OBJECTIVES: Recent value-based frameworks, such as ASCO, provide algorithms to compute the relative value of oncology therapies. Such algorithms are typically based on a weighted combination of endpoint scores that provide an avenue to optimize the value outcome. As such, this study aims to: 1) propose a platform to incorporate ASCO’s evolved methodology to compute net health benefit (NHB), 2) assess sensitivity of underlying clinical endpoints, and 3) provide real world examples of how the platform may be employed. METHODS: A Monte Carlo simulation-based platform was developed to compute a baseline distribution of NHB scores of the advanced disease setting, and conduct sensitivity analysis of endpoints. The endpoints spanned three key drug attributes: Clinical Benefits including Hazard Ratio (HR), median Overall Survival (mOS), HR for Disease progression (HR DP), median Progression-free Survival (mPFS), Response Rate (RR)), Drug Toxicity (T), and bonus attributes including Tail of the curve (ToC), Palliation (P), QoL (Q), and Treatment-free Interval (TFI). Further for each endpoint combinations, associated clinical trial costs were estimated. NHB scores were then compared against those costs to determine optimal endpoint combinations for clinical trials of three example oncology drugs. RESULTS: While the NHB scores trace a bell curve with a maximum score of 140, the median scores across all endpoints were in the 17-70 range. The levers with the highest sensitivity were HR and mOS. Most importantly, holding clinical benefit score constant, Toxicity has higher sensitivity vs Palliation or Treatment-free Interval. CONCLUSIONS: While clinical benefit is the primary driver for the NHB score, multiple combinations of clinical endpoints were possible to achieve the same NHB score, suggesting that if the criteria of success includes optimizing NHB, certain clinical endpoints may be prioritized to achieve relatively lower oncology trial costs. This platform may be extended to additional value-based frameworks such as ICER in future studies.