Primary endpoint acceptance and its association with benefit ratings and market access in German health technology assessments: a retrospective cohort study

Abstract

Background

In Germany, the Health Technology Assessment process, governed by the Federal Joint Committee (G-BA), evaluates relative efficacy and safety of newly European Medicines Agency (EMA)-approved drugs against the standard of care using patient-relevant endpoints. This study aimed to investigate the impact of the acceptance of the primary endpoint on the likelihood of an additional benefit and patient access in assessments that fulfilled criteria for study design and comparator for both authorities.

Methods

We examined whether the acceptance of the primary endpoint by G-BA resulted in a higher likelihood of a positive benefit rating and a reduced risk of market withdrawal by pharmaceutical companies of drugs that underwent a benefit assessment between 2011 and 2023, in which an RCT with an adequate comparator according to G-BA served as evidence base. These objectives were tested using a stratified Mantel–Haenszel test (stratified for oncological and non-oncological benefit assessments).

Findings

In 433 benefit assessments, G-BA accepted the primary endpoints used by the EMA in only 238 cases (54.97%). Acceptance was linked to a higher probability of a positive rating, compared to assessments without (OR 2.64, 0.95 confidence interval [1.74; 4.00]). Furthermore, assessments with accepted primary endpoints were less often associated with market withdrawal (OR 0.70 [0.40; 1.24]). Of note, 152 benefit assessments (35.10%) resulted in a non-positive rating despite using a study cohort with evidence that should cover both evaluation scopes, indicating a potential discrepancy between EMA and G-BA in the assessments of new drugs.

Interpretation

The study highlights a significant harmonisation gap between EMA and G-BA in the evaluation of innovative drugs, affecting benefit ratings and potentially market access. The acceptance of primary endpoints by G-BA is crucial for positive ratings. However, the differing evaluation of high-quality evidence can also be attributed to other critical factors, such as equivalent consideration of other study endpoints and slicing of the study population.

Funding

None. GM, MG, and AH are employed by Takeda. The authors are solely responsible for the content.