Role of ovarian cancer-initiating cells in high-grade serous ovarian carcinogenesis

Abstract

A subpopulation of tumor cells known as ovarian cancer initiating cells (OCICs) have been shown to be the cells that propagate the tumor phenotype in ovarian cancer. Studies have showed that a very small population (100) of these cells is sufficient to induce a tumor phenotype; while a large quantity of tumor cells (5 X 105) are required to induce such a phenotype. In this study we studied the functional changes in genes expressed in the OCIC phenotype which were important for such efficient propagation of cancers. To enable this analysis, we generated mRNA expression and DNA methylation profiles of OCICs and compared them with those of tumor and normal ovarian surface epithelial cells. We identified four pathways which regulated most of the observed changes and were predicted to be important factors in distinguishing the OCICs from tumors and normal cells. The gene signatures for these pathways were analyzed by unsupervised clustering in order to determine the similarities of OCICs with respect to tumor and normal samples. We further believed that the OCICs can be used as indicators towards the genesis and progression of early events in the ovarian cancers. In light of this, we considered two hypotheses which are currently addressing the genesis of ovarian cancer. The first hypothesis proposed ovarian surface epithelial cells to be cells of origin of the ovarian cancer while the other proposed the fallopian tube cells to be contributing the cell of origin for these cancers. It is also believed that these two cells can be reciprocal cells of origin for the cancer phenotype. In order to test these hypotheses, we integrated the in-house dataset with a public domain fallopian tube gene expression data. The integration of the results obtained from these analyses provided better understanding of the early events in ovarian carcinogenesis.