Abstract Colorectal cancer (CRC) is one of the cancers most highly affected by diet. Recommendations for reducing risk of CRC include weight management, eating plentiful plant-based high fiber foods, and limiting intakes of red meats and ultra-processed foods. In addition, an increased proportion of monounsaturated and omega-3 fats in the diet is beneficial for promoting weight management and reducing inflammation. Despite the evidence supporting the importance of diet and the demonstrated success for eliciting dietary changes in research settings, simple methods that can be easily implemented in medical settings are lacking. We developed a bespoke mHealth app (MyBestGI), incorporating an autonomy-supporting, self-regulatory approach to behavior change. The program promotes identification of the benefits of healthy eating on well-being. Two versions of the program are being tested in a 12-month, 3-arm randomized trial that seeks to recruit 240 people at increased risk for CRC. Participants are randomized to either 1) a control group that receives written information on cancer preventive diets, 2) a treatment group that receives a MyBestGI app version for logging 4 food groups associated with increased risks of CRC (red meats, processed meats, added sugars and refined grains), or 3) a treatment group that receives a MyBestGI app version for logging the same 4 food groups to limit plus 7 food groups to encourage in personalized quantities. The two treatment groups also receive a user manual, behavior-oriented, biweekly text messages, and supportive coaching calls. Logging food groups is requested at least 3 per week for 3 months, and 3 per week every other week for the next 9 months. The app displays logging results relative to individualized targets and has features that prompt reflecting on the user’s own results and planning for future eating. The primary endpoints are weight loss and improvement in a dietary cancer prevention score. Recruitment began in June 2023, and the study is enrolling 8-10 participants per month. The drop-out rate is lower than expected. Compliance with study procedures and app use is high. Responses to the end-of-day reflection questions within the app show that 92% of users are neutral, happy, or very happy with their dietary behavior for the day. For the 15 participants who had 12 weeks of logging data available, foods were logged for an average of 38 days (range 20-75) which is slightly higher than the requested logging of 36 days over 12 weeks. Review of the support calls for protocol fidelity indicates participants are enthusiastic about the MyBestGI program, making changes in their eating, and are internalizing the autonomy-supporting beliefs promoted in the program. These early results suggest users are engaging with the MyBestGI app at a very high level, making use of the provided app features, and enjoying the program. The MyBestGI approach therefore has excellent potential to support dietary change towards a cancer preventive diet in a format that is facile to implement in high-risk populations. Citation Format: Zora Djuric, Michelle Segar, Ananda Sen, Reema Kadri, Rob Adwere-Boamah, Juno Orr, Katherine Poore, Samara Rifkin, Lorraine Buis. Multimodality dietary intervention for colorectal cancer prevention: The MyBestGI randomized trial [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr LB375.