{"id":34350,"topic":"peptides","source":"Nature","title":"Evaluation of the neurotrophic peptide mixture in pathogenetic therapy of patients with Parkinson’s disease - Nature","url":"https://www.nature.com/articles/s41531-026-01270-6","url_hash":"7e0ce62838f04795aa99c83894f940f4f9c5b7fa","author":"","summary":"<a href=\"https://news.google.com/rss/articles/CBMiX0FVX3lxTE4tTGM3aWNjbjR5czF1Zm9XMVR4YzM5R0JjREZKTmkyckY3MVRhVFhPbEZySnlydnVXMmU5VHc2REVQQVZIZExPNHNXSTBEVVgtZW96TWVWY0pmMmc0T0sw?oc=5\" target=\"_blank\">Evaluation of the neurotrophic peptide mixture in pathogenetic therapy of patients with Parkinson’s disease</a>&nbsp;&nbsp;<font color=\"#6f6f6f\">Nature</font>","content":"Abstract\nThis exploratory, single-group, open-label study investigated 17 patients with Parkinson’s disease (PD) using a pre-post design. Motor and non-motor outcomes were assessed through clinical scales, biochemical and genetic analyses, and machine learning models (Gradient Boosting Machines, Random Forests). After treatment with a neurotrophic peptide mixture, improvements were observed in daily activity (16%), cognition (11%), depression (10% reduction), and reactive anxiety (23% reduction). Biological changes included a 45% increase in platelet δ-granules, higher mitochondrial counts, elevated gene expression (notably BDNF in women, p = 0.046), and modulation of oxidative stress markers (17% reduction in TBARS, 30% increase in GSH). Machine learning identified BDNF and PINK1 expression, along with MOCA and MMSE scores, as key predictors of UPDRS improvement. These findings suggest that neurotrophic peptide therapy may influence clinical, structural, and molecular domains in PD. Larger, controlled trials are warranted to confirm therapeutic potential and clarify associations with cognitive and neurotrophic parameters.\nSimilar content being viewed by others\nIntroduction\nParkinson’s disease is a neurodegenerative condition typically observed in older individuals, characterized by a gradual onset and progressive nature. The initial manifestation often includes tremors, followed by symptoms such as slowed movement (bradykinesia) and muscle stiffness (rigidity)1. The disease is linked to the degeneration of dopaminergic neurons in the substantia nigra, the presence of alpha-synuclein aggregates within neuronal cells (Lewy bodies and Lewy neurites), and subsequent pathology in regions like the basal forebrain, amygdala, and medial temporal lobe structures, progressing to cortical areas in advanced stages2.\nMitochondrial abnormalities and dysfunction are recognized as key contributors to Parkinson’s disease pathogenesis in a subset of patients3. Some data was obtained on the availability correlation of the functional activity of mitochondria in lymphocytes, which displays multisystem mitochondrial failure and indicates of energy dysfunction in general4. The spectrum of mitochondrial dysfunctions in PD includes dysfunction of the respiratory chain (inhibition of the NADH-oxidase oxidation pathway), hyperproduction of ATP, disruption of Ca-regulatory systems, opening of the mitochondrial pore, perturbations of mitochondrial dynamics, and dysregulation of mitochondrial clearance5.\nAlthough oxidative stress plays a significant role in the pathogenesis of PD, therapy based on the use of exogenous antioxidants has not been shown to be highly effective. An alternative strategy that can counteract the damaging effects of free radicals and restore cellular redox balance is the activation of endogenous antioxidant enzymes present in the central neural system, such as superoxide dismutase, catalase, and glutathione peroxidase. The transcription of these cytoprotective proteins is regulated by the transcription factor Nrf2, which plays a leading role in the regulation of cellular redox status and is the main regulator of the cellular response to oxidative stress through the induction of antioxidant and detoxifying enzymes and proteins6. This is consistent with our previous data, which showed that lipid peroxidation was significantly increased while reduced glutathione (GSH) and glutathione-dependent antioxidant enzymes activity markedly decreased in blood of patients with PD as well as in rat brain under modeling of PD7. Various activators of the Nrf2/ARE signaling pathway, including peptides, have shown their effectiveness in both in vitro and in vivo models of neurological disorders8.\nThe Neurotrophic peptide mixture is peptidergic mixture with neurotrophic-like properties, made from a mixture of low molecular weight, porcine-derived peptides and free amino acids. The pharmaceutical preparation includes the peptide fragments: nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), ciliary neurotrophic factor (CNTF), enkephalins, orexin, and P219.\nThe use of the Neurotrophic peptide mixture significantly reduces neurological deficits, improves cognitive abilities, and reduces the volume of the infarction zone during stroke10,11. The Neurotrophic peptide mixture has an organ-specific multimodal effect on the brain, i.e., it provides metabolic regulation, neuroprotection, functional neuromodulation, and neurotrophic activity12. The Neurotrophic peptide mixture protects neurons from the damaging effects of lactic acidosis, prevents the formation of free radicals13, increases survival and prevents neuronal death under conditions of hypoxia and ischemia, and reduces the damaging neurotoxic effect of glutamate14.\nOver the past decade, compelling evidence has emerged neuroprotective properties of the Neurotrophic peptide mixture, explaining its multi-level impact on both the pathological cascade of damage and on subsequent recovery processes15. The ability of the Neurotrophic peptide mixture to influence the processes of neuroprotection and neuroplasticity was proven in a series of studies15,16,17 on a model of Alzheimer’s disease in transgenic mice for the amyloid precursor protein (mThy-1-hAPP751 tg). Using a model of Alzheimer’s disease, it was shown that the Neurotrophic peptide mixture, through the regulation of the synthesis and transport of amyloid precursor protein and the deposition of β-amyloid in synapses, has a synaptotrophic effect, which plays a primary role in the processes of neuroplasticity, which are extremely important in the acquisition of new skills and the maintenance of existing ones18.\nIncreasing studies have demonstrated that the Neurotrophic peptide mixture administration can promote recovery of motor function, improve early brain injury, and decrease hippocampal neuronal death19. In recent clinical studies, the Neurotrophic peptide mixture improved overall outcomes after moderate to severe traumatic brain injury and was also safe and better for early rehabilitation patients after ischemic stroke20. Lu et al. reported that the Neurotrophic peptide mixture can improve neurological dysfunction and brain damage, relieve neuroinflammation and prevent apoptosis after brain injury21.\nBased on the multifaceted mechanism of action of the Neurotrophic peptide mixture, including its effect on the processes of neurogenesis, neuroplasticity and neuroprotection, the drug currently has three main indications for use - stroke, dementia and traumatic brain injury10,22,23.\nLuchanina et al. demonstrated for the first time the effect of Neurotrophic peptide mixture on motor activity and muscle tone in patients with PD treated with levodopa13. It was established that the Neurotrophic peptide mixture improved the engraftment of a DOPA transplant from stem cells in the substantia nigra24 and reduced the severity of symptoms in model animals12.\nThe current exploratory study dataset comprised clinical, cognitive, biochemical, and demographic information from PD patients undergoing the Neurotrophic peptide mixture therapy. Key features included cognitive assessments (MMSE, MOCA), neurotrophic markers (BDNF, PINK1), oxidative stress indicators (SOD, DJ1), and demographic factors (age, sex). The primary outcome was the change in UPDRS scores, serving as an objective measure of therapeutic response to the Neurotrophic peptide mixture. Also, in this study, machine learning models were employed to predict UPDRS score changes as a key indicator of the Neurotrophic peptide mixture’s therapeutic effectiveness, offering a data-driven evaluation of treatment response.\nResults\nClinical and neurophysiological assessment of the effectiveness of course treatment\nA course of the Neurotrophic peptide mixture administration against the background of basic antiparkinsonian therapy had a beneficial effect on the condition of the majority of patients, primarily on their activity in everyday life and the ability to self-care. Objectively, this effect was expressed in a significant decrease in scores of all UPDRS subscales (~16% on average) (Fig. 1).\nBoth clinically central endpoints improved significantly (p < 0.01). Secondary measures (UPDRS I and UPDRS II) generally trend in the same direction; effect sizes are provided without formal inference.\nA beneficial impact was observed following administration of a regimen of the Neurotrophic peptide mixture on a spectrum of non-motor dysfunctions concomitant with the motor manifestations of Parkinson’s disease (PD). Participants exhibited statistically significant enhancements in cognitive abilities as assessed by specialized instruments including MMSE, FAB, and MoCA (~11% on average). Moreover, there was a notable reduction in scores indicative of depression measured via the Beck Depression Inventory scale (~10% on average), as well as in reactive anxiety assessed by the Spielberger-Hanin scale (~23% on average). Notably, levels of personal anxiety remained consistent relative to baseline measurements. As a result, all observed Secondary outcomes are trended in the direction of improvement (Fig. 2).\nFor instrumental assessment of the Neurotrophic peptide mixture effect on motor functions, the Motor tempo test showed higher sensitivity. As a result, the significant decrease in the average, minimum and maximum time of motor reactions was recorded (Fig. 3).\nThose changes are more characteristic of the reactions of the right hand, while for the left hand only the average value of the motor tempo decreased. It can be assumed that this asymmetry of effects is associated with the sidedness of the pathological process, since a significant part of patients (76% of the study group) demonstrated predominantly right-sided symptoms of PD. These findings show a valuable positive impact of Neurotrophic peptide mixture on neurophysiological parameters.\nAt the same time, the latent period of a simple sensorimotor reaction did not change during a course of the Neurotrophic peptide mixture administration (Fig. 3).\nChanges of the H-reflex and M-response after administration of a regimen of Сerebrolysin\nIn the context of administering the Neurotrophic peptide mixture, there was an elevation observed in the threshold for triggering the H-reflex of the gastrocnemius muscle, increasing from 5 [3; 9] mA to 6 [5.8; 18] mA (Fig. 4). Notably, there were no alterations noted in the latency period or power attributes of the H-reflex, nor in the parameters of the M-response, compared to their baseline values.\nPotentials P300\nAnalysis of endogenous components revealed that in the studied group of patients with PD, there was a delay in the formation of the P3 wave (median 434 ms, interquartile range 386–512 ms in the temporal area of the neocortex) compared to the corresponding age norm of 390–410 ms. After the course of the Neurotrophic peptide mixture, a statistically significant increase in the latency period of the P3 component in the central-parietal lead was recorded, up to 394 [304; 446] ms (Fig. 5). There were no significant changes in the latency period of the N2 component and the amplitudes of the N2 and P3 components.\nThe delay in the formation of the P3 wave is a characteristic feature of several neurodegenerative diseases, particularly Alzheimer’s and Parkinson’s diseases. For patients with borderline PD and moderate cognitive impairments, an increase in the latency period of P3 has been described, with a reverse correlation between this parameter and the volume of gray matter in the globus pallidus, dysfunction of which leads to symptoms such as bradykinesia and rigidity25. Improvement in this parameter following the course of the Neurotrophic peptide mixture suggests that a more pronounced effect on the state of higher cortical functions can be expected in patients with increased risk of developing or progressing cognitive impairments.\nCorrelation analysis of clinical and neurophysiological parameters after a course of treatment with the Neurotrophic peptide mixture\nThe degree of improvement in several non-motor characteristics (cognitive functions assessed by the FAB scale, amplitude of cognitive evoked potential P3, and signs of depression measured by the Beck scale) was greater in individuals who initially received a higher daily dose of levodopa (Figs. 6–8).\nFor indicators of cognitive functions (MMSE, P3 cognitive evoked potential amplitude), a similar relationship with the time of onset of the disease was also found: more pronounced positive effects of the Neurotrophic peptide mixture treatment were observed in patients with an earlier onset of PD and a longer duration of the disease (Figs. 6, 8).\nAt the same time, a more pronounced reduction in reaction time was characteristic of individuals with lower initial ratings of PD severity on the UPDRS scale and lower levels of anxiety. Such correlation relationships are found both when directly assessing the time of motor reactions in the Motor tempo and Sensorimotor reaction tests, and when determining the latent periods of the H-reflex and M-response (Figs. 9, 10).\nThe use of the Neurotrophic peptide mixture in the treatment course improves the general condition of patients with Parkinson’s disease, as evidenced by a significant decrease in the quantitative indicators of all UPDRS subscales, attenuation of affective disorders such as anxiety and depression, and improvement in cognitive function. Against the background of the Neurotrophic peptide mixture administration, dynamic characteristics of simple motor reactions improve, and the threshold for the emergence of the H-reflex increases, indicating facilitation of supraspinal control processes over the activity of motor neurons. In general, the results show a strong tendency to improvement in different parameters.\nMitochondrial dysfunction before and after treatment with the Neurotrophic peptide mixture in patients with Parkinson’s disease\nThe conducted studies revealed significant differences in the structural organization of platelets in individuals with PD examined before the start of the standard treatment process. Electron microscopic and morphometric study of platelets in patients with PD revealed that platelets were mainly represented by senescent cells. They had a reduced number of dense δ granules, which was 5.1 ± 0.4 per one cell (in healthy individuals - 8.4 ± 0.5 units per one cell, p < 0.05). However, the number of transparent α-granules turned out to be increased from 5.3 ± 0.2 up to 10.3 ± 0.6 per cell. In addition, the latter often had large sizes (>100 nm) and contained a significant amount of protein debris (Fig. 11).\nThe single mitochondria that were found in the cells (2–4 per cell) were structurally quite preserved and had an average diameter of 0.48 ± 0.02 µm, not significantly different from those determined in the control group (0.41 ± 0.04 μm).\nSignificant abnormalities were found in the ultrastructure of mitochondria in platelets of patients. In addition to increased mitochondrial membrane permeability (outer and/or inner, crystas-forming), the formation of septated organelles was observed. The mechanisms and causes of their formation are still unexplored. It is assumed that septated mitochondria can accumulate Са2+26. Since it is known that the accumulation of Са2+ and the synthesis of ATP are alternative processes associated with the transfer of electrons, an increase in the Ca-accumulating ability of mitochondria may indicate a decrease in the synthesis of ATP in them. In addition, the appearance of septated mitochondria may indicate partial cell necrosis. These ultrastructural changes are evidence of the formation of mitochondrial dysfunction in platelets.\nThe use of the Neurotrophic peptide mixture in the complex therapy of patients with CP was accompaniedby reliable dynamics of the morphofunctional state of platelets (Table 1). The use of the Neurotrophic peptide mixture led to a significant (by 45%) increase in the number of δ-granules. No significant changes in the number of α granules were detected; only a tendency to its decrease was observed. An increase in the number of δ-granules indicates an increase in serotonin reserves. Probably, one of the possible mechanisms of the neuroprotective effect of the Neurotrophic peptide mixture is realized in this way. In addition, the Ca2+ -accumulating capacity in thrombocytes is relatively reduced, which should contribute to the improvement of ATP synthesis.\nDespite the small number of mitochondria in platelets, a slight increase in their number was observed - from 2 to 4 organelles before treatment to 5–7 organelles after treatment with the Neurotrophic peptide mixture, which, in combination with the absence of detection of septated organelles, may indicate a decrease in the severity of mitochondrial dysfunction.\nThus, treatment with the Neurotrophic peptide mixture improves the morphofunctional state of platelets in patients with PD, it increases serotonin reserves, reduces Ca2+ -accumulating capacity, reduces the manifestations of mitochondrial dysfunction, and exhibits mild antithrombotic properties.\nEffect of the Neurotrophic peptide mixture treatment on PARKIN gene expression in patients with Parkinson’s disease\nFor the whole group of tested individuals, taken together, no correlation of the expression of the measured genes was shown using the Spearman test. At the same time, when dividing by sex, multidirectional differences in gene expression before and after therapy with the drug Neurotrophic peptide mixture were found (Table 2).\nIn particular, the group of men was characterized by a decrease in the expression of the BDNF gene, and in women the expression increased after the use of the drug (Fig. 12).\nIt is important to note that when comparing the results of gene expression with cognitive parameters, an increase in BDNF in women (but not in men) was also accompanied by an increase in the MMSE score (Fig. 13).\nNo statistically significant difference was observed in expression of other three genes (DJ-1, PINK1, TNFalfa) before and after treatment with the Neurotrophic peptide mixture (Figs. 14–16).\nThe correlation analysis shows higher values between MMSE and PINK1 expression, FAB a and DJ1, FAB a and PINK1, MOCA a and DJ1, MOCA a and PINK1, MOCA and PINK1 for men (Table 3).\nFor women, the correlation analysis shows strong values of clinically central endpoints UPDRS III and UPDRS S with almost all studied genes expression rates (Table 4).\nEffect of the Neurotrophic peptide mixture treatment on the state of the prooxidant-antioxidant system in patients with Parkinson’s disease\nThe intensification of prooxidant processes in blood plasma and erythrocytes of patients with moderate PD stages was registered. It was established a significant rise (by 40%) in the content of the secondary products of lipid peroxidation in blood plasma as well as a marked increase (by 50%) of the H2O2 formation in erythrocytes of patients with PD compared to healthy participants (Tables 5 and 6). Our assessment of the antioxidant protective effectiveness was largely based on the changes in activity of antioxidant enzymes – SOD and catalase, which represent the first line of such defense, providing detoxification of reactive oxygen metabolites27. In patients with PD, we have found an increase in the activities of SOD and catalase by 21% and 69%, respectively in comparison with healthy subjects. In addition, we analyzed the glutathione cycle, which is one of the main intracellular mechanisms for maintaining redox homeostasis28. In our study, the concentration of GSH was significantly decreased (by 58%) in erythrocytes of PD patients as well as GPx activity though this change was not statistically significant (by 15%) compared to healthy controls. We found that CBL administration has moderate antioxidant effects in PD patients by decreasing the TBARS accumulation (by 17%) in blood plasma as well as H2O2 content in erythrocytes (by 12%) compared with the CBL-untreated patients. These effects of CBL were accompanied by a decrease in the activities of SOD (by 11%) and catalase (by 20%) in blood of patients with PD compared with the CBL-untreated ones. In CBL-treated patients, there was registered an increase in the GSH concentration (by 30%) and GPx activity (by 15%) in erythrocytes compared with PD patients though this change was not statistically significant compared to the CBL-untreated patients (Tables 5 and 6).\nUPDRS score change prediction using multiple machine learning models\nAmong the models, GBMs demonstrated the highest predictive accuracy, achieving an RMSE of 18.59, MSE of 345.82, and MAE of 15.39. The top-performing RF model exhibited slightly higher RMSE (21.99) and MAE (17.78), underscoring the superior capability of GBMs to capture complex patterns in the data.\nThe boxplot of feature importance across the top 10 models provided valuable insights into which factors most significantly influenced the prediction of UPDRS score changes. Below are the most impactful features, sorted by median scaled importance: Fig. 17\n- \n                    BDNF Change (Median Importance: 0.98): Changes in Brain-Derived Neurotrophic Factor (BDNF) were the most influential predictor of UPDRS score change. BDNF’s high importance suggests that neurotrophic support may play a central role in the Neurotrophic peptide mixture’s therapeutic effects, potentially enhancing neuroplasticity and recovery. \n- \n                    MOCA (Median Importance: 0.62): The Montreal Cognitive Assessment (MOCA), a measure of cognitive function, was ranked as a highly predictive feature, highlighting the relevance of cognitive improvement as a response indicator for the Neurotrophic peptide mixture therapy. \n- \n                    BDNF (after treatment) (Median Importance: 0.61): Post-treatment levels of BDNF were also significant, indicating that maintaining or boosting neurotrophic factors is associated with positive therapeutic outcomes in PD patients. \n- \n                    MMSE, S (Median Importance: 0.58): The Mini-Mental State Examination (MMSE) score also emerged as a key predictor, reinforcing the importance of cognitive assessments in evaluating the Neurotrophic peptide mixture’s efficiency. \n- \n                    PINK1 (after treatment) (Median Importance: 0.55): PINK1, a mitochondrial health marker, was another significant predictor, suggesting that the Neurotrophic peptide mixture’s potential benefits may extend to cellular and mitochondrial health. \nThe boxplot analysis revealed that features related to neurotrophic support (BDNF) and cognitive function (MOCA, MMSE) showed the most consistent importance across models, as evidenced by their high median importance values and narrow interquartile ranges. In contrast, features with wider variability, such as TBA-active products and Age, had broader interquartile ranges, indicating that their contribution to predictive accuracy may depend on model-specific interactions or particular patient subgroups.