Association of the tumor microenvironment collagen score and immunoscore with colon cancer lymph node metastasis

Le résumé fourni par la source

BACKGROUND: In clinical practice, lymph node status has an important impact on colon cancer (CC) management and treatment. The role of the tumor microenvironment collagen score and immunoscore in colon cancer lymph node metastasis remains unknown. METHODS: A total of 249 CC patients who underwent laparoscopic-assisted D3 lymphadenectomy from June 2016 to May 2019 were included. The patients' clinicopathological data were collected retrospectively. A total of 142 collagen features were extracted by multiphoton imaging and collagen quantification. A collagen score was constructed using a LASSO logistic regression model. Antibodies against CD3 and CD8 were used for immunostaining. The immunoscore was constructed based on the mean densities of CD3 + and CD8 + T cells both in the tumor center and invasion margin on imaging. RESULTS: The lymph node metastasis rate among colon cancer patients was 42.2% (105/249). The multivariate analysis indicated that lymphatic invasion (OR: 3.892, 95% CI: 1.784-8.491, p = 0.001), vascular invasion (OR, 3.234, 95% CI: 1.544-6.776); p = 0.002), mucus adenocarcinoma and signet-ring cell carcinoma (OR: 2.990, 95% CI: 1.413-6.328, p = 0.004), the collagen score (OR: 6.304, 95% CI: 2.145-18.527, p = 0.001) and the immunoscore [intermediate group (OR, 2.473; 95% CI, 1.192-5.130; p = 0.015); low group (OR, 5.877; 95% CI, 2.423-14.257; p < 0.01)] were independent risk factors for colon cancer lymph node metastasis. The newly developed model comprising these five independent predictors showed good discrimination with an AUROC of 0.809 (95% CI: 0.755-0.862). The new model performed significantly better than the traditional clinicopathological model [AUROC: 0.715 (95% CI: 0.649-0.780), p < 0.001]. CONCLUSIONS: The tumor microenvironment collagen score and immunoscore are associated with colon cancer lymph node metastasis.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.