OBJECTIVES : In clinical trials surrogate endpoints are oftentimes used as substitute for patient-relevant endpoints. However, in the early benefit assessment of pharmaceuticals in Germany and in order be able to provide patient-relevant evidence a validation of surrogate endpoints is critical. The aim of this study was to assess the utilization, methodology and acceptance of surrogate validations in AMNOG benefit dossiers.

METHODS : Dossiers for benefit assessment were retrieved from the homepage of the Federal Joint Committee (G-BA). It was assessed whether the pharmaceutical companies performed a surrogate validation and if a surrogate parameter was considered for benefit assessment by the Institute for Quality and Efficiency in Health Care (IQWiG) and later accepted by the G-BA.

RESULTS : In total 397 AMNOG dossiers were analyzed. In 105 (26.4%) dossiers a surrogate validation was presented. Of all attempted surrogate validations, 36 (34.3%) were rejected and 51 (48.6%) were not further commented (e. g. due to limitations in study design). In total, 18 validations (17.1% of all attempted validations) were accepted as valid. Validations were only accepted in infectious indications (94.4%) and oncological diseases (5.6%). Accepted endpoints were (sustained) virologic response for reduced risk for hepatocellular carcinoma as well as AIDS-defining diseases/death, CD4 cell count for AIDS-defining diseases/death and disease-free survival (DFS) for overall survival. In most cases the pharmaceutical companies referred to prior benefit assessments. Only for DFS a comprehensive surrogate validation including systematic literature search and statistical calculations was performed.

CONCLUSIONS : Even though the methodological demands by IQWiG are high, in one quarter of submitted dossiers pharmaceutical companies attempted to validate a surrogate parameter as a patient-relevant endpoint. Overall, the acceptance of validations is possible; however, the success rate is small (17%). Most accepted surrogate validations were within infectious diseases and their validation referred to prior benefit assessments.