\nDiscussion\nThe findings of this exploratory study suggest that the Neurotrophic peptide mixture has a multifaceted impact on various aspects of PD pathology, particularly regarding platelet ultrastructure, gene expression, neurophysiological parameters, and oxidative stress markers. These findings align with existing literature, which underscores the neuroprotective and neurorestorative properties of the Neurotrophic peptide mixture.\nImprovement in motor functions, particularly in the Motor tempo test, was more pronounced in patients with predominantly right-sided symptoms of PD. This asymmetry in response could be related to the lateralization of the pathological process. Additionally, the elevation in the threshold for triggering the H-reflex suggests modifications in spinal cord excitability, which could translate to improvements in motor function. The study reported statistically significant enhancements in cognitive abilities, as measured by the MMSE, FAB, and MoCA scales. Additionally, there was a notable reduction in depression scores (BDI) and reactive anxiety (Spielberger-Hanin scale), although personal anxiety levels remained unchanged. These findings highlight the potential of the Neurotrophic peptide mixture in ameliorating both motor and non-motor symptoms of PD. The improvements in motor function, cognitive abilities, and emotional well-being observed in this study suggest that the Neurotrophic peptide mixture has a broad therapeutic potential.\nThe study also observed significant neurophysiological improvements post-Neurotrophic peptide mixture treatment. The latency period of the P3 component, a marker of cognitive processing, was significantly reduced. This improvement in P3 latency indicates enhanced cortical function and suggests that the Neurotrophic peptide mixture may have beneficial effects on cognitive processing speed and efficiency in PD patients. These neurophysiological benefits align with the observed cognitive improvements, further supporting the potential of the Neurotrophic peptide mixture in mitigating cognitive decline in PD.\nIn PD patients, platelets exhibited several signs of senescence, including a reduction in dense δ-granules and an increase in α-granules29. The number of δ-granules per cell was significantly lower in PD patients (5.1 ± 0.4) compared to healthy individuals (8.4 ± 0.5). The administration of the Neurotrophic peptide mixture significantly increased the number of δ-granules in platelets by 45%, indicating enhanced serotonin reserves, which could contribute to the neuroprotective effects observed clinically. This increase in δ-granules likely reflects a normalization of platelet storage function, which is crucial given the role of serotonin in mood and neuroprotection. Despite no significant change in the number of α-granules, the trend towards a decrease might indicate an alteration in platelet activation status.\nAdditionally, the Neurotrophic peptide mixture reduced the calcium-accumulating capacity of platelets, potentially improving ATP synthesis and overall platelet function. This reduction in calcium accumulation is crucial as it indicates a shift away from the pathological state observed in untreated PD platelets. The observed improvements in platelet ultrastructure and function may translate to better overall cellular health, reflecting the systemic benefits of the Neurotrophic peptide mixture treatment beyond its direct neural effects. The enhancement in serotonin reserves and mitochondrial function underscores the multifaceted role of the Neurotrophic peptide mixture in addressing both neurological and peripheral manifestations of PD.\nThe Neurotrophic peptide mixture treatment appears to modulate BDNF gene expression in a sex-specific manner in patients with Parkinson’s disease, with women showing an increase and men showing a decrease in expression levels. This modulation is accompanied by improved cognitive function in women, highlighting the potential of the Neurotrophic peptide mixture as a therapeutic option that may benefit female PD patients more significantly. The stable expression of DJ-1, PINK1, and TNFα genes suggests that the primary neuroprotective mechanisms of the Neurotrophic peptide mixture in PD may involve pathways other than those mediated by these genes. These findings emphasize the need for further research to explore the underlying mechanisms of sex differences in PD treatment responses and to optimize therapeutic strategies accordingly.\nThe excessive production and release of ROS in PD occurs as a result of the combined action of mitochondrial dysfunction, dopamine autooxidation, and neuroinflammation30. In present study, the Neurotrophic peptide mixture demonstrated a moderate antioxidant effect, reducing the levels of lipid peroxidation products and H2O2 in erythrocytes. This reduction was accompanied by a decrease in SOD and catalase activities, and an increase in GSH concentration and GPx activity. These changes indicate a restoration of redox balance and a reduction in oxidative stress, which are crucial in the pathology of PD. One of the most promising effects of CBL on OS developing in blood and tissues in PD could be its ability to decrease intracellular α-synuclein aggregation in neurons, mimicking the action of classical NTFs and growth factors as well as of NTFs belonged to a novel family CDNF/MANF31. The ability of the Neurotrophic peptide mixture to modulate the antioxidant system suggests its potential role in protecting neuronal cells from oxidative damage and subsequent apoptosis.\nThe findings underscore the central role of cognitive and neurotrophic markers in predicting UPDRS score changes as an indicator of the Neurotrophic peptide mixture efficacy. The high importance of BDNF aligns with previous research on neuroprotective therapies, where neurotrophic support has been linked to improved neuronal survival and plasticity. The inclusion of mitochondrial markers, such as PINK1, suggests that the Neurotrophic peptide mixture may exert beneficial effects at the cellular level, addressing mitochondrial dysfunction—a known factor in PD pathology.\nThe boxplot analysis provided further clarity on the stability and reliability of each feature’s importance. Features with high median importance and narrow interquartile ranges, like BDNF and MOCA, indicate consistent predictive power across models, making them robust indicators of therapeutic response. The broader variability observed in other features, such as oxidative stress markers and demographic factors, suggests that while they may contribute to specific aspects of disease progression, they are less central to predicting therapeutic efficacy. The superior performance of GBMs in capturing non-linear relationships reinforces the value of using advanced machine learning techniques to analyze complex clinical datasets. These findings support the potential of the Neurotrophic peptide mixture as a multifaceted therapeutic option, with the capacity to improve cognitive and neurotrophic parameters in PD patients. The results of the present study indicate an antioxidant action of the Neurotrophic peptide mixture in patients with moderate stages of Parkinson’s disease. This action is mainly based on the decrease of the prooxidant events in peripheral blood of PD patients as well as on the prevention of thiol redox defects and reducing an imbalance in oxidant–antioxidant status.\nOverall, statistically significant improvements and consistent correlations were observed in the key clinical endpoints (UPDRS Part III and UPDRS S), while secondary and exploratory outcomes demonstrated patterns suggestive of mechanistic links that warrant confirmation in larger studies.\nInterpretation of its results is limited by several factors, including the small sample size, lack of a control group receiving placebo in addition to standard therapy, and a relatively short duration of the observation period. Since the study is preliminary, exploratory in nature and aimed at collecting primary data on the response to the drug, only one group of patients was formed, which allows identifying primary trends and assessing the feasibility of future larger studies. Future research should focus on larger cohorts and longer treatment durations to confirm these findings and to explore the long-term effects of neurotrophic peptide mixture on gene expression and cognitive function in PD patients. Particular attention should be paid to investigating sex differences in response to the drug in future studies.\nThe results of this exploratory study provide a preliminary rationale for investigating the Neurotrophic peptide mixture as an adjunct therapy in Parkinson’s disease. During the treatment period, changes were observed in platelet ultrastructure, gene expression, neurophysiological parameters, and oxidative stress markers. These findings, along with shifts in cognitive and neurotrophic markers, may suggest a multidimensional biological response.\nHowever, these observations must be interpreted with extreme caution. Given the limited sample size and the absence of a control group, no causal inferences can be drawn. Therefore, larger, placebo-controlled clinical trials are warranted to determine if these initial observations can be substantiated, to investigate potential sex-specific and long-term effects, and to explore biomarkers that may help elucidate the mechanisms of the Neurotrophic peptide mixture in PD.\nMethods\nPatient information\nThe exploratory study included 17 patients with Parkinson’s disease with an established diagnosis according to the inclusion-exclusion criteria of the British Brain Bank32, aged 45–74 years with disease stage 2.0–3.0 (2.9 ± 0.3, according to Hoehn and Yahr scale). Written informed consent was obtained from all participants.\nThe neurotrophic peptide mixture was prescribed against the background of basic stable doses of antiparkinsonian therapy (Levodopa-containing drugs), which were used during 28–30 days before starting the course of neurotrophic peptide mixture and during the entire study period. No dose adjustments were allowed during the study, and medication dose was not included as a covariate in the statistical analysis. The median daily Levodopa dose was about 500 mg, it was individually optimized depending on disease stage and the patient’s condition. The Neurotrophic peptide mixture dosage regimen was 20 ml (215.2 mg/ml) intravenously, once a day for 10 days. Infusions were carried out in the first half of the day.\nEthics statement and study design\nThe Ethics Committee of the Clinical Department of the Institute of Gerontology named after D.F. Chebotariov of the National Academy of Medical Sciences of Ukraine approved the protocol of current research (Protocol No. 4, 09/06/2022). Written informed consent was obtained from all participants prior to enrollment.\nThe study was conducted in accordance with the principles of the Declaration of Helsinki (2000) and the European Council Convention on Human Rights and Biomedicine (2005), and relevant national standards for the participation in the research were taken into account.\nClinical assessment\nCharacterization of the stages of PD was carried out using the Hoehn and Yahr scale. To assess the dynamics of motor disorders, the standardized international Unified Parkinson’s Disease Rating Scale (UPDRS) was used. For a comprehensive assessment of the state of non-motor functions, the following clinical scales were used:\n- \n                    Mini-Mental State Examination (MMSE); \n- \n                    Test battery for assessing frontal dysfunction (Frontal Assessment Battery, FAB); \n- \n                    Montreal Cognitive Assessment (MoCA); \n- \n                    Beck Depression Inventory (BDI); \n- \n                    Spielberger-Hanin Anxiety Scale. \nSensorimotor reaction and motor tempo\nMotor tempo was determined using a computer program as the time between consecutive presses of the same finger on two keys spaced 20 cm apart on the keyboard. The patient was instructed to press the keys at the fastest possible tempo. The duration of each test was 20 s. Testing was conducted for each hand separately, and then the data were averaged.\nThe latent period of simple sensorimotor reaction was determined as the interval between the moment of presentation of a visual signal on the computer monitor and the moment the patient pressed the key in response to the signal. Geometric shapes (circle, square, oval) of red and white colors were used as signals, with 20 signals presented at intervals of 2–4 s. Testing was conducted for each hand separately, and the data were averaged.\nElectroneuromyography\nThe technique of registering the H-reflex, as a reflex response of muscles to the stimulation of sensory fibers of a mixed nerve followed by monosynaptic activation of spinal motoneurons and nerve motor fibers, was performed using disposable surface recording electrodes placed in the projection of the posterior group of muscles of the calf, with stimulation of the tibial nerve in the popliteal fossa. The dynamics of the amplitudes of H- and M-responses (mV) with increasing strength of nerve electrical stimulation (mA), the threshold for the occurrence of the H-reflex (mA), the latency period of the H-reflex (ms), and the ratio of the maximum amplitude of H max/M max (%) were evaluated.\nEvoked potentials P300\nEEG was recorded monopolarly from 21 leads (Fp1, Fp2, Fpz, F3, F4, Fz, C3, C4, Cz, P3, P4, Pz, T3, T4, F7, F8, T5, T6, O1, O2, Oz). The combined ear electrode was used as a reference electrode. The interelectrode resistance was less than 5 kOhm, the signal sampling frequency was 500 Hz, and the high- and low-pass filters were 0.3 Hz and 35 Hz, respectively. Endogenous cognitive evoked potentials N2 and P3 were recorded according to the oddball paradigm using sound stimuli with a frequency of 1000 and 2000 Hz and a duration of 10 ms.\nPlatelet and mitochondrial structure with Electron microscopy\nBlood plasma enriched with platelets containing leukocytes was obtained by centrifugation of whole blood obtained from the ulnar vein of the subjects at room temperature for 15 min at 120 g in a T-30 centrifuge. The plasma was carefully separated from the sediment and centrifuged at 2000 g for 20 min using a Vortecs Combispin FVL-2400N minicentrifuge33.\nElectron microscopic studies of platelets and mitochondria were carried out according to the generally accepted method with double fixation with glutaraldehyde and OSO4, dehydration with alcohols of increasing concentrations and subsequent embedding in Epon. Ultrathin sections with a thickness of 40–60 nm, contrasted with uranyl acetate and lead citrate, were examined using an electron microscope PEM-125K (Ukraine).\nMorphometric calculations of platelets’ structure were performed using the Image Tool (USA) on 130–150 fields for each group of subjects. The number of optically dense small σ-granules and optically transparent large α-granules was determined.\nPBMC isolation and quantitative PCR\nBlood collection, PBMC isolation, RNA extraction and cDNA synthesis were performed by standard methods. The collection of blood for further cell isolation and qPCR was made on 0 and 11th day (after a 10-day course of neurotrophic peptide mixture treatment). Briefly, patients’ blood was collected in 4 ml vacutainers with EDTA. Density gradient centrifugation of diluted whole blood layered over a density gradient medium was used for PBMC isolation. PBMC were gradually frozen at a temperature of −20 °C, and then transferred to the temperature of −80 °C for long-term storage. RNA was isolated from cells using the Direct-zol RNA Miniprep Kit (Zymo Research, USA) following the manufacturer’s protocols. The isolated RNA was treated with DNase to remove DNA residues that could give a false positive signal for the expression of target genes. After this, reverse transcription was carried out using a commercial kit ZymoScript RT PreMix (Zymo Research, USA), guided by the manufacturer’s recommendations.\nSamples of this cDNA were used for further qPCR-RT (quantitative real-time polymerase chain reaction), which was used to measure the expression of the DJ-1, PINK1, BDNF and TNFα genes. To identify a suitable internal control for RT-qPCR data normalization, we empirically tested the expression stability of three candidate reference genes (rps18, gusb, and beta-actin). The sequences of all used primers are listed in the Supplementary Information. Analysis of the raw Cqt values revealed that both beta-actin and gusb were the most stably expressed genes, showing minimal variation and a strong positive correlation with each other. In contrast, rps18 expression was found to be significantly more variable across our samples and was therefore excluded. Although both beta-actin and gusb were suitable candidates, we selected beta-actin for normalization due to its slightly higher expression stability.\nThe sequences of all used primers are listed in the Supplementary Information. The PCR reaction mix was prepared using a commercial kit of reagents for RT-PCR (HOT FIREPol® EvaGreen® qPCR Mix Plus, Solis BioDyne, Estonia) - reverse transcription and polymerase chain reaction with the addition of betaine (Sigma, USA) at a final concentration of 1 M. To determine differences in mRNA expression of Parkin genes, Repeated Measures Anova was used.\nAccumulation curves were generated using Opticon Monitor 3 software. After amplification was completed, raw gene product amplification data were used to calculate Cy0.\nMeasurement of oxidative stress biomarkers\nSecondary products of lipid peroxidation, the activity of antioxidant enzymes superoxide dismutase (SOD) and catalase (CAT) were measured in blood plasma. In red blood cells – the content of hydrogen peroxide (H2O2); the state of the glutathione system was assessed by the content of reduced glutathione (GSH) and the activity of the enzyme glutathione peroxidase (GP).\nFor erythrocyte isolation, samples were centrifuged at 8000 rpm for 15 min to separate plasma and red blood cells. Using centrifugation, erythrocytes were washed twice of plasma with cold sterile physiological sodium chloride solution. Erythrocyte hemolysate was obtained by adding distilled water to washed erythrocytes in a 1:1 ratio. The hemoglobin (Hb) content was also measured (Hemoglobin Assay kit MAK115, Sigma-Aldrich, St. Louis, MO, USA).\nDetermination of thiobarbituric acid reactive substances (TBARS) levels: TBARS were isolated by boiling plasma with the thiobarbituric acid reagent (0.5% 2-thiobarbituric acid/10% trichloroacetic acid/0.63 mM hydrochloric acid) and measuring the absorbance at 532 nm. The results are expressed as µM/ml using a molar extinction coefficient 1.56 * 105 mm−1 cm−1.\nCatalase activity measurement: The method is based on the ability of molybdenum salts to form a colored complex with hydrogen peroxide, which was determined at a wavelength of 410 nm. Enzyme activity (μM/min/L) was calculated using a molar extinction coefficient of 22.2 * 103 mm−1cm−1.\nSuperoxide dismutase (SOD) activity was determined spectrophotometrically. The method is based on the ability of SOD to inhibit the autoxidation of adrenaline at pH 10.2. The results were expressed as specific activity of the enzyme in units per ml of plasma. One unit of SOD activity is defined as the amount of protein causing 50% inhibition of the conversion rate of adrenaline to adrenochrome under specified conditions.\nThe activity of selenium-dependent glutathione peroxidase (GPx) was determined by recording the rate of NADPH oxidation in the presence of reduced glutathione at a wavelength of 340 nm. Enzyme activity was expressed in nM NADPH/min/ml.\nThe amount of reduced glutathione (μM/ml) was determined using Elman’s reagent (5.5’-dithio-bis-nitrobenzoic acid in absolute ethanol). The change in extinction was recorded at a wavelength of 412 nm every minute for 5 minutes.\nHydrogen peroxide content was studied using the FOX method, which is based on the peroxide-mediated oxidation of Fe2 + , followed by the reaction of Fe3+ with xylenol orange. Absorbance of the Fe3 + -xylenol orange complex was determined at a wavelength of 560 nm. The amount of H2O2 (μM/mgHb) was calculated using the H2O2 standard curve for each independent experiment.\nMachine learning prediction of UPDRS score change\nTo model UPDRS score changes, H2O’s AutoML framework was employed34, allowing for the training, tuning, and evaluation of multiple Machine Learning models, including Gradient Boosting Machines (GBMs) and Random Forests (RFs). AutoML facilitated hyperparameter optimization, ensuring robust and accurate predictions.\n- \n                    Gradient Boosting Machines (GBMs): GBMs iteratively build a series of decision trees, where each subsequent tree focuses on correcting errors made by its predecessors. This sequential approach enables GBMs to capture intricate, non-linear relationships within the data, making them particularly suitable for analyzing how clinical and biochemical factors impact UPDRS score changes and, consequently, therapeutic response35. \n- \n                    Random Forests (RFs): RFs are ensemble models that combine multiple decision trees, each trained on a random subset of the data. Known for their robustness and reduced susceptibility to overfitting, RFs provide valuable insights into feature importance by averaging across numerous trees36. This model serves as a stable baseline to complement the more flexible GBMs. \nThe models were evaluated using the following metrics to quantify prediction accuracy:\n- \n                    Root Mean Squared Error (RMSE): Captures the average deviation between predicted and actual values. \n- \n                    Mean Squared Error (MSE): Measures the variance of predictions relative to actual outcomes. \n- \n                    Mean Absolute Error (MAE): Reflects the average absolute deviation between predicted and observed values. \nThese metrics allowed for a thorough assessment of each model’s ability to predict UPDRS score changes, highlighting the Neurotrophic peptide mixture’s potential impact on disease progression.\nFeature importance was evaluated using the top 10-performing models from a set of 50 trained with H2O AutoML. Each feature’s importance was measured by its “scaled importance,” which represents its relative contribution to prediction accuracy. The models were trained using two-fold cross-validation, and feature importance was extracted during the validation stage.\nData from the top 10 models were aggregated, and features were sorted by their median feature importance. It is worth noting that while building a generalized model on such a small dataset is not feasible, the features identified with the highest median importance could be considered potential markers in future studies. The results were visualized using a boxplot (Fig. 16) to highlight the median importance and variability of each feature.