Abstract
Digital twins in clinical care promise to explore treatment possibilities through patient-informed computational models before implementation. FDA recognizes model-informed drug development as the use of biological and statistical models drawing on preclinical and clinical data to support regulatory decisions including dose selection, trial design, and safety investigation. Evidence supports narrowly scoped models addressing specific clinical questions rather than complete human simulations. Usefulness depends on whether the model improves decisions beyond available information, identifies data insufficiencies, and demonstrates validated performance beyond its construction dataset. Critical considerations include population applicability, transparency regarding predictions and limitations, and documented protocols for when patient reports diverge from model outputs. The most promising near-term applications may emerge from modest, well-defined successes: answering specific questions more reliably, selecting measurements more intelligently, or facilitating uncertainty discussion. Trust requires clarity about what the model predicts, for whom, over what timeframe, with documented accuracy evidence. Responsibility for interpretation must remain visible to care teams. Executives should distinguish marketing language about "complete digital twins" from evidence-supported model-informed applications with clearly defined purposes and validated performance in relevant populations.
Read the complete research: Could Your Digital Twin Help Choose a Treatment?.