Prediction of Differential Pharmacologic Response in Chronic Pain Using Functional Neuroimaging Biomarkers and Support Vector Machine Algorithm - An Exploratory Study.

Abstract

OBJECTIVE\nThere is increasing demand for prediction of chronic pain treatment outcomes using machine-learning models, in order to improve suboptimal pain management. In this exploratory study, we used baseline brain functional connectivity patterns from chronic pain patients suffering from Fibromyalgia (FM) to predict whether a patient would respond differentially to either milnacipran or pregabalin, two FDA-approved drugs for the treatment of FM.\n\n\nMETHODS\nFM patients participated in two separate double-blind placebo controlled crossover studies of milnacipran (N=15) and pregabalin (N=13), respectively. Functional magnetic resonance imaging during rest was collected before treatment to measure intrinsic functional brain connectivity in several brain regions involved in pain processing. A support vector machine algorithm was used to classify FM patients into responders, defined as those patients with a 20% or greater improvement in clinical pain, to either milnacipran or pregabalin.\n\n\nRESULTS\nConnectivity patterns involving the posterior cingulate and dorsolateral prefrontal cortex individually classified pregabalin responders versus milnacipran responders with 77% accuracy. Performance of this classification improved when both posterior cingulate and dorsolateral prefrontal connectivity patterns were combined, resulting in a 92% classification accuracy. These results were not related to confounding factors including head motion, scanner sequence, or hardware status. Connectivity patterns failed to differentiate drug non-responders across the two studies.\n\n\nCONCLUSION\nBrain functional connectivity patterns used in a machine learning framework differentially predicted clinical response to pregabalin and milnacipran in chronic pain patients. These findings highlight the promise of machine learning in pain prognosis and treatment prediction.