The analysis of the PI3K/AKT pathway in human ovarian cancer with a combination of genomic and genetic approaches

Ovarian cancer is the fifth most common cancer among American women. Multiple genetic changes are responsible for ovarian cancer development and progression. Identification of these genetic aberrations can facilitate diagnosis and treatment of ovarian cancer. Here we applied Array Comparative Genomic Hybridization for whole-genome analysis of DNA copy number alterations in 93 late-stage primary ovarian tumors and 15 ovarian cell lines. We detected both previously reported as well as novel copy number alterations and refined the large regions of known aberrations into smaller distinct ones with a 1Mb resolution. Most frequent copy number gains were found at 3q and 8q (>40%), and most frequent copy number losses were found at 17q (43%). High-level amplifications and homozygous deletions harboring potential cancer genes were also detected. Pathway analysis of DNA copy number profiles revealed that the PI3K/AKT signaling pathway was most frequently targeted for copy number aberrations in ovarian cancer. Survival analysis discovered that copy number gains of two major components of the PI3K/AKT pathway, PIK3CA and AKT2, were associated with decreased patient survival time. Pathway analysis with gene expression data confirmed that the PI3K/AKT pathway was the most significantly deregulated signaling pathway in ovarian cancer cells. By systematically evaluating the PI3K/AKT pathway, we identified AKT2 as a favorable drug target. We also discovered that genetic alterations of the PI3K/AKT pathway components could serve as molecular markers to predict drug response, which could be combined with targeted therapeutic to optimize drug discovery for ovarian cancer. In summary, our study provided a comprehensive profiling of ovarian tumor genome and revealed insights of important signaling pathways in ovarian cancer; we demonstrated the capability of using copy number profiles to assist and predict prognosis, and furthermore, we identified a favorable drug target and suggested important factors (such as genetic background) that needed to be included in drug development of ovarian cancer.