OBJECTIVES: With the introduction of AMNOG in January 2011, an early benefit assessment (EBA) was required for new medicines in Germany. EBAs are based on the additional therapeutic benefit of a drug on patient-relevant endpoints (PREs). We compared the acceptance of PREs for oncology in regulatory trials, and in EBAs conducted by German health technology assessment (HTA) bodies. METHODS: EBAs on oncology drugs and the respective regulatory trials were reviewed. The Federal Joint Committee (G-BA) website was used to obtain manufacturers’ value dossiers, Institute for Quality and Efficiency in Health Care (IQWiG) assessments, and G-BA resolutions. Acceptance of endpoints in the dimensions of mortality, morbidity and quality of life (QoL) by HTA bodies, IQWiG and G-BA, were compared to those accepted for regulatory trials. Data on endpoints used in regulatory trials were obtained from the manufacturers’ value dossiers. RESULTS: Overall survival (OS) and measures of disease morbidity, such as progression-free survival (PFS), were generally accepted in regulatory trials. OS was accepted by IQWiG and G-BA as a mortality endpoint for evaluating additional benefit. Widely accepted morbidity endpoints such as PFS were not deemed patient-relevant by IQWiG and G‑BA. In general, QoL questionnaires used in regulatory trials were accepted by the HTA bodies, although minor variability between questionnaires led to some exclusions from the HTA evaluations and the obtained QoL data revealed a number of missing values. CONCLUSIONS: HTA and regulatory bodies largely agree on the acceptance of mortality and QoL endpoints typically evaluated in oncology. Considerable variability was observed in the acceptance of PREs in morbidity. Evaluating additional benefit based only on mortality and QoL endpoints underestimate the potential value of new drugs. Multiple endpoints, which capture all three dimensions, should be evaluated in regulatory trials and accepted by IQWiG and G‑BA to confirm patient-relevant additional benefit.