182 adults aged 18-65 years meeting DSM-5 criteria for major depressive disorder (confirmed by SCID-5) and classified as treatment-resistant (failure of ≥2 adequate antidepressant trials) were randomised 1:1 to iTBS (600 pulses/session) or standard 10 Hz rTMS (3,000 pulses/session), both delivered to the left DLPFC at 120% resting motor threshold, 5 days/week for 4 weeks (20 sessions total). Randomisation used a computerised random-number generator with a permuted-block design. Treating research nurses were unblinded; raters and participants were blinded to allocation. The primary outcome was change in HDRS-17 from baseline to Week 4, analysed by intention-to-treat using a linear mixed-effects model, with a pre-specified non-inferiority margin of 3 HDRS-17 points. Secondary outcomes included response and remission rates, CGI, GAD-7, BAI, the TMS Patient Satisfaction Questionnaire, protocol preference, and feasibility (missed sessions, early withdrawal). Follow-up continued to Week 8.
