Fibroblast Growth Factor 9 as a Potential Biomarker for Schizophrenia

Preclinical and clinical studies have suggested that fibroblast growth factor (FGF) system contributed to the onset and development of schizophrenia (SCZ). However, there was no strong clinical evidence to link an individual FGF with SCZ. In this study, we aim to measure blood FGF9 levels in the patients with SCZ with and/or without medication, and test whether FGF9 has a potential to be a biomarker for SCZ. We recruited 130 patients with SCZ and 111 healthy individuals, and the ELISA and qRT-PCR assays were used to measure serum FGF9 levels in the participants. ELISA assay demonstrated that serum FGF9 protein levels were dramatically reduced in first-episode, drug-free patients, but not in chronically medicated patients when compared to healthy control subjects. Further analysis showed that treatment of the first-episode, drug-free SCZ patients with antipsychotics for 8 weeks significantly increased the serum FGF9 levels. In addition, we found that blood FGF9 mRNA levels were significantly lower in first-onset SCZ patients than controls. Under the receiver operating characteristic curve, the optimal cutoff values for FGF9 protein level as an indicator for diagnosis of drug-free SCZ patients was projected to be 166.4 pg/ml, which yielded a sensitivity of 0.955 and specificity of 0.86, and the area under the curve was 0.973 (95% CI, 0.954-0.993). Furthermore, FGF9 had good performance to discriminate between drug-free SCZ patients and chronically medicated patients, the optimal cutoff value for FGF9 concentration was projected to be 165.035 pg/ml with a sensitivity of 0.86 and specificity of 0.919, and the AUC was 0.968 (95% CI, 0.944, 0.991). Taken together, our results for the first time demonstrated the dysregulation of FGF9 in SCZ, and FGF9 has the potential to be served as a biomarker for SCZ.