\nStatistical analysis\nHypotheses about the effects of the drug were tested using the Wilcoxon (for related samples) and Mann-Whitney (for independent groups) tests. The relationship between the investigated indicators was determined in the form of Spearman’s rank correlation coefficients.\nStatistical processing of gene expression data was performed using Statistica 8.0 software (StatSoft Inc., USA). For statistical analysis, Spearman’s non-parametric test was used to investigate the correlation between parameters. The difference in expression of the studied genes before and after treatment was determined using the Wilcoxon t-test (Wilcoxon signed-rank test).\nThe results of the study of the state of the prooxidant-antioxidant system were processed statistically using the Origin program. The significance of differences between comparison groups was determined by the method of analysis of variance (ANOVA) followed by the Bonferroni test (post-hoc test).\nOutcomes and hierarchy\nFor this exploratory study, outcomes were grouped into three hierarchical levels.\n- \n                    Clinically central endpoints were prespecified as the UPDRS III (motor score) and the UPDRS Total score, as the primary efficacy measures, as these are the most widely accepted, clinically relevant, and objective indicators of motor impairment and overall disease severity in Parkinson’s disease. \n- \n                    Secondary clinical endpoints included UPDRS I (non-motor experiences of daily living) and UPDRS II (motor experiences of daily living), and validated scales of non-motor function: MMSE, MoCA, FAB (cognitive measures), and BDI and Spielberger–Hanin Anxiety Scale (mood and affective measures). These outcomes were considered supportive of the clinically central endpoints. \n- \n                    Exploratory outcomes included additional non-motor measures, neurophysiological parameters (P300 latency and amplitude), and molecular/structural biomarkers (gene expression, platelet ultrastructure, mitochondrial morphology, and oxidative stress markers). These outcomes were included to generate mechanistic insights into the potential modes of action of the Neurotrophic peptide mixture. \nFormal hypothesis testing was restricted to the clinically central endpoints. Secondary and exploratory outcomes are presented with descriptive statistics and effect sizes. To minimize Type I error, no multiple-testing correction was applied outside of the clinically central endpoints.\nCorrelation analyses were performed using Spearman’s rank correlation coefficient (ρ). For correlations involving clinically central endpoints, p-values are reported. Correlations among secondary and exploratory outcomes are reported without formal inferential testing, to avoid Type I error inflation.\nData availability\nThe datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.\nReferences\n- Alexoudi, A., Alexoudi, I. & Gatzonis, S. Parkinson’s disease pathogenesis, evolution and alternative pathways: A review. Rev. Neurol. (Paris) 174, 699–704 (2018). \n- Chung, S. J. et al. The Pattern of Striatal Dopamine Depletion as a Prognostic Marker in De Novo Parkinson Disease. Clin. Nucl. Med 201843, 787–792 (2018). \n- Moon, H. E. & Paek, S. H. Mitochondrial Dysfunction in Parkinson’s Disease. Exp. Neurobiol. 24, 103–116 (2015). \n- Mansur, A. et al. Characterization of 3 PET tracers for quantification of mitochondrial andsynaptic function in healthy human brain:18F-BCPP-EF,11C-SA-4503, and11C-UCB-J. J. Nucl. Med 61, 96–103 (2020). \n- Simonenkov, A. P. & Kluyev, V. M. Syndrome serotonin deficiency. M.: BINOM, 96-97 (2013). \n- Mauricio, R. et al. Tackling gaps in developing life-changing treatments for dementia. Alzheimers Dement 5, 241–253 (2019). \n- Gonchar, O. O., Klymenko, O. O., Drevytska, T. I., Bratus, L. V. & Mankovska, I. M. Oxidative stress in rat heart mitochondria under a rotenone model of Parkinson’ disease: a corrective effect of capicor treatment. Ukr. Biochem J. 93, 21–30 (2021). \n- Schneider-Garces, N. J. et al. Span, CRUNCH, and beyond: working memory capacity and the aging brain. J. Cogn. Neurosci. 22, 655–669 (2010). \n- Fiani, B. et al. Cerebrolysin for stroke, neurodegeneration, and traumatic brain injury: review of the literature and outcomes. Neurol. Sci. 42, 1345–1353 (2021). \n- Brainin, M. Cerebrolysin: a multi-target drug for recovery after stroke. Expert Rev. Neurother. 18, 681–687 (2018). \n- Requeejoe, C. et al. Nanodelivery of Cerebrolysin and rearing enriched environment induce neuroprotective effects in a preclinical rat model of Parkinson’s disease. Mol Neurobiol. Springer Science+Business Media, LLC (2017). \n- Neveen, A. et al. A promising therapeutic potential of Cerebrolysin in the 6-OHDA rat model of Parkinson's disease. Life Sci 155, 174–179 (2016). \n- Luchanina, E., Karaban, I. & Berezetskaya, N. Diagnosis of Parkinsonas disease by electrophysiological methods. Diagn. Rehabil. Parkinson. Dis. 2, 27–58 (2011). \n- Eder, P., Reinprecht, I., Schreiner, E., Skofitsch, G. & Windisch, M. Increased density of glutamate receptor subunit 1 due to Cerebrolysin treatment: an immunohistochemical study on aged rats. Histochem J. 33, 605–612 (2001). \n- Stepanichev, M. et al. Effects of cerebrolysin on nerve growth factor system in the aging rat brain. Restor. Neurol. Neurosci. 35, 571–581 (2017). \n- Veinbergs, I., Mante, M., Mallory, M. & Masliah, E. Neurotrophic effects of Cerebrolysin in animal models of excitotoxicity. J. Neural Transm. Suppl. 59, 273–280 (2000). \n- Alvarez, X. A. et al. Neuroimmunotrophic effects of cerebrolysin in an animal model of hippocampal degeneration induced by β-amyloid. In: Korczyn, A. D., ed. International Congress on Vascular Dementia 2017. Bologna, Italy: Monduzzi Editore, 233–237 (1999). \n- Gavrilova, S. I. & Alvarez, A. Cerebrolysin in the therapy of mild cognitive impairment and dementia due to Alzheimeras disease: 30 years of clinical use. Med Res Rev. 41, 1–29 (2020). \n- DeBoer, S. R. et al. Enhanced spontaneous motor recovery after stroke in mice treated with cerebrolysin. Neurorehabil Neural Repair 35, 525–533 (2021). \n- Muresanu, D. F. et al. Efficacy and safety of cerebrolysin in neurorecovery after moderate-severe traumatic brain injury: results from the CAPTAIN II trial. Neurol. Sci. 41, 1171–1181 (2020). \n- Lu, W. et al. Cerebrolysin alleviates early brain injury after traumatic brain injury by inhibiting neuroinflammation and apoptosis via TLR signaling pathway. Acta Cir. Bras. 37, e370605 (2022). \n- Chen, C. C., Wei, S. T., Tsaia, S. C., Chen, X. X. & Cho, D. Y. Cerebrolysin enhances cognitive recovery of mild traumatic brain injury patients: double-blind, placebo-controlled, randomized study. Br. J. Neurosurg. 27, 803–807 (2013). \n- Cui, S. Cerebrolysin for vascular dementia. Cochrane Database Syst Rev. 11, 19–20 (2019). \n- Rochenstein, E. et al. Neuropeptide treatment with cerebrolysin enhances the survival of grafted neural stem cells in a synuclein transgenic model of Parkinson’s disease. J. Exp. Neurosci. 9, 131–140 (2015). \n- Hünerli, D. et al. Mild cognitive impairment in Parkinson’s disease is associated with decreased P300 amplitude and reduced putamen volume. Clin. Neurophysiol. 130, 1208–1217 (2019). \n- Alvarez, X. A., Lombardi, V. R., Fernández-Novoa, L., García, M. & Sampedro, C. Cerebrolysin reduces microglial activation in vivo and in vitro: a potential mechanism of neuroprotection. J. Neural Transm. 59, 281–292 (2000). \n- Kulinskiĭ, V. I. & Kolesnichenko, L. S. Glutathione system. I. Synthesis, transport, glutathione transferases, glutathione peroxidases. Biomed. Khim 55, 255–277 (2009). \n- Zenkov, N. K., Lankin, V. Z. & Menshikova, E. B. Oxidative stress: The biochemical and pathophysiological aspects. M.: MAIK. Science / Interperiodica 4, 352–361 (2001). \n- Park, S. M. et al. Evidence that alpha-synuclein functions as a negative regulator of Ca(++)-dependent alpha-granule release from human platelets. Blood 100, 2506–2514 (2002). \n- Luo, Y., Hoffer, A., Hoffer, B. & Qi, X. Mitochondria: A therapeutic target for Parkinson’s disease?. Int J. Mol. Sci. 16, 20704–20730 (2015). \n- Voutilainen, M. H., Arumae, U., Airavaara, M. & Saarma, M. Therapeutic potential of the endoplasmic reticulum located and secreted CDNF/MANF family of neurotrophic factors in Parkinson’s disease. FEBS Lett. 589, 3739–3748 (2015). \n- Hughes, A. J., Daniel, S. E., Kilford, L. & Lees, A. J. Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: A clinico-pathological study of 100 cases. J. Neurol. Neurosurg. Psychiatry 55, 181–184 (1992). \n- Filippova, O. I., Koloskov, A. V. & Stolitsa, A. A. Methods for studying the functional activity of platelets (literature review). Transfusiology 13, 493–514 (2012). \n- LeDell, E. & Posse, C. H2O AutoML: Scalable Automatic Machine Learning. J. Open Source Softw. 5, 1–10 (2020). \n- Chen, J. & Guestrin, C. XGBoost: A Scalable Tree Boosting System. Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery Data Mining, 785–794 (2016). \n- Liaw, A. & Wiener, M. Classification and regression by random Forest. R. N. 2, 18–22 (2002). \nAcknowledgements\nThis work was supported by the National Academy of Medical Sciences, Ukraine (Grant # 0123U100771).\nAuthor information\nAuthors and Affiliations\nContributions\nD.K. took part in writing of the main text of manuscript, coordinated the work of scientific laboratories in their experiments and results, contributed to the conceptualization of the project, I.K. was the lead scientist of part connected to clinical assessment, data analysis, project conception and organization, N.K., N.M. and S.K. performed and analyzed electroencephalograms and elecroneuromyograms of patients, collected patient data regarding sensorimotor reactions and motor tempo, K.R. participated in the preparation of the literature review, implemented the methodology of morphological studies, processing the data of the section about the morphological parameters. O.G. examined markers of oxidative stress and made substantial contributions to the analysis of data as well as participated in a critical review of the manuscript, I.M. helped with the preparation of the manuscript and analysis of data. O.Z. wrote an abstract, implemented the main methodology for gene expression measurement, analysis and interpretation of data, O.B. and K.M. were involved to gene expression experiments and analysis of qPCR-RT data, as well as collection of biological material, S.S. provided the results of deep analysis of obtained data using multiple machine learning models, performed feature importance analysis, T.P. contributed to the administration of project, wrote and edited the main manuscript text. All authors reviewed the manuscript.\nCorresponding author\nEthics declarations\nCompeting interests\nThe authors declare no competing interests.\nAdditional information\nPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.\nRights and permissions\nOpen Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.\nAbout this article\nCite this article\nKrasnienkov, D., Karaban, I., Karasevych, N. et al. Evaluation of the neurotrophic peptide mixture in pathogenetic therapy of patients with Parkinson’s disease. npj Parkinsons Dis. 12, 55 (2026). https://doi.org/10.1038/s41531-026-01270-6\n- Received: \n- Accepted: \n- Published: \n- Version of record: \n- DOI: https://doi.org/10.1038/s41531-026-01270-6","image_url":"https://media.springernature.com/m685/springer-static/image/art%3A10.1038%2Fs41531-026-01270-6/MediaObjects/41531_2026_1270_Fig1_HTML.png","lang":"en","published_at":"2026-01-23T08:00:00+00:00","fetched_at":"2026-07-12T01:15:08+00:00","status":"read","starred":0,"extract_state":"ok","summary_auto":"Abstract\nThis exploratory, single-group, open-label study investigated 17 patients with Parkinson’s disease (PD) using a pre-post design. Motor and non-motor outcomes were assessed through clinical scales, biochemical and genetic analyses, and machine learning models (Gradient Boosting Machines, Random Forests).","cluster_id":null,"extract_retries":0,"extract_error":null,"contract_version":"news_item.v1","format_contract_version":"news_item_formats.v1","dedup_url":"https://www.nature.com/articles/s41531-026-01270-6","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 58549 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":58549,"summary_length":313,"usable_text_length":58549,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":58549,"summary_length":313}},"news_item":{"id":34350,"canonical_url":"https://www.nature.com/articles/s41531-026-01270-6","source_url":"https://www.nature.com/articles/s41531-026-01270-6","title":"Evaluation of the neurotrophic peptide mixture in pathogenetic therapy of patients with Parkinson’s disease - Nature","source_name":"Nature","author":null,"published_at":"2026-01-23T08:00:00+00:00","locale":"en","topic":"peptides","tags":[],"rss_summary":"<a href=\"https://news.google.com/rss/articles/CBMiX0FVX3lxTE4tTGM3aWNjbjR5czF1Zm9XMVR4YzM5R0JjREZKTmkyckY3MVRhVFhPbEZySnlydnVXMmU5VHc2REVQQVZIZExPNHNXSTBEVVgtZW96TWVWY0pmMmc0T0sw?oc=5\" target=\"_blank\">Evaluation of the neurotrophic peptide mixture in pathogenetic therapy of patients with Parkinson’s disease</a>&nbsp;&nbsp;<font color=\"#6f6f6f\">Nature</font>","full_text":"Abstract\nThis exploratory, single-group, open-label study investigated 17 patients with Parkinson’s disease (PD) using a pre-post design. Motor and non-motor outcomes were assessed through clinical scales, biochemical and genetic analyses, and machine learning models (Gradient Boosting Machines, Random Forests). After treatment with a neurotrophic peptide mixture, improvements were observed in daily activity (16%), cognition (11%), depression (10% reduction), and reactive anxiety (23% reduction). Biological changes included a 45% increase in platelet δ-granules, higher mitochondrial counts, elevated gene expression (notably BDNF in women, p = 0.046), and modulation of oxidative stress markers (17% reduction in TBARS, 30% increase in GSH). Machine learning identified BDNF and PINK1 expression, along with MOCA and MMSE scores, as key predictors of UPDRS improvement. These findings suggest that neurotrophic peptide therapy may influence clinical, structural, and molecular domains in PD. Larger, controlled trials are warranted to confirm therapeutic potential and clarify associations with cognitive and neurotrophic parameters.\nSimilar content being viewed by others\nIntroduction\nParkinson’s disease is a neurodegenerative condition typically observed in older individuals, characterized by a gradual onset and progressive nature. The initial manifestation often includes tremors, followed by symptoms such as slowed movement (bradykinesia) and muscle stiffness (rigidity)1. The disease is linked to the degeneration of dopaminergic neurons in the substantia nigra, the presence of alpha-synuclein aggregates within neuronal cells (Lewy bodies and Lewy neurites), and subsequent pathology in regions like the basal forebrain, amygdala, and medial temporal lobe structures, progressing to cortical areas in advanced stages2.\nMitochondrial abnormalities and dysfunction are recognized as key contributors to Parkinson’s disease pathogenesis in a subset of patients3. Some data was obtained on the availability correlation of the functional activity of mitochondria in lymphocytes, which displays multisystem mitochondrial failure and indicates of energy dysfunction in general4. The spectrum of mitochondrial dysfunctions in PD includes dysfunction of the respiratory chain (inhibition of the NADH-oxidase oxidation pathway), hyperproduction of ATP, disruption of Ca-regulatory systems, opening of the mitochondrial pore, perturbations of mitochondrial dynamics, and dysregulation of mitochondrial clearance5.\nAlthough oxidative stress plays a significant role in the pathogenesis of PD, therapy based on the use of exogenous antioxidants has not been shown to be highly effective. An alternative strategy that can counteract the damaging effects of free radicals and restore cellular redox balance is the activation of endogenous antioxidant enzymes present in the central neural system, such as superoxide dismutase, catalase, and glutathione peroxidase. The transcription of these cytoprotective proteins is regulated by the transcription factor Nrf2, which plays a leading role in the regulation of cellular redox status and is the main regulator of the cellular response to oxidative stress through the induction of antioxidant and detoxifying enzymes and proteins6. This is consistent with our previous data, which showed that lipid peroxidation was significantly increased while reduced glutathione (GSH) and glutathione-dependent antioxidant enzymes activity markedly decreased in blood of patients with PD as well as in rat brain under modeling of PD7. Various activators of the Nrf2/ARE signaling pathway, including peptides, have shown their effectiveness in both in vitro and in vivo models of neurological disorders8.\nThe Neurotrophic peptide mixture is peptidergic mixture with neurotrophic-like properties, made from a mixture of low molecular weight, porcine-derived peptides and free amino acids. The pharmaceutical preparation includes the peptide fragments: nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), ciliary neurotrophic factor (CNTF), enkephalins, orexin, and P219.\nThe use of the Neurotrophic peptide mixture significantly reduces neurological deficits, improves cognitive abilities, and reduces the volume of the infarction zone during stroke10,11. The Neurotrophic peptide mixture has an organ-specific multimodal effect on the brain, i.e., it provides metabolic regulation, neuroprotection, functional neuromodulation, and neurotrophic activity12. The Neurotrophic peptide mixture protects neurons from the damaging effects of lactic acidosis, prevents the formation of free radicals13, increases survival and prevents neuronal death under conditions of hypoxia and ischemia, and reduces the damaging neurotoxic effect of glutamate14.\nOver the past decade, compelling evidence has emerged neuroprotective properties of the Neurotrophic peptide mixture, explaining its multi-level impact on both the pathological cascade of damage and on subsequent recovery processes15. The ability of the Neurotrophic peptide mixture to influence the processes of neuroprotection and neuroplasticity was proven in a series of studies15,16,17 on a model of Alzheimer’s disease in transgenic mice for the amyloid precursor protein (mThy-1-hAPP751 tg). Using a model of Alzheimer’s disease, it was shown that the Neurotrophic peptide mixture, through the regulation of the synthesis and transport of amyloid precursor protein and the deposition of β-amyloid in synapses, has a synaptotrophic effect, which plays a primary role in the processes of neuroplasticity, which are extremely important in the acquisition of new skills and the maintenance of existing ones18.\nIncreasing studies have demonstrated that the Neurotrophic peptide mixture administration can promote recovery of motor function, improve early brain injury, and decrease hippocampal neuronal death19. In recent clinical studies, the Neurotrophic peptide mixture improved overall outcomes after moderate to severe traumatic brain injury and was also safe and better for early rehabilitation patients after ischemic stroke20. Lu et al. reported that the Neurotrophic peptide mixture can improve neurological dysfunction and brain damage, relieve neuroinflammation and prevent apoptosis after brain injury21.\nBased on the multifaceted mechanism of action of the Neurotrophic peptide mixture, including its effect on the processes of neurogenesis, neuroplasticity and neuroprotection, the drug currently has three main indications for use - stroke, dementia and traumatic brain injury10,22,23.\nLuchanina et al. demonstrated for the first time the effect of Neurotrophic peptide mixture on motor activity and muscle tone in patients with PD treated with levodopa13. It was established that the Neurotrophic peptide mixture improved the engraftment of a DOPA transplant from stem cells in the substantia nigra24 and reduced the severity of symptoms in model animals12.\nThe current exploratory study dataset comprised clinical, cognitive, biochemical, and demographic information from PD patients undergoing the Neurotrophic peptide mixture therapy. Key features included cognitive assessments (MMSE, MOCA), neurotrophic markers (BDNF, PINK1), oxidative stress indicators (SOD, DJ1), and demographic factors (age, sex). The primary outcome was the change in UPDRS scores, serving as an objective measure of therapeutic response to the Neurotrophic peptide mixture. Also, in this study, machine learning models were employed to predict UPDRS score changes as a key indicator of the Neurotrophic peptide mixture’s therapeutic effectiveness, offering a data-driven evaluation of treatment response.\nResults\nClinical and neurophysiological assessment of the effectiveness of course treatment\nA course of the Neurotrophic peptide mixture administration against the background of basic antiparkinsonian therapy had a beneficial effect on the condition of the majority of patients, primarily on their activity in everyday life and the ability to self-care. Objectively, this effect was expressed in a significant decrease in scores of all UPDRS subscales (~16% on average) (Fig. 1).\nBoth clinically central endpoints improved significantly (p < 0.01). Secondary measures (UPDRS I and UPDRS II) generally trend in the same direction; effect sizes are provided without formal inference.\nA beneficial impact was observed following administration of a regimen of the Neurotrophic peptide mixture on a spectrum of non-motor dysfunctions concomitant with the motor manifestations of Parkinson’s disease (PD). Participants exhibited statistically significant enhancements in cognitive abilities as assessed by specialized instruments including MMSE, FAB, and MoCA (~11% on average). Moreover, there was a notable reduction in scores indicative of depression measured via the Beck Depression Inventory scale (~10% on average), as well as in reactive anxiety assessed by the Spielberger-Hanin scale (~23% on average). Notably, levels of personal anxiety remained consistent relative to baseline measurements. As a result, all observed Secondary outcomes are trended in the direction of improvement (Fig. 2).\nFor instrumental assessment of the Neurotrophic peptide mixture effect on motor functions, the Motor tempo test showed higher sensitivity. As a result, the significant decrease in the average, minimum and maximum time of motor reactions was recorded (Fig. 3).\nThose changes are more characteristic of the reactions of the right hand, while for the left hand only the average value of the motor tempo decreased. It can be assumed that this asymmetry of effects is associated with the sidedness of the pathological process, since a significant part of patients (76% of the study group) demonstrated predominantly right-sided symptoms of PD. These findings show a valuable positive impact of Neurotrophic peptide mixture on neurophysiological parameters.\nAt the same time, the latent period of a simple sensorimotor reaction did not change during a course of the Neurotrophic peptide mixture administration (Fig. 3).