To evaluate the impact of structured exercise interventions on circulating serum biomarkers related to neuroplasticity and inflammation in individuals with Parkinson’s disease(PD).

Exercise is a non-pharmacological therapy in PD with benefits beyond motor symptoms and can modulate neuroplasticity and inflammation. Brain-derived neurotrophic factor(BDNF) supports synaptic plasticity and dopaminergic neuronal survival, lower circulating BDNF correlates with PD severity, making exercise-related changes clinically meaningful. Circulating biomarkers including BDNF, GDNF and inflammatory cytokines like IL-6, TNF-α, and IL-10 index these adaptations, yet reported effects are heterogeneous and methodologically limited, warranting a systematic review and meta-analysis.

PubMed, Embase, and the Cochrane Library were searched from inception to September 2025 for clinical studies in adults with PD comparing any structured exercise intervention to a non-exercise control, lasting ≥2 weeks. Primary outcomes were circulating neuroplasticity biomarkers and inflammatory biomarkers. Two reviewers independently screened studies and extracted data. Pooled mean differences(MD) with 95% confidence intervals(CIs) were calculated using random-effects models in RevMan5.4, and heterogeneity assessed with I² values.

Twelve studies with 513 participants were included. Exercise significantly increased circulating BDNF versus control(MD 0.53; 95%CI 0.47-0.60; p<0.001;I²=0%), indicating a robust and homogeneous neuroplasticity effect. IL-10 (MD 11.52; 95%CI -11.38-34.42; p=0.32;I²=89.9%) and IL-6 (MD 5.56; 95%CI -17.92-6.79; p=0.38;I²=83%) did not differ significantly, with substantial heterogeneity. No significant effects were observed for GDNF(MD −37.24; 95%CI −217.15-142.68;p=0.69;I²=45%), IGF-1(MD 5.39; 95%CI −32.23-43.01; p=0.78;I²=0%), IL-1β (MD 0.27; 95%CI −0.30-0.84;p=0.35;I²=0%), TNF-α(MD −0.17; 95%CI −0.54-0.20;p=0.37;I²=0%), or VEGF(MD −22.22; 95%CI −110.96-66.53; p=0.62;I²=0%). Sensitivity analyses (leave-one-out) yielded stable pooled estimates, and funnel plots suggested low publication bias for BDNF and consistent null effects for other outcomes.

Exercise significantly elevates circulating BDNF in PD, supporting neuroplasticity. Other biomarkers show no consistent or significant changes, indicating limited systemic inflammatory modulation. Future trials should standardize assays and sampling, prespecify modality and intensity, and power biomarker endpoints to clarify anti-inflammatory effects.