ROLE OF REAL-WORLD EVIDENCE IN US FDA SAFETY LABEL MODIFICATIONS - INSIGHTS FROM CASE STUDIES
Author(s)
Saifuddin Kharawala, MD1, Deepali Mittal, MD1, Pankdeep Chhabra, MD1, Paul Gandhi, MD2, Monique Martin, PharmD1.
1Red Nucleus, Greater London, United Kingdom, 2Red Nucleus, Yardley, PA, USA.
1Red Nucleus, Greater London, United Kingdom, 2Red Nucleus, Yardley, PA, USA.
OBJECTIVES: To gain insights into FDA’s perspectives on using real-world evidence (RWE) to support changes to safety warnings.
METHODS: We conducted searches of bibliographic databases and grey literature to identify precedents of publicly-documented FDA decisions to modify label safety warnings (2015-2026) based at least partly on RWE, e.g., claims data, electronic health records, registries, or FDA-Sentinel. We extracted data on evidence sources, study design, regulatory context, and RWE contribution for five selected cases.
RESULTS: Though a small sample, the cases provide important insights. In some, existing warnings were removed, and in others, safety warnings were added. The main outcomes are summarized below. Safety warnings overturned: (1) GLP-1 receptor agonists: An existing suicidal ideation and behavior (SIB) risk warning was removed based on a meta-analysis [N=107,910], an FDA-Sentinel retrospective cohort study [N= 2,243,138], and published observational studies (2026). (2) Varenicline: An existing boxed warning on serious neuropsychiatric events including SIB was removed based on results from an FDA-mandated Phase 4 study [n=8144] (2016). (3) Clozapine: An existing risk evaluation and mitigation strategy program (for severe neutropenia risk) was removed based on a literature review, FDA Adverse Event Reporting System (FAERS) and observational study data, which together captured 20 years of data (2015). New safety warnings added: (1) Montelukast: A new boxed warning on neuropsychiatric events including SIB was added, based on FAERS data, an FDA-Sentinel study [N=457,377], and published animal studies (2020). (2) Recombinant zoster vaccine: A Guillain-Barré Syndrome warning was added based on post-marketing surveillance and a Medicare claims database study [N= 3,729,863 vaccinations] (2021).
CONCLUSIONS: RWE submitted to the FDA for post-marketing label reviews should be based on large datasets and/or collected over an extended time period. Regardless, sponsors should consider that the FDA review may result in safety warnings being added, rather than relaxed, if the new data warrant this.
METHODS: We conducted searches of bibliographic databases and grey literature to identify precedents of publicly-documented FDA decisions to modify label safety warnings (2015-2026) based at least partly on RWE, e.g., claims data, electronic health records, registries, or FDA-Sentinel. We extracted data on evidence sources, study design, regulatory context, and RWE contribution for five selected cases.
RESULTS: Though a small sample, the cases provide important insights. In some, existing warnings were removed, and in others, safety warnings were added. The main outcomes are summarized below. Safety warnings overturned: (1) GLP-1 receptor agonists: An existing suicidal ideation and behavior (SIB) risk warning was removed based on a meta-analysis [N=107,910], an FDA-Sentinel retrospective cohort study [N= 2,243,138], and published observational studies (2026). (2) Varenicline: An existing boxed warning on serious neuropsychiatric events including SIB was removed based on results from an FDA-mandated Phase 4 study [n=8144] (2016). (3) Clozapine: An existing risk evaluation and mitigation strategy program (for severe neutropenia risk) was removed based on a literature review, FDA Adverse Event Reporting System (FAERS) and observational study data, which together captured 20 years of data (2015). New safety warnings added: (1) Montelukast: A new boxed warning on neuropsychiatric events including SIB was added, based on FAERS data, an FDA-Sentinel study [N=457,377], and published animal studies (2020). (2) Recombinant zoster vaccine: A Guillain-Barré Syndrome warning was added based on post-marketing surveillance and a Medicare claims database study [N= 3,729,863 vaccinations] (2021).
CONCLUSIONS: RWE submitted to the FDA for post-marketing label reviews should be based on large datasets and/or collected over an extended time period. Regardless, sponsors should consider that the FDA review may result in safety warnings being added, rather than relaxed, if the new data warrant this.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
HPR9
Topic
Health Policy & Regulatory
Topic Subcategory
Approval & Labeling
Disease
Mental Health (including addiction), No Additional Disease & Conditions/Specialized Treatment Areas