Purpose: To establish an objective histologic grading system of venous invasion.
Methods: A total of 229 patients with pT3 and pT4 colorectal cancer who underwent curative surgery with lymph node dissection were retrospectively analyzed. Potential prognosis-related characteristics of venous invasion, including the number of venous invasion, morphologic type of venous invasion, maximum size of veins invaded, and location of venous vessel involved were evaluated on elastica van Gieson stained sections.
Results: The relapse-free survival curves between the venous-invasion-positive group and the negative group were significantly different (5 y survival rates were 73.4% and 92.2%, respectively, P=0.001). When patients were divided into 3 groups according to the average number of venous invasions observed in a glass slide [G0 (none), G1 (positive but <4), and G2 (4 or more)], there was a significant difference in the survival rate among the 3 groups [5 y survival rates were 92.2%, 77.8%, and 56.4%, respectively, P=0.008 (G0 vs. G1), P=0.017 (G1 vs. G2)]. The postoperative recurrence rate was 10.8% in the G0 patients, whereas it was 32.5% in the G1 and 51.7% in the G2 patients [P=0.0007 (G0 vs. G1), P=0.047 (G1 vs. G2)]. Multivariate analysis showed the number of venous invasions [hazard ratio (HR) 2.72, P=0.027], depth of invasion (HR 2.26, P=0.014), and lymph node metastasis (HR 2.43, P=0.008) were independent prognostic factors.
Conclusions: Three ranked tumor grading system based on the number of venous invasion in a glass slide with elastica van Gieson staining could be an objective and important treatment index for colorectal cancer patients.