Neoplasm Reports in Food and Drug Administration Adverse Event Reporting System Following Angiotensin Receptor Blocker Recalls

BACKGROUND

A worldwide voluntary recall of valsartan in July 2018 due to the potential carcinogen N-nitrosodimethylamine received extensive media and public attention. This was followed by more Food and Drug Administration (FDA) recalls regarding other contaminated ARB (angiotensin receptor blocker) products. Our study investigated the association between the FDA recalls and ARB neoplasm adverse events (AEs) reported to the FDA adverse event reporting system.

METHODS

In this cross-sectional study, data were retrospectively collected from the FDA adverse event reporting system database from January 2015 to December 2019. Reporting odds ratios (RORs) were estimated to detect signals of association between ARBs (valsartan, irbesartan, and losartan) and reported neoplasm AEs using negative (amoxicillin and sertraline) and positive (omeprazole and ranitidine) control exposures. The χ² was used to compare categorical variables.

RESULTS

A total of 2 181 524 AEs, including 10 461 nonmetastatic neoplasm AEs were analyzed. Monthly RORs (95% CI) of valsartan-associated neoplasms versus controls (ROR*: valsartan/negative exposures; ROR†: valsartan/omeprazole; and ROR‡: valsartan/ranitidine) showed the highest signals after the recall date in July 2018 (7.64 [4.78-12.19]*; 4.77 [3.36-6.79]†; 4.13 [2.50-6.84]‡) and August 2018 (7.87 [5.19-11.94]*; 5.65 [4.12-7.75]†; and 7.20 [4.46-11.63]‡). In contrast, the highest cancer signals for the irbesartan and losartan recalls detected in March 2019 (4.80*; 4.06†; and 3.38‡) and April 2019 (3.63*; 3.69†; and 2.52‡) respectively, were lower. One-year postrecall reported neoplasm AEs were ≈2-fold higher for valsartan than irbesartan (OR, 1.77 [95% CI, 1.47-2.13], P<0.0001) and losartan (OR, 2.07 [95% CI, 1.85-2.32], P<0.0001). Although all ARBs had the same nitrosamine contamination, we found 1-year postrecall versus prerecall cancer signals for valsartan were 3-fold higher versus control exposures, while the changes in RORs for irbesartan and losartan were only 20-30% higher.

CONCLUSIONS

Significantly more postrecall neoplasms were reported for valsartan, with higher valsartan-associated cancer signals compared with irbesartan and losartan, although they all contained the same carcinogenic contaminant. Extensive media coverage of the FDA valsartan recall may have alarmed patients and generated these abrupt, biologically infeasible cancer signals.