Schizophrenia is increasingly recognized as a neurodevelopmental disease with an additional degenerative component, comprising cognitive decline and loss of cortical gray matter. We hypothesized that a neuroprotective/neurotrophic add-on strategy, recombinant human erythropoietin (rhEPO) in addition to stable antipsychotic medication, may be able to improve cognitive function even in chronic schizophrenic patients. Therefore, we designed a double-blind, placebo-controlled, randomized, multicenter, proof-of-principle (phase II) study. This study had a total duration of 2 years and an individual duration of 12 weeks with an additional safety visit at 16 weeks. Chronic schizophrenic men (N = 39) with defined cognitive deficit (≥1 s.d. below normal in the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS)), stable medication and disease state, were treated for 3 months with a weekly short (15 min) intravenous infusion of 40 000 IU rhEPO (N = 20) or placebo (N = 19). Main outcome measure was schizophrenia-relevant cognitive function at week 12. The neuropsychological test set (RBANS subtests delayed memory, language-semantic fluency, attention and Wisconsin Card Sorting Test (WCST-64) - perseverative errors) was applied over 2 days at baseline, 2 weeks, 4 weeks and 12 weeks of study participation. Both placebo and rhEPO patients improved in all evaluated categories. Patients receiving rhEPO showed a significant improvement over placebo patients in schizophrenia-related cognitive performance (RBANS subtests, WCST-64), but no effects on psychopathology or social functioning. Also, a significant decline in serum levels of S100B, a glial damage marker, occurred upon rhEPO. The fact that rhEPO is the first compound to exert a selective and lasting beneficial effect on cognition should encourage new treatment strategies for schizophrenia.