Anterior midcingulate cortex activity during pain catastrophizing predicts and sequentially tracks reduction in pain interference after cognitive behavioral therapy in fibromyalgia

ABSTRACT: Fibromyalgia (FM) is a chronic pain disorder characterized by widespread pain, fatigue, and functional disability. Cognitive behavioral therapy (CBT) has been shown to alleviate FM symptoms, but the neural mechanisms underlying CBT's effectiveness in reducing functional impact remain unclear. In this study, we investigated the temporal dynamics of clinical and neural responses to CBT. We conducted a longitudinal randomized controlled neuroimaging trial with 98 patients with FM assigned to an 8-week CBT or education intervention. Functional MRI (fMRI) data were collected at baseline (PRE), midtreatment (MID, at 4 weeks), and post-treatment (POST) while participants engaged in task-evoked pain catastrophizing during the fMRI scan session. Clinical outcomes, including pain interference and catastrophizing, were assessed at all time points. The CBT group demonstrated early (MID-PRE, P < 0.05) and late (POST-MID, P < 0.005) reductions in pain interference and catastrophizing. Anterior midcingulate cortex (aMCC) activity during pain catastrophizing task tracked symptom improvement over time and predicted early vs late improvements over the 8-week course of CBT. Moreover, machine learning models indicated that baseline aMCC activation was a more accurate predictor of treatment response than baseline levels of patient-reported pain and catastrophizing. This study provides evidence that brain activity during pain catastrophizing plays a critical role in tracking and predicting symptom improvements after CBT in FM. These findings highlight the potential of neuroimaging biomarkers for optimizing patient-specific treatment strategies in chronic pain management.