How would you accelerate the preclinical, nonclinical, and clinical testing to get a drug on the market faster?

Pharmaceutical companies test thousands of compounds on the bench before they get a potential candidate for a blockbuster drug which generates more than $1 billion in revenue annually and recovers the company’s R&D investment.

How would you accelerate the preclinical, nonclinical, and clinical testing to get a drug on the market faster?

There are several ways I believe that the drug development process can be sped up. The multi-step procedure has held up for decades, despite its notoriously low success rates, which leads me to believe that the more effective way to speed up the process is to implement innovative methods and technology, rather than alter the process itself.

Drug discovery is a data-driven experiment. From its discovery to market, each drug comes with a history of a massive data base composed of high-resolution medical images, genomic profiles, metabolites, molecular structure, biological information, and subject records. The overwhelming amount of data can be challenging for researchers to keep up with, and no doubt researchers spend a significant amount of time in organizing, analyzing and storing the data that they have collected. With machine learning and deep learning, AI can correlate, assimilate and connect existing data more efficiently to discover patterns, make connections and effectively store the data pools. Pharmaceutical giant Pfizer uses the IBM Watson for Drug Discover cloud-based platform, which compromises 25 million Medline article abstracts and one million medical journal articles. Given the fact that a researcher can typically read only between 200 to 300 articles per year, having machine algorithm do the reading would dramatically increase the efficiently of the process. 

Finding new compounds during drug discovery is time consuming and difficult, for the possible combinations and arrangements are endless. Such research requires medical data on molecular structures, proteins, metabolites, and genes. Processing this huge amount of information is incredibly time consuming. Development and implementation of AI techniques that can process the same information much faster would speed up the drug development process drastically. In fact, pharmaceutical companies are in the midst of discovering such techniques. Exscientia’s AI platform encodes deep-rooted knowledge for compound design and assessment, to screen compounds in cells or animal models. By comparing the results of a newly designed compound with the anticipated performance and with other molecules, researchers are able to evolve compound designs to help drug discovery. Exscientia can rapidly synthesize and assay small batches of compounds, which could help refine the models being developed and evolve the designs (Brazil).

I’ve only discussed how artificial intelligence can be applied during drug discovery to accelerate the process. However, artificial intelligence can be applied in the many other stages of drug development. Clinical trials, for example, require more than three thousand subjects throughout phase I to III. Most companies use recruitment firms to find subjects and by examining individual medical records, which not only is not an efficient process but also takes a lot of time. Machine learning could be used during this process by developing a model that includes the medical information of a subject to build an inclusion/exclusion criteria that will speed up the evaluation process.


[1] The Drug Development Process:

[2] Tufts CSDD R&D Cost Study Now Published:

[3] Discovery and Development:

[4] AI Provides New Insights for Accelerated Drug Development:

[5] AI in Pharma and Biomedicine – Analysis of the Top 5 Global Drug Companies:

[6] What if AI could take your research to the next level?:

[7] Machine Learning Drug Discovery Applications – Pfizer, Roche, GSK, and More:

[8] Artificial Intelligence: will it change the way drugs are discovered?:

Categories: Clinical