Artificial Intelligence Application in Early Phase Drug Development
Trends in Early Phase Drug Development
It’s the start of a new year, and our thoughts are turning to future trends in the early phase drug development landscape. The realm of new drug testing, where we ensure the safety of new treatments that can change people’s lives, is always evolving, and there’s no shortage of exciting technology and capabilities on the horizon.
Artificial Intelligence
Artificial intelligence, or AI, has been on the radar screen as a possible tool to improve drug development, and clinical trial design and conduct for some time. Many major pharmaceutical companies are using AI in the discovery phase. In 2019, venture capital funding for AI in drug development and clinical trials reached $5.2 billion1, a slight decline over 2018.
Natural Language Processing
AI researchers are also finding practical application in the field of clinical trials, with a particularly relevant branch of AI called natural language processing, or NLP. NLP makes it possible for computers to analyze the written and spoken word. It would enable computer analysis and decision making from structured and unstructured data from medical records, relevant guidelines, real-world data, and other sources that could potentially enhance the quality and efficiency of clinical trials in many ways.
Medical data, like most information, is increasingly digitized and captured in electronic format. Electronic health record databases and wearable devices (among others) capture and store huge amounts of data. An NLP algorithm can be trained to spot terms, synonyms, or abbreviations with a common meaning that doctors may use interchangeably. With this ability to identify information across a vast number of files, quickly, there is potential to save billions of dollars, to speed up medical advances and to expand access to experimental treatments.
Participant Recruitment with NLP
For example, the participant recruitment process is often the most time-consuming and expensive step of a clinical trial. Use of AI with NLP capability could allow algorithms to search large numbers of medical files for people who would be eligible to participate in a given clinical trial, and reduce the timeline and cost of participant recruiting by the CRO.
To further facilitate recruitment, efforts are underway to make it possible for computers to interpret the descriptions of clinical trials. The inclusion and exclusion criteria are commonly written in plain text. In order for medical facilities to search their patient databases for people that meet the eligibility criteria, these criteria must first be translated into a standardized, coded query format that the database can understand. Chunhua Weng, a biomedical informaticist at Columbia University in New York City, and her colleagues built an open-source web tool called Criteria2Query that uses NLP to enable researchers and administrators to search databases for potential trial participants without needing to know a database query language.
Clinical Trial Design with AI
Another area in which AI is being applied is in clinical trial design. Every clinical trial has a protocol that exactly describes how the trial will be conducted. When designing a trial, researchers depend on information from numerous sources, including other similar studies, previous clinical data, and regulatory guidelines. During the trial, if a problem occurs that leads to amending the protocol, it can lead to lengthy delays, and significantly increase the trial cost. Drug development is quicker, and less costly, when protocols are well designed. AI-powered software can process all the necessary input far faster than a human, and can also collate more data than a person could read.
Our Solutions
Enhancing the efficiency of early phase drug development is always top of mind for us at Altasciences. While the application of AI technology is tested, and its use matures, we continue to excel with our tried and true approaches for trial design and recruitment. Our integrated, efficient processes, that ensure preclinical and clinical researchers have access to all our sponsors’ information in a timely fashion, help us make timely decisions and delivery quality results.
Contact one of our experts today!
