Abstract 2315: AI-enabled precision oncology era: Advanced and interactive interpretation of next-gneneration sequencing (NGS) reports

Abstract Background: Next-generation sequencing (NGS) has been widely adopted in clinical practice, but there are still unmet needs among physicians regarding the interpretation and application of NGS reports. These needs include selecting multi-targeted and combination therapies, keeping up with evolving treatment options and clinical trial results, and incorporating personalized medicine based on real clinical cases. Methods: To address these challenges, we propose using large language model (LLM) to analyze and interpret clinical questions, perform semantic searches in a high-quality database, and provide customized prompts and logical outputs. Our knowledge base includes diagnostic and treatment details for 600+ tumor subtypes, 620, 000+ entries, 2, 000+ drug details, and 440, 000+ clinical trials. The database is updated automatically using a fact engine and medical literature sources. Personalized gene interpretation is incorporated using pretraining, cosine similarity, and context learning prompts. Results: Our tool, named "SmartMTB", enables personalized interpretation of genomic testing reports for precision oncology. Users input the latest clinical features using a selection-based or question-and-answer approach. The tool identifies the treatment stage and genomic variations, provides information on mutation frequencies, sensitivity and resistance of targeted and immunotherapies, and priority selection strategies for multi-gene mutations. It matches similar clinical cases and provides real-time updates on treatment options and clinical trial literature. It also facilitates patient matching for clinical trials and considers economic factors. Conclusion: We present a groundbreaking artificial intelligence (AI) tool in cancer care that revolutionizes precision oncology. This AI system, trained on extensive datasets, swiftly and accurately analyzes and interprets complex genomic and clinical data. It enables clinicians to make informed decisions about personalized treatment strategies, significantly improving patient outcomes in cancer therapy. Citation Format: Hui Chen, Zanmei Xu, Lijuan Chen, Mingmin Wang, Peng Zhang, Fei Pang, Kai Wang. AI-enabled precision oncology era: Advanced and interactive interpretation of next-gneneration sequencing (NGS) reports abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts) ; 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84 (6Suppl): Abstract nr 2315.