Inference Of Signaling Network Perturbations Associated With PI3K Pathway Mutations In Endometrial Cancer

Abstract

Introduction: Cancer is a complex heterogeneous disease that results from accumulation of genomic aberrations reflected in signaling network alterations. Identifying signaling network signatures involved in carcinogenesis greatly facilitates the development of rational targeted therapy. Endometrial cancer is the most common gynecologic cancer but treatment options for advanced or recurrent disease are limited and ineffective. The phosphatidylinositol 3’-kinase (PI3K) pathway has been implicated in the etiology of endometrial cancer and therefore represents an ideal therapeutic target. Through sequencing on 222 endometrial cancer patient samples, we have demonstrated frequent mutations in multiple PI3K family members including PTEN (44%), PIK3CA (p110alpha catalytic subunit of PI3K; 40%) and PIK3R1 (p85alpha regulatory subunit; 20%). Of particular interest, PIK3CA or PIK3R1 mutations co-exist with PTEN heterozygous mutations at frequencies higher than expected in contrast to most cancers, indicating the likelihood that they cooperate with PTEN aberrations in the pathophysiology of endometrial cancer. High-throughput reverse-phase protein array show that PTEN protein loss is a dominant effect resulting in PI3K pathway activation and that PIK3CA or PIK3R1 mutations mimic PTEN loss in tumors where PTEN is heterozygously mutated and PTEN protein is retained, suggesting that PIK3CA or PIK3R1 mutations might be co-selected with heterozygous PTEN mutations to compensate for incomplete loss of PTEN protein. Methods: In this study, correlation-based Relevance Networks are built after categorization of the patient samples into subgroups based on PI3K pathway mutation status (single mutation or co-mutations). Results: The results elucidate signaling network perturbations associated with different mutation and co-mutation patterns in endometrial cancer. Conclusion: Identifying signaling network signatures downstream of the mutations will aid the development of effective molecular markers and therapeutic targeting in patients with different PI3K pathway aberrations.