Leveraging Omics-Based Computational Approaches to Identify and Validate Novel Therapeutic Candidates for Endometriosis

ABSTRACT – PROJECT 3 Endometriosis is a common, estrogen-dependent, inflammatory disorder that causes debilitating chronic pelvic pain including severe dysmenorrhea and dyspareunia, infertility, and a reduced quality of life for 176 million women and teens worldwide. Treatment of endometriosis-associated pain is mainly surgical and/or medical. Surgical removal of disease results in 50% relapse of pain within 2-5 years. Medical treatments, largely unchanged over decades, comprise nonsteroidal anti-inflammatory drugs (NSAIDs) and hormones that lower estrogen levels or oppose its actions, and result in variable symptom relief. The development and availability of large-scale genomic, transcriptomic, and other molecular profiling technologies, in combination with the deployment of the network concept of drug targets and the power of phenotypic screening, provide an unprecedented opportunity to advance rational drug repurposing and data-driven development of drug combinations. The goal of Project 3 is to leverage endometriosis transcriptomics data combined with publicly available drug screening data and apply a computational drug-repurposing pipeline to identify single agent and combination therapies from existing drugs based on expression reversal perturbing molecular networks away from disease-associated cellular dysfunction, and validate select drugs in human endometrial cells in vitro and an animal model of endometriosis pain. In Aim 1, we will use transcriptomic-based computational drug- repurposing to identify potential new single agent and combination therapeutics based on expression reversal leveraging public transcriptomics data. Our hypothesis is that the inverse expression profiles between the drug repositioning candidates and the disease signatures will result in therapeutic predictions. In Aim 2, we will determine the capacity of compounds of interest (COIs) to inhibit inflammatory signaling responses in primary human immune and endometrial cells through the use of an ex vivo high-throughput mass-tag barcoding assay. We hypothesize that the most promising COIs identified in silico (Aim 1) will improve endometriosis symptoms by inhibiting pro-inflammatory signaling responses in endometrial and/or immune cells. Finally, in Aim 3 we will determine the efficacy of compounds of interest to alleviate pain in a preclinical endometriosis model. We hypothesize that the COIs identified for the treatment of endometriosis will alter the endometriotic microenvironment to alleviate pain. We anticipate this study will serve as the basis for studies on newly discovered novel targets and drug-repurposing as well as functional validation in endometrial tissue as well as testing in preclinical models, and if successful, clinical trials for endometriosis-associated pain in women. We hope that this novel approach will change the paradigm of “one size fits all” hormonal treatment for endometriosis-associated pelvic pain and expand therapeutic options to new therapies and established therapies repurposed to improve the lives of millions of affected women and teens and expand the research pipeline in this space.