Problems of using artificial intelligence technologies in clinical drug research

The use of artificial intelligence (AI) technologies in clinical trials (CT) of medicines has led to a significant reduction in time and financial costs for the development of new medicines. However, their widespread adoption faces a number of unresolved issues related to data quality, regulation, ethics, and security.

The purpose of this study was to assess the current state of AI application in CT and identify key issues hindering its widespread adoption.

Results and discussion. The key applications of AI in CT are analyzed: the use of large language models (TrialGPT, Elsa) for data analysis, the creation of digital patient counterparts and synthetic controls for CT modeling, as well as predictive analytics to optimize research design and risk assessment. It has been shown that these methods can improve the efficiency of patient selection, predict outcomes with a high area under ROC score (up to 92.7 %), and accelerate drug development. However, serious limitations have been identified: the dependence of the quality of models on the representativeness of data, the risks associated with imperfect algorithms, as well as the lack of specialized regulatory regulation and standards, including in the Russian Federation. The problems of data confidentiality, obtaining informed consent, and determining responsibility for harm caused by the use of AI are highlighted.

Conclusions. The widespread use of AI to increase the efficiency and personalization of CT, its scaling requires solving complex regulatory, legal and ethical challenges.

1. Jayatunga M.K.P., Ayers M., Bruens L., Jayanth D., Meier C. How successful are AI-discovered drugs in clinical trials? A first analysis and emerging lessons. Drug Discov. Today. 2024; 29(6):104009. doi: 10.1016/j.drudis.2024.104009

2. Thorlund K., Dron L., Park J.J.H., Mills E.J. Synthetic and external controls in clinical trials: A primer for researchers. Clin. Epidemiol. 2020;(12):457–467. doi: 10.2147/CLEP.S242097

3. Clinical trials gain intelligence. Nat. Biotechnol. 2025;43(11):1017–1018. doi: 10.1038/s41587-025-02754-1

4. Sharma A., Kosasih E., Zhang J., Brintrup A., Calinescu A. Digital twins: state-of-the-art theory and practice, challenges, and open research questions. J. Ind. Inf. Integr. 2022;(30):100383. doi: 10.1016/j.jii.2022.100383

5. Chopra H., Annu, Shin D.K., Munjal K., Priyanka, Dhama K., Emran T.B. Revolutionizing clinical trials: the role of AI in accelerating medical breakthroughs. Int. J. Surg. 2023;109(12):4211–4220. doi: 10.1097/JS9.0000000000000705

6. Khan O.F., Alvi M., Salluzzi M., Bridges R., Riskas J., Haider S.A., Sivan V., Samorodova O., Hennessy J., Shah D., Baldauf-Lenschen F., Predicting overall survival (OS) differences and treatment duration from longitudinal AI-driven imaging-based prognostication (IPRO) and first-line (1L) treatment status among patients with advanced non-small cell lung cancer (aNSCLC). J. Thorac. Oncol. 2023;20(3):S83. doi: 10.1016/S1556-0864(25)00311-9

7. Wang Z., Zhou Y., Zhang Y., Mo Y.K., Wang Y. XMR: an explainable multimodal neural network for drug response prediction. Front. Bioinform. 2023;(3):1164482. doi: 10.3389/fbinf.2023.1164482

8. Lecuelle J., Truntzer C., Basile D., Laghi L., Greco L., Ilie A., Rageot D., Emile J.F., Bibeau F., Taïeb J., Derangere V., Lepage C., Ghiringhelli F. Machine learning evaluation of immune infiltrate through digital tumor score allows prediction of survival outcome in a pooled analysis of three international stage III colon cancer cohorts. EBioMedicine. 2024;(105):105207. doi: 10.1016/j.ebiom.2024.105207

9. Fang G., Annis I.E., Elston-Lafata J., Cykert S. Applying machine learning to predict real-world individual treatment effects: insights from a virtual patient cohort. J. Am. Med. Inform. Assoc. 2019;26(10):977– 988. doi: 10.1093/jamia/ocz036

10. Seo S., Kim Y., Han H.J., Son W.C., Hong Z.Y., Sohn I., Shim J., Hwang C. Predicting successes and failures of clinical trials using outer product-based convolutional neural networks. Front. Pharmacol. 2021;(12):670670. doi: 10.3389/fphar.2021.670670

