Investigating the Safety Profile of Fast-Track COVID-19 Drugs Using the FDA Adverse Event Reporting System Database: A Comparative Observational Study

Abstract

Background: The US Food and Drug Administration (US FDA) granted emergency use authorization (EUA) for multiple coronavirus disease 2019 (COVID-19) drugs as a medical countermeasure during the COVID-19 pandemic. Despite these drugs' fast-track nature, concerns persist regarding their efficacy and potential adverse effects. Thus, the continuous surveillance and understanding of these drugs' safety profiles are crucial in such scenarios.

Objective: Using the FDA Adverse Event Reporting System (FAERS) database, we aimed to compare the adverse drug reactions (ADRs) of four fast-track COVID-19 drugs to explore the potential of real-world data for providing prompt feedback in clinical settings.

Methods: To evaluate the post-marketing safety of fast-track COVID-19 drugs, we descriptively evaluated the ADRs of four COVID-19 drugs (bebtelovimab, molnupiravir, nirmatrelvir/ritonavir, and remdesivir) using FAERS data reported from January 2020 to June 2022. We examined FAERS case records of COVID-19 drugs reported as the "primary suspect drug" as a case group and the records of other drugs as the control. "Serious adverse drug reactions (SADRs)" were defined based on FDA guidelines. Using reporting odds ratios, disproportionality analysis was conducted to determine significant signals for ADRs related to each of the four drugs compared with those of others, both at the preferred term (PT) and system organ class (SOC) levels. To explore the occurrence of reporting each serious outcome reported to the four drugs, we fitted logistic regression models, adjusting for age and sex.

Results: During the study period, 5 248 221 cases were submitted to FAERS, including 17 275 cases of the four COVID-19 drugs: bebtelovimab (532 cases), molnupiravir (1106 cases), nirmatrelvir/ritonavir (9217 cases), and remdesivir (6420 cases). A total of 64, 46, 116, and 207 PTs with significant disproportionality were identified for each drug, respectively. "Infusion-related reaction" (18.4%), "diarrhea" (7.4%), "dysgeusia" (11.4%), and "increased alanine aminotransferase" (14.5%) were the most frequently reported SADRs for bebtelovimab, molnupiravir, nirmatrelvir/ritonavir, and remdesivir, respectively. Among the 27 SOCs, statistically significant signals were observed in 10, 3, 0, and 8 SOCs for bebtelovimab, molnupiravir, nirmatrelvir/ritonavir, and remdesivir, respectively. Remdesivir showed a higher occurrence for the reporting of death or life-threatening ADRs compared with the control (adjusted odds ratio (OR) = 2.44, 95% confidence interval (CI) = 2.23-2.59; adjusted OR = 1.82, 95% CI = 1.64-2.02, respectively).

Conclusions: We identified potential ADRs associated with COVID-19 drugs and provided insights into their real-world safety. This study demonstrated that real-world data and real-time safety reviews could be effective methods for the timely detection of ADR signals of drugs that have received fast-track approval, as exemplified by COVID-19 drugs. These findings underscore the importance of the continued surveillance, efficient data processing, and establishment of automated pipelines for real-time safety reviews.

Keywords: COVID‐19; FDA adverse event reporting system; adverse drug reaction; post‐marketing surveillance; real‐world data.