Prospective validation of a lymphocyte infiltration prognostic test in stage III colon cancer patients treated with adjuvant FOLFOX

Background: The prognostic value of lymphocyte infiltration (LI) of colorectal carcinoma (CC) has been demonstrated by several groups. However, no validated test is currently available for clinical practice. We previously described an automated and reproducible method for testing LI and aimed to validate it for clinical use. Patients and methods: According to National Institutes of Health criteria, we designed a prospective validation of this biomarker in patients included in the PETACC8 phase III study. Primary objective was to compare percentage of patients alive and without recurrence at 2 years in patients with high versus low LI (\#NCT02364024). Associations of LI with patient recurrence and survival were analysed, and multivariable models were adjusted for treatment and relevant factors. Automated testing of LI was performed on virtual slides without access to clinical data. Results: Among the 1220 CC patients enrolled, LI was high, low and not evaluable in 241 (19.8%), 790 (64.8%) and 189 (15.5%), respectively. Primary objective was met with a 2-year recurrence rate of 14.4% versus 21.1% in patients with high and low LI, respectively (p = 0.02). Patients with high LI also had better disease free survival (DFS) and overall survival (OS). Tumour stage, grade, RAS status and BRAF status were with LI the only prognostic markers in multivariable analysis for OS. Subgroup analyses revealed that high LI had better DFS and OS in mismatch repair (MMR) proficient patients, and in patients without RAS mutation, but not in MMR deficient and RAS mutated patients. Conclusion: Although this is the first validation with high level of evidence (IIB) of the prognostic value of a LI test in colon cancers, it still needs to be confirmed in independent series of colon cancer patients. (C) 2017 Elsevier Ltd. All rights reserved.