A “Backward” Bayesian Method for Determination of Criteria for Making Go/No-Go Decisions in the Early Phases

Decision-Making Criteria and Methods for Initiating Late-Stage Clinical Trials in Drug Development from a Multi-Stakeholder Perspective: A Scoping Review

The decision-making process in drug development involves "go/no-go" decisions, particularly at the transition from early to late-stage trials. While the decisions are solely made by drug developers, they must take into account the perspectives of multiple stakeholders-such as regulatory agencies, HTA bodies, payers, patients, and ethics committees-to ensure well-informed and robust decision-making. These perspectives influence key considerations, including resource allocation, risk mitigation, regulatory compliance, etc. To support this process, quantitative methodologies, including Bayesian and hybrid frequentist-Bayesian approaches, have been introduced to improve decision-making. However, these methodologies often do not fully account for the diverse priorities and needs of all stakeholders. This scoping review examines criteria and methods used in decision-making at the phase II to III transition, with a focus on broadening the probability of success (PoS) concept beyond efficacy alone. Our review explores PoS for different success definitions, such as regulatory approval, market access, financial viability, and competitive performance. Key themes include decision criteria selection, trial design optimization, utility-based approaches, financial metrics, and multi-stakeholder considerations in decision-making. Our findings highlight both the limitations of current methodologies and potential paths forward, including the integration of real-world data (RWD) and advanced analytics. This work complements a companion manuscript by Cetinyurek-Yavuz et al. (2025) providing a detailed review of PoS methodologies focused solely on efficacy, specifically PoS for achieving statistical significance in phase III studies, including definitions, terminologies, and analytical approaches. Together, these studies provide a foundation for advancing late-stage trial decisions toward a more balanced, data-driven, and stakeholder-aligned approach.

Bayesian multivariate probability of success using historical data with type I error rate control

In clinical trials, it is common to have multiple clinical outcomes (e.g., coprimary endpoints or a primary and multiple secondary endpoints). It is often desirable to establish efficacy in at least one of multiple clinical outcomes, which leads to a multiplicity problem. In the frequentist paradigm, the most popular methods to correct for multiplicity are typically conservative. Moreover, despite guidance from regulators, it is difficult to determine the sample size of a future study with multiple clinical outcomes. In this article, we introduce a Bayesian methodology for multiple testing that asymptotically guarantees type I error control. Using a seemingly unrelated regression model, correlations between outcomes are specifically modeled, which enables inference on the joint posterior distribution of the treatment effects. Simulation results suggest that the proposed Bayesian approach is more powerful than the method of Holm (1979), which is commonly utilized in practice as a more powerful alternative to the ubiquitous Bonferroni correction. We further develop multivariate probability of success, a Bayesian method to robustly determine sample size in the presence of multiple outcomes.