myTomorrows, a global health technology company working to help patients discover and access treatments, announced today the launch of the beta version of TrialSearch AI.
The first-of-its-kind artificial intelligence (AI) tool will help to broaden physician insights into clinical trials and expanded access programs (EAPs) available in myTomorrows' database, which is curated for people looking for investigational pre-approval options and sourced from public registries. Physicians can use the tool to simplify the search process by quickly and accurately identifying opportunities for suitable pre-approval treatments, based on a patient’s profile. myTomorrows is currently accepting requests from physicians to trial the tool.
“Our team of researchers and developers designed TrialSearch AI to bring greater speed, transparency, and accuracy to the complex, time-consuming task of manually searching for clinical trials” said Michel van Harten, MD, CEO at myTomorrows. “As a physician who cared for patients diagnosed with cancer, I know first-hand the importance of providing actionable information that can help physicians, patients, and caregivers discover and access potentially life-saving treatments.”
Latest research shows that LLM AI technology can be used reliably in clinical trial searches and can reduce pre-screening time by up to 90%.
Physicians can enter a basic, de-identified patient profile into the tool, such as the patient’s condition, age, sex, and country of residence, and other medical information related to patient history in a free-text form. This input is used to identify relevant clinical trial eligibility criteria and to provide a list of clinical trials and, if suitable, expanded access programs, that may be appropriate for the patient.
Physicians will then complete a series of minimal manual checks to ensure correctness of the tool’s results. The physician can transparently check what criteria were identified, and if the analysis of the tool is correct to the best of their clinical knowledge. This ensures that physicians are still at the very center of the patient’s treatment discovery journey and no decisions are made without them. Utilizing TrialSearch AI significantly reduces pre-screening checking time, research shows by up to 90%, for treating physicians. By using myTomorrows’ end-to-end platform, physicians can then refer patients to clinical trial sites meaning they are able to search, review and refer in a matter of minutes.
TrialSearch AI operates on the GPT series of LLM AI models developed by OpenAI. Utilizing LLM, myTomorrows searches worldwide for possible treatment pathways, giving physicians an unbiased and comprehensive breakdown of pre-approval options for which patients may be eligible. The tool will then filter exclusion criteria based on physician input of information describing a patient’s condition and profile.
In an article pre-published on arXiv, myTomorrows’ investigators showed that a tool incorporating GPT-3.5, can efficiently assist physicians in pre-screening clinical trials, as tested in 10 (fictive) patient profiles across 10 different diseases. These findings underpin both the validity of TrialSearch AI and the anticipated value it offers to healthcare providers to find potential pre-approval options quicker and easier, therefore potentially furthering patient access to clinical trials and other pre-approval options.
For more information about TrialSearch AI, or to join the wait list to trial the tool, visit webpage.