IMPORTANCE At present, no reliable predictors exist to distinguish future responders from nonresponders to treatment during the first episode of psychosis. Among potential neuroimaging predictors of treatment response, gyrification represents an important marker of the integrity of normal cortical development that may characterize, already at illness onset, a subgroup of patients with particularly poor outcome. OBJECTIVE To determine whether patients with first-episode psychosis who do not respond to 12 weeks of antipsychotic treatment already have significant gyrification defects at illness onset. DESIGN Case-control study with 12 weeks' longitudinal follow-up to determine treatment response. SETTING Secondary psychiatric services in an inner-city area (South London, England). PARTICIPANTS A total of 126 subjects, including 80 patients presenting with first-episode psychosis and 46 healthy controls. Patients were scanned at the outset and received various antipsychotic medications in a naturalistic clinical setting. They were followed up for 12 weeks and classified as responders or nonresponders if they reached criteria for symptom remission, evaluated with the Psychiatric and Personal History Schedule. OBSERVATION Patients were exposed to naturalistic antipsychotic treatment for 12 weeks following a magnetic resonance imaging scan. MAIN OUTCOMES AND MEASURES Cortical gyrification was assessed using local gyrification index in a vertexwise fashion across the entire cortical surface with correction for multiple testing using permutation analysis. Differences in local gyrification index were assessed between responders, nonresponders, and healthy controls. The effect of diagnosis (affective vs nonaffective psychosis) on the local gyrification index was also investigated in responders and nonresponders. RESULTS Patients with first-episode psychosis showed a significant reduction in gyrification (hypogyria) across multiple brain regions compared with healthy controls. Interestingly, nonresponders showed prominent hypogyria at bilateral insular, left frontal, and right temporal regions when compared with responders (all clusters significant at P < .05). These effects were present for both affective and nonaffective psychoses. CONCLUSIONS AND RELEVANCE Gyrification appears to be a useful predictor of antipsychotic treatment response. Early neurodevelopmental aberrations may predict unfavorable prognosis in psychosis, irrespective of the existing diagnostic boundaries.