Objective: The purpose of this study was to characterize the functional brain changes involved in THC modulation of chronic neuropathic pain. Methods: Fifteen patients with chronic radicular neuropathic pain participated in a randomized, double-blind, placebo-controlled trial employing a counterbalanced, within-subjects design. Pain assessments and functional resting state brain scans were performed at baseline and after sublingual THC administration. We examined functional connectivity of the anterior cingulate cortex (ACC) and pain related network dynamics using graph theory measures. Results: THC significantly reduced patients’ pain compared to placebo. THC-induced analgesia was correlated with a reduction in functional connectivity between the anterior cingulate cortex (ACC) and the sensorimotor cortex. Moreover, the degree of reduction was predictive of the response to THC. Graph theory analyses of local measures demonstrated reduction in network connectivity in areas involved in pain processing, and specifically in the dorsolateral prefrontal cortex (DLPFC), which were correlated with individual pain reduction. Conclusions: These results suggest that the ACC and DLPFC, two major cognitive-emotional modulation areas, and their connections to somatosensory areas, are functionally involved in the analgesic effect of THC in chronic pain. This effect may therefore be mediated through induction of functional disconnection between regulatory high-order affective regions and the sensorimotor cortex. Moreover, baseline functional connectivity between these brain areas may serve as a predictor for the extent of pain relief induced by THC.