\nChanges of the H-reflex and M-response after administration of a regimen of Сerebrolysin\nIn the context of administering the Neurotrophic peptide mixture, there was an elevation observed in the threshold for triggering the H-reflex of the gastrocnemius muscle, increasing from 5 [3; 9] mA to 6 [5.8; 18] mA (Fig. 4). Notably, there were no alterations noted in the latency period or power attributes of the H-reflex, nor in the parameters of the M-response, compared to their baseline values.\nPotentials P300\nAnalysis of endogenous components revealed that in the studied group of patients with PD, there was a delay in the formation of the P3 wave (median 434 ms, interquartile range 386–512 ms in the temporal area of the neocortex) compared to the corresponding age norm of 390–410 ms. After the course of the Neurotrophic peptide mixture, a statistically significant increase in the latency period of the P3 component in the central-parietal lead was recorded, up to 394 [304; 446] ms (Fig. 5). There were no significant changes in the latency period of the N2 component and the amplitudes of the N2 and P3 components.\nThe delay in the formation of the P3 wave is a characteristic feature of several neurodegenerative diseases, particularly Alzheimer’s and Parkinson’s diseases. For patients with borderline PD and moderate cognitive impairments, an increase in the latency period of P3 has been described, with a reverse correlation between this parameter and the volume of gray matter in the globus pallidus, dysfunction of which leads to symptoms such as bradykinesia and rigidity25. Improvement in this parameter following the course of the Neurotrophic peptide mixture suggests that a more pronounced effect on the state of higher cortical functions can be expected in patients with increased risk of developing or progressing cognitive impairments.\nCorrelation analysis of clinical and neurophysiological parameters after a course of treatment with the Neurotrophic peptide mixture\nThe degree of improvement in several non-motor characteristics (cognitive functions assessed by the FAB scale, amplitude of cognitive evoked potential P3, and signs of depression measured by the Beck scale) was greater in individuals who initially received a higher daily dose of levodopa (Figs. 6–8).\nFor indicators of cognitive functions (MMSE, P3 cognitive evoked potential amplitude), a similar relationship with the time of onset of the disease was also found: more pronounced positive effects of the Neurotrophic peptide mixture treatment were observed in patients with an earlier onset of PD and a longer duration of the disease (Figs. 6, 8).\nAt the same time, a more pronounced reduction in reaction time was characteristic of individuals with lower initial ratings of PD severity on the UPDRS scale and lower levels of anxiety. Such correlation relationships are found both when directly assessing the time of motor reactions in the Motor tempo and Sensorimotor reaction tests, and when determining the latent periods of the H-reflex and M-response (Figs. 9, 10).\nThe use of the Neurotrophic peptide mixture in the treatment course improves the general condition of patients with Parkinson’s disease, as evidenced by a significant decrease in the quantitative indicators of all UPDRS subscales, attenuation of affective disorders such as anxiety and depression, and improvement in cognitive function. Against the background of the Neurotrophic peptide mixture administration, dynamic characteristics of simple motor reactions improve, and the threshold for the emergence of the H-reflex increases, indicating facilitation of supraspinal control processes over the activity of motor neurons. In general, the results show a strong tendency to improvement in different parameters.\nMitochondrial dysfunction before and after treatment with the Neurotrophic peptide mixture in patients with Parkinson’s disease\nThe conducted studies revealed significant differences in the structural organization of platelets in individuals with PD examined before the start of the standard treatment process. Electron microscopic and morphometric study of platelets in patients with PD revealed that platelets were mainly represented by senescent cells. They had a reduced number of dense δ granules, which was 5.1 ± 0.4 per one cell (in healthy individuals - 8.4 ± 0.5 units per one cell, p < 0.05). However, the number of transparent α-granules turned out to be increased from 5.3 ± 0.2 up to 10.3 ± 0.6 per cell. In addition, the latter often had large sizes (>100 nm) and contained a significant amount of protein debris (Fig. 11).\nThe single mitochondria that were found in the cells (2–4 per cell) were structurally quite preserved and had an average diameter of 0.48 ± 0.02 µm, not significantly different from those determined in the control group (0.41 ± 0.04 μm).\nSignificant abnormalities were found in the ultrastructure of mitochondria in platelets of patients. In addition to increased mitochondrial membrane permeability (outer and/or inner, crystas-forming), the formation of septated organelles was observed. The mechanisms and causes of their formation are still unexplored. It is assumed that septated mitochondria can accumulate Са2+26. Since it is known that the accumulation of Са2+ and the synthesis of ATP are alternative processes associated with the transfer of electrons, an increase in the Ca-accumulating ability of mitochondria may indicate a decrease in the synthesis of ATP in them. In addition, the appearance of septated mitochondria may indicate partial cell necrosis. These ultrastructural changes are evidence of the formation of mitochondrial dysfunction in platelets.\nThe use of the Neurotrophic peptide mixture in the complex therapy of patients with CP was accompaniedby reliable dynamics of the morphofunctional state of platelets (Table 1). The use of the Neurotrophic peptide mixture led to a significant (by 45%) increase in the number of δ-granules. No significant changes in the number of α granules were detected; only a tendency to its decrease was observed. An increase in the number of δ-granules indicates an increase in serotonin reserves. Probably, one of the possible mechanisms of the neuroprotective effect of the Neurotrophic peptide mixture is realized in this way. In addition, the Ca2+ -accumulating capacity in thrombocytes is relatively reduced, which should contribute to the improvement of ATP synthesis.\nDespite the small number of mitochondria in platelets, a slight increase in their number was observed - from 2 to 4 organelles before treatment to 5–7 organelles after treatment with the Neurotrophic peptide mixture, which, in combination with the absence of detection of septated organelles, may indicate a decrease in the severity of mitochondrial dysfunction.\nThus, treatment with the Neurotrophic peptide mixture improves the morphofunctional state of platelets in patients with PD, it increases serotonin reserves, reduces Ca2+ -accumulating capacity, reduces the manifestations of mitochondrial dysfunction, and exhibits mild antithrombotic properties.\nEffect of the Neurotrophic peptide mixture treatment on PARKIN gene expression in patients with Parkinson’s disease\nFor the whole group of tested individuals, taken together, no correlation of the expression of the measured genes was shown using the Spearman test. At the same time, when dividing by sex, multidirectional differences in gene expression before and after therapy with the drug Neurotrophic peptide mixture were found (Table 2).\nIn particular, the group of men was characterized by a decrease in the expression of the BDNF gene, and in women the expression increased after the use of the drug (Fig. 12).\nIt is important to note that when comparing the results of gene expression with cognitive parameters, an increase in BDNF in women (but not in men) was also accompanied by an increase in the MMSE score (Fig. 13).\nNo statistically significant difference was observed in expression of other three genes (DJ-1, PINK1, TNFalfa) before and after treatment with the Neurotrophic peptide mixture (Figs. 14–16).\nThe correlation analysis shows higher values between MMSE and PINK1 expression, FAB a and DJ1, FAB a and PINK1, MOCA a and DJ1, MOCA a and PINK1, MOCA and PINK1 for men (Table 3).\nFor women, the correlation analysis shows strong values of clinically central endpoints UPDRS III and UPDRS S with almost all studied genes expression rates (Table 4).\nEffect of the Neurotrophic peptide mixture treatment on the state of the prooxidant-antioxidant system in patients with Parkinson’s disease\nThe intensification of prooxidant processes in blood plasma and erythrocytes of patients with moderate PD stages was registered. It was established a significant rise (by 40%) in the content of the secondary products of lipid peroxidation in blood plasma as well as a marked increase (by 50%) of the H2O2 formation in erythrocytes of patients with PD compared to healthy participants (Tables 5 and 6). Our assessment of the antioxidant protective effectiveness was largely based on the changes in activity of antioxidant enzymes – SOD and catalase, which represent the first line of such defense, providing detoxification of reactive oxygen metabolites27. In patients with PD, we have found an increase in the activities of SOD and catalase by 21% and 69%, respectively in comparison with healthy subjects. In addition, we analyzed the glutathione cycle, which is one of the main intracellular mechanisms for maintaining redox homeostasis28. In our study, the concentration of GSH was significantly decreased (by 58%) in erythrocytes of PD patients as well as GPx activity though this change was not statistically significant (by 15%) compared to healthy controls. We found that CBL administration has moderate antioxidant effects in PD patients by decreasing the TBARS accumulation (by 17%) in blood plasma as well as H2O2 content in erythrocytes (by 12%) compared with the CBL-untreated patients. These effects of CBL were accompanied by a decrease in the activities of SOD (by 11%) and catalase (by 20%) in blood of patients with PD compared with the CBL-untreated ones. In CBL-treated patients, there was registered an increase in the GSH concentration (by 30%) and GPx activity (by 15%) in erythrocytes compared with PD patients though this change was not statistically significant compared to the CBL-untreated patients (Tables 5 and 6).\nUPDRS score change prediction using multiple machine learning models\nAmong the models, GBMs demonstrated the highest predictive accuracy, achieving an RMSE of 18.59, MSE of 345.82, and MAE of 15.39. The top-performing RF model exhibited slightly higher RMSE (21.99) and MAE (17.78), underscoring the superior capability of GBMs to capture complex patterns in the data.\nThe boxplot of feature importance across the top 10 models provided valuable insights into which factors most significantly influenced the prediction of UPDRS score changes. Below are the most impactful features, sorted by median scaled importance: Fig. 17\n- \n                    BDNF Change (Median Importance: 0.98): Changes in Brain-Derived Neurotrophic Factor (BDNF) were the most influential predictor of UPDRS score change. BDNF’s high importance suggests that neurotrophic support may play a central role in the Neurotrophic peptide mixture’s therapeutic effects, potentially enhancing neuroplasticity and recovery. \n- \n                    MOCA (Median Importance: 0.62): The Montreal Cognitive Assessment (MOCA), a measure of cognitive function, was ranked as a highly predictive feature, highlighting the relevance of cognitive improvement as a response indicator for the Neurotrophic peptide mixture therapy. \n- \n                    BDNF (after treatment) (Median Importance: 0.61): Post-treatment levels of BDNF were also significant, indicating that maintaining or boosting neurotrophic factors is associated with positive therapeutic outcomes in PD patients. \n- \n                    MMSE, S (Median Importance: 0.58): The Mini-Mental State Examination (MMSE) score also emerged as a key predictor, reinforcing the importance of cognitive assessments in evaluating the Neurotrophic peptide mixture’s efficiency. \n- \n                    PINK1 (after treatment) (Median Importance: 0.55): PINK1, a mitochondrial health marker, was another significant predictor, suggesting that the Neurotrophic peptide mixture’s potential benefits may extend to cellular and mitochondrial health. \nThe boxplot analysis revealed that features related to neurotrophic support (BDNF) and cognitive function (MOCA, MMSE) showed the most consistent importance across models, as evidenced by their high median importance values and narrow interquartile ranges. In contrast, features with wider variability, such as TBA-active products and Age, had broader interquartile ranges, indicating that their contribution to predictive accuracy may depend on model-specific interactions or particular patient subgroups.\nDiscussion\nThe findings of this exploratory study suggest that the Neurotrophic peptide mixture has a multifaceted impact on various aspects of PD pathology, particularly regarding platelet ultrastructure, gene expression, neurophysiological parameters, and oxidative stress markers. These findings align with existing literature, which underscores the neuroprotective and neurorestorative properties of the Neurotrophic peptide mixture.\nImprovement in motor functions, particularly in the Motor tempo test, was more pronounced in patients with predominantly right-sided symptoms of PD. This asymmetry in response could be related to the lateralization of the pathological process. Additionally, the elevation in the threshold for triggering the H-reflex suggests modifications in spinal cord excitability, which could translate to improvements in motor function. The study reported statistically significant enhancements in cognitive abilities, as measured by the MMSE, FAB, and MoCA scales. Additionally, there was a notable reduction in depression scores (BDI) and reactive anxiety (Spielberger-Hanin scale), although personal anxiety levels remained unchanged. These findings highlight the potential of the Neurotrophic peptide mixture in ameliorating both motor and non-motor symptoms of PD. The improvements in motor function, cognitive abilities, and emotional well-being observed in this study suggest that the Neurotrophic peptide mixture has a broad therapeutic potential.\nThe study also observed significant neurophysiological improvements post-Neurotrophic peptide mixture treatment. The latency period of the P3 component, a marker of cognitive processing, was significantly reduced. This improvement in P3 latency indicates enhanced cortical function and suggests that the Neurotrophic peptide mixture may have beneficial effects on cognitive processing speed and efficiency in PD patients. These neurophysiological benefits align with the observed cognitive improvements, further supporting the potential of the Neurotrophic peptide mixture in mitigating cognitive decline in PD.\nIn PD patients, platelets exhibited several signs of senescence, including a reduction in dense δ-granules and an increase in α-granules29. The number of δ-granules per cell was significantly lower in PD patients (5.1 ± 0.4) compared to healthy individuals (8.4 ± 0.5). The administration of the Neurotrophic peptide mixture significantly increased the number of δ-granules in platelets by 45%, indicating enhanced serotonin reserves, which could contribute to the neuroprotective effects observed clinically. This increase in δ-granules likely reflects a normalization of platelet storage function, which is crucial given the role of serotonin in mood and neuroprotection. Despite no significant change in the number of α-granules, the trend towards a decrease might indicate an alteration in platelet activation status.\nAdditionally, the Neurotrophic peptide mixture reduced the calcium-accumulating capacity of platelets, potentially improving ATP synthesis and overall platelet function. This reduction in calcium accumulation is crucial as it indicates a shift away from the pathological state observed in untreated PD platelets. The observed improvements in platelet ultrastructure and function may translate to better overall cellular health, reflecting the systemic benefits of the Neurotrophic peptide mixture treatment beyond its direct neural effects. The enhancement in serotonin reserves and mitochondrial function underscores the multifaceted role of the Neurotrophic peptide mixture in addressing both neurological and peripheral manifestations of PD.\nThe Neurotrophic peptide mixture treatment appears to modulate BDNF gene expression in a sex-specific manner in patients with Parkinson’s disease, with women showing an increase and men showing a decrease in expression levels. This modulation is accompanied by improved cognitive function in women, highlighting the potential of the Neurotrophic peptide mixture as a therapeutic option that may benefit female PD patients more significantly. The stable expression of DJ-1, PINK1, and TNFα genes suggests that the primary neuroprotective mechanisms of the Neurotrophic peptide mixture in PD may involve pathways other than those mediated by these genes. These findings emphasize the need for further research to explore the underlying mechanisms of sex differences in PD treatment responses and to optimize therapeutic strategies accordingly.\nThe excessive production and release of ROS in PD occurs as a result of the combined action of mitochondrial dysfunction, dopamine autooxidation, and neuroinflammation30. In present study, the Neurotrophic peptide mixture demonstrated a moderate antioxidant effect, reducing the levels of lipid peroxidation products and H2O2 in erythrocytes. This reduction was accompanied by a decrease in SOD and catalase activities, and an increase in GSH concentration and GPx activity. These changes indicate a restoration of redox balance and a reduction in oxidative stress, which are crucial in the pathology of PD. One of the most promising effects of CBL on OS developing in blood and tissues in PD could be its ability to decrease intracellular α-synuclein aggregation in neurons, mimicking the action of classical NTFs and growth factors as well as of NTFs belonged to a novel family CDNF/MANF31. The ability of the Neurotrophic peptide mixture to modulate the antioxidant system suggests its potential role in protecting neuronal cells from oxidative damage and subsequent apoptosis.\nThe findings underscore the central role of cognitive and neurotrophic markers in predicting UPDRS score changes as an indicator of the Neurotrophic peptide mixture efficacy. The high importance of BDNF aligns with previous research on neuroprotective therapies, where neurotrophic support has been linked to improved neuronal survival and plasticity. The inclusion of mitochondrial markers, such as PINK1, suggests that the Neurotrophic peptide mixture may exert beneficial effects at the cellular level, addressing mitochondrial dysfunction—a known factor in PD pathology.\nThe boxplot analysis provided further clarity on the stability and reliability of each feature’s importance. Features with high median importance and narrow interquartile ranges, like BDNF and MOCA, indicate consistent predictive power across models, making them robust indicators of therapeutic response. The broader variability observed in other features, such as oxidative stress markers and demographic factors, suggests that while they may contribute to specific aspects of disease progression, they are less central to predicting therapeutic efficacy. The superior performance of GBMs in capturing non-linear relationships reinforces the value of using advanced machine learning techniques to analyze complex clinical datasets. These findings support the potential of the Neurotrophic peptide mixture as a multifaceted therapeutic option, with the capacity to improve cognitive and neurotrophic parameters in PD patients. The results of the present study indicate an antioxidant action of the Neurotrophic peptide mixture in patients with moderate stages of Parkinson’s disease. This action is mainly based on the decrease of the prooxidant events in peripheral blood of PD patients as well as on the prevention of thiol redox defects and reducing an imbalance in oxidant–antioxidant status.\nOverall, statistically significant improvements and consistent correlations were observed in the key clinical endpoints (UPDRS Part III and UPDRS S), while secondary and exploratory outcomes demonstrated patterns suggestive of mechanistic links that warrant confirmation in larger studies.\nInterpretation of its results is limited by several factors, including the small sample size, lack of a control group receiving placebo in addition to standard therapy, and a relatively short duration of the observation period. Since the study is preliminary, exploratory in nature and aimed at collecting primary data on the response to the drug, only one group of patients was formed, which allows identifying primary trends and assessing the feasibility of future larger studies. Future research should focus on larger cohorts and longer treatment durations to confirm these findings and to explore the long-term effects of neurotrophic peptide mixture on gene expression and cognitive function in PD patients. Particular attention should be paid to investigating sex differences in response to the drug in future studies.\nThe results of this exploratory study provide a preliminary rationale for investigating the Neurotrophic peptide mixture as an adjunct therapy in Parkinson’s disease. During the treatment period, changes were observed in platelet ultrastructure, gene expression, neurophysiological parameters, and oxidative stress markers. These findings, along with shifts in cognitive and neurotrophic markers, may suggest a multidimensional biological response.\nHowever, these observations must be interpreted with extreme caution. Given the limited sample size and the absence of a control group, no causal inferences can be drawn. Therefore, larger, placebo-controlled clinical trials are warranted to determine if these initial observations can be substantiated, to investigate potential sex-specific and long-term effects, and to explore biomarkers that may help elucidate the mechanisms of the Neurotrophic peptide mixture in PD.\nMethods\nPatient information\nThe exploratory study included 17 patients with Parkinson’s disease with an established diagnosis according to the inclusion-exclusion criteria of the British Brain Bank32, aged 45–74 years with disease stage 2.0–3.0 (2.9 ± 0.3, according to Hoehn and Yahr scale). Written informed consent was obtained from all participants.\nThe neurotrophic peptide mixture was prescribed against the background of basic stable doses of antiparkinsonian therapy (Levodopa-containing drugs), which were used during 28–30 days before starting the course of neurotrophic peptide mixture and during the entire study period. No dose adjustments were allowed during the study, and medication dose was not included as a covariate in the statistical analysis. The median daily Levodopa dose was about 500 mg, it was individually optimized depending on disease stage and the patient’s condition. The Neurotrophic peptide mixture dosage regimen was 20 ml (215.2 mg/ml) intravenously, once a day for 10 days. Infusions were carried out in the first half of the day.