Reducing Uncertainty In Early Oral Drug Development: Decisions That Benefit From Predictive Insights

In early oral solid dose development, some of the most consequential formulation decisions are made when data are thinnest and materials are scarcest. Poor solubility characterization alone accounts for over 50% of IND failures, stability-related issues contribute to more than 40% of NDA delays, and early-stage inefficiencies can represent up to 30% of total development costs. Sequential, empirical experimentation remains necessary, but it carries real risk when applied without predictive direction.

This paper examines five decision points where AI and ML-driven modeling can meaningfully reduce uncertainty before significant API is consumed: solubility enhancement strategy, FIH dose and exposure projection, API–formulation compatibility, stability as a design input, and scale-up feasibility. Two case studies illustrate what this looks like in practice, including a compaction simulation study that required 45 grams of API while preserving 30 kilograms of material and accelerating validation by three months. Explore the full paper to see how predictive formulation insights are being applied across early development programs.