Background: Negative symptoms of schizophrenia respond poorly to medication. rTMS targeting the dorsolateral prefrontal cortex (DLPFC) is promising but variably effective, partly due to targeting inaccuracy. In this study, we evaluated the efficacy and safety of MRI-guided rTMS for negative symptoms of schizophrenia. Methods: In this single-center, randomized, assessor-blinded, sham-controlled trial, 100 patients were assigned 1:1 to MRI-guided neuronavigated active rTMS or sham for 4 weeks (20 sessions). Individualized left DLPFC targets were defined on structural MRI. The primary outcome was the change in PANSS negative factor score at week 4. Secondary outcomes were SANS, MoCA, and PSP. Adverse events were recorded. Repeated-measures ANOVA was used for primary and secondary clinical outcomes, and exploratory correlation analyses were performed for stimulation deviation measures. Results: Active rTMS reduced negative symptoms more than sham (ΔPANSS-N: −6.5±3.0 vs. −2.7±2.9; P <0.001), with a significant time×group interaction ( P =0.001). MoCA and PSP improved versus sham ( P =0.01; P =0.005). Treatment was well tolerated with mild, transient adverse events. Exploratory analyses showed that smaller stimulation-to-target deviations were associated with greater PANSS improvement ( r =−0.42, P =0.002), and symptom reduction correlated with cognitive gains ( r =0.41, P =0.003). Conclusions: Active MRI-guided rTMS was more effective than sham stimulation in reducing negative symptoms in patients with schizophrenia over 4 weeks and was well tolerated. Secondary improvements in cognitive and functional measures were observed, but these findings should be interpreted as supportive outcomes. Exploratory analyses related to stimulation deviation were hypothesis-generating only. However, because depressive symptoms were not systematically assessed in this trial, it cannot be excluded that part of the observed improvement in negative symptoms may reflect changes in depressive symptomatology.
Guo et al. (Mon,) studied this question.