Objective: Brain-derived neurotrophic factor (BDNF) and S100B are involved in brain plasticity processes and their serum levels have been demonstrated to be altered in patients with psychoses. This study aimed to identify subgroups of patients with psychotic disorders across diagnostic boundaries that show a specific symptom profile or response to treatment with antipsychotics, by measuring serum levels of BDNF and S100B. Methods: The study sample consisted of 58 patients with DSM-IV psychotic disorders. Comprehensive Assessment of Symptoms and History (CASH), Positive and Negative Syndrome Scale (PANSS) and Clinical Global Impression scale for severity and improvement (CGI-S/CGI-I) were applied at baseline and after 6 weeks of antipsychotic treatment. At both time points, serum levels of BDNF and S100B were measured and compared with a matched control sample. Results: Baseline BDNF and S100B levels were significantly lower in patients as compared with controls and did not change significantly during treatment. Dividing the patient sample according to baseline biochemical parameters (low and high 25% and middle 50%), no differences in symptom profiles or outcome were found with respect to BDNF. However, the subgroups with low and high S100B levels had higher PANSS scores than the middle subgroup. In addition, the high subgroup still showed significantly more negative symptoms after treatment, whereas the low subgroup showed more positive symptoms compared with the other subgroups. Conclusion: Serum levels of BDNF and S100B are lowered in patients with psychotic disorders across diagnostic boundaries. The differences between high and low S100B subgroups suggest a relationship between S100B, symptom dimensions and treatment response, irrespective of diagnostic categories.