Review proposes a staged roadmap for digital twins in drug evaluation and flags practical and regulatory challenges

The review states that improved computational resources and AI development allow current DT to integrate multi-omics data via hybrid mechanism- and data-driven models, enabling personalized simulation of patients, organs, and cells. Compared with understandings that treat DT mainly as an engineering or manufacturing concept, this work locates DT's capability boundary explicitly in multi-omics integration and multi-scale (patient, organ, cell) personalized simulation. This judgment comes from the review abstract's summary statement, reflecting the authors' synthesis of the field; the abstract provides no specific datasets, sample sizes, or controlled experiments.

The review argues that these capabilities make DT application in drug evaluation and drug repurposing discovery possible. It extends DT's application scenario from general modeling and simulation to the specific and critical drug-development stage of drug evaluation and drug repurposing discovery. The abstract phrases this as possibility ('makes ... possible'), a directional judgment without specific drug cases or quantified effects.

The authors review current and emerging DT applications across the drug evaluation continuum and propose a staged development roadmap. Rather than scattered application reports, this work organizes evidence from a 'continuum' perspective and offers a staged advancement path, providing the field with a structured framework. This is a review-type synthesis and framework proposal; the abstract does not list the roadmap's specific stages or criteria.

The authors emphasize that practical and regulatory issues have emerged amid rapid development, and identify pivotal challenges that must be addressed to realize DT's full potential in drug evaluation. It expands the discussion from technical feasibility to implementation and regulatory dimensions, signaling that conditions beyond technology also determine whether DT can truly be used in drug evaluation. The abstract summarizes these as 'pivotal challenges' without enumerating them or detailing regulatory provisions.

This work is positioned as a review and roadmap for the drug evaluation setting, intended for researchers, drug developers, and regulatory-oriented readers interested in how DT can be phased into drug R&D. It offers an organization of the application landscape and an advancement path rather than validation results for a specific drug or model; readers can use it to locate which roadmap stage their work occupies and which practical and regulatory conditions remain to be addressed.

Because the loaded text is incomplete, containing only the abstract plus acknowledgements and metadata, the roadmap's specific stages, the itemized challenges, the cited application cases, and regulatory provisions are not presented, so their coverage and argumentative detail cannot be judged here. Readers concerned with concrete implementation conditions and regulatory pathways for DT in drug evaluation still need to consult the full text; moreover, the abstract's 'possible' is a directional judgment whose real-world feasibility awaits support from the original evidence.