Beyond H&E: CDX2 immunohistochemistry in tumor budding assessment of colorectal carcinoma.

Abstract

BACKGROUND AND OBJECTIVES: Tumor budding (TB) phenomenon is recognized as one of the important adverse prognostic factors of colorectal carcinoma (CRC) and is currently assessed following to the International Tumor Budding Consensus Conference (ITBCC) recommendations on routine Hematoxylin and Eosin (H&E)-stained slides. However, TB evaluation may become difficult in cases showing inflammatory infiltrates, stromal desmoplasia, or glandular fragmentation. Immunohistochemical (IHC) markers, including caudal type homeobox 2 (CDX2), may improve the identification of isolated tumor buds in such situations. PATIENTS, MATERIALS AND METHODS: The retrospective study included 170 CRC cases selected from the archives of a single regional pathology center. TB was evaluated on both H&E-stained and CDX2-immunostained slides according to a modified ITBCC-based grading system including an additional Bd0 category for cases lacking identifiable tumor buds. Tumor buds were counted within a standardized field area. TB distribution was evaluated in the proximal, middle, and distal tumor regions. RESULTS: Low-grade budding lesions predominated in all evaluated tumor regions on both H&E and CDX2 assessment. Regional variation in TB distribution was observed along the invasive tumor front, supporting the presence of spatial heterogeneity in CRC. Comparative analysis of TB evaluation on H&E-stained preparations and on CDX2 immunolabeled preparations revealed the existence of differences between the two methods, especially in cases with low or intermediate budding activity. CDX2 immunostaining facilitated the identification of additional tumor buds, leading to the reclassification of a number of cases to higher budding categories (intermediary and high-grade budding lesions), especially in difficult cases, characterized by intense stromal reaction, inflammation or glandular fragmentation. CONCLUSIONS: The present study demonstrated that IHC labeling should be considered primarily as a detection tool for accurate TB assessment, but other antibodies should be tested to increase the accuracy of tumor bud identification.