Objective: To investigate the effects of sleep electroencephalogram-modulated repetitive transcranial magnetic stimulation (SEM-rTMS) and conventional rTMS (C-rTMS) on the activity of hypothalamic-pituitary-adrenal (HPA) axis in patients with depression. Methods: In a double-blind, randomized controlled trial, 164 patients diagnosed with depression were randomized to treatment with SEM-rTMS (n=57), C-rTMS (n=55) or sham rTMS (n=52) for 30 min every day for 10 d. Before and after treatment plasma concentrations of adrenocorticotropic hormone (ACTH) and cortisol (CORT) were measured, and the 24-item Hamilton Depression Rating Scale (HAMD-24) was used for assessment. Results: The HAMD-24 scores and plasma ACTH and CORT concentrations of these depressive patients before treatment were significantly different from those of the normal control group (P0.05). The HAMD-24 scores and plasma ACTH and CORT concentrations in the SEM-r TMS group and conventional rTMS group were decreased significantly (P<0.05). There was a significant positive correlation between the HAMD-24 scores and plasma ACTH (n=240, r=0.105, P=0.048) and CORT concentrations (n=240, r=0.126, P=0.023) in the patients with depression before and after treatment. Conclusion: The antidepressant effect of rTMS, including SEM-rTMS, may be related to its decreasing HPA axis activity. (This trail was registered. No: ChiCTR-TRC-00000465)