Unlocking New Therapeutic Potentials: An MIS-Driven Framework for Repurposing FDA-Approved Drugs
TL;DR
By enabling researchers to resume trials at advanced stages (Phase II/III), this model offers a strategic pathway to reduce R&D expenditures from billions to millions and accelerate market entry by 5–7 years.
Abstract
As pharmaceutical Research and Development (R&D) faces the mounting pressures of Eroom’s Law, characterized by a doubling of drug development costs every nine years, traditional de novo drug discovery has become an increasingly unsustainable model. In 2026, the average cost to bring a New Molecular Entity (NME) to market stands at $2.6 billion, with timelines exceeding 12 years and clinical success rates hovering near 10%. This paper proposes a transformative Management Information System (MIS)-driven framework for drug repurposing to mitigate these inefficiencies. The framework utilizes a multi-layered digital architecture to identify new therapeutic indications for FDA-approved and failed-but-safe drug candidates. The Data Integration Layer aggregates heterogeneous datasets from DrugBank, PubChem, and ClinicalTrials.gov, creating a centralized knowledge base. The Analytical Layer employs Artificial Intelligence (AI) for Transcriptomic Signature Matching—identifying drugs that inversely mirror disease gene expressions—and in-silico Molecular Docking to simulate physical binding affinities. A Decision Support System (DSS) then filters these results using weighted scoring algorithms, safety hazard filters, and economic "Phase Leap" estimations. By enabling researchers to resume trials at advanced stages (Phase II/III), this model offers a strategic pathway to reduce R&D expenditures from billions to millions and accelerate market entry by 5–7 years. The integration of a continuous feedback loop ensures that validation results refine the AI models, establishing a self-evolving intelligence system for modern pharmacology.