A genomic expression signature model for cervical carcinogenesis by molecular profiling of microdissected epithelial and stromal cells Free

2977

Human cervical cancer arises from graded epithelial precursor lesions known as cervical intraepithelial neoplasia (CIN) at the epithelial-stromal cell interface suggesting the molecular events in this microenvironment are critical to carcinogenesis. Cervical transformation follows a well conserved histological pattern of progression from normal epithelial cells to preinvasive CIN 1, 2, and 3 and finally to invasive squamous cell carcinoma (SCCA providing an intriguing system to use genomics to identify the early events in cervical cell transformation).

This was done by microdissection of histologically normal and abnormal epithelial and stromal cells (stromal cells were predominantly fibroblasts cells >75%) combined with a linear mixed model for analysis of gene expression patterns, accommodating for age and race. The number of patients and samples, collection methods, and microarrays (>3000 microarrays) make this one of the most comprehensive studies to date investigating the neoplastic transitions. Bioinformatic analysis identified expression pathways that coincided with histopathologic transitions from normal to cervical intraepithelial (CIN 1-3) and cancer that was used to develop the first in vivo model of early events in cervical transformation.

This model identifies three distinct genomic signatures at the epithelial-stromal interface: 1) an early CIN 1 pro-proliferative / immune suppression early genomic signature; 2) an intermediate CIN 2 pro-angiogenic stromal/epithelial interaction genomic signature; and 3) a late CIN 3 pro-invasive genomic signature. In addition, this work suggests that both neoplastic and the surrounding stromal cells individually express of pro-angiogenic factors suggesting a potential cooperative cross talk communication between stromal and epithelial cells in the tumor microenvironment to induce an angiogenic switch, as proposed by Hanahan and Folkman.

This study is the first genomic approach to determine grade-specific changes in gene expression along the entire neoplastic spectrum of human cervical carcinogenesis. It also suggests that pre-malignant epithelial cells must overcome a series of micro-environmental stresses to obtain a transformed phenotype including cell cycle checkpoints, host immune surveillance systems, nutrient shortages, and lastly cell overcrowding. While this work validates many pre-existing ideas for transformation in vitro, the real novelty is the identification of the order of these events in vivo and their association with specific CIN transitions from normal to transformed epithelial cells.