Improving tumour budding evaluation in colon cancer by extending the assessment area in colectomy specimens

Aims: It is recommended that tumour budding in colon cancer be counted on haematoxylin and eosin-stained sections in a hotspot area of 0.785 mm² with a ×20 microscope objective. However, tumour buds may be difficult to visualise on haematoxylin and eosin-stained sections, and counting in such a limited area may result in overestimation in cases with focal budding. The aim of this study was to assess the contributions of various factors to improving tumour budding risk stratification: increasing the number of fields counted, using cytokeratin immunostaining, and recording proliferation, the apoptotic index and the emperipoletic index in tumour buds.

Methods and results: We created an exploratory series composed of 172 cases of colon cancer in all stages, and we analysed the survival probability in a second cohort of 158 stage I-II patients. According to our results, counting of budding in 10 fields was the only factor that was significantly correlated with disease-free survival probability in stage I-II patients [hazard ratio (HR) for high versus low grade of 7.64, 95% confidence interval (CI) 5.54-27.92, P = 0.01; HR for intermediate versus low grade of 3.02, 95% CI 1.54-26.72, P = 0.04). Emperipolesis was frequently observed in tumour buds, whereas the mitotic index and the apoptotic index were extremely low. Although cytokeratin immunostaining increased interobserver concordance, it did not improve the accuracy of tumour budding grading.

Conclusions: According to our results, counting in 10 fields significantly enhanced the budding grade risk stratification in colon cancer patients, and cytokeratin immunostaining could be reserved as a complementary technique for challenging cases with an inflammatory infiltrate and/or striking fibrosis.