Go No Go

The purpose of this project is to develop an interactive R package and Shiny application designed to streamline Phase 2 go/no-go decision-making by leveraging longitudinal dose-response modeling. This innovative tool consolidates multiple evidence sources—such as MBMA, PBPK, and QSP models—to evaluate the probability of success relative to predefined Minimum Acceptable and Target Values. By enabling users to assess predictive probabilities, visualize complex model outputs, and simulate various scenarios, the app aims to support informed, data-driven clinical decisions, ultimately enhancing the robustness and confidence of go/no-go determinations in drug development.

Project Description: an interactive R package and Shiny application that facilitates Phase 2 go/no-go decision-making based on longitudinal dose-response modeling. The app integrates multiple sources of evidence—including MBMA, PBPK and QSP model outputs—to assess probability of success against predefined Minimum Acceptable Value and Target Value. Users can evaluate predictive probabilities of success under various assumptions, visualize model outputs, and simulate alternative scenarios to inform robust clinical decisions.

Tools and Skills Students will Use and Learn: Implementation of R package and shiny app, nlme, JAGS/STAN

Preference for Student Profile: Proficiency in R and JAGS/STAN