11. Lei D., Qin K., Li W., Pinaya W.H.L., Tallman M.J., Patino L.R., Strawn J.R., Fleck D., Klein C.C., Lui S., Gong Q., Adler C.M., Mechelli A., Sweeney J.A., DelBello M.P. Brain morphometric features predict medication response in youth with bipolar disorder: a prospective randomized clinical trial. Psychol. Med. 2023;53(9):4083–4093. doi: 10.1017/s0033291722000757

12. Ferdowsi S., Knafou J., Borissov N., Vicente Alvarez D., Mishra R., Amini P., Teodoro D. Deep learning-based risk prediction for interventional clinical trials based on protocol design: A retrospective study. Patterns. 2023;4(3):100689. doi: 10.1016/j.patter.2023.100689

13. Wang Z., Xiao C., Sun J. AutoTrial: Prompting language models for clinical trial design. arXiv. 2023:2305.11366 doi: 10.48550/arXiv.2305.11366

14. Cross J.L., Choma M.A., Onofrey J.A. Bias in medical AI: implications for clinical decision-making. PLOS Dig. Health. 2024;3(11):e0000651. doi: 10.1371/journal.pdig.0000651

15. Teodoro D., Naderi N., Yazdani A., Zhang B., Bornet A. A scoping review of artificial intelligence applications in clinical trial risk assessment. NPJ Digit. Med. 2025;8(1):486. doi: 10.1038/s41746-025-01886-7

16. Predetermined change control plans for machine learning-enabled medical devices: guiding principles. Digital Gov. Available at: https://www.gov.uk/government/publications/predetermined-change-control-plans-for-machine-learning-enabled-medical-devices-guiding-principles (date accessed: 01.09.2025).

17. Ibrahim H., Liu X., Rivera S.C., Moher D., Chan A.W., Sydes M.R., Calvert M.J., Denniston A.K. Reporting guidelines for clinical trials of artificial intelligence interventions: the SPIRIT-AI and CONSORT-AI guidelines. Trials. 202;22(1):11. doi: 10.1186/s13063-020-04951-6

18. Foote H.P., Hong C., Anwar M., Borentain M., Bugin K., Dreyer N., Fessel J., Goyal N., Hanger M., Hernandez A.F., Hornik C.P., Jackman J.G., Lindsay A.C., Matheny M.E., Ozer K., Seidel J., Stockbridge N., Embi P.J., Lindsell C.J. Embracing Generative Artificial Intelligence in Clinical Research and Beyond: Opportunities, Challenges, and Solutions. JACC Adv. 2025;4(3):101593. doi: 10.1016/j.jacadv.2025.101593

19. Convention for the Protection of Human Rights and Dignity in Relation to the Application of Biology and Medicine: Convention on Human Rights and Biomedicine (Oviedo, April 4, 1997). [In Russian]. Available at: https://rm.coe.int/168007d004

20. The World Medical Association Declaration of Helsinki: Guidelines for Physicians Conducting Biomedical Research Involving Human Subjects. Adopted by the 18th World Medical Assembly, Helsinki, Finland, June 1964 and revised by the 29th World Medical Assembly, Tokyo, Japan, October 1975, the 35th World Medical Assembly, Venice, Italy, October 1983, the 41st World Medical Assembly, Hong Kong, September 1989, and the 48th General Assembly, Somerset West, South Africa, October 1996. [In Russian]. Available at: https://rassep.ru/sovet-po-etike/manifesty/helsinskaya-deklaratsiya/

21. Federal Law of Russian Federation “On Circulation of Medicinal Products” dated April 12, 2010 No. 61-FZ. [In Russian]. Available at: http://government.ru/docs/all/99466/.

22. Decision of the Council of the Eurasian Economic Commission dated November 3, 2016 No. 79 “On approval of good clinical practice rules of the Eurasian Economic Union”. [In Russian]. Available at: https://roszdravnadzor.gov.ru/drugs/controllslp/documents/70579.

23. Melikhov O.G., Tsepova E.A. Main elements of clinical study agreement. Remedium. 2015;(11):54–58. [In Russian].