Integration of Electronic Health Record and Molecular Data for Precision Cancer Screening and Surveillance

Recent advances in understanding the environmental and genetic factors that contribute to colorectal cancer have not yet been fully integrated into healthcare practices. For example, the current approach to colonoscopy screening primarily relies on a person’s age, treating everyone the same without considering their unique risk factors. Additionally, follow-up colonoscopies after polyp removal often depend on specific characteristics of polyp tissue, which may not effectively predict the risk of future cancer. This can lead to lower-risk patients undergoing unnecessary tests, while higher-risk patients might not receive the timely follow-up they need.

To address these challenges, our research team has developed a long-term study involving patients who have undergone multiple colonoscopy exams at Partners HealthCare. We are collecting detailed health data from electronic health records (EHR) and enhancing this information with advanced natural language processing algorithms. This data is then combined with information from the state cancer registry and the Partners Biobank, which contains genetic data. We are also collecting tissue samples for further analysis.

By integrating this comprehensive information, we aim to discover new biomarkers for early cancer detection, test prediction models in real-world clinical settings, and develop more precise tools for assessing individual risk. This will help improve cancer screening and surveillance for everyone.