The German AMNOG healthcare reform includes a mandatory early-benefit-assessment (EBA) at launch. As per German social code, EBA is based on registration trials and includes evaluation of the patient-relevant effect of the new medicines compared to an appropriate comparator as defined by the Federal Joint Committee (G-BA). Current EBA decisions released have unveiled issues regarding the acceptance of some patient-relevant endpoints as G-BA and IQWiG are grading the endpoints, focusing on overall survival as the preferred endpoint in oncology.A taskforce of experienced German outcomes research, medical, health-technology assessment and biostatistics researchers in industry was appointed. After agreement on core assumptions, a draft position was prepared. Input on iterative versions was solicited from a panel of reviewers from industry and external stakeholders.Distinctive features of registration trials in oncology need to be considered when these studies form basis for EBA, especially in cancer-indications with long post-progression survival; and with several consecutive therapeutic options available post-progression. Ethical committees, caregivers and patients often demand cross-over-designs diluting the treatment-effect on overall survival. Regulatory authorities require evaluation of morbidity-related study endpoints including survival of patients without their disease getting worse (i.e., progression-free survival). Also, progression requires treatment-changes, another strong indicator for its relevance to patients.Based on specific guidelines and clinical trial programs that were developed to be consistent with regulatory guidance, endpoints in oncology are thoroughly evaluated in terms of their patient-relevance. This extensive knowledge and experience should be fully acknowledged during EBA when assessing the patient-relevant benefit of innovative medicines in oncology.JEL codesD61; H51; I18.

determination. 22 drugs met NICE's EOL-SPP for the indication for which they were being appraised. Twelve of these drugs had the EOL-SPP criterion applied to the only indication for which they were licensed. The EOL-SPP criterion was applied to the cumulative populations of ten drugs which had marketing authorization for more than one indication. The seven drugs that did not meet the EOL-SPP criterion all had individual indications which were within the number of what is considered acceptable (Յ7,000), but had total cumulative populations that were greater. Two STAs in particular stand out. The appraisal committee accepted that panitumumab met the EOL-SPP criterion for its current indication but noted that the EMA recommended a marketing extension which would raise the expected patient population to 10,000. In its final appraisal determination for abiraterone NICE overturned its original decision that the drug did not meet the EOL-SPP criterion, even though it noted that abiraterone may be recommended for a marketing extension for a greater patient population. CONCLUSIONS: There is no evidence to suggest NICE applies the EOL-SPP to the cumulative populations of currently licensed indications plus potential future indications. Estimates of the quality-adjusted life-years (QALYs) gained and incremental cost per QALY gained were then compared the assessments of the Amélioration du Service Médical Rendu (ASMR) made by the Haute Autorité de Santé (HAS) in France for the same drugs in the same clinical indications. RESULTS: In the UK, the estimates of QALYs gained ranged from 0.018 to 1.85 and estimates of incremental cost-per QALY from £1800 to £458,000. The estimate of incremental cost per QALY was a good predictor of the level of restriction imposed on the use of the drug concerned. Patient access schemes, which normally imply price reductions, were proposed in 45% of cases. In France, the distribution of ASMRs was 1, 16%; 2, 8%; 3, 21%; 4, 24%; 5, 24%; and uncategorized/ non-reimbursed, 8%. Since ASMRs of 4 and above signify minor or no improvement over existing therapy, these ratings imply that, in around half the cases, the drugs concerned would face price controls. Overall, the assessments of value added in the two jurisdictions produced very similar results. A superior ASMR rating was a good predictor of both higher QALYs gained and a lower incremental cost per QALY. CONCLUSIONS: We conclude that, despite the contrasting approaches employed in France and the UK for assessing the value added by new drugs, the overall assessments of value added produced very similar results. However, the implications of these assessments for patient access to, and prices of, anticancer drugs in the two jurisdictions require further investigation.