Improved Prediction of Colon Cancer Recurrence Using the ColoNode mRNA Test: A 5-Year Survival Analysis.

Twenty-five percent of colon cancer patients having curative surgery will recur. Most of them will die from distant metastasis partly due to shortcomings of routine histopathological analysis, that is, undetected tumor cells in the regional lymph node field. We have developed ColoNode, an mRNA test in which five biomarkers are quantified and related to a house-keeping gene. Thereby, four different risk groups are defined from low to very high-risk of recurrence being highly valuable as various indicators of the risk of occult tumor cell dissemination and their aggressiveness. Here we report results of the 5-year survival analysis of a prospective study involving nine Swedish hospitals comparing ColoNode with routine histopathology. The study involves 186 colon cancer patients. The following four conclusions can be drawn: (1) ColoNode surpassed histopathology by identifying all 20 patients who were found by histopathology and an additional seven patients who recurred. Multivariate analysis provided a hazard ratio of 6.7 (p = 0.002) while histopathology lost its significance (HR: 1.3; p = 0.60). Only perineural invasion remained as an independent risk factor apart from ColoNode, with a hazard ratio of 2.7 (p = 0.038). (2) Patients belonging to medium-risk group apparently benefited from adjuvant chemotherapy treatment. (3) A total of 113 patients (≈60%) belonged to the low-risk group and can be considered cured by surgery alone. (4) Four patients were missed by both ColoNode and histopathology. Possible reasons for this are discussed. The results suggest that the ColoNode test should be added to the arsenal of prognostic tools, maybe even replacing histopathological lymph node analysis.