Related Experiment Video
Updated: Feb 23, 2026

Colorectal Cancer Cell Surface Protein Profiling Using an Antibody Microarray and Fluorescence Multiplexing
Published on: September 25, 2011
Poorly Differentiated Clusters in Colorectal Adenocarcinomas Share Biological Similarities with Micropapillary
Mineui Hong1, Jeong Won Kim2, Mi Kyung Shin1
1Department of Pathology, Hallym University Kangnam Sacred Heart Hospital, Hallym University College of Medicine, Seoul, Korea.
Abstract:
In colorectal carcinoma, poorly differentiated clusters (PDCs) are a poor prognostic indicator and show morphological continuity and behavioral similarities to micropapillary patterns (MPPs) as well as tumor buds (TBs). Epithelial-mesenchymal transition (EMT) and inhibition of cancer-stromal interactions may contribute to the development of PDCs. To clarify the biological nature of PDCs, we examined immunohistochemical stainings for β-catenin, Ki-67, E-cadherin, epithelial cell adhesion molecule (EpCAM), MUC1, and epithelial membrane antigen (EMA), which are associated with EMT and cancer-stromal interactions. The expression frequencies and patterns of PDCs, TBs, and differentiated neoplastic glands from the tumor center (TC) were compared. In the study group (117 cases), the nuclear β-catenin staining index was higher in PDCs (37.3%) and TBs (43.3%) than in neoplastic glands from TC (8.9%, P < 0.001). The mean Ki-67 labeling index in TC was 71.5%, whereas it was decreased in PDCs (31.2%) and TBs (10.2%, P < 0.001). E-cadherin and EpCAM displayed a tendency to be found along the cell membrane in TC samples (91.5% and 92.3%, respectively), whereas they showed loss of membranous staining in PDC (44.4% and 36.8%, respectively) and TB samples (60.7% and 68.4%, respectively). An inside-out pattern for MUC1 and EMA was frequently observed in PDC (48.7% and 45.3%, respectively) and TB samples (46.2% and 45.3%, respectively), but not in TC samples. Our data demonstrate that there is a pathogenetic overlap among PDCs, TBs, and MPPs and suggest that they might represent sequential growth patterns that branch from common biological processes such as dedifferentiation and alteration in cancer-stromal interactions.

