Lithium and quetiapine are effective treatments for bipolar disorder, but their potential neuroprotective effects in humans remain unclear. A single blinded equivalence randomized controlled maintenance trial was conducted in a prospective cohort of first-episode mania (FEM) patients (n=26) to longitudinally compare the putative protective effects of lithium and quetapine on grey and white matter volume. A healthy control sample was also collected (n=20). Using structural MRI scans, voxel-wise grey and white matter volumes at baseline and changes over time in response to treatment were investigated. Patients were assessed at three time points (baseline, 3 and 12-month follow-up), whereas healthy controls were assessed at two time points (baseline and 12-month follow-up). Patients were randomized to lithium (serum level 0.6 mmol l−1, n=20) or quetiapine (flexibly dosed up to 800 mg per day, n=19) monotherapy. At baseline, compared with healthy control subjects, patients with FEM showed reduced grey matter in the orbitofrontal cortex, anterior cingulate, inferior frontal gyrus and cerebellum. In addition, patients had reduced internal capsule white matter volume bilaterally (t1,66>3.20, P<0.01). Longitudinally, there was a significant treatment × time effect only in the white matter of the left internal capsule (F2,112=8.54, P<0.01). Post hoc testing showed that, compared with baseline, lithium was more effective than quetiapine in slowing the progression of white matter volume reduction after 12 months (t1,24=3.76, P<0.01). Our data support the role of lithium but not quetiapine therapy in limiting white matter reduction early in the illness course after FEM.