Analysis of data-driven physician engagement and effects on patient recruitment for clinical trial targeting rare gene mutations.

e23286 Background: Advancements in sequencing technology have revolutionized the detection of rare mutations underlying the root cause of genetic diseases. For instance, specific rare point mutations in TP53 found in approximately 2% of all tumors (Duffy, 2022), has emerged as an actionable target, with precision treatments now under development (Dumbrava, 2022). However, conducting clinical trials for treatments targeting diseases with rare germline and/or somatic mutations presents significant challenges, including difficulties in detecting rare mutations sites (Sloan, 2018; Sornkom, 2021) and in identifying eligible patients within small, geographically dispersed populations (Kempf, 2018). Based on our previous work (Liu, 2024), we hypothesize that real-time data-driven physician engagement can accelerate patient recruitment for clinical trials involving drugs targeting rare mutations. Methods: Unstructured real-world data from biomarker testing reports were collected from 526 laboratories across the US. Diaceutics proprietary Generative Artificial Intelligence Lab Language Model (genAI LLM) was used here to capture positive results for the selected mutation sites. The linked requesting physicians were identified as candidates to receive a personalized digital engagement. The digital engagement content provides context to the rare mutation along with ongoing clinical trials. The goal of digital engagement is to increase clinical awareness of potential treatment options for patients with specific rare mutation sites, helping physicians to connect eligible patients with the appropriate treatments. Results: Within the first six months of follow-up, 129 positive testing results for the specific rare point mutation of TP53 in solid tumors were captured collectively. Based on this data, 121 physicians with eligible patients were selected for targeted digital engagement. Out of the 121 engagement emails sent, 118 were successfully delivered, with 36 physicians (~30.5%) engaged with the communication by opening and reading the email. This successful engagement may ultimately lead to the enrollment of actionable patients into matched clinical trials. Conclusions: This study highlights the significance of data-driven, physician-focused digital engagement in delivering targeted messages during critical treatment decision windows. Specifically, it addresses the challenge of patient recruitment for clinical trials involving rare mutations, ultimately facilitating the transition from positive biomarker testing results to the corresponding treatment.