OBJECTIVES: Overall survival (OS) for cancer drugs is an important clinical endpoint to payers. This study examined oncology reimbursement recommendations by CADTH’s pan-Canadian Oncology Drug Review (pCODR) for products that that lacked statistically significant OS data and the factors that may have influenced these recommendations.

METHODS: All publicly available pCODR final recommendation reports were identified from 1 January 2016 to 31 December 2017 from which dates, decisions, appraisal outcomes, and clinical endpoints were extracted.

RESULTS: Forty-one recommendations were issued by pCODR; 27 (66%) were positive (ie, with or without clinical criteria and/or conditions) and 14 (34%) were negative (ie, uncertain or insufficient clinical benefit and/or uncertain cost effectiveness). Among positive recommendations, 10 (37%) had statistically significant OS data and 17 (63%) did not report or have statistically significant OS data. Among negative recommendations, 7 (50%) had statistically significant OS data and 7 (50%) did not assess or have statistically significant OS data. Four (24%) positive recommendations without statistically significant OS data noted crossover as a potential confounder of OS. Additional reasons noted for positive recommendations without statistically significant OS data were: limited treatment options/unmet need, and demonstration of clinical benefit in progression-free survival (PFS) and/or other clinical endpoints.

CONCLUSIONS: This study highlights that positive pCODR recommendations may be made in the absence of a clear OS benefit, likely due to crossover as a confounder and/or presence of strong PFS data. Further research may determine additional factors contributing to positive and negative pCODR recommendations despite unavailable OS data.