Novel Artificial Intelligence Tool May Help Patients with Cancer Find, Understand Clinical Trials

Data from an ongoing trial presented at the 2021 American Society of Clinical Oncology Gastrointestinal Cancers Symposium demonstrated that patients with gastrointestinal cancers benefitted from using an artificial intelligence-based search tool to find and understand clinical trials.

“Patients have limited access to an understanding of clinical trials and the online search tools that we have available – not only for patients, but even for providers – can be very hard to navigate,” study author Dr. Pashtoon Kasi, a clinical assistant professor of internal medicine, hematology, oncology and blood and marrow transplantation at the University of Iowa Health Care, said during a virtual presentation of the data.

Trial enrollment may become more representative of diverse populations if the understanding of, and access to, clinical trials is improved, according to the study authors. In fact, they wrote that that lack of diversity in clinical trials has been an issue that can result in “sometimes devastating medical consequences.”

READ MORE: Patients Appear Receptive to Use of Artificial Intelligence for Skin Cancer Screening

As a result, the authors aimed to assess a novel clinical trial search tool in 29 patients with gastrointestinal cancers — who represented a diverse population in terms of ethnicity, education and understanding of clinical trials — and see if patient experience improved with the tool, compared to clinicaltrials.gov (which is traditionally used to search for trials).

The novel tool, cancertrialsearch.com, helped match patients with gastrointestinal cancers to clinical trials using seven different criteria: trial sponsor, trial phase, tumor type, cancer stage, across solid tumors, by genetic mutations, and by previous treatment.

After utilizing cancertrialsearch.com, patients were asked to complete a 20-minute survey evaluating both clinicaltrials.gov and the novel tool. Using a five-point scoring system (the Likert scale), patients rated aspects of usability for each platform.

“We took GI clinical trials as an example of how this would be applicable to several other tumor types,” Kasi said. “The trial information is restructured based on key eligibility differentiators, [and] is patient-focused in terms of providing questionnaires that have been based on eligibility differentiators, which could be further personalized, and a value for institutions as well as patient advocacy groups.”

READ MORE: Artificial Intelligence May Accurately Predict Short-Term Mortality in Patients with Cancer

At baseline, the average rating for patients’ understanding of clinical trials was 3.1. Patient associations and the internet were their main sources of information about trials. Finding clinical trials was easier for patients using cancertrialsearch.com, compared to not using the tool (3.7 vs. 2.7, respectively).

Additionally, the authors noted that the tool improved patients’ understanding of information presented (3.8 vs. 2.6); and directionally provided more clarity on how to enroll in trials (4.2 vs. 3.7). Ultimately, these factors led to improved patient satisfaction (3.4 vs. 2.3).

Cancertrialsearch.com also could lead to better connection among principal investigators, patients and caregivers, advocacy groups, according to the study authors.

“In summary, with this cancertrialsearch.com tool, personalizing and simplifying client information can lead to better patient comprehension, as well as ease of access and navigation, leading to better patient satisfaction and an increased trial pool,” Kasi said. “This could mean broader and more diverse patient pool.”

A version of this story appeared on OncLive® as “AI-Based Tool May Help Patients Find GI Cancer Clinical Trials.”

For more news on cancer updates, research and education, don’t forget to subscribe to CURE®’s newsletters here.