\nEthics statement and study design\nThe Ethics Committee of the Clinical Department of the Institute of Gerontology named after D.F. Chebotariov of the National Academy of Medical Sciences of Ukraine approved the protocol of current research (Protocol No. 4, 09/06/2022). Written informed consent was obtained from all participants prior to enrollment.\nThe study was conducted in accordance with the principles of the Declaration of Helsinki (2000) and the European Council Convention on Human Rights and Biomedicine (2005), and relevant national standards for the participation in the research were taken into account.\nClinical assessment\nCharacterization of the stages of PD was carried out using the Hoehn and Yahr scale. To assess the dynamics of motor disorders, the standardized international Unified Parkinson’s Disease Rating Scale (UPDRS) was used. For a comprehensive assessment of the state of non-motor functions, the following clinical scales were used:\n- \n                    Mini-Mental State Examination (MMSE); \n- \n                    Test battery for assessing frontal dysfunction (Frontal Assessment Battery, FAB); \n- \n                    Montreal Cognitive Assessment (MoCA); \n- \n                    Beck Depression Inventory (BDI); \n- \n                    Spielberger-Hanin Anxiety Scale. \nSensorimotor reaction and motor tempo\nMotor tempo was determined using a computer program as the time between consecutive presses of the same finger on two keys spaced 20 cm apart on the keyboard. The patient was instructed to press the keys at the fastest possible tempo. The duration of each test was 20 s. Testing was conducted for each hand separately, and then the data were averaged.\nThe latent period of simple sensorimotor reaction was determined as the interval between the moment of presentation of a visual signal on the computer monitor and the moment the patient pressed the key in response to the signal. Geometric shapes (circle, square, oval) of red and white colors were used as signals, with 20 signals presented at intervals of 2–4 s. Testing was conducted for each hand separately, and the data were averaged.\nElectroneuromyography\nThe technique of registering the H-reflex, as a reflex response of muscles to the stimulation of sensory fibers of a mixed nerve followed by monosynaptic activation of spinal motoneurons and nerve motor fibers, was performed using disposable surface recording electrodes placed in the projection of the posterior group of muscles of the calf, with stimulation of the tibial nerve in the popliteal fossa. The dynamics of the amplitudes of H- and M-responses (mV) with increasing strength of nerve electrical stimulation (mA), the threshold for the occurrence of the H-reflex (mA), the latency period of the H-reflex (ms), and the ratio of the maximum amplitude of H max/M max (%) were evaluated.\nEvoked potentials P300\nEEG was recorded monopolarly from 21 leads (Fp1, Fp2, Fpz, F3, F4, Fz, C3, C4, Cz, P3, P4, Pz, T3, T4, F7, F8, T5, T6, O1, O2, Oz). The combined ear electrode was used as a reference electrode. The interelectrode resistance was less than 5 kOhm, the signal sampling frequency was 500 Hz, and the high- and low-pass filters were 0.3 Hz and 35 Hz, respectively. Endogenous cognitive evoked potentials N2 and P3 were recorded according to the oddball paradigm using sound stimuli with a frequency of 1000 and 2000 Hz and a duration of 10 ms.\nPlatelet and mitochondrial structure with Electron microscopy\nBlood plasma enriched with platelets containing leukocytes was obtained by centrifugation of whole blood obtained from the ulnar vein of the subjects at room temperature for 15 min at 120 g in a T-30 centrifuge. The plasma was carefully separated from the sediment and centrifuged at 2000 g for 20 min using a Vortecs Combispin FVL-2400N minicentrifuge33.\nElectron microscopic studies of platelets and mitochondria were carried out according to the generally accepted method with double fixation with glutaraldehyde and OSO4, dehydration with alcohols of increasing concentrations and subsequent embedding in Epon. Ultrathin sections with a thickness of 40–60 nm, contrasted with uranyl acetate and lead citrate, were examined using an electron microscope PEM-125K (Ukraine).\nMorphometric calculations of platelets’ structure were performed using the Image Tool (USA) on 130–150 fields for each group of subjects. The number of optically dense small σ-granules and optically transparent large α-granules was determined.\nPBMC isolation and quantitative PCR\nBlood collection, PBMC isolation, RNA extraction and cDNA synthesis were performed by standard methods. The collection of blood for further cell isolation and qPCR was made on 0 and 11th day (after a 10-day course of neurotrophic peptide mixture treatment). Briefly, patients’ blood was collected in 4 ml vacutainers with EDTA. Density gradient centrifugation of diluted whole blood layered over a density gradient medium was used for PBMC isolation. PBMC were gradually frozen at a temperature of −20 °C, and then transferred to the temperature of −80 °C for long-term storage. RNA was isolated from cells using the Direct-zol RNA Miniprep Kit (Zymo Research, USA) following the manufacturer’s protocols. The isolated RNA was treated with DNase to remove DNA residues that could give a false positive signal for the expression of target genes. After this, reverse transcription was carried out using a commercial kit ZymoScript RT PreMix (Zymo Research, USA), guided by the manufacturer’s recommendations.\nSamples of this cDNA were used for further qPCR-RT (quantitative real-time polymerase chain reaction), which was used to measure the expression of the DJ-1, PINK1, BDNF and TNFα genes. To identify a suitable internal control for RT-qPCR data normalization, we empirically tested the expression stability of three candidate reference genes (rps18, gusb, and beta-actin). The sequences of all used primers are listed in the Supplementary Information. Analysis of the raw Cqt values revealed that both beta-actin and gusb were the most stably expressed genes, showing minimal variation and a strong positive correlation with each other. In contrast, rps18 expression was found to be significantly more variable across our samples and was therefore excluded. Although both beta-actin and gusb were suitable candidates, we selected beta-actin for normalization due to its slightly higher expression stability.\nThe sequences of all used primers are listed in the Supplementary Information. The PCR reaction mix was prepared using a commercial kit of reagents for RT-PCR (HOT FIREPol® EvaGreen® qPCR Mix Plus, Solis BioDyne, Estonia) - reverse transcription and polymerase chain reaction with the addition of betaine (Sigma, USA) at a final concentration of 1 M. To determine differences in mRNA expression of Parkin genes, Repeated Measures Anova was used.\nAccumulation curves were generated using Opticon Monitor 3 software. After amplification was completed, raw gene product amplification data were used to calculate Cy0.\nMeasurement of oxidative stress biomarkers\nSecondary products of lipid peroxidation, the activity of antioxidant enzymes superoxide dismutase (SOD) and catalase (CAT) were measured in blood plasma. In red blood cells – the content of hydrogen peroxide (H2O2); the state of the glutathione system was assessed by the content of reduced glutathione (GSH) and the activity of the enzyme glutathione peroxidase (GP).\nFor erythrocyte isolation, samples were centrifuged at 8000 rpm for 15 min to separate plasma and red blood cells. Using centrifugation, erythrocytes were washed twice of plasma with cold sterile physiological sodium chloride solution. Erythrocyte hemolysate was obtained by adding distilled water to washed erythrocytes in a 1:1 ratio. The hemoglobin (Hb) content was also measured (Hemoglobin Assay kit MAK115, Sigma-Aldrich, St. Louis, MO, USA).\nDetermination of thiobarbituric acid reactive substances (TBARS) levels: TBARS were isolated by boiling plasma with the thiobarbituric acid reagent (0.5% 2-thiobarbituric acid/10% trichloroacetic acid/0.63 mM hydrochloric acid) and measuring the absorbance at 532 nm. The results are expressed as µM/ml using a molar extinction coefficient 1.56 * 105 mm−1 cm−1.\nCatalase activity measurement: The method is based on the ability of molybdenum salts to form a colored complex with hydrogen peroxide, which was determined at a wavelength of 410 nm. Enzyme activity (μM/min/L) was calculated using a molar extinction coefficient of 22.2 * 103 mm−1cm−1.\nSuperoxide dismutase (SOD) activity was determined spectrophotometrically. The method is based on the ability of SOD to inhibit the autoxidation of adrenaline at pH 10.2. The results were expressed as specific activity of the enzyme in units per ml of plasma. One unit of SOD activity is defined as the amount of protein causing 50% inhibition of the conversion rate of adrenaline to adrenochrome under specified conditions.\nThe activity of selenium-dependent glutathione peroxidase (GPx) was determined by recording the rate of NADPH oxidation in the presence of reduced glutathione at a wavelength of 340 nm. Enzyme activity was expressed in nM NADPH/min/ml.\nThe amount of reduced glutathione (μM/ml) was determined using Elman’s reagent (5.5’-dithio-bis-nitrobenzoic acid in absolute ethanol). The change in extinction was recorded at a wavelength of 412 nm every minute for 5 minutes.\nHydrogen peroxide content was studied using the FOX method, which is based on the peroxide-mediated oxidation of Fe2 + , followed by the reaction of Fe3+ with xylenol orange. Absorbance of the Fe3 + -xylenol orange complex was determined at a wavelength of 560 nm. The amount of H2O2 (μM/mgHb) was calculated using the H2O2 standard curve for each independent experiment.\nMachine learning prediction of UPDRS score change\nTo model UPDRS score changes, H2O’s AutoML framework was employed34, allowing for the training, tuning, and evaluation of multiple Machine Learning models, including Gradient Boosting Machines (GBMs) and Random Forests (RFs). AutoML facilitated hyperparameter optimization, ensuring robust and accurate predictions.\n- \n                    Gradient Boosting Machines (GBMs): GBMs iteratively build a series of decision trees, where each subsequent tree focuses on correcting errors made by its predecessors. This sequential approach enables GBMs to capture intricate, non-linear relationships within the data, making them particularly suitable for analyzing how clinical and biochemical factors impact UPDRS score changes and, consequently, therapeutic response35. \n- \n                    Random Forests (RFs): RFs are ensemble models that combine multiple decision trees, each trained on a random subset of the data. Known for their robustness and reduced susceptibility to overfitting, RFs provide valuable insights into feature importance by averaging across numerous trees36. This model serves as a stable baseline to complement the more flexible GBMs. \nThe models were evaluated using the following metrics to quantify prediction accuracy:\n- \n                    Root Mean Squared Error (RMSE): Captures the average deviation between predicted and actual values. \n- \n                    Mean Squared Error (MSE): Measures the variance of predictions relative to actual outcomes. \n- \n                    Mean Absolute Error (MAE): Reflects the average absolute deviation between predicted and observed values. \nThese metrics allowed for a thorough assessment of each model’s ability to predict UPDRS score changes, highlighting the Neurotrophic peptide mixture’s potential impact on disease progression.\nFeature importance was evaluated using the top 10-performing models from a set of 50 trained with H2O AutoML. Each feature’s importance was measured by its “scaled importance,” which represents its relative contribution to prediction accuracy. The models were trained using two-fold cross-validation, and feature importance was extracted during the validation stage.\nData from the top 10 models were aggregated, and features were sorted by their median feature importance. It is worth noting that while building a generalized model on such a small dataset is not feasible, the features identified with the highest median importance could be considered potential markers in future studies. The results were visualized using a boxplot (Fig. 16) to highlight the median importance and variability of each feature.\nStatistical analysis\nHypotheses about the effects of the drug were tested using the Wilcoxon (for related samples) and Mann-Whitney (for independent groups) tests. The relationship between the investigated indicators was determined in the form of Spearman’s rank correlation coefficients.\nStatistical processing of gene expression data was performed using Statistica 8.0 software (StatSoft Inc., USA). For statistical analysis, Spearman’s non-parametric test was used to investigate the correlation between parameters. The difference in expression of the studied genes before and after treatment was determined using the Wilcoxon t-test (Wilcoxon signed-rank test).\nThe results of the study of the state of the prooxidant-antioxidant system were processed statistically using the Origin program. The significance of differences between comparison groups was determined by the method of analysis of variance (ANOVA) followed by the Bonferroni test (post-hoc test).\nOutcomes and hierarchy\nFor this exploratory study, outcomes were grouped into three hierarchical levels.\n- \n                    Clinically central endpoints were prespecified as the UPDRS III (motor score) and the UPDRS Total score, as the primary efficacy measures, as these are the most widely accepted, clinically relevant, and objective indicators of motor impairment and overall disease severity in Parkinson’s disease. \n- \n                    Secondary clinical endpoints included UPDRS I (non-motor experiences of daily living) and UPDRS II (motor experiences of daily living), and validated scales of non-motor function: MMSE, MoCA, FAB (cognitive measures), and BDI and Spielberger–Hanin Anxiety Scale (mood and affective measures). These outcomes were considered supportive of the clinically central endpoints. \n- \n                    Exploratory outcomes included additional non-motor measures, neurophysiological parameters (P300 latency and amplitude), and molecular/structural biomarkers (gene expression, platelet ultrastructure, mitochondrial morphology, and oxidative stress markers). These outcomes were included to generate mechanistic insights into the potential modes of action of the Neurotrophic peptide mixture. \nFormal hypothesis testing was restricted to the clinically central endpoints. Secondary and exploratory outcomes are presented with descriptive statistics and effect sizes. To minimize Type I error, no multiple-testing correction was applied outside of the clinically central endpoints.\nCorrelation analyses were performed using Spearman’s rank correlation coefficient (ρ). For correlations involving clinically central endpoints, p-values are reported. Correlations among secondary and exploratory outcomes are reported without formal inferential testing, to avoid Type I error inflation.\nData availability\nThe datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.\nReferences\n- Alexoudi, A., Alexoudi, I. & Gatzonis, S. Parkinson’s disease pathogenesis, evolution and alternative pathways: A review. Rev. Neurol. (Paris) 174, 699–704 (2018). \n- Chung, S. J. et al. The Pattern of Striatal Dopamine Depletion as a Prognostic Marker in De Novo Parkinson Disease. Clin. Nucl. Med 201843, 787–792 (2018). \n- Moon, H. E. & Paek, S. H. Mitochondrial Dysfunction in Parkinson’s Disease. Exp. Neurobiol. 24, 103–116 (2015). \n- Mansur, A. et al. Characterization of 3 PET tracers for quantification of mitochondrial andsynaptic function in healthy human brain:18F-BCPP-EF,11C-SA-4503, and11C-UCB-J. J. Nucl. Med 61, 96–103 (2020). \n- Simonenkov, A. P. & Kluyev, V. M. Syndrome serotonin deficiency. M.: BINOM, 96-97 (2013). \n- Mauricio, R. et al. Tackling gaps in developing life-changing treatments for dementia. Alzheimers Dement 5, 241–253 (2019). \n- Gonchar, O. O., Klymenko, O. O., Drevytska, T. I., Bratus, L. V. & Mankovska, I. M. Oxidative stress in rat heart mitochondria under a rotenone model of Parkinson’ disease: a corrective effect of capicor treatment. Ukr. Biochem J. 93, 21–30 (2021). \n- Schneider-Garces, N. J. et al. Span, CRUNCH, and beyond: working memory capacity and the aging brain. J. Cogn. Neurosci. 22, 655–669 (2010). \n- Fiani, B. et al. Cerebrolysin for stroke, neurodegeneration, and traumatic brain injury: review of the literature and outcomes. Neurol. Sci. 42, 1345–1353 (2021). \n- Brainin, M. Cerebrolysin: a multi-target drug for recovery after stroke. Expert Rev. Neurother. 18, 681–687 (2018). \n- Requeejoe, C. et al. Nanodelivery of Cerebrolysin and rearing enriched environment induce neuroprotective effects in a preclinical rat model of Parkinson’s disease. Mol Neurobiol. Springer Science+Business Media, LLC (2017). \n- Neveen, A. et al. A promising therapeutic potential of Cerebrolysin in the 6-OHDA rat model of Parkinson's disease. Life Sci 155, 174–179 (2016). \n- Luchanina, E., Karaban, I. & Berezetskaya, N. Diagnosis of Parkinsonas disease by electrophysiological methods. Diagn. Rehabil. Parkinson. Dis. 2, 27–58 (2011). \n- Eder, P., Reinprecht, I., Schreiner, E., Skofitsch, G. & Windisch, M. Increased density of glutamate receptor subunit 1 due to Cerebrolysin treatment: an immunohistochemical study on aged rats. Histochem J. 33, 605–612 (2001). \n- Stepanichev, M. et al. Effects of cerebrolysin on nerve growth factor system in the aging rat brain. Restor. Neurol. Neurosci. 35, 571–581 (2017). \n- Veinbergs, I., Mante, M., Mallory, M. & Masliah, E. Neurotrophic effects of Cerebrolysin in animal models of excitotoxicity. J. Neural Transm. Suppl. 59, 273–280 (2000). \n- Alvarez, X. A. et al. Neuroimmunotrophic effects of cerebrolysin in an animal model of hippocampal degeneration induced by β-amyloid. In: Korczyn, A. D., ed. International Congress on Vascular Dementia 2017. Bologna, Italy: Monduzzi Editore, 233–237 (1999). \n- Gavrilova, S. I. & Alvarez, A. Cerebrolysin in the therapy of mild cognitive impairment and dementia due to Alzheimeras disease: 30 years of clinical use. Med Res Rev. 41, 1–29 (2020). \n- DeBoer, S. R. et al. Enhanced spontaneous motor recovery after stroke in mice treated with cerebrolysin. Neurorehabil Neural Repair 35, 525–533 (2021). \n- Muresanu, D. F. et al. Efficacy and safety of cerebrolysin in neurorecovery after moderate-severe traumatic brain injury: results from the CAPTAIN II trial. Neurol. Sci. 41, 1171–1181 (2020). \n- Lu, W. et al. Cerebrolysin alleviates early brain injury after traumatic brain injury by inhibiting neuroinflammation and apoptosis via TLR signaling pathway. Acta Cir. Bras. 37, e370605 (2022). \n- Chen, C. C., Wei, S. T., Tsaia, S. C., Chen, X. X. & Cho, D. Y. Cerebrolysin enhances cognitive recovery of mild traumatic brain injury patients: double-blind, placebo-controlled, randomized study. Br. J. Neurosurg. 27, 803–807 (2013). \n- Cui, S. Cerebrolysin for vascular dementia. Cochrane Database Syst Rev. 11, 19–20 (2019). \n- Rochenstein, E. et al. Neuropeptide treatment with cerebrolysin enhances the survival of grafted neural stem cells in a synuclein transgenic model of Parkinson’s disease. J. Exp. Neurosci. 9, 131–140 (2015). \n- Hünerli, D. et al. Mild cognitive impairment in Parkinson’s disease is associated with decreased P300 amplitude and reduced putamen volume. Clin. Neurophysiol. 130, 1208–1217 (2019). \n- Alvarez, X. A., Lombardi, V. R., Fernández-Novoa, L., García, M. & Sampedro, C. Cerebrolysin reduces microglial activation in vivo and in vitro: a potential mechanism of neuroprotection. J. Neural Transm. 59, 281–292 (2000). \n- Kulinskiĭ, V. I. & Kolesnichenko, L. S. Glutathione system. I. Synthesis, transport, glutathione transferases, glutathione peroxidases. Biomed. Khim 55, 255–277 (2009). \n- Zenkov, N. K., Lankin, V. Z. & Menshikova, E. B. Oxidative stress: The biochemical and pathophysiological aspects. M.: MAIK. Science / Interperiodica 4, 352–361 (2001). \n- Park, S. M. et al. Evidence that alpha-synuclein functions as a negative regulator of Ca(++)-dependent alpha-granule release from human platelets. Blood 100, 2506–2514 (2002). \n- Luo, Y., Hoffer, A., Hoffer, B. & Qi, X. Mitochondria: A therapeutic target for Parkinson’s disease?. Int J. Mol. Sci. 16, 20704–20730 (2015). \n- Voutilainen, M. H., Arumae, U., Airavaara, M. & Saarma, M. Therapeutic potential of the endoplasmic reticulum located and secreted CDNF/MANF family of neurotrophic factors in Parkinson’s disease. FEBS Lett. 589, 3739–3748 (2015). \n- Hughes, A. J., Daniel, S. E., Kilford, L. & Lees, A. J. Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: A clinico-pathological study of 100 cases. J. Neurol. Neurosurg. Psychiatry 55, 181–184 (1992). \n- Filippova, O. I., Koloskov, A. V. & Stolitsa, A. A. Methods for studying the functional activity of platelets (literature review). Transfusiology 13, 493–514 (2012). \n- LeDell, E. & Posse, C. H2O AutoML: Scalable Automatic Machine Learning. J. Open Source Softw. 5, 1–10 (2020). \n- Chen, J. & Guestrin, C. XGBoost: A Scalable Tree Boosting System. Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery Data Mining, 785–794 (2016). \n- Liaw, A. & Wiener, M. Classification and regression by random Forest. R. N. 2, 18–22 (2002). \nAcknowledgements\nThis work was supported by the National Academy of Medical Sciences, Ukraine (Grant # 0123U100771).\nAuthor information\nAuthors and Affiliations\nContributions\nD.K. took part in writing of the main text of manuscript, coordinated the work of scientific laboratories in their experiments and results, contributed to the conceptualization of the project, I.K. was the lead scientist of part connected to clinical assessment, data analysis, project conception and organization, N.K., N.M. and S.K. performed and analyzed electroencephalograms and elecroneuromyograms of patients, collected patient data regarding sensorimotor reactions and motor tempo, K.R. participated in the preparation of the literature review, implemented the methodology of morphological studies, processing the data of the section about the morphological parameters. O.G. examined markers of oxidative stress and made substantial contributions to the analysis of data as well as participated in a critical review of the manuscript, I.M. helped with the preparation of the manuscript and analysis of data. O.Z. wrote an abstract, implemented the main methodology for gene expression measurement, analysis and interpretation of data, O.B. and K.M. were involved to gene expression experiments and analysis of qPCR-RT data, as well as collection of biological material, S.S. provided the results of deep analysis of obtained data using multiple machine learning models, performed feature importance analysis, T.P. contributed to the administration of project, wrote and edited the main manuscript text. All authors reviewed the manuscript.\nCorresponding author\nEthics declarations\nCompeting interests\nThe authors declare no competing interests.