Thursday, March 22, 2012: 10:45 a.m. - 12:15 p.m.
Presentation Type: Oral Session
Objectives: The present study aimed at gaining 3D quantitative information about cancer cell proliferation activity and microinvasion mode of 30 cases of oral squamous cell carcinoma (SCC) arising at the border of tongue from Japanese patients. Methods: The use of archival human tissues in this study was performed with the approval of the Institutional Review Board. Based on the pathodiagnostic survey on HE-stained sections, tissue core biopsies of 3mm in diameter were dissected using a paraffin tissue microarray. Serially-cut sections (4µm thick, 100 sections per individual cases) were dual-immunostained using cytokeratin antibodies for parenchyma labeling and Ki-67 for cell proliferation assessment. In the current protocol, cytokeratin-positive parenchyma (yellow) and Ki67-positive nuclei (blue) were differentially labeled with Vector stain to facilitate discrimination of parenchyma/stroma and cytoplasm/nucleus by means of RGB color segmentation procedures. All immunolabeled serial sections were digitized with virtual microscopy (NanoZoomer, Hamamatsu Photonics). Image registration and segmentation for 3D reconstruction were automated with the aid of Ratoc TRI-SRF2 software to avoid operator-dependent subjectivity. Results: The reconstructed 3D allowed us to conduct morphometry on 2 to 3 mm3 in total tissue volume and 105/mm3 nuclei, making it possible to accumulate quantitative information regarding parenchyma volume, nucleus number, and Ki67-positive/negative ratio. Histomorphometric analysis also proved the migration and sequestration of cancer cells into the surrounding stroma, in a form of single cells or coherent multi-cellular foci, indicative of the diversity of cancer invasion in the proper microenvironment. The unique advantage of the 3D morphometry is the differential diagnosis of the continuative and/or segmental cancer architecture at the infiltration front where 2D histology showed small invasive foci into tongue muscle. Conclusions: The results support the potentiality of histology-based 3D reconstruction in understanding of cancer biology and pathological diagnosis of tumor aggressiveness and malignancy. Supported by Grants-in-Aid from JSPS of Japan.This abstract is based on research that was funded entirely or partially by an outside source: Supported by Grants-in-Aid from JSPS of Japan
Keywords: Carcinogenesis, Digital image analysis, Human, Pathology and Tongue