A Comprehensive Toolbox to Facilitate Quantitative Decision Science in Drug Development: A web-based R shiny application GOahead

TL;DR

GOahead is a web-based R shiny application that provides a comprehensive toolbox for real-time, quantitative decision-making in drug development, supporting Go/No-Go decisions across various statistical frameworks.

Contribution

It introduces a versatile, user-friendly platform that integrates multiple decision criteria and statistical methods for improved drug development decision processes.

Findings

Supports Bayesian and frequentist frameworks

Includes multiple arms design with MCP-MOD

Enables prospective planning and implementation

Abstract

Decision-making is critical at each stage of drug development and making informed and transparent Go/No-Go decisions require a sound quantitative decision framework. We designed and implemented GOahead, a comprehensive web-based tool to improve how statisticians and collaborators could prospectively plan and implement the selected Go/No-Go decision approach in real-time. In the paper, we conducted a comprehensive overview of dual-criterion and confidence interval-based approaches to enable quantitative decision-making. illustrative examples are demonstrated for single and two arms designs in both Bayesian and frequentist frameworks, multiple arms design with MCP-MOD is also demonstrated. GOahead can be found on shinyapps server.