\nAdditional information\nPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.\nRights and permissions\nOpen Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.\nAbout this article\nCite this article\nKrasnienkov, D., Karaban, I., Karasevych, N. et al. Evaluation of the neurotrophic peptide mixture in pathogenetic therapy of patients with Parkinson’s disease. npj Parkinsons Dis. 12, 55 (2026). https://doi.org/10.1038/s41531-026-01270-6\n- Received: \n- Accepted: \n- Published: \n- Version of record: \n- DOI: https://doi.org/10.1038/s41531-026-01270-6","excerpt":"Abstract\nThis exploratory, single-group, open-label study investigated 17 patients with Parkinson’s disease (PD) using a pre-post design. Motor and non-motor outcomes were assessed through clinical scales, biochemical and genetic analyses, and machine learning models (Gradient Boosting Machines, Random Forests).","extraction":{"state":"ok","confidence":0.9,"error":null,"explanation":"High confidence: full text extraction produced 58549 characters.","diagnostics_url":"/api/diagnose?url=https%3A//www.nature.com/articles/s41531-026-01270-6","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 58549 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":58549,"summary_length":313,"usable_text_length":58549,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":58549,"summary_length":313}}},"display_formats":["compact","card","full","digest_section","json"]},"daily_stack_record":{"title":"Evaluation of the neurotrophic peptide mixture in pathogenetic therapy of patients with Parkinson’s disease - Nature","url":"https://www.nature.com/articles/s41531-026-01270-6","summary":"Abstract\nThis exploratory, single-group, open-label study investigated 17 patients with Parkinson’s disease (PD) using a pre-post design. Motor and non-motor outcomes were assessed through clinical scales, biochemical and genetic analyses, and machine learning models (Gradient Boosting Machines, Random Forests).","source":"Nature","date":"2026-01-23T08:00:00+00:00","content":"Abstract\nThis exploratory, single-group, open-label study investigated 17 patients with Parkinson’s disease (PD) using a pre-post design. Motor and non-motor outcomes were assessed through clinical scales, biochemical and genetic analyses, and machine learning models (Gradient Boosting Machines, Random Forests). After treatment with a neurotrophic peptide mixture, improvements were observed in daily activity (16%), cognition (11%), depression (10% reduction), and reactive anxiety (23% reduction). Biological changes included a 45% increase in platelet δ-granules, higher mitochondrial counts, elevated gene expression (notably BDNF in women, p = 0.046), and modulation of oxidative stress markers (17% reduction in TBARS, 30% increase in GSH). Machine learning identified BDNF and PINK1 expression, along with MOCA and MMSE scores, as key predictors of UPDRS improvement. These findings suggest that neurotrophic peptide therapy may influence clinical, structural, and molecular domains in PD. Larger, controlled trials are warranted to confirm therapeutic potential and clarify associations with cognitive and neurotrophic parameters.\nSimilar content being viewed by others\nIntroduction\nParkinson’s disease is a neurodegenerative condition typically observed in older individuals, characterized by a gradual onset and progressive nature. The initial manifestation often includes tremors, followed by symptoms such as slowed movement (bradykinesia) and muscle stiffness (rigidity)1. The disease is linked to the degeneration of dopaminergic neurons in the substantia nigra, the presence of alpha-synuclein aggregates within neuronal cells (Lewy bodies and Lewy neurites), and subsequent pathology in regions like the basal forebrain, amygdala, and medial temporal lobe structures, progressing to cortical areas in advanced stages2.\nMitochondrial abnormalities and dysfunction are recognized as key contributors to Parkinson’s disease pathogenesis in a subset of patients3. Some data was obtained on the availability correlation of the functional activity of mitochondria in lymphocytes, which displays multisystem mitochondrial failure and indicates of energy dysfunction in general4. The spectrum of mitochondrial dysfunctions in PD includes dysfunction of the respiratory chain (inhibition of the NADH-oxidase oxidation pathway), hyperproduction of ATP, disruption of Ca-regulatory systems, opening of the mitochondrial pore, perturbations of mitochondrial dynamics, and dysregulation of mitochondrial clearance5.\nAlthough oxidative stress plays a significant role in the pathogenesis of PD, therapy based on the use of exogenous antioxidants has not been shown to be highly effective. An alternative strategy that can counteract the damaging effects of free radicals and restore cellular redox balance is the activation of endogenous antioxidant enzymes present in the central neural system, such as superoxide dismutase, catalase, and glutathione peroxidase. The transcription of these cytoprotective proteins is regulated by the transcription factor Nrf2, which plays a leading role in the regulation of cellular redox status and is the main regulator of the cellular response to oxidative stress through the induction of antioxidant and detoxifying enzymes and proteins6. This is consistent with our previous data, which showed that lipid peroxidation was significantly increased while reduced glutathione (GSH) and glutathione-dependent antioxidant enzymes activity markedly decreased in blood of patients with PD as well as in rat brain under modeling of PD7. Various activators of the Nrf2/ARE signaling pathway, including peptides, have shown their effectiveness in both in vitro and in vivo models of neurological disorders8.\nThe Neurotrophic peptide mixture is peptidergic mixture with neurotrophic-like properties, made from a mixture of low molecular weight, porcine-derived peptides and free amino acids. The pharmaceutical preparation includes the peptide fragments: nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), ciliary neurotrophic factor (CNTF), enkephalins, orexin, and P219.\nThe use of the Neurotrophic peptide mixture significantly reduces neurological deficits, improves cognitive abilities, and reduces the volume of the infarction zone during stroke10,11. The Neurotrophic peptide mixture has an organ-specific multimodal effect on the brain, i.e., it provides metabolic regulation, neuroprotection, functional neuromodulation, and neurotrophic activity12. The Neurotrophic peptide mixture protects neurons from the damaging effects of lactic acidosis, prevents the formation of free radicals13, increases survival and prevents neuronal death under conditions of hypoxia and ischemia, and reduces the damaging neurotoxic effect of glutamate14.\nOver the past decade, compelling evidence has emerged neuroprotective properties of the Neurotrophic peptide mixture, explaining its multi-level impact on both the pathological cascade of damage and on subsequent recovery processes15. The ability of the Neurotrophic peptide mixture to influence the processes of neuroprotection and neuroplasticity was proven in a series of studies15,16,17 on a model of Alzheimer’s disease in transgenic mice for the amyloid precursor protein (mThy-1-hAPP751 tg). Using a model of Alzheimer’s disease, it was shown that the Neurotrophic peptide mixture, through the regulation of the synthesis and transport of amyloid precursor protein and the deposition of β-amyloid in synapses, has a synaptotrophic effect, which plays a primary role in the processes of neuroplasticity, which are extremely important in the acquisition of new skills and the maintenance of existing ones18.\nIncreasing studies have demonstrated that the Neurotrophic peptide mixture administration can promote recovery of motor function, improve early brain injury, and decrease hippocampal neuronal death19. In recent clinical studies, the Neurotrophic peptide mixture improved overall outcomes after moderate to severe traumatic brain injury and was also safe and better for early rehabilitation patients after ischemic stroke20. Lu et al. reported that the Neurotrophic peptide mixture can improve neurological dysfunction and brain damage, relieve neuroinflammation and prevent apoptosis after brain injury21.\nBased on the multifaceted mechanism of action of the Neurotrophic peptide mixture, including its effect on the processes of neurogenesis, neuroplasticity and neuroprotection, the drug currently has three main indications for use - stroke, dementia and traumatic brain injury10,22,23.\nLuchanina et al. demonstrated for the first time the effect of Neurotrophic peptide mixture on motor activity and muscle tone in patients with PD treated with levodopa13. It was established that the Neurotrophic peptide mixture improved the engraftment of a DOPA transplant from stem cells in the substantia nigra24 and reduced the severity of symptoms in model animals12.\nThe current exploratory study dataset comprised clinical, cognitive, biochemical, and demographic information from PD patients undergoing the Neurotrophic peptide mixture therapy. Key features included cognitive assessments (MMSE, MOCA), neurotrophic markers (BDNF, PINK1), oxidative stress indicators (SOD, DJ1), and demographic factors (age, sex). The primary outcome was the change in UPDRS scores, serving as an objective measure of therapeutic response to the Neurotrophic peptide mixture. Also, in this study, machine learning models were employed to predict UPDRS score changes as a key indicator of the Neurotrophic peptide mixture’s therapeutic effectiveness, offering a data-driven evaluation of treatment response.\nResults\nClinical and neurophysiological assessment of the effectiveness of course treatment\nA course of the Neurotrophic peptide mixture administration against the background of basic antiparkinsonian therapy had a beneficial effect on the condition of the majority of patients, primarily on their activity in everyday life and the ability to self-care. Objectively, this effect was expressed in a significant decrease in scores of all UPDRS subscales (~16% on average) (Fig. 1).\nBoth clinically central endpoints improved significantly (p < 0.01). Secondary measures (UPDRS I and UPDRS II) generally trend in the same direction; effect sizes are provided without formal inference.\nA beneficial impact was observed following administration of a regimen of the Neurotrophic peptide mixture on a spectrum of non-motor dysfunctions concomitant with the motor manifestations of Parkinson’s disease (PD). Participants exhibited statistically significant enhancements in cognitive abilities as assessed by specialized instruments including MMSE, FAB, and MoCA (~11% on average). Moreover, there was a notable reduction in scores indicative of depression measured via the Beck Depression Inventory scale (~10% on average), as well as in reactive anxiety assessed by the Spielberger-Hanin scale (~23% on average). Notably, levels of personal anxiety remained consistent relative to baseline measurements. As a result, all observed Secondary outcomes are trended in the direction of improvement (Fig. 2).\nFor instrumental assessment of the Neurotrophic peptide mixture effect on motor functions, the Motor tempo test showed higher sensitivity. As a result, the significant decrease in the average, minimum and maximum time of motor reactions was recorded (Fig. 3).\nThose changes are more characteristic of the reactions of the right hand, while for the left hand only the average value of the motor tempo decreased. It can be assumed that this asymmetry of effects is associated with the sidedness of the pathological process, since a significant part of patients (76% of the study group) demonstrated predominantly right-sided symptoms of PD. These findings show a valuable positive impact of Neurotrophic peptide mixture on neurophysiological parameters.\nAt the same time, the latent period of a simple sensorimotor reaction did not change during a course of the Neurotrophic peptide mixture administration (Fig. 3).\nChanges of the H-reflex and M-response after administration of a regimen of Сerebrolysin\nIn the context of administering the Neurotrophic peptide mixture, there was an elevation observed in the threshold for triggering the H-reflex of the gastrocnemius muscle, increasing from 5 [3; 9] mA to 6 [5.8; 18] mA (Fig. 4). Notably, there were no alterations noted in the latency period or power attributes of the H-reflex, nor in the parameters of the M-response, compared to their baseline values.\nPotentials P300\nAnalysis of endogenous components revealed that in the studied group of patients with PD, there was a delay in the formation of the P3 wave (median 434 ms, interquartile range 386–512 ms in the temporal area of the neocortex) compared to the corresponding age norm of 390–410 ms. After the course of the Neurotrophic peptide mixture, a statistically significant increase in the latency period of the P3 component in the central-parietal lead was recorded, up to 394 [304; 446] ms (Fig. 5). There were no significant changes in the latency period of the N2 component and the amplitudes of the N2 and P3 components.\nThe delay in the formation of the P3 wave is a characteristic feature of several neurodegenerative diseases, particularly Alzheimer’s and Parkinson’s diseases. For patients with borderline PD and moderate cognitive impairments, an increase in the latency period of P3 has been described, with a reverse correlation between this parameter and the volume of gray matter in the globus pallidus, dysfunction of which leads to symptoms such as bradykinesia and rigidity25. Improvement in this parameter following the course of the Neurotrophic peptide mixture suggests that a more pronounced effect on the state of higher cortical functions can be expected in patients with increased risk of developing or progressing cognitive impairments.\nCorrelation analysis of clinical and neurophysiological parameters after a course of treatment with the Neurotrophic peptide mixture\nThe degree of improvement in several non-motor characteristics (cognitive functions assessed by the FAB scale, amplitude of cognitive evoked potential P3, and signs of depression measured by the Beck scale) was greater in individuals who initially received a higher daily dose of levodopa (Figs. 6–8).\nFor indicators of cognitive functions (MMSE, P3 cognitive evoked potential amplitude), a similar relationship with the time of onset of the disease was also found: more pronounced positive effects of the Neurotrophic peptide mixture treatment were observed in patients with an earlier onset of PD and a longer duration of the disease (Figs. 6, 8).\nAt the same time, a more pronounced reduction in reaction time was characteristic of individuals with lower initial ratings of PD severity on the UPDRS scale and lower levels of anxiety. Such correlation relationships are found both when directly assessing the time of motor reactions in the Motor tempo and Sensorimotor reaction tests, and when determining the latent periods of the H-reflex and M-response (Figs. 9, 10).\nThe use of the Neurotrophic peptide mixture in the treatment course improves the general condition of patients with Parkinson’s disease, as evidenced by a significant decrease in the quantitative indicators of all UPDRS subscales, attenuation of affective disorders such as anxiety and depression, and improvement in cognitive function. Against the background of the Neurotrophic peptide mixture administration, dynamic characteristics of simple motor reactions improve, and the threshold for the emergence of the H-reflex increases, indicating facilitation of supraspinal control processes over the activity of motor neurons. In general, the results show a strong tendency to improvement in different parameters.\nMitochondrial dysfunction before and after treatment with the Neurotrophic peptide mixture in patients with Parkinson’s disease\nThe conducted studies revealed significant differences in the structural organization of platelets in individuals with PD examined before the start of the standard treatment process. Electron microscopic and morphometric study of platelets in patients with PD revealed that platelets were mainly represented by senescent cells. They had a reduced number of dense δ granules, which was 5.1 ± 0.4 per one cell (in healthy individuals - 8.4 ± 0.5 units per one cell, p < 0.05). However, the number of transparent α-granules turned out to be increased from 5.3 ± 0.2 up to 10.3 ± 0.6 per cell. In addition, the latter often had large sizes (>100 nm) and contained a significant amount of protein debris (Fig. 11).\nThe single mitochondria that were found in the cells (2–4 per cell) were structurally quite preserved and had an average diameter of 0.48 ± 0.02 µm, not significantly different from those determined in the control group (0.41 ± 0.04 μm).\nSignificant abnormalities were found in the ultrastructure of mitochondria in platelets of patients. In addition to increased mitochondrial membrane permeability (outer and/or inner, crystas-forming), the formation of septated organelles was observed. The mechanisms and causes of their formation are still unexplored. It is assumed that septated mitochondria can accumulate Са2+26. Since it is known that the accumulation of Са2+ and the synthesis of ATP are alternative processes associated with the transfer of electrons, an increase in the Ca-accumulating ability of mitochondria may indicate a decrease in the synthesis of ATP in them. In addition, the appearance of septated mitochondria may indicate partial cell necrosis. These ultrastructural changes are evidence of the formation of mitochondrial dysfunction in platelets.\nThe use of the Neurotrophic peptide mixture in the complex therapy of patients with CP was accompaniedby reliable dynamics of the morphofunctional state of platelets (Table 1). The use of the Neurotrophic peptide mixture led to a significant (by 45%) increase in the number of δ-granules. No significant changes in the number of α granules were detected; only a tendency to its decrease was observed. An increase in the number of δ-granules indicates an increase in serotonin reserves. Probably, one of the possible mechanisms of the neuroprotective effect of the Neurotrophic peptide mixture is realized in this way. In addition, the Ca2+ -accumulating capacity in thrombocytes is relatively reduced, which should contribute to the improvement of ATP synthesis.\nDespite the small number of mitochondria in platelets, a slight increase in their number was observed - from 2 to 4 organelles before treatment to 5–7 organelles after treatment with the Neurotrophic peptide mixture, which, in combination with the absence of detection of septated organelles, may indicate a decrease in the severity of mitochondrial dysfunction.\nThus, treatment with the Neurotrophic peptide mixture improves the morphofunctional state of platelets in patients with PD, it increases serotonin reserves, reduces Ca2+ -accumulating capacity, reduces the manifestations of mitochondrial dysfunction, and exhibits mild antithrombotic properties.\nEffect of the Neurotrophic peptide mixture treatment on PARKIN gene expression in patients with Parkinson’s disease\nFor the whole group of tested individuals, taken together, no correlation of the expression of the measured genes was shown using the Spearman test. At the same time, when dividing by sex, multidirectional differences in gene expression before and after therapy with the drug Neurotrophic peptide mixture were found (Table 2).\nIn particular, the group of men was characterized by a decrease in the expression of the BDNF gene, and in women the expression increased after the use of the drug (Fig. 12).\nIt is important to note that when comparing the results of gene expression with cognitive parameters, an increase in BDNF in women (but not in men) was also accompanied by an increase in the MMSE score (Fig. 13).\nNo statistically significant difference was observed in expression of other three genes (DJ-1, PINK1, TNFalfa) before and after treatment with the Neurotrophic peptide mixture (Figs. 14–16).\nThe correlation analysis shows higher values between MMSE and PINK1 expression, FAB a and DJ1, FAB a and PINK1, MOCA a and DJ1, MOCA a and PINK1, MOCA and PINK1 for men (Table 3).\nFor women, the correlation analysis shows strong values of clinically central endpoints UPDRS III and UPDRS S with almost all studied genes expression rates (Table 4).\nEffect of the Neurotrophic peptide mixture treatment on the state of the prooxidant-antioxidant system in patients with Parkinson’s disease\nThe intensification of prooxidant processes in blood plasma and erythrocytes of patients with moderate PD stages was registered. It was established a significant rise (by 40%) in the content of the secondary products of lipid peroxidation in blood plasma as well as a marked increase (by 50%) of the H2O2 formation in erythrocytes of patients with PD compared to healthy participants (Tables 5 and 6). Our assessment of the antioxidant protective effectiveness was largely based on the changes in activity of antioxidant enzymes – SOD and catalase, which represent the first line of such defense, providing detoxification of reactive oxygen metabolites27. In patients with PD, we have found an increase in the activities of SOD and catalase by 21% and 69%, respectively in comparison with healthy subjects. In addition, we analyzed the glutathione cycle, which is one of the main intracellular mechanisms for maintaining redox homeostasis28. In our study, the concentration of GSH was significantly decreased (by 58%) in erythrocytes of PD patients as well as GPx activity though this change was not statistically significant (by 15%) compared to healthy controls. We found that CBL administration has moderate antioxidant effects in PD patients by decreasing the TBARS accumulation (by 17%) in blood plasma as well as H2O2 content in erythrocytes (by 12%) compared with the CBL-untreated patients. These effects of CBL were accompanied by a decrease in the activities of SOD (by 11%) and catalase (by 20%) in blood of patients with PD compared with the CBL-untreated ones. In CBL-treated patients, there was registered an increase in the GSH concentration (by 30%) and GPx activity (by 15%) in erythrocytes compared with PD patients though this change was not statistically significant compared to the CBL-untreated patients (Tables 5 and 6).\nUPDRS score change prediction using multiple machine learning models\nAmong the models, GBMs demonstrated the highest predictive accuracy, achieving an RMSE of 18.59, MSE of 345.82, and MAE of 15.39. The top-performing RF model exhibited slightly higher RMSE (21.99) and MAE (17.78), underscoring the superior capability of GBMs to capture complex patterns in the data.