Use of Real-World Data in Regulatory and Reimbursement Submissions for Personalized Medicines in Europe

OBJECTIVES: Real-world data (RWD) are increasingly used in regulatory submissions to demonstrate efficacy and safety for novel drugs. Personalized drugs are often innovative cancer or rare disease drugs, which might increase the use of RWD in the absence of controlled trials. In this study, we aimed to investigate the use of RWD in EMA and subsequent NICE and HAS reimbursement submissions of personalized medicines.

METHODS: Personalized medicines were identified through FDA Progress and Outlook Reports from 2017 to 2020. Details on EMA approval and use of RWD were extracted from EMA initial authorisation documents. In case RWD was used, we reviewed NICE and HAS appraisal documents for similar use of RWD.

RESULTS: Of the 62 EMA-approved personalized medicines, 13 (21%) presented RWD in the EMA submission; either as comparator arm for a single-arm trial (n=11; 85%), to support RCT efficacy data (n=1; 7.7%) or to inform the natural history (n=1; 7.7%). Of the 13 EMA submissions using RWD, 8 of the 10 available NICE submissions and 10 of the 13 HAS submissions used RWD. In most cases, NICE accepted the RWD but noted that the magnitude of the treatment effect remained uncertain in submissions using RWD as comparator evidence for single-arm trials, especially for cancer drugs. Of the 10 HAS submissions including RWD, 7 used RWD for indirect comparisons, 3 for descriptive purposes (natural history of disease and early use); the main concern was the quality of evidence and lack of a descriptive or quantitative bias assessment.

CONCLUSIONS: RWD was used in one-fifth of 62 EMA submissions for personalized medicines; mostly to provide comparative evidence for single-arm trial efficacy data. The same RWD was mostly used subsequently in NICE and HAS submissions. Although accepted, further data collection was requested to reduce the remaining uncertainty around the treatment effect.