P105: A Post-Authorization Safety Study of the Effectiveness of Additional Risk Minimization Measures for Luspatercept: Challenges
Senior Manager, Global Risk Management
Bristol-Myers Squibb Company United States
The objective of this case study is to present some of the challenges encountered during the development and conduct of a Post-Authorization Safety Study (PASS, EUPAS 42016) requested by the European Medicines Agency (EMA) Pharmacovigilance Risk Assessment Committee (PRAC).
A major pharmaceutical company is conducting a PASS to evaluate the effectiveness of additional risk minimization measures (aRMMs) for luspatercept. The study team collated the challenges encountered during study start-up and execution along with the various mitigation strategies employed.
The first challenge was the timing of the PRAC request, as at this point BMS was undergoing integration by harmonization of processes and systems with another company (Celgene). Due to the regulatory timeline requested by PRAC, the study protocol development had to be accomplished at the same time the integration of the companies was still ongoing. This created uncertainty in sourcing internal funds for the study, required additional efforts to adapt to procedural requirements that were harmonized, and made it a challenge to identify team members to develop the study protocol and related materials.
The second challenge was that PRAC, in their response to the initial protocol submission, requested several changes to the survey design, including 1) the marketing authorization holder (MAH) should provide a specific list of countries (ensuring geographical representation) and detailed milestones for luspatercept launch and aRMMs distribution, 2) the timeframe should be narrowed for survey execution and that additional strategies should be implemented to reduce any possible bias, and 3) a comparison should be conducted between early and late respondents as well as between respondents and non-respondents to assess the impact of selection bias.
The third challenge was due to the survey response rate, which was lower than expected. The study protocol stated an expected 2.5% response rate, based on published literature; however, after fielding the survey in the first two countries, the average response rate was only 1.3%. This low response rate suggested that we might not collect the needed number of responses (200). Therefore, the team explored ways of increasing response rate, including increasing the number of healthcare professionals (HCPs) that would be invited to participate.
Team integrations typically enhance knowledge transfer among scientists. Both companies had well-defined risk management departments, including scientists with previous experience with such studies. This combination of experience allowed for stronger cross-functional workstreams in developing procedural documents and scientific collaboration on protocol development and study start-up.
We addressed the PRAC response based on industry best practices and our experience. Regarding listing the specific countries, we noted that the exact list of the countries would depend on future local market product approvals of the additional educational materials, their implementation by the various NCAs, and the timing of achieving full reimbursement. Therefore, we could only propose a list of potential countries. Regarding the target population for the survey, we proposed to include HCPs who intended to prescribe luspatercept and had received the aRMMs materials. This type of recruitment strategy should not create a selection bias, as the target audience of the materials is anticipated to cover all potential prescribers in the local markets. Regarding conducting the early/late comparison, such an analysis may be uninformative, since the survey is only for 3 months, and most HCP responses are to the initial survey invitation.
Other ways of increasing the response rate were explored, including increasing the number of reminders for the survey, increasing the amount of time allowed for survey completion, and maximizing the target population size. The team decided on exploring only measures that would not require a protocol amendment. However, some of these measures, such as increasing target population size, have increased the overall cost of the study.
This case study adds to the body of knowledge on the challenges of developing and executing a PASS for the assessment of the effectiveness of aRMMs.
Co-authors: S. Carou-Keenan, Z. Zakaite, S. Kaehler, I. Kim, and M. L. Chan-Lis