Predictors of response to accelerated rTMS in the treatment of treatment-resistant depression

Accelerated repetitive transcranial magnetic stimulation (rTMS) is a promising treatment for treatment-resistant depression (TRD). We aimed to investigate the existence of clinical predictive factors in response to accelerated rTMS in the treatment of TRD. In total, 119 TRD patients who received accelerated rTMS were included in this study. The stimulation protocol was 15 Hz stimulation over the the left dorsolateral prefrontal cortex. The protocol consisted of 25 sessions, each session lasting 30 min for a total of 3000 pulses. Five sessions were applied per day for 5 consecutive days. At baseline (T0), day 5 (immediately after treatment) (T1), 4 weeks after treatment (T2), depression severity was evaluated using the 17-item Hamilton Depression Rating Scale (HAMD-17), cognitive function was evaluated using Wisconsin Card Sorting Test (WCST), the intensity of suicidal ideation was evaluated using the Columbia-Suicide Severity Rating Scale (C-SSRS). Systemic immune-inflammation index (SII) was calculated at T0 and T2. The HAMD-17 scores, WCST performance, the C-SSRS scores at T1 and T2 were improved from T0 (P < 0.01). The SII at T2 was lower than at T0 (P < 0.01). The response rates at T1 and T2 were 57.98% (69/119) and 48.74% (58/119), respectively. The results of binary logistic analysis showed that shorter course of depression, two failed antidepressant trials, no history of ECT treatment, and lower levels of SII were predictive factors for accelerated rTMS treatment response at T1 and T2 (P < 0.05), while not having a history of hospitalization was a predictive factor for response at T2 (P < 0.05) but not at T1 (P > 0.05). Based on ROC curve analysis, the optimal cut-off values of SII for discriminating responders from non-responders at T1 and T2 were < 478.56 and < 485.03, respectively. The AUC of SII at T0 predicting response for T1 and T2 were 0.729 and 0.797. We found several clinical predictors of better responses to the accelerated rTMS. Identifying clinical predictors of response is relevant to personalize and adapt rTMS protocols in TRD patients.