Preclinical and clinical studies gave evidence that lithium could be useful in the treatment of Alzheimer's disease (AD). One possible mechanism of action might be the induction of neurotrophins. Recently, we found a significant increase of brain-derived neurotrophic factor (BDNF) serum levels in AD patients treated with lithium and a significant decrease of ADAS Cog sum scores in comparison to placebo-treated patients. In another previous study we have shown that glial cell line-derived neurotrophic factor (GDNF) levels in CSF of patients with early AD are increased most probably due to an upregulated expression in CNS as an adaptive process of the impaired brain to enhance neurotrophic support at least in early stages of disease. Here we assessed the influence of a lithium treatment on GDNF serum and cerebrospinal fluid (CSF) concentrations in a subset of a greater sample recruited for a randomized, single-blinded, placebo-controlled, parallel-group multicenter 10-week study, investigating the efficacy of lithium treatment in AD patients. We found a significant negative correlation of lithium concentration in serum with GDNF concentration in CSF at the end of treatment (r = -0.585, p = 0.036) and with the difference of GDNF concentration in CSF before and after treatment (r = - 0.755, p = 0.003). However, we could not show a difference in GDNF concentrations between the patients after the treatment with lithium or placebo (serum, mean ± standard deviation: 434.3 ± 117.9 pg/ml versus 543.8 ± 250.0 pg/ml, p = 0.178; CSF, 62.3 ± 37.4 pg/ml versus 72.8 ± 43.9 pg/ml, p = 0.511). The findings of the present investigation indicated that beneficial effects of the lithium treatment might reduce the necessity of enhanced GDNF expression in the CNS in early AD.