\nThe boxplot of feature importance across the top 10 models provided valuable insights into which factors most significantly influenced the prediction of UPDRS score changes. Below are the most impactful features, sorted by median scaled importance: Fig. 17\n- \n                    BDNF Change (Median Importance: 0.98): Changes in Brain-Derived Neurotrophic Factor (BDNF) were the most influential predictor of UPDRS score change. BDNF’s high importance suggests that neurotrophic support may play a central role in the Neurotrophic peptide mixture’s therapeutic effects, potentially enhancing neuroplasticity and recovery. \n- \n                    MOCA (Median Importance: 0.62): The Montreal Cognitive Assessment (MOCA), a measure of cognitive function, was ranked as a highly predictive feature, highlighting the relevance of cognitive improvement as a response indicator for the Neurotrophic peptide mixture therapy. \n- \n                    BDNF (after treatment) (Median Importance: 0.61): Post-treatment levels of BDNF were also significant, indicating that maintaining or boosting neurotrophic factors is associated with positive therapeutic outcomes in PD patients. \n- \n                    MMSE, S (Median Importance: 0.58): The Mini-Mental State Examination (MMSE) score also emerged as a key predictor, reinforcing the importance of cognitive assessments in evaluating the Neurotrophic peptide mixture’s efficiency. \n- \n                    PINK1 (after treatment) (Median Importance: 0.55): PINK1, a mitochondrial health marker, was another significant predictor, suggesting that the Neurotrophic peptide mixture’s potential benefits may extend to cellular and mitochondrial health. \nThe boxplot analysis revealed that features related to neurotrophic support (BDNF) and cognitive function (MOCA, MMSE) showed the most consistent importance across models, as evidenced by their high median importance values and narrow interquartile ranges. In contrast, features with wider variability, such as TBA-active products and Age, had broader interquartile ranges, indicating that their contribution to predictive accuracy may depend on model-specific interactions or particular patient subgroups.\nDiscussion\nThe findings of this exploratory study suggest that the Neurotrophic peptide mixture has a multifaceted impact on various aspects of PD pathology, particularly regarding platelet ultrastructure, gene expression, neurophysiological parameters, and oxidative stress markers. These findings align with existing literature, which underscores the neuroprotective and neurorestorative properties of the Neurotrophic peptide mixture.\nImprovement in motor functions, particularly in the Motor tempo test, was more pronounced in patients with predominantly right-sided symptoms of PD. This asymmetry in response could be related to the lateralization of the pathological process. Additionally, the elevation in the threshold for triggering the H-reflex suggests modifications in spinal cord excitability, which could translate to improvements in motor function. The study reported statistically significant enhancements in cognitive abilities, as measured by the MMSE, FAB, and MoCA scales. Additionally, there was a notable reduction in depression scores (BDI) and reactive anxiety (Spielberger-Hanin scale), although personal anxiety levels remained unchanged. These findings highlight the potential of the Neurotrophic peptide mixture in ameliorating both motor and non-motor symptoms of PD. The improvements in motor function, cognitive abilities, and emotional well-being observed in this study suggest that the Neurotrophic peptide mixture has a broad therapeutic potential.\nThe study also observed significant neurophysiological improvements post-Neurotrophic peptide mixture treatment. The latency period of the P3 component, a marker of cognitive processing, was significantly reduced. This improvement in P3 latency indicates enhanced cortical function and suggests that the Neurotrophic peptide mixture may have beneficial effects on cognitive processing speed and efficiency in PD patients. These neurophysiological benefits align with the observed cognitive improvements, further supporting the potential of the Neurotrophic peptide mixture in mitigating cognitive decline in PD.\nIn PD patients, platelets exhibited several signs of senescence, including a reduction in dense δ-granules and an increase in α-granules29. The number of δ-granules per cell was significantly lower in PD patients (5.1 ± 0.4) compared to healthy individuals (8.4 ± 0.5). The administration of the Neurotrophic peptide mixture significantly increased the number of δ-granules in platelets by 45%, indicating enhanced serotonin reserves, which could contribute to the neuroprotective effects observed clinically. This increase in δ-granules likely reflects a normalization of platelet storage function, which is crucial given the role of serotonin in mood and neuroprotection. Despite no significant change in the number of α-granules, the trend towards a decrease might indicate an alteration in platelet activation status.\nAdditionally, the Neurotrophic peptide mixture reduced the calcium-accumulating capacity of platelets, potentially improving ATP synthesis and overall platelet function. This reduction in calcium accumulation is crucial as it indicates a shift away from the pathological state observed in untreated PD platelets. The observed improvements in platelet ultrastructure and function may translate to better overall cellular health, reflecting the systemic benefits of the Neurotrophic peptide mixture treatment beyond its direct neural effects. The enhancement in serotonin reserves and mitochondrial function underscores the multifaceted role of the Neurotrophic peptide mixture in addressing both neurological and peripheral manifestations of PD.\nThe Neurotrophic peptide mixture treatment appears to modulate BDNF gene expression in a sex-specific manner in patients with Parkinson’s disease, with women showing an increase and men showing a decrease in expression levels. This modulation is accompanied by improved cognitive function in women, highlighting the potential of the Neurotrophic peptide mixture as a therapeutic option that may benefit female PD patients more significantly. The stable expression of DJ-1, PINK1, and TNFα genes suggests that the primary neuroprotective mechanisms of the Neurotrophic peptide mixture in PD may involve pathways other than those mediated by these genes. These findings emphasize the need for further research to explore the underlying mechanisms of sex differences in PD treatment responses and to optimize therapeutic strategies accordingly.\nThe excessive production and release of ROS in PD occurs as a result of the combined action of mitochondrial dysfunction, dopamine autooxidation, and neuroinflammation30. In present study, the Neurotrophic peptide mixture demonstrated a moderate antioxidant effect, reducing the levels of lipid peroxidation products and H2O2 in erythrocytes. This reduction was accompanied by a decrease in SOD and catalase activities, and an increase in GSH concentration and GPx activity. These changes indicate a restoration of redox balance and a reduction in oxidative stress, which are crucial in the pathology of PD. One of the most promising effects of CBL on OS developing in blood and tissues in PD could be its ability to decrease intracellular α-synuclein aggregation in neurons, mimicking the action of classical NTFs and growth factors as well as of NTFs belonged to a novel family CDNF/MANF31. The ability of the Neurotrophic peptide mixture to modulate the antioxidant system suggests its potential role in protecting neuronal cells from oxidative damage and subsequent apoptosis.\nThe findings underscore the central role of cognitive and neurotrophic markers in predicting UPDRS score changes as an indicator of the Neurotrophic peptide mixture efficacy. The high importance of BDNF aligns with previous research on neuroprotective therapies, where neurotrophic support has been linked to improved neuronal survival and plasticity. The inclusion of mitochondrial markers, such as PINK1, suggests that the Neurotrophic peptide mixture may exert beneficial effects at the cellular level, addressing mitochondrial dysfunction—a known factor in PD pathology.\nThe boxplot analysis provided further clarity on the stability and reliability of each feature’s importance. Features with high median importance and narrow interquartile ranges, like BDNF and MOCA, indicate consistent predictive power across models, making them robust indicators of therapeutic response. The broader variability observed in other features, such as oxidative stress markers and demographic factors, suggests that while they may contribute to specific aspects of disease progression, they are less central to predicting therapeutic efficacy. The superior performance of GBMs in capturing non-linear relationships reinforces the value of using advanced machine learning techniques to analyze complex clinical datasets. These findings support the potential of the Neurotrophic peptide mixture as a multifaceted therapeutic option, with the capacity to improve cognitive and neurotrophic parameters in PD patients. The results of the present study indicate an antioxidant action of the Neurotrophic peptide mixture in patients with moderate stages of Parkinson’s disease. This action is mainly based on the decrease of the prooxidant events in peripheral blood of PD patients as well as on the prevention of thiol redox defects and reducing an imbalance in oxidant–antioxidant status.\nOverall, statistically significant improvements and consistent correlations were observed in the key clinical endpoints (UPDRS Part III and UPDRS S), while secondary and exploratory outcomes demonstrated patterns suggestive of mechanistic links that warrant confirmation in larger studies.\nInterpretation of its results is limited by several factors, including the small sample size, lack of a control group receiving placebo in addition to standard therapy, and a relatively short duration of the observation period. Since the study is preliminary, exploratory in nature and aimed at collecting primary data on the response to the drug, only one group of patients was formed, which allows identifying primary trends and assessing the feasibility of future larger studies. Future research should focus on larger cohorts and longer treatment durations to confirm these findings and to explore the long-term effects of neurotrophic peptide mixture on gene expression and cognitive function in PD patients. Particular attention should be paid to investigating sex differences in response to the drug in future studies.\nThe results of this exploratory study provide a preliminary rationale for investigating the Neurotrophic peptide mixture as an adjunct therapy in Parkinson’s disease. During the treatment period, changes were observed in platelet ultrastructure, gene expression, neurophysiological parameters, and oxidative stress markers. These findings, along with shifts in cognitive and neurotrophic markers, may suggest a multidimensional biological response.\nHowever, these observations must be interpreted with extreme caution. Given the limited sample size and the absence of a control group, no causal inferences can be drawn. Therefore, larger, placebo-controlled clinical trials are warranted to determine if these initial observations can be substantiated, to investigate potential sex-specific and long-term effects, and to explore biomarkers that may help elucidate the mechanisms of the Neurotrophic peptide mixture in PD.\nMethods\nPatient information\nThe exploratory study included 17 patients with Parkinson’s disease with an established diagnosis according to the inclusion-exclusion criteria of the British Brain Bank32, aged 45–74 years with disease stage 2.0–3.0 (2.9 ± 0.3, according to Hoehn and Yahr scale). Written informed consent was obtained from all participants.\nThe neurotrophic peptide mixture was prescribed against the background of basic stable doses of antiparkinsonian therapy (Levodopa-containing drugs), which were used during 28–30 days before starting the course of neurotrophic peptide mixture and during the entire study period. No dose adjustments were allowed during the study, and medication dose was not included as a covariate in the statistical analysis. The median daily Levodopa dose was about 500 mg, it was individually optimized depending on disease stage and the patient’s condition. The Neurotrophic peptide mixture dosage regimen was 20 ml (215.2 mg/ml) intravenously, once a day for 10 days. Infusions were carried out in the first half of the day.\nEthics statement and study design\nThe Ethics Committee of the Clinical Department of the Institute of Gerontology named after D.F. Chebotariov of the National Academy of Medical Sciences of Ukraine approved the protocol of current research (Protocol No. 4, 09/06/2022). Written informed consent was obtained from all participants prior to enrollment.\nThe study was conducted in accordance with the principles of the Declaration of Helsinki (2000) and the European Council Convention on Human Rights and Biomedicine (2005), and relevant national standards for the participation in the research were taken into account.\nClinical assessment\nCharacterization of the stages of PD was carried out using the Hoehn and Yahr scale. To assess the dynamics of motor disorders, the standardized international Unified Parkinson’s Disease Rating Scale (UPDRS) was used. For a comprehensive assessment of the state of non-motor functions, the following clinical scales were used:\n- \n                    Mini-Mental State Examination (MMSE); \n- \n                    Test battery for assessing frontal dysfunction (Frontal Assessment Battery, FAB); \n- \n                    Montreal Cognitive Assessment (MoCA); \n- \n                    Beck Depression Inventory (BDI); \n- \n                    Spielberger-Hanin Anxiety Scale. \nSensorimotor reaction and motor tempo\nMotor tempo was determined using a computer program as the time between consecutive presses of the same finger on two keys spaced 20 cm apart on the keyboard. The patient was instructed to press the keys at the fastest possible tempo. The duration of each test was 20 s. Testing was conducted for each hand separately, and then the data were averaged.\nThe latent period of simple sensorimotor reaction was determined as the interval between the moment of presentation of a visual signal on the computer monitor and the moment the patient pressed the key in response to the signal. Geometric shapes (circle, square, oval) of red and white colors were used as signals, with 20 signals presented at intervals of 2–4 s. Testing was conducted for each hand separately, and the data were averaged.\nElectroneuromyography\nThe technique of registering the H-reflex, as a reflex response of muscles to the stimulation of sensory fibers of a mixed nerve followed by monosynaptic activation of spinal motoneurons and nerve motor fibers, was performed using disposable surface recording electrodes placed in the projection of the posterior group of muscles of the calf, with stimulation of the tibial nerve in the popliteal fossa. The dynamics of the amplitudes of H- and M-responses (mV) with increasing strength of nerve electrical stimulation (mA), the threshold for the occurrence of the H-reflex (mA), the latency period of the H-reflex (ms), and the ratio of the maximum amplitude of H max/M max (%) were evaluated.\nEvoked potentials P300\nEEG was recorded monopolarly from 21 leads (Fp1, Fp2, Fpz, F3, F4, Fz, C3, C4, Cz, P3, P4, Pz, T3, T4, F7, F8, T5, T6, O1, O2, Oz). The combined ear electrode was used as a reference electrode. The interelectrode resistance was less than 5 kOhm, the signal sampling frequency was 500 Hz, and the high- and low-pass filters were 0.3 Hz and 35 Hz, respectively. Endogenous cognitive evoked potentials N2 and P3 were recorded according to the oddball paradigm using sound stimuli with a frequency of 1000 and 2000 Hz and a duration of 10 ms.\nPlatelet and mitochondrial structure with Electron microscopy\nBlood plasma enriched with platelets containing leukocytes was obtained by centrifugation of whole blood obtained from the ulnar vein of the subjects at room temperature for 15 min at 120 g in a T-30 centrifuge. The plasma was carefully separated from the sediment and centrifuged at 2000 g for 20 min using a Vortecs Combispin FVL-2400N minicentrifuge33.\nElectron microscopic studies of platelets and mitochondria were carried out according to the generally accepted method with double fixation with glutaraldehyde and OSO4, dehydration with alcohols of increasing concentrations and subsequent embedding in Epon. Ultrathin sections with a thickness of 40–60 nm, contrasted with uranyl acetate and lead citrate, were examined using an electron microscope PEM-125K (Ukraine).\nMorphometric calculations of platelets’ structure were performed using the Image Tool (USA) on 130–150 fields for each group of subjects. The number of optically dense small σ-granules and optically transparent large α-granules was determined.\nPBMC isolation and quantitative PCR\nBlood collection, PBMC isolation, RNA extraction and cDNA synthesis were performed by standard methods. The collection of blood for further cell isolation and qPCR was made on 0 and 11th day (after a 10-day course of neurotrophic peptide mixture treatment). Briefly, patients’ blood was collected in 4 ml vacutainers with EDTA. Density gradient centrifugation of diluted whole blood layered over a density gradient medium was used for PBMC isolation. PBMC were gradually frozen at a temperature of −20 °C, and then transferred to the temperature of −80 °C for long-term storage. RNA was isolated from cells using the Direct-zol RNA Miniprep Kit (Zymo Research, USA) following the manufacturer’s protocols. The isolated RNA was treated with DNase to remove DNA residues that could give a false positive signal for the expression of target genes. After this, reverse transcription was carried out using a commercial kit ZymoScript RT PreMix (Zymo Research, USA), guided by the manufacturer’s recommendations.\nSamples of this cDNA were used for further qPCR-RT (quantitative real-time polymerase chain reaction), which was used to measure the expression of the DJ-1, PINK1, BDNF and TNFα genes. To identify a suitable internal control for RT-qPCR data normalization, we empirically tested the expression stability of three candidate reference genes (rps18, gusb, and beta-actin). The sequences of all used primers are listed in the Supplementary Information. Analysis of the raw Cqt values revealed that both beta-actin and gusb were the most stably expressed genes, showing minimal variation and a strong positive correlation with each other. In contrast, rps18 expression was found to be significantly more variable across our samples and was therefore excluded. Although both beta-actin and gusb were suitable candidates, we selected beta-actin for normalization due to its slightly higher expression stability.\nThe sequences of all used primers are listed in the Supplementary Information. The PCR reaction mix was prepared using a commercial kit of reagents for RT-PCR (HOT FIREPol® EvaGreen® qPCR Mix Plus, Solis BioDyne, Estonia) - reverse transcription and polymerase chain reaction with the addition of betaine (Sigma, USA) at a final concentration of 1 M. To determine differences in mRNA expression of Parkin genes, Repeated Measures Anova was used.\nAccumulation curves were generated using Opticon Monitor 3 software. After amplification was completed, raw gene product amplification data were used to calculate Cy0.\nMeasurement of oxidative stress biomarkers\nSecondary products of lipid peroxidation, the activity of antioxidant enzymes superoxide dismutase (SOD) and catalase (CAT) were measured in blood plasma. In red blood cells – the content of hydrogen peroxide (H2O2); the state of the glutathione system was assessed by the content of reduced glutathione (GSH) and the activity of the enzyme glutathione peroxidase (GP).\nFor erythrocyte isolation, samples were centrifuged at 8000 rpm for 15 min to separate plasma and red blood cells. Using centrifugation, erythrocytes were washed twice of plasma with cold sterile physiological sodium chloride solution. Erythrocyte hemolysate was obtained by adding distilled water to washed erythrocytes in a 1:1 ratio. The hemoglobin (Hb) content was also measured (Hemoglobin Assay kit MAK115, Sigma-Aldrich, St. Louis, MO, USA).\nDetermination of thiobarbituric acid reactive substances (TBARS) levels: TBARS were isolated by boiling plasma with the thiobarbituric acid reagent (0.5% 2-thiobarbituric acid/10% trichloroacetic acid/0.63 mM hydrochloric acid) and measuring the absorbance at 532 nm. The results are expressed as µM/ml using a molar extinction coefficient 1.56 * 105 mm−1 cm−1.\nCatalase activity measurement: The method is based on the ability of molybdenum salts to form a colored complex with hydrogen peroxide, which was determined at a wavelength of 410 nm. Enzyme activity (μM/min/L) was calculated using a molar extinction coefficient of 22.2 * 103 mm−1cm−1.\nSuperoxide dismutase (SOD) activity was determined spectrophotometrically. The method is based on the ability of SOD to inhibit the autoxidation of adrenaline at pH 10.2. The results were expressed as specific activity of the enzyme in units per ml of plasma. One unit of SOD activity is defined as the amount of protein causing 50% inhibition of the conversion rate of adrenaline to adrenochrome under specified conditions.\nThe activity of selenium-dependent glutathione peroxidase (GPx) was determined by recording the rate of NADPH oxidation in the presence of reduced glutathione at a wavelength of 340 nm. Enzyme activity was expressed in nM NADPH/min/ml.\nThe amount of reduced glutathione (μM/ml) was determined using Elman’s reagent (5.5’-dithio-bis-nitrobenzoic acid in absolute ethanol). The change in extinction was recorded at a wavelength of 412 nm every minute for 5 minutes.\nHydrogen peroxide content was studied using the FOX method, which is based on the peroxide-mediated oxidation of Fe2 + , followed by the reaction of Fe3+ with xylenol orange. Absorbance of the Fe3 + -xylenol orange complex was determined at a wavelength of 560 nm. The amount of H2O2 (μM/mgHb) was calculated using the H2O2 standard curve for each independent experiment.\nMachine learning prediction of UPDRS score change\nTo model UPDRS score changes, H2O’s AutoML framework was employed34, allowing for the training, tuning, and evaluation of multiple Machine Learning models, including Gradient Boosting Machines (GBMs) and Random Forests (RFs). AutoML facilitated hyperparameter optimization, ensuring robust and accurate predictions.\n- \n                    Gradient Boosting Machines (GBMs): GBMs iteratively build a series of decision trees, where each subsequent tree focuses on correcting errors made by its predecessors. This sequential approach enables GBMs to capture intricate, non-linear relationships within the data, making them particularly suitable for analyzing how clinical and biochemical factors impact UPDRS score changes and, consequently, therapeutic response35. \n- \n                    Random Forests (RFs): RFs are ensemble models that combine multiple decision trees, each trained on a random subset of the data. Known for their robustness and reduced susceptibility to overfitting, RFs provide valuable insights into feature importance by averaging across numerous trees36. This model serves as a stable baseline to complement the more flexible GBMs. \nThe models were evaluated using the following metrics to quantify prediction accuracy:\n- \n                    Root Mean Squared Error (RMSE): Captures the average deviation between predicted and actual values. \n- \n                    Mean Squared Error (MSE): Measures the variance of predictions relative to actual outcomes. \n- \n                    Mean Absolute Error (MAE): Reflects the average absolute deviation between predicted and observed values. \nThese metrics allowed for a thorough assessment of each model’s ability to predict UPDRS score changes, highlighting the Neurotrophic peptide mixture’s potential impact on disease progression.