Multicenter International Study of the Consensus Immunoscore for the Prediction of Relapse and Survival in Early-Stage Colon Cancer

Simple Summary This work determines the predictive value of the consensus Immunoscore in 1885 patients with AJCC/UICC-TNM stage I-II Colon Cancer (CC) from North American, European, and Asian care centers. Herein, we demonstrate that the immunity of early-stage CC patients, more than cancer cell-associated parameters, predicts outcome for stage I/II patients. Similar results were found for high-risk patients defined based on parameters such as the grade of differentiation, T4 stage, venous emboli, lymphatic invasion or perineural invasion (VELIPI). Within these pathological risk subgroups, the consensus Immunoscore accurately identifies early-stage CC patients with different clinical outcome, without treatment bias. Thus, the Immunoscore reliably diagnoses low immune cell infiltrated patients at risk of relapse, that would benefit from a more frequent and detailed medical monitoring. The Immunoscore is a patient classification method that can guide treatment decisions, through the quantification of CD3+ and cytotoxic CD8+ T-lymphocytes densities within the tumor and its invasive margin.Abstract Background: The prognostic value of Immunoscore was evaluated in Stage II/III colon cancer (CC) patients, but it remains unclear in Stage I/II, and in early-stage subgroups at risk. An international Society for Immunotherapy of Cancer (SITC) study evaluated the pre-defined consensus Immunoscore in tumors from 1885 AJCC/UICC-TNM Stage I/II CC patients from Canada/USA (Cohort 1) and Europe/Asia (Cohort 2). METHODS: Digital-pathology is used to quantify the densities of CD3+ and CD8+ T-lymphocyte in the center of tumor (CT) and the invasive margin (IM). The time to recurrence (TTR) was the primary endpoint. Secondary endpoints were disease-free survival (DFS), overall survival (OS), prognosis in Stage I, Stage II, Stage II-high-risk, and microsatellite-stable (MSS) patients. RESULTS: High-Immunoscore presented with the lowest risk of recurrence in both cohorts. In Stage I/II, recurrence-free rates at 5 years were 78.4% (95%-CI, 74.4-82.6), 88.1% (95%-CI, 85.7-90.4), 93.4% (95%-CI, 91.1-95.8) in low, intermediate and high Immunoscore, respectively (HR (Hi vs. Lo) = 0.27 (95%-CI, 0.18-0.41); p < 0.0001). In Cox multivariable analysis, the association of Immunoscore to outcome was independent (TTR: HR (Hi vs. Lo) = 0.29, (95%-CI, 0.17-0.50); p < 0.0001) of the patient's gender, T-stage, sidedness, and microsatellite instability-status (MSI). A significant association of Immunoscore with survival was found for Stage II, high-risk Stage II, T4N0 and MSS patients. The Immunoscore also showed significant association with TTR in Stage-I (HR (Hi vs. Lo) = 0.07 (95%-CI, 0.01-0.61); P = 0.016). The Immunoscore had the strongest (69.5%) contribution chi(2) for influencing survival. Patients with a high Immunoscore had prolonged TTR in T4N0 tumors even for patients not receiving chemotherapy, and the Immunoscore remained the only significant parameter in multivariable analysis. CONCLUSION: In early CC, low Immunoscore reliably identifies patients at risk of relapse for whom a more intensive surveillance program or adjuvant treatment should be considered.