Long-term prognostic value of apoptotic index in colorectal cancer: a 5-year multicentre cohort study

Abstract

Despite TNM (Tumour, Node, Metastasis) classification being the gold standard for estimating the prognosis of early-stage colorectal cancer, significant variability in long-term survival persists among patients within the same TNM stage, underscoring the importance of the disease's biological heterogeneity and the need for novel markers. This study investigates the determinants of 5-year mortality in patients with colon or rectal cancer through the analysis of 448 diagnostic tumour samples from a prospective multicentre cohort. We assessed sociodemographic, clinical, and pathological data, as well as the apoptotic index (AI) measured by the terminal deoxynucleotidyl transferase dUTP nick-end labelling (TUNEL) technique. Overall survival was the primary outcome, and Cox regression was used to estimate the hazard ratio (HR). Multivariate 5-year survival analysis identified the highest risk associated with TNM stages IV [p<0.001, HR 12.06, confidence interval (CI) 5.75-25.31] and III (p<0.001, HR 3.52, CI 1.88-6.62), followed by an AI >1.8% (p<0.001, HR 2.16, CI 1.46-3.20), male biological sex (p<0.05, HR 1.58, CI 1.05-2.37), tumour location on the right colon (p<0.024, HR 1.55, CI 1.06-2.27), and age (p<0.001, HR 1.05, CI 1.04-1.07). Our findings underscore the long-term prognostic value of a high AI as a determinant of poor prognosis in colorectal cancer and highlight the need to refine conventional prognostic markers to enable more precise risk stratification.

Keywords: 5-year survival; TUNEL technique; apoptosis; biomarker; colon cancer; prognosis; rectal cancer.