Neural and psychophysical predictors of treatment response to transcranial direct current stimulation and mindfulness-based meditation for knee osteoarthritis pain

AimWe investigated the heterogeneity of treatment effects in transcranial direct current stimulation (tDCS) with mindfulness-based meditation (MBM) and within each individual study group (tDCS alone, MBM alone, and sham) among individuals with symptomatic knee osteoarthritis. We also explored participant characteristics underlying this heterogeneity. MethodsThis secondary analysis drew on a double-blind, randomized, sham-controlled, phase II, parallel-group trial in which 200 participants were assigned to one of four groups: (1) active tDCS + active MBM, (2) active tDCS + sham MBM, (3) sham tDCS + active MBM, or (4) sham tDCS + sham MBM. Participants received ten 20-minute tDCS sessions (active or sham) administered concurrently with MBM (active or sham). Latent class growth analysis was used to identify subgroups with distinct treatment response trajectories (responders vs. non-responders) based on changes in clinical pain (Numeric Rating Scale) from baseline to post-intervention. Generalized linear models were then applied to determine baseline factors associated with participants response classification, including demographic, clinical, and psychological characteristics; quantitative sensory testing battery; and pain-related cortical hemodynamic activity measured using functional near-infrared spectroscopy (fNIRS) in response to punctate and thermal stimuli. ResultsResponders in the active tDCS + active MBM and active tDCS + sham MBM groups demonstrated greater improvements in clinical pain from baseline to post-intervention than non-responders (p < 0.001). In the active tDCS + active MBM group, greater cortical activation in the fNIRS channel S06-D06 of the left somatosensory cortex in response to punctate stimuli, identifying as white, and lower conditioned pain modulation (reflecting less efficient endogenous pain modulation), were significantly associated with being responders (p < 0.05). In the active tDCS + sham MBM group, younger age and lower heat pain tolerance at the knee were significantly associated with being responders (p < 0.05). No clear response patterns were observed in the remaining groups. ConclusionFactors underlying heterogeneity of treatment effects, including somatosensory cortical activation and pain modulatory profiles, may provide preliminary insights to inform the development of personalized neuromodulation (stimulation) protocols.