Microinvasion of liver metastases from colorectal cancer: predictive factors and application for determining clinical target volume.

OBJECTIVES: This study evaluates the microscopic characteristics of liver metastases from colorectal cancer (LMCRC) invasion and provides a reference for expansion from gross tumor volume (GTV) to clinical targeting volume (CTV). METHODS: Data from 129 LMCRC patients treated by surgical resection at our hospital between January 2008 and September 2009 were collected for study. Tissue sections used for pathology and clinical data were reviewed. Patient information used for the study included gender, age, original tumor site, number of tumors, tumor size, levels of carcinoembryonic antigen (CEA) and carbohydrate antigen 199 (CA199), synchronous or metachronous liver metastases, and whether patients received chemotherapy. The distance of liver microinvasion from the tumor boundary was measured microscopically by two senior pathologists. RESULTS: Of 129 patients evaluated, 81 (62.8 %) presented microinvasion distances from the tumor boundary ranging between 1.0 - 7.0 mm. A GTV-to-CTV expansion of 5, 6.7, or 7.0 mm was required to provide a 95, 99, or 100 % probability, respectively, of obtaining clear resection margins by microscopic observation. The extent of invasion was not related to gender, age, synchronous or metachronous liver metastases, tumor size, CA199 level, or chemotherapy. The extent of invasion was related to original tumor site, CEA level, and number of tumors. A scoring system was established based on the latter three positive predictors. Using this system, an invasion distance less than 3 mm was measured in 93.4 % of patients with a score of ≤1 point, but in only 85.7 % of patients with a score of ≤2 points. CONCLUSIONS: The extent of tumor invasion in our LMCRC patient cohort correlated with original tumor site, CEA level, and number of tumors. These positive predictors may potentially be used as a scoring system for determining GTV-to-CTV expansion.