OBJECTIVES: Since 2013, manufacturers are required to submit a pharmacoeconomic evaluation to the French HTA for all innovative products claiming an important added clinical benefit with a substantial impact on healthcare expenditures. The objective of this analysis was to review the French HTA decisions and provide a summary of the methodological concerns (MCs) reported in chronic lymphocytic leukaemia (CLL) indications.

METHODS: All available opinions for CLL indications published up to March 2020 were reviewed. Methodological choices, specific indications, nature of the request and MCs were extracted.

RESULTS: A total of eight published opinions were identified and reviewed. The cost-effectiveness was not demonstrated in any of them and only two did not raise major concerns. Overall, 49 MCs were identified: 12 major, 13 important, and 24 minor. Almost half (24/49) of MCs were related to modelling and methodological choices with a non-negligible impact on the overall decision: 10 major, 9 important and 5 minor MCs. The ten major MCs were related to uncertainty and limitations pertaining to the efficacy data (data immaturity, etc.), comparators choices (lack of relevant comparators, etc.), studied population (missing subgroups, etc.) and utilities (non-conformity of the estimations method, etc.). The nine important MCs were caused by the modelling assumptions and their poor discussion as well as by the weak external and internal validity of the model. Most of the minor concerns were related to a lack of clarity in the presentation of the results and the methodological choices, miscalculations, or the absence of required scenario analyses.

CONCLUSIONS: This review highlighted the lack of available and mature clinical data in CLL, inciting manufacturers to consider an important number of modelling assumptions. Considering assumptions should be associated with a robust rationale, sensitivity analyses and discussion to ensure that the French HTA can properly assess results considering the modelling-related uncertainty.