The European Medicines Agency (EMA) conditional marketing authorisation (CMA) allows earlier access to medicines for serious or rare diseases on limited evidence, leaving uncertainty about their effectiveness. This study examined whether lack-of-effic...
The European Medicines Agency (EMA) conditional marketing authorisation (CMA) allows earlier access to medicines for serious or rare diseases on limited evidence, leaving uncertainty about their effectiveness. This study examined whether lack-of-efficacy (LoE) reports for such medicines could serve as early warning signals before withdrawal using the WHO VigiBase.
We examined four medicines that received CMA after 2016 and were subsequently withdrawn for LoE, comparing them with other drugs reported to VigiBase during the same periods. The reporting odds ratios (RORs) with 95% confidence intervals were calculated from case/non-case data, with additional sensitivity and time-series analyses performed.
In the main analysis, crizanlizumab had a higher proportion of LoE reports than the comparator, with an ROR of 1.59 (95% CI : 1.31–1.92), showing statistically significant disproportionality. This elevation remained above 1.0 in multiple sensitivity settings, including exclusion of vaccines and reports with missing data. In the time-series analysis, crizanlizumab entered the early detection band in 2021 and 2022, with yearly RORs of 1.40 and 1.87, respectively, indicating two consecutive years of increased LoE disproportionality before withdrawal. By contrast, the remaining three products showed no evidence of increased lack-of-efficacy (LoE) disproportionality in either the main or time-series analyses. When the comparator was restricted to anticancer agents, belantamab mafodotin (ROR 0.62, 95% CI: 0.42–0.92) and olaratumab (ROR 1.11, 95% CI: 0.53–2.34) showed higher ROR estimates than in the main analysis, but these increases were neither statistically significant nor signal-detectable for LoE disproportionality.
The main contribution of this study is that it is the first to directly link LoE signals detected in VigiBase to EMA CMA products whose authorisation was subsequently withdrawn specifically for lack of efficacy, thereby exploring the potential role of LoE disproportionality as a regulatory early warning indicator.
In conclusion, these findings suggest that LoE-focused disproportionality analysis in VigiBase may identify emerging effectiveness concerns for some medicines, such as crizanlizumab, prior to withdrawal. However, because such signals were not consistent across all withdrawn products, further studies using refined comparators and diverse signal-detection algorithms are needed to support their regulatory application.