BACKGROUND

In recent years, an association between dipeptidyl peptidase-4 (DPP-4) inhibitors and bullous pemphigoid has been detected in pharmacovigilance studies in European and Asian countries; however, no pharmacovigilance data have been published yet in the USA.

OBJECTIVE

The objective of this study was to examine the relationship between bullous pemphigoid and DPP-4 inhibitors and other oral diabetes mellitus medications in the FDA Adverse Event Reporting System (FAERS).

METHODS

Case/non-case analyses were performed in the FAERS using data from 2006 to 2020 to examine the reporting odds ratio (ROR) signal for bullous pemphigoid for all classes of oral diabetes medications. These analyses were performed under multiple conditions to control for bias: (1) comparison to all other drugs in the FAERS; (2) comparison to other diabetes medications; and (3) comparison to all other diabetes medications where only a single agent was implicated.

RESULTS

A statistically significant ROR for bullous pemphigoid was found for DPP-4 inhibitors under all conditions: (1) 109.79 (95% confidence interval [CI] 101.61-118.62); (2) 74.46 (95% CI 60.58-91.52); and (3) 35.94 (95% CI 27.91-46.28). A larger signal was seen for non-US Food and Drug Administration (FDA)-approved (anagliptin, vildagliptin, teneligliptin) vs FDA-approved DPP-4 inhibitors (alogliptin, linagliptin, saxagliptin, sitagliptin), likely because of an overestimation of the ROR for non-FDA-approved drugs. The largest signal was seen under conditions 1 and 2 with vildagliptin (1) 1022.83 (95% CI 909.45-1150.35) and (2) 158.84 (95% CI 127.01-198.66) followed by anagliptin (1) 628.63 (95% CI 221.36-1785.24) and (2) 60.64 (95% CI 20.98-175.26), alogliptin, teneligliptin, linagliptin, sitagliptin, and saxagliptin. Under condition 3, the largest signal was seen with linagliptin 122.25 (95% CI 93.96-159.07). Both metformin and the sulfonylureas had a significant ROR under condition 2 [3.42 (95% CI 3.01-3.89) and 2.07 (95% CI 1.66-2.57) respectively]; however, this association was not present under condition 3 as only confounded cases occurred, and a large majority of reported cases had concurrent exposure to a DPP-4 inhibitor.

CONCLUSIONS

Our findings support an association between DPP-4 inhibitors and bullous pemphigoid. This association was maintained under controls to limit bias and falsely elevated signal, including controlling for disease state and cases with multiple drug exposures. Non-FDA-approved DPP-4 inhibitors had a larger ROR compared with FDA-approved DPP-4 inhibitors, likely owing to fewer reported adverse effects overall for non-FDA-approved drugs in FAERS.