Advances and Challenges in Prognosis Prediction Systems for Non-metastatic Colorectal Cancer

Colorectal cancer is one of the most common malignancies in the world, and patients without distant metastasis have a better prognosis. As some patients in this group are still at high risk of recurrence, accurate prognosis prediction and further treatment are the keys to improve the clinical cure rate of colorectal cancer. TNM staging, a widely used prognostic system in clinical practice, has some limitations: the system only includes three indicators, and uses a simple linear method to classify patients, while ignoring the objectivity of prognosis as a complex nonlinear phenomenon. At the same time, the system has never integrated or improved the newly discovered prognostic indicators, like clinical information, pathological information, molecular markers and immune markers. We speculate the above information may be used as supplementary factors to combine with the traditional prognosis prediction system, so as to improve the accuracy of prognosis assessment. This article systematically reviews the research progress in this field and proposes a possible way to build a new generation of prognosis prediction system for non-metastatic colorectal cancer.