Spinal cord injury (SCI) is a devastating condition with no available treatment to date, affecting millions of people worldwide and reducing dramatically their life quality and expectancy.

Despite the obvious motor and sensory impairments, SCIs are responsible for multiple secondary medical complication such as neuropathic pain, sexual function, bladder and bowel function, autonomic dysreflexia, and more. The prevalence of neuropathic pain in SCI patients ranges between 50%-80% within a year post injury and is always placed at the top three symptoms as to the level it impedes their life quality. Despite recent advancements in our understanding of pain mechanisms, an efficient and reliable treatment regimen for neuropathic pain after SCI does not exist. The efficacy of various treatment schemes is highly variable, and they are often presented with considerable side effects with the most important being the development of opioid addiction. Recent reports suggest the early diagnosis of neuropathic pain as the most effective “treatment” since neuropathic pain symptoms can become unmanageable if not targeted early. Thus, discoveries of biomarkers for prediction of neuropathic pain development and non-opioid drugs are the most important research problems in the field of SCI and neuropathic pain.

Our proposal has two main goals. First, to develop a prognostic model for neuropathic pain development 6 and/or 12 months after SCI using acutely data collected during the hospital stay after the SCI. Second, to identify systemic gene expression patterns in the white blood cells (WBC) of chronic SCI patients that are enriched in those who experience neuropathic pain and use these patterns to computationally predict and validate in vitro of known chemical non-opioid compounds that target them. To achieve that we propose three specific aims:

In Aim 1, we will take advantage of the large database of acute human SCI data assembled by our team during our prospective clinical study (TRACK-SCI). We have enrolled over 200 SCI patients, and the study is still ongoing. The clinical database consists of up to 22,000 acute care data for each patient, expression levels of 17,000 genes in white blood cells within 24 hours post injury, and over 500 unique blood tests (blood counts, metabolic and biochemical panels, etc.) across all patients during their hospital stay. Following previous successful attempts of our team we will apply machine learning approaches to harness these large datasets and create a mathematical model that predicts the development of neuropathic pain at 6 and/or 12 months post SCI. Our findings will be validated using additional data collected at three other TRACK-SCI sites (University of California San Francisco, Fresno, Ohio State University, and University of Washington).

In Aim 2, we will use RNAseq data from WBCs of chronic SCI patients and using gene co- expression network analysis we will identify gene modules that are highly enriched in the subgroup of SCI patients that experience neuropathic pain. These gene modules will be used as input in the CMap platform developed by the BROAD Institute which predicts chemical compounds that can reverse the expression patterns.

Lastly in Aim 3, we will validate the top ranked compounds predicted from CMap in an in vitro human primary leukocyte culture. WBCs from chronic SCI patients experiencing neuropathic pain will be isolated, cultivated, and treated with the predicted compounds for 24 hours. At 24 hours, total RNA extracted the gene expression levels will be measured by qPCR. Compounds that confirm the CMap prediction and reverse the expression pattern will be great candidates for further studies and even clinical trials in the future.