Abstract The article discusses the evaluation of the Neuropeptide effectiveness in the pathogenetic therapy of patients with Parkinson's disease. Multiple motor and non-motor functions in PD patients treated with the Neuropeptide are investigated, using clinical and biochemical assessments, cell and genetic analysis and Machine learning models, including Gradient Boosting Machines and Random Forests. The Neuropeptide administration resulted in improved daily activity, cognitive functions, reduced depression, and anxiety in Parkinson's patients. Significant changes were observed in platelet ultrastructure (45% increase in number of δ-granules, and an increase in number of mitochondria from 2–4 to 5–7 per cell), gene expression (increase in BDNF gene expression in women (p = 0.046) associated with an improvement in their MMSE scores, significant correlations between MOCA scores and DJ-1 expression), in oxidative stress markers (reduction of TBARS by 17% and H2O2 by 12%, accompanied by increasing of GSH concentration by 30% and GPx activity by 15%). Applying GBMs in capturing non-linear relationships reinforced the value of using advanced machine learning techniques to analyze complex clinical datasets. The findings support the potential of the Neuropeptide as a multifaceted therapeutic option, with capacity to improve cognitive and neurotrophic parameters in PD patients.
Krasnienkov et al. (Fri,) studied this question.