Deploying an AI-based online search tool to increase patients’ access to and understanding of solid tumor GI clinical trials.

Abstract

456 Background: The number and complexity of clinical trials has been growing, making finding and accessing trials increasingly challenging. Additionally, the goal of representative trial populations remains elusive with sometimes devastating medical consequences. A key barrier to increasing recruitment efficiency is patients’ low awareness of clinical trials as an option. They rarely hear about trials from their oncologists, unless they are based at a research center, which introduces trial population biases by including only patients with good healthcare access. And even if patients decide to search for information on trials themselves, the information is difficult to understand. There are existing trial search tools, but they have multiple shortcomings: complex navigation, limited search functionality, clinical trial data is too complex for non-specialists to understand, and lack of clarity on next steps. Methods: Access to trials via a novel trial search tool (www.ancora.ai) which uses AI to restructure trial information was the main focus of this study. For increased GI cancer specificity, 6 key GI cancers were included into the AI model and information was restructured based on key eligibility differentiators identified (see data structure table). Research participants (xx GI oncology patients) were recruited with the support of patient associations (e.g. The Colon Club) leveraging social media. They were asked to complete a 20 minute survey evaluating both an established tool (clinicaltrials.gov) and the novel tool to find trials relevant for them. They had to assess different aspects of the tools’ usability on a 5-level Likert scale. Survey results were analyzed in MS Excel, using built-in statistical functions (mean, standard deviation, student’s t-test). Results: Please note the survey is ongoing and results will be updated. Current survey respondents had a base level of awareness of clinical trials (3.1±1.2), with patient associations and the internet as leading information sources. The novel tool seemed to make it easier to find trials (3.7±0.9 vs. 2.7±1.3), showed an improvement in ease of understanding information presented (3.8±1.1 vs. 2.6±1.3) and directionally provided more clarity on what next steps towards trial enrollment would be (4.2±0.8 vs. 3.7±1.4). Overall, this led to higher patient satisfaction (3.4±1.1 vs. 2.3±0.5). Conclusions: Novel patient-focused, AI-driven clinical trial support tools can unlock trial access for all patients. Democratizing trial information will not only increase trial accrual, but also patient satisfaction while reducing disparities. [Table: see text]