Clinical trials are an essential step in formulating a new drug and each year the number of trials continue to grow. The overall success rate for clinical trials are around 14% (U.S. Food and Drug Administration, 2019), with the progression from Phase II to Phase III being the least successful. According the findings of Wong, Siah and Lo (2018), the probability of a successful transition from Phase I to Phase II 66.4%, from Phase II to Phase is 58.3% and Phase III to approval is 59.0% (Wong, Siah and Lo, 2018)
In 2018, the US Food and Drug Administration (FDA) approved 59 new drugs (U.S. Food and Drug Administration, 2019), the highest figure to date. When looking at the procedural aspects of a trial for example: trial design, patient recruitment and patient monitoring, the introduction of Artificial Intelligence (AI) can potentially streamline the process in a variety of ways, particularly by reducing the incidence of human errors which will increase cost and decrease time efficiency.
The ability to use wearable technology to gather participants real-time diagnostic information, can expedite data collection. Which will provide constant, accurate and honest feedback without potential bias, as a result of self-reporting (Harrer et al., 2019). According to a pilot study carried out by Bain et al (2017), the use of AI allowed for real-time monitoring of pills taken and resulted in a greater number of patients adhering to the medication regimen including the identification of patients with irregular behaviour (e.g. missed doses).
The adaptability of AI would allow it to recognize certain trends over the course of a trial and identify where a trial could be going wrong, which could potentially increase the success rates of future clinical trials. The potential for AI is vast and by gradually incorporating its use into different aspects of clinical trials, it could inspire new possibilities for the future of drug development.
For more about the application of AI in the pharmaceutical field : https://blog.blackswan-analysis.co.uk/artificial-intelligence-a-new-era-of-data-driven-medicine
Bain, EE., Shafner, L., Walling, DP., Othman, AA., Chuang-Stein, C., Hinkle, J. and Hanina, A. (2017) Use of a Novel Artificial Intelligence Platform on Mobile Devices to Assess Dosing Compliance in a Phase 2 Clinical Trial in Subjects With Schizophrenia, JMIR Mhealth Uhealth 2017;5(2)
Harrer, S., Shah, P., Antony, B. and Hu, J. (2019). Artificial Intelligence for Clinical Trial Design. Trends in Pharmacological Sciences, 40(8), pp.577-591.
U.S. Food and Drug Administration. (2019). Novel Drug Approvals for 2018. Available at: https://www.fda.gov/drugs/new-drugs-fda-cders-new-molecular-entities-and-new-therapeutic-biological-products/novel-drug-approvals-2018#top [Accessed: 26 Sep. 2019].
Wong, C., Siah, K. and Lo, A. (2018). Estimation of clinical trial success rates and related parameters. Biostatistics, 20(2), pp.273-286.