Predictive Value of CD3, CD8, CD45RO Expression in Stage II/III Colorectal Cancer

Abstract

This article describes how colorectal carcinomas (CRC) are the fourth most frequently diagnosed cancer. However, despite advances in specific treatment, approximately 40% to 50% of patients have disease recurrence following potentially curative surgery, highlighting the demand for improvement of staging, treatment and prognosis. Inflammatory infiltration composed of lymphocytes is a common feature found in neoplasms and studies have indicated an association between tumor immune infiltrates and favorable disease outcomes. The aim of this article is to assess the association of tumor infiltrating lymphocytes and DNA damage response expression levels with disease outcome and survival in stage II and III CRC. The authors examined the prognostic value of immunological biomarkers using 5-year follow-up data to train an Artificial Neural Network (ANN) designed for the prediction of personalized risk. Data for three biomarkers which were found significant were subsequently used to train the ANN which predicted 5-year survival with 72.9% sensitivity and 68.8% specificity.