Late-life depression (LLD) is a growing worldwide problem due to demographic changes, with limited treatment options due to high rates of pharmacotherapy adverse effects, accessibility of psychotherapy, and tolerability of electroconvulsive therapy. Novel neuromodulation techniques, such as repetitive transcranial magnetic stimulation (rTMS), may overcome these limitations. The objective of this study is to determine the efficacy, tolerability, and cognitive effects of high-dose deep rTMS in LLD. In this study we randomized older adults between 60 and 85 years old with major depressive disorder (MDD) to sham or active deep rTMS (H1 coil, 6012 pulses, 18 Hz, 120% of resting motor threshold) delivered over the dorsolateral and ventrolateral prefrontal cortex 5 days per week over 4 weeks. Our primary outcome was remission of depression in an intention-to-treat analysis. We also assessed change in cognitive functioning with rTMS treatment and tolerability based on adverse effects. Fifty-two participants were randomized to active (n = 25) or sham H1 coil (n = 27). Remission rate was significantly higher with active than sham rTMS (40.0% vs 14.8%) with a number needed to treat of 4.0 (95% CI: 2.1-56.5). There was no change on any measure of executive function and no serious adverse events. Adverse effect profiles were similar between active and sham rTMS, except for reports of pain being significantly more common in the active condition (16.0% vs 0%). High-dose deep rTMS appears to be safe, well tolerated, and efficacious in the treatment of LLD.