Hiring in life sciences? Share your open positions with our professional community. Read more Close

Advertisement

Addressing Eroom's Law: The Shift Toward Clinically Driven Artificial Intelligence in Drug Discovery.

Created on 13 Nov 2026

Authors

Jin-Zang He, Zi-Xiang Huo, Chen-Yu Wang, Yan-Hong Ye, Zheng-Jie Shi, Jia-Lu Sun, Ze-Yu Meng, Na-Na Wang, Han-Yue Ming, Sen Xu, Jian-Li Meng, Shou-Ni Wang

Published in

Current computer-aided drug design. Nov 11, 2026. Epub Nov 11, 2026.

Abstract

Pharmaceutical Research and Development (R&D) is characterized by high costs, lengthy timelines, and frequent clinical failure. Eroom's Law describes the historical decline in new drug approvals per unit of R&D expenditure. Artificial Intelligence-driven Drug Discovery (AIDD) offers new approaches to target identification, molecular design, and property prediction, but improved computational performance alone does not establish clinical benefit.
This narrative review examines how AIDD can facilitate transitioning from target identification to molecular design and clinical evaluation.
It discusses graph neural networks, structure prediction, and generative models alongside pharmacological and clinical evidence. Four dimensions structure the discussion: target structure and dynamics; subcellular localization and target accessibility; pharmacokinetics and binding kinetics; and clinical factors. They provide a qualitative framework for identifying evidence gaps, not a mathematical model or predictive tool for drug efficacy. A complementary target-molecule-clinical framework illustrates how experimental and clinical evidence can inform earlier stages of drug development.
Clinical examples, including rentosertib and GTAEXS617, emphasize distinguishing design objectives, preclinical findings, and clinical observations. Rentosertib's phase IIa trial provides an early clinical signal, although its limited sample size and duration preclude conclusions regarding sustained benefit. Training-data bias, data leakage, limited interpretability, synthetic accessibility, and uncertain generalizability beyond the training domain remain important concerns. This review acknowledges selective coverage of the literature and study-design heterogeneity.
Assessing whether AIDD can improve overall drug-development efficiency requires comparative evidence on development time, cost, safety, and clinical success. Experimental validation and longitudinal clinical follow-up remain essential for evaluating these outcomes.

PMID:
43023444
Bibliographic data and abstract were imported from PubMed on 13 Nov 2026.

Read full publication at:
Please sign in to see all details.

Advertisement

Stats

  • Community rating n/a 0 votes
  • Reviewers' rating n/a 0 votes
  • Your rating

1-terrible, 9-excellent. How would you rate this publication? Sign in in to submit your rating.

  • Recommendations n/a n/a positive of 0 vote(s)
  • Views 8
  • Comments 0

Recommended by

  • No recommendations yet.

Post a comment

You need to be signed in to post comments. You can sign in here.

Comments

There are no comments yet.

Advertisement