POST-MARKETING PHARMACOVIGILANCE ANALYSIS OF UBLITUXIMAB USING FAERS DATABASE AND LITERATURE MINING
Author(s)
Susmita R. Riya, MSc1, Mohammad Al-Mamun, PhD2;
1West Virginia University, Pharmaceutical Systems and Policy, Morgantown, WV, USA, 2Binghamton University, Systems Science and Industrial Engineering, Binghamton, NY, USA
1West Virginia University, Pharmaceutical Systems and Policy, Morgantown, WV, USA, 2Binghamton University, Systems Science and Industrial Engineering, Binghamton, NY, USA
OBJECTIVES: Ublituximab was approved in late 2022 to treat relapsing-multiple sclerosis. To date, the drug has limited real-world safety data. This study aimed to evaluate the post-marketing adverse events (AEs) associated with ublituximab utilizing the FDA Adverse Event Reporting System (FAERS) database and natural language processing (NLP) analysis with published literatures and reports.
METHODS: This study utilized AE data from 2022Q4 to 2025Q3. Disproportionality analyses were performed using Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), Bayesian Confidence Propagation Neural Network (BCPNN) and Multi-Item Gamma Poisson Shrinker (MGPS). A MedDRA preferred term (PT) was classified as a strong safety signal only if it met all four criteria for ROR, PRR, BCPNN and MGPS. A natural-language-processing (NLP) pipeline was applied to ublituximab clinical trials, case reports, reviews, and prescribing information to identify mentions of FAERS strong-signal PTs and summarize their document frequency.
RESULTS: Several strongly flagged AEs from FAERS, including infusion-related reactions (n = 1007, ROR = 182.45, 95% CI 167.91-198.26), upper respiratory tract infection (n = 22, ROR = 3.88, 95% CI 2.55, 5.91) were also reported in the NLP-screened published literature appearing in 87% and 50% of included publications, respectively. In contrast, other FAERS detected signals with high disproportionality such as peroneal nerve palsy (n= 4, ROR=8.23 95%CI 3.08, 22.00), ocular icterus (n=3, ROR=9.50 95% CI 3.05,29.55) were not reported in the published literature.
CONCLUSIONS: The study highlighted potential under-reported safety concerns emerging from post-marketing surveillance, underscoring the importance of continued real-world safety monitoring, especially using newer methods such as NLP.
METHODS: This study utilized AE data from 2022Q4 to 2025Q3. Disproportionality analyses were performed using Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), Bayesian Confidence Propagation Neural Network (BCPNN) and Multi-Item Gamma Poisson Shrinker (MGPS). A MedDRA preferred term (PT) was classified as a strong safety signal only if it met all four criteria for ROR, PRR, BCPNN and MGPS. A natural-language-processing (NLP) pipeline was applied to ublituximab clinical trials, case reports, reviews, and prescribing information to identify mentions of FAERS strong-signal PTs and summarize their document frequency.
RESULTS: Several strongly flagged AEs from FAERS, including infusion-related reactions (n = 1007, ROR = 182.45, 95% CI 167.91-198.26), upper respiratory tract infection (n = 22, ROR = 3.88, 95% CI 2.55, 5.91) were also reported in the NLP-screened published literature appearing in 87% and 50% of included publications, respectively. In contrast, other FAERS detected signals with high disproportionality such as peroneal nerve palsy (n= 4, ROR=8.23 95%CI 3.08, 22.00), ocular icterus (n=3, ROR=9.50 95% CI 3.05,29.55) were not reported in the published literature.
CONCLUSIONS: The study highlighted potential under-reported safety concerns emerging from post-marketing surveillance, underscoring the importance of continued real-world safety monitoring, especially using newer methods such as NLP.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
P23
Topic
Real World Data & Information Systems
Topic Subcategory
Health & Insurance Records Systems
Disease
SDC: Neurological Disorders