onset of drug-induced AAV was 9 months (IQR, 1-36 months).

Drug-induced AAV was considered serious in 98.1% (n=472/481) of the cases and proved fatal in 8.9% (n=43/481) of the cases. Hydralazine, propylthiouracil, thiamazole, sofosbuvir, minocycline, carbimazole, mirabegron, and nintedanib were associated with the highest disproportional reporting of an association with AAV in the VigiBase.

Limitations of the current analysis included notoriety bias (ie, increased number of reports following safety alerts in the media), missing data (eg, on the delay between the introduction of a drug and the onset of AAV being available in <10% of patients), and bias associated with the retrospective declarative design of the study.

The researchers concluded, “Specific prescriber attention to these drugs is needed, as well as experimental and observational studies to confirm these epidemiologic data.”

Reference

Deshayes S, Dolladille C, Dumont A, et al. A worldwide pharmacoepidemiologic update on drug-induced antineutrophil cytoplasmic antibody-associated vasculitis in the era of targeted therapies. Arthritis Rheumatol. 2022;74(1):134-139. doi:10.1002/art.41902