BACKGROUND: Cortisol and inflammatory markers have been increasingly reported as abnormal at psychosis onset. The main aim of our study was to investigate the ability of these biomarkers to predict treatment response at 12 weeks follow-up in first episode psychosis. METHODS: In a longitudinal study, we collected saliva and blood samples in 68 first episode psychosis patients (and 57 controls) at baseline and assessed response to clinician-led antipsychotic treatment after 12 weeks. Moreover, we repeated biological measurements in 39 patients at the same time we assessed the response. Saliva samples were collected at multiple time points during the day to measure diurnal cortisol levels and cortisol awakening response (CAR); interleukin (IL)-1β, IL-2, IL-4, IL-6, IL-8, IL-10, tumor necrosis factor-α, and interferon-γ (IFN-γ) levels were analyzed from serum samples. Patients were divided into Non-Responders (n = 38) and Responders (n = 30) according to the Remission symptom criteria of the Schizophrenia Working Group Consensus. RESULTS: At first onset, Non-Responders had markedly lower CAR (d = 0.6, P = .03) and higher IL-6 and IFN-γ levels (respectively, d = 1.0, P = .003 and d = 0.9, P = .02) when compared with Responders. After 12 weeks, Non-Responders show persistent lower CAR (P = .01), and higher IL-6 (P = .04) and IFN-γ (P = .05) when compared with Responders. Comparison with controls show that these abnormalities are present in both patients groups, but are more evident in Non-Responders. CONCLUSIONS: Cortisol and inflammatory biomarkers at the onset of psychosis should be considered as possible predictors of treatment response, as well as potential targets for the development of novel therapeutic agents.