Abstract 7093: Prediction of overall survival in stage II and III colon cancer through machine learning of rapidly-acquired proteomics

Abstract The use of adjuvant therapy in stage II-III colon cancer (CC) has been controversial for decades. No markers or models now is satisfactory enough to be adopted widely by the medical community. Here, we analyzed the proteomics of paraffin-embedded tissue (FFPE) biopsy samples from 230 CC patients (stages II-III) with up to 10 years of survival by using pressure cycling technology (PCT) and data-independent acquisition (DIA) mass spectrometry. Using machine learning, we established a novel and practical protein-based clinical classification system for CC prognosis which was further verified in an independent validation cohort. Using the nine protein-based clinical model, we improved the area under the curve (AUC) value to 0.926 in the training cohort and 0.872 in the validation cohort. Our promising model will be a potential approach to prognostication to aid in rational follow-up schedule-making and risk-adaptive individualized therapies. Citation Format: Kailun Xu, Xiaoyang Yin, Tiannan Guo, Shu Zheng, Yingkuan Shao. Prediction of overall survival in stage II and III colon cancer through machine learning of rapidly-acquired proteomics [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(6_Suppl):Abstract nr 7093.