Abstract 1039: Bridging gaps in cancer care: Utilizing large language models for accessible dietary recommendations

Abstract Background: Weight management is directly linked to cancer recurrence and survival; however, oncology nutritional counseling is not typically covered by insurance, creating a disparity among patients without nutritional education and easy food access. Novel ways of imparting personalized nutrition advice are imperative to address this issue. Large language models offer a promising and sustainable path toward dietary advice tailored to individual needs. Objective: This study aimed to assess the capacity of large language models to offer personalized dietary advice to patients with breast cancer. Methods: Thirty-one prompt templates were designed to evaluate the dietary recommendations generated by ChatGPT and Gemini with variations within eight categorical variables, including cancer stage, comorbidity, location, culture, age, dietary guideline, budget, and store. Seven prompts were selected for response from four board-certified dietitians. The responses were then evaluated based on qualitative observations of the large language model-generated grocery lists and meal plans and a quantitative comparison of the calories and macronutrients of large language model- and dietitian-generated meal plans to the Acceptable Macronutrient Distribution Ranges and United States Department of Agriculture daily calorie standards. Results: Large language models generated personalized grocery lists and meal plans, adapting to location, culture, and budget, but not age, disease stage, comorbidities, or dietary guidelines. Gemini provided more comprehensive responses, including visuals and specific prices. While dietitian-generated diets offered more adherent total daily calorie contents to the United States Department of Agriculture’s estimated calorie needs, ChatGPT and Gemini offered more adherent macronutrient ratios to the Acceptable Macronutrient Distribution Range. Conclusions: Large language models can provide personalized dietary advice to cancer patients who may lack access to this care. Recommendations apply to patients with variable food access, socioeconomic means, and cultural preferences and can be a tool to increase health equity. Citation Format: Julia A. Logan, Nicole L. Simone, Sriya Sadhu, Cameo Hazlewood, Sara Burke, Hamidreza Nourzadeh, Yevgeniy Vinogradskiy, Wookjin Choi, Adam P. Dicker, Melissa Denton, Caroline Black, Corey Ciaverelli, Jacqueline Stulb, Amy Leader. Bridging gaps in cancer care: Utilizing large language models for accessible dietary recommendations [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1039.