OBJECTIVES: The unmet need for oncology therapies is substantial. The introduction of innovative, high-cost treatments, coupled with mounting budgetary pressures, will necessitate value trade-offs across cancer types. Defining value will be critical to informing decision-making. Relative value assessment was conducted using a diverse set of outcome metrics and treatment costs for a variety of metastatic cancers.   METHODS: A review of sources published between 1/1/2000-11/30/2015 identified outcomes data for approved treatments for metastatic cancers. Data were extracted or derived for median and mean OS, milestone survival rates, and other outcomes metrics. Relative outcomes were compared across treatments with and without consideration of costs.  In cost-outcome analyses, the cost per given survival benefit identified agents with the highest relative value. RESULTS: Survival metric results varied by agent within cancer type. For instance, when used for treating NSCLC, nivolumab yielded the highest improvement in mean OS (4.1 months, pre-treated squamous disease) and 1-year survival rate (10.8 percentage points, pre-treated non-squamous disease), whereas afatinib yielded the highest median OS improvement (4.1 months, 1^st-line EGFR Del19 and L858R mutants). For treatment of prostate cancer, abiraterone yielded the highest improvement in 1-year survival rate (14.4 percentage points, pre-treated), while enzalutamide yielded the highest median OS improvement (4.8 months, pre-treated) and sipuleucel-T the highest mean OS improvement (3.6 months, 1^st-line). Results of cost-value analyses varied with the applied metric as some agents achieved a higher cost-value with some metrics but not with others.  CONCLUSIONS: Valuing oncology therapies based on improvements in median OS yields distinct results versus other survival metrics. Unlike other metrics, median OS fails to account for long-term survival benefits. Other frameworks developed for assessing the value of oncology therapies (e.g., ASCO, ESMO) are driven by median OS/PFS benefits and may not adequately value long-term survival, which is the ultimate goal of cancer treatment.