\nFeature importance was evaluated using the top 10-performing models from a set of 50 trained with H2O AutoML. Each feature’s importance was measured by its “scaled importance,” which represents its relative contribution to prediction accuracy. The models were trained using two-fold cross-validation, and feature importance was extracted during the validation stage.\nData from the top 10 models were aggregated, and features were sorted by their median feature importance. It is worth noting that while building a generalized model on such a small dataset is not feasible, the features identified with the highest median importance could be considered potential markers in future studies. The results were visualized using a boxplot (Fig. 16) to highlight the median importance and variability of each feature.\nStatistical analysis\nHypotheses about the effects of the drug were tested using the Wilcoxon (for related samples) and Mann-Whitney (for independent groups) tests. The relationship between the investigated indicators was determined in the form of Spearman’s rank correlation coefficients.\nStatistical processing of gene expression data was performed using Statistica 8.0 software (StatSoft Inc., USA). For statistical analysis, Spearman’s non-parametric test was used to investigate the correlation between parameters. The difference in expression of the studied genes before and after treatment was determined using the Wilcoxon t-test (Wilcoxon signed-rank test).\nThe results of the study of the state of the prooxidant-antioxidant system were processed statistically using the Origin program. The significance of differences between comparison groups was determined by the method of analysis of variance (ANOVA) followed by the Bonferroni test (post-hoc test).\nOutcomes and hierarchy\nFor this exploratory study, outcomes were grouped into three hierarchical levels.\n- \n                    Clinically central endpoints were prespecified as the UPDRS III (motor score) and the UPDRS Total score, as the primary efficacy measures, as these are the most widely accepted, clinically relevant, and objective indicators of motor impairment and overall disease severity in Parkinson’s disease. \n- \n                    Secondary clinical endpoints included UPDRS I (non-motor experiences of daily living) and UPDRS II (motor experiences of daily living), and validated scales of non-motor function: MMSE, MoCA, FAB (cognitive measures), and BDI and Spielberger–Hanin Anxiety Scale (mood and affective measures). These outcomes were considered supportive of the clinically central endpoints. \n- \n                    Exploratory outcomes included additional non-motor measures, neurophysiological parameters (P300 latency and amplitude), and molecular/structural biomarkers (gene expression, platelet ultrastructure, mitochondrial morphology, and oxidative stress markers). These outcomes were included to generate mechanistic insights into the potential modes of action of the Neurotrophic peptide mixture. \nFormal hypothesis testing was restricted to the clinically central endpoints. Secondary and exploratory outcomes are presented with descriptive statistics and effect sizes. To minimize Type I error, no multiple-testing correction was applied outside of the clinically central endpoints.\nCorrelation analyses were performed using Spearman’s rank correlation coefficient (ρ). For correlations involving clinically central endpoints, p-values are reported. Correlations among secondary and exploratory outcomes are reported without formal inferential testing, to avoid Type I error inflation.\nData availability\nThe datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.\nReferences\n- Alexoudi, A., Alexoudi, I. & Gatzonis, S. Parkinson’s disease pathogenesis, evolution and alternative pathways: A review. Rev. Neurol. (Paris) 174, 699–704 (2018). \n- Chung, S. J. et al. The Pattern of Striatal Dopamine Depletion as a Prognostic Marker in De Novo Parkinson Disease. Clin. Nucl. Med 201843, 787–792 (2018). \n- Moon, H. E. & Paek, S. H. Mitochondrial Dysfunction in Parkinson’s Disease. Exp. Neurobiol. 24, 103–116 (2015). \n- Mansur, A. et al. Characterization of 3 PET tracers for quantification of mitochondrial andsynaptic function in healthy human brain:18F-BCPP-EF,11C-SA-4503, and11C-UCB-J. J. Nucl. Med 61, 96–103 (2020). \n- Simonenkov, A. P. & Kluyev, V. M. Syndrome serotonin deficiency. M.: BINOM, 96-97 (2013). \n- Mauricio, R. et al. Tackling gaps in developing life-changing treatments for dementia. Alzheimers Dement 5, 241–253 (2019). \n- Gonchar, O. O., Klymenko, O. O., Drevytska, T. I., Bratus, L. V. & Mankovska, I. M. Oxidative stress in rat heart mitochondria under a rotenone model of Parkinson’ disease: a corrective effect of capicor treatment. Ukr. Biochem J. 93, 21–30 (2021). \n- Schneider-Garces, N. J. et al. Span, CRUNCH, and beyond: working memory capacity and the aging brain. J. Cogn. Neurosci. 22, 655–669 (2010). \n- Fiani, B. et al. Cerebrolysin for stroke, neurodegeneration, and traumatic brain injury: review of the literature and outcomes. Neurol. Sci. 42, 1345–1353 (2021). \n- Brainin, M. Cerebrolysin: a multi-target drug for recovery after stroke. Expert Rev. Neurother. 18, 681–687 (2018). \n- Requeejoe, C. et al. Nanodelivery of Cerebrolysin and rearing enriched environment induce neuroprotective effects in a preclinical rat model of Parkinson’s disease. Mol Neurobiol. Springer Science+Business Media, LLC (2017). \n- Neveen, A. et al. A promising therapeutic potential of Cerebrolysin in the 6-OHDA rat model of Parkinson's disease. Life Sci 155, 174–179 (2016). \n- Luchanina, E., Karaban, I. & Berezetskaya, N. Diagnosis of Parkinsonas disease by electrophysiological methods. Diagn. Rehabil. Parkinson. Dis. 2, 27–58 (2011). \n- Eder, P., Reinprecht, I., Schreiner, E., Skofitsch, G. & Windisch, M. Increased density of glutamate receptor subunit 1 due to Cerebrolysin treatment: an immunohistochemical study on aged rats. Histochem J. 33, 605–612 (2001). \n- Stepanichev, M. et al. Effects of cerebrolysin on nerve growth factor system in the aging rat brain. Restor. Neurol. Neurosci. 35, 571–581 (2017). \n- Veinbergs, I., Mante, M., Mallory, M. & Masliah, E. Neurotrophic effects of Cerebrolysin in animal models of excitotoxicity. J. Neural Transm. Suppl. 59, 273–280 (2000). \n- Alvarez, X. A. et al. Neuroimmunotrophic effects of cerebrolysin in an animal model of hippocampal degeneration induced by β-amyloid. In: Korczyn, A. D., ed. International Congress on Vascular Dementia 2017. Bologna, Italy: Monduzzi Editore, 233–237 (1999). \n- Gavrilova, S. I. & Alvarez, A. Cerebrolysin in the therapy of mild cognitive impairment and dementia due to Alzheimeras disease: 30 years of clinical use. Med Res Rev. 41, 1–29 (2020). \n- DeBoer, S. R. et al. Enhanced spontaneous motor recovery after stroke in mice treated with cerebrolysin. Neurorehabil Neural Repair 35, 525–533 (2021). \n- Muresanu, D. F. et al. Efficacy and safety of cerebrolysin in neurorecovery after moderate-severe traumatic brain injury: results from the CAPTAIN II trial. Neurol. Sci. 41, 1171–1181 (2020). \n- Lu, W. et al. Cerebrolysin alleviates early brain injury after traumatic brain injury by inhibiting neuroinflammation and apoptosis via TLR signaling pathway. Acta Cir. Bras. 37, e370605 (2022). \n- Chen, C. C., Wei, S. T., Tsaia, S. C., Chen, X. X. & Cho, D. Y. Cerebrolysin enhances cognitive recovery of mild traumatic brain injury patients: double-blind, placebo-controlled, randomized study. Br. J. Neurosurg. 27, 803–807 (2013). \n- Cui, S. Cerebrolysin for vascular dementia. Cochrane Database Syst Rev. 11, 19–20 (2019). \n- Rochenstein, E. et al. Neuropeptide treatment with cerebrolysin enhances the survival of grafted neural stem cells in a synuclein transgenic model of Parkinson’s disease. J. Exp. Neurosci. 9, 131–140 (2015). \n- Hünerli, D. et al. Mild cognitive impairment in Parkinson’s disease is associated with decreased P300 amplitude and reduced putamen volume. Clin. Neurophysiol. 130, 1208–1217 (2019). \n- Alvarez, X. A., Lombardi, V. R., Fernández-Novoa, L., García, M. & Sampedro, C. Cerebrolysin reduces microglial activation in vivo and in vitro: a potential mechanism of neuroprotection. J. Neural Transm. 59, 281–292 (2000). \n- Kulinskiĭ, V. I. & Kolesnichenko, L. S. Glutathione system. I. Synthesis, transport, glutathione transferases, glutathione peroxidases. Biomed. Khim 55, 255–277 (2009). \n- Zenkov, N. K., Lankin, V. Z. & Menshikova, E. B. Oxidative stress: The biochemical and pathophysiological aspects. M.: MAIK. Science / Interperiodica 4, 352–361 (2001). \n- Park, S. M. et al. Evidence that alpha-synuclein functions as a negative regulator of Ca(++)-dependent alpha-granule release from human platelets. Blood 100, 2506–2514 (2002). \n- Luo, Y., Hoffer, A., Hoffer, B. & Qi, X. Mitochondria: A therapeutic target for Parkinson’s disease?. Int J. Mol. Sci. 16, 20704–20730 (2015). \n- Voutilainen, M. H., Arumae, U., Airavaara, M. & Saarma, M. Therapeutic potential of the endoplasmic reticulum located and secreted CDNF/MANF family of neurotrophic factors in Parkinson’s disease. FEBS Lett. 589, 3739–3748 (2015). \n- Hughes, A. J., Daniel, S. E., Kilford, L. & Lees, A. J. Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: A clinico-pathological study of 100 cases. J. Neurol. Neurosurg. Psychiatry 55, 181–184 (1992). \n- Filippova, O. I., Koloskov, A. V. & Stolitsa, A. A. Methods for studying the functional activity of platelets (literature review). Transfusiology 13, 493–514 (2012). \n- LeDell, E. & Posse, C. H2O AutoML: Scalable Automatic Machine Learning. J. Open Source Softw. 5, 1–10 (2020). \n- Chen, J. & Guestrin, C. XGBoost: A Scalable Tree Boosting System. Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery Data Mining, 785–794 (2016). \n- Liaw, A. & Wiener, M. Classification and regression by random Forest. R. N. 2, 18–22 (2002). \nAcknowledgements\nThis work was supported by the National Academy of Medical Sciences, Ukraine (Grant # 0123U100771).\nAuthor information\nAuthors and Affiliations\nContributions\nD.K. took part in writing of the main text of manuscript, coordinated the work of scientific laboratories in their experiments and results, contributed to the conceptualization of the project, I.K. was the lead scientist of part connected to clinical assessment, data analysis, project conception and organization, N.K., N.M. and S.K. performed and analyzed electroencephalograms and elecroneuromyograms of patients, collected patient data regarding sensorimotor reactions and motor tempo, K.R. participated in the preparation of the literature review, implemented the methodology of morphological studies, processing the data of the section about the morphological parameters. O.G. examined markers of oxidative stress and made substantial contributions to the analysis of data as well as participated in a critical review of the manuscript, I.M. helped with the preparation of the manuscript and analysis of data. O.Z. wrote an abstract, implemented the main methodology for gene expression measurement, analysis and interpretation of data, O.B. and K.M. were involved to gene expression experiments and analysis of qPCR-RT data, as well as collection of biological material, S.S. provided the results of deep analysis of obtained data using multiple machine learning models, performed feature importance analysis, T.P. contributed to the administration of project, wrote and edited the main manuscript text. All authors reviewed the manuscript.\nCorresponding author\nEthics declarations\nCompeting interests\nThe authors declare no competing interests.\nAdditional information\nPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.\nRights and permissions\nOpen Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.\nAbout this article\nCite this article\nKrasnienkov, D., Karaban, I., Karasevych, N. et al. Evaluation of the neurotrophic peptide mixture in pathogenetic therapy of patients with Parkinson’s disease. npj Parkinsons Dis. 12, 55 (2026). https://doi.org/10.1038/s41531-026-01270-6\n- Received: \n- Accepted: \n- Published: \n- Version of record: \n- DOI: https://doi.org/10.1038/s41531-026-01270-6","confidence":0.9,"diagnostics_url":"/api/diagnose?url=https%3A//www.nature.com/articles/s41531-026-01270-6","quality_bucket":"high","failure_kind":"none","retryable":false,"quality_reason":"High confidence: full text extraction produced 58549 characters.","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 58549 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":58549,"summary_length":313,"usable_text_length":58549,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":58549,"summary_length":313}},"tags":[]},"fallback_formats":["markdown","json","html"],"actions":{"read":"/item/34350","export_markdown":"/api/items/34350/export?format=markdown","export_json":"/api/items/34350/export?format=json","diagnose":"/api/diagnose?url=https%3A//www.nature.com/articles/s41531-026-01270-6"},"formats":{"full":{"id":34350,"title":"Evaluation of the neurotrophic peptide mixture in pathogenetic therapy of patients with Parkinson’s disease - Nature","url":"https://www.nature.com/articles/s41531-026-01270-6","source":"Nature","author":null,"published_at":"2026-01-23T08:00:00+00:00","locale":"en","topic":"peptides","tags":[],"excerpt":"Abstract\nThis exploratory, single-group, open-label study investigated 17 patients with Parkinson’s disease (PD) using a pre-post design. Motor and non-motor outcomes were assessed through clinical scales, biochemical and genetic analyses, and machine learning models (Gradient Boosting Machines, Random Forests).","full_text":"Abstract\nThis exploratory, single-group, open-label study investigated 17 patients with Parkinson’s disease (PD) using a pre-post design. Motor and non-motor outcomes were assessed through clinical scales, biochemical and genetic analyses, and machine learning models (Gradient Boosting Machines, Random Forests). After treatment with a neurotrophic peptide mixture, improvements were observed in daily activity (16%), cognition (11%), depression (10% reduction), and reactive anxiety (23% reduction). Biological changes included a 45% increase in platelet δ-granules, higher mitochondrial counts, elevated gene expression (notably BDNF in women, p = 0.046), and modulation of oxidative stress markers (17% reduction in TBARS, 30% increase in GSH). Machine learning identified BDNF and PINK1 expression, along with MOCA and MMSE scores, as key predictors of UPDRS improvement. These findings suggest that neurotrophic peptide therapy may influence clinical, structural, and molecular domains in PD. Larger, controlled trials are warranted to confirm therapeutic potential and clarify associations with cognitive and neurotrophic parameters.\nSimilar content being viewed by others\nIntroduction\nParkinson’s disease is a neurodegenerative condition typically observed in older individuals, characterized by a gradual onset and progressive nature. The initial manifestation often includes tremors, followed by symptoms such as slowed movement (bradykinesia) and muscle stiffness (rigidity)1. The disease is linked to the degeneration of dopaminergic neurons in the substantia nigra, the presence of alpha-synuclein aggregates within neuronal cells (Lewy bodies and Lewy neurites), and subsequent pathology in regions like the basal forebrain, amygdala, and medial temporal lobe structures, progressing to cortical areas in advanced stages2.\nMitochondrial abnormalities and dysfunction are recognized as key contributors to Parkinson’s disease pathogenesis in a subset of patients3. Some data was obtained on the availability correlation of the functional activity of mitochondria in lymphocytes, which displays multisystem mitochondrial failure and indicates of energy dysfunction in general4. The spectrum of mitochondrial dysfunctions in PD includes dysfunction of the respiratory chain (inhibition of the NADH-oxidase oxidation pathway), hyperproduction of ATP, disruption of Ca-regulatory systems, opening of the mitochondrial pore, perturbations of mitochondrial dynamics, and dysregulation of mitochondrial clearance5.\nAlthough oxidative stress plays a significant role in the pathogenesis of PD, therapy based on the use of exogenous antioxidants has not been shown to be highly effective. An alternative strategy that can counteract the damaging effects of free radicals and restore cellular redox balance is the activation of endogenous antioxidant enzymes present in the central neural system, such as superoxide dismutase, catalase, and glutathione peroxidase. The transcription of these cytoprotective proteins is regulated by the transcription factor Nrf2, which plays a leading role in the regulation of cellular redox status and is the main regulator of the cellular response to oxidative stress through the induction of antioxidant and detoxifying enzymes and proteins6. This is consistent with our previous data, which showed that lipid peroxidation was significantly increased while reduced glutathione (GSH) and glutathione-dependent antioxidant enzymes activity markedly decreased in blood of patients with PD as well as in rat brain under modeling of PD7. Various activators of the Nrf2/ARE signaling pathway, including peptides, have shown their effectiveness in both in vitro and in vivo models of neurological disorders8.\nThe Neurotrophic peptide mixture is peptidergic mixture with neurotrophic-like properties, made from a mixture of low molecular weight, porcine-derived peptides and free amino acids. The pharmaceutical preparation includes the peptide fragments: nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), ciliary neurotrophic factor (CNTF), enkephalins, orexin, and P219.\nThe use of the Neurotrophic peptide mixture significantly reduces neurological deficits, improves cognitive abilities, and reduces the volume of the infarction zone during stroke10,11. The Neurotrophic peptide mixture has an organ-specific multimodal effect on the brain, i.e., it provides metabolic regulation, neuroprotection, functional neuromodulation, and neurotrophic activity12. The Neurotrophic peptide mixture protects neurons from the damaging effects of lactic acidosis, prevents the formation of free radicals13, increases survival and prevents neuronal death under conditions of hypoxia and ischemia, and reduces the damaging neurotoxic effect of glutamate14.\nOver the past decade, compelling evidence has emerged neuroprotective properties of the Neurotrophic peptide mixture, explaining its multi-level impact on both the pathological cascade of damage and on subsequent recovery processes15. The ability of the Neurotrophic peptide mixture to influence the processes of neuroprotection and neuroplasticity was proven in a series of studies15,16,17 on a model of Alzheimer’s disease in transgenic mice for the amyloid precursor protein (mThy-1-hAPP751 tg). Using a model of Alzheimer’s disease, it was shown that the Neurotrophic peptide mixture, through the regulation of the synthesis and transport of amyloid precursor protein and the deposition of β-amyloid in synapses, has a synaptotrophic effect, which plays a primary role in the processes of neuroplasticity, which are extremely important in the acquisition of new skills and the maintenance of existing ones18.\nIncreasing studies have demonstrated that the Neurotrophic peptide mixture administration can promote recovery of motor function, improve early brain injury, and decrease hippocampal neuronal death19. In recent clinical studies, the Neurotrophic peptide mixture improved overall outcomes after moderate to severe traumatic brain injury and was also safe and better for early rehabilitation patients after ischemic stroke20. Lu et al. reported that the Neurotrophic peptide mixture can improve neurological dysfunction and brain damage, relieve neuroinflammation and prevent apoptosis after brain injury21.\nBased on the multifaceted mechanism of action of the Neurotrophic peptide mixture, including its effect on the processes of neurogenesis, neuroplasticity and neuroprotection, the drug currently has three main indications for use - stroke, dementia and traumatic brain injury10,22,23.\nLuchanina et al. demonstrated for the first time the effect of Neurotrophic peptide mixture on motor activity and muscle tone in patients with PD treated with levodopa13. It was established that the Neurotrophic peptide mixture improved the engraftment of a DOPA transplant from stem cells in the substantia nigra24 and reduced the severity of symptoms in model animals12.\nThe current exploratory study dataset comprised clinical, cognitive, biochemical, and demographic information from PD patients undergoing the Neurotrophic peptide mixture therapy. Key features included cognitive assessments (MMSE, MOCA), neurotrophic markers (BDNF, PINK1), oxidative stress indicators (SOD, DJ1), and demographic factors (age, sex). The primary outcome was the change in UPDRS scores, serving as an objective measure of therapeutic response to the Neurotrophic peptide mixture. Also, in this study, machine learning models were employed to predict UPDRS score changes as a key indicator of the Neurotrophic peptide mixture’s therapeutic effectiveness, offering a data-driven evaluation of treatment response.\nResults\nClinical and neurophysiological assessment of the effectiveness of course treatment\nA course of the Neurotrophic peptide mixture administration against the background of basic antiparkinsonian therapy had a beneficial effect on the condition of the majority of patients, primarily on their activity in everyday life and the ability to self-care. Objectively, this effect was expressed in a significant decrease in scores of all UPDRS subscales (~16% on average) (Fig. 1).\nBoth clinically central endpoints improved significantly (p < 0.01). Secondary measures (UPDRS I and UPDRS II) generally trend in the same direction; effect sizes are provided without formal inference.\nA beneficial impact was observed following administration of a regimen of the Neurotrophic peptide mixture on a spectrum of non-motor dysfunctions concomitant with the motor manifestations of Parkinson’s disease (PD). Participants exhibited statistically significant enhancements in cognitive abilities as assessed by specialized instruments including MMSE, FAB, and MoCA (~11% on average). Moreover, there was a notable reduction in scores indicative of depression measured via the Beck Depression Inventory scale (~10% on average), as well as in reactive anxiety assessed by the Spielberger-Hanin scale (~23% on average). Notably, levels of personal anxiety remained consistent relative to baseline measurements. As a result, all observed Secondary outcomes are trended in the direction of improvement (Fig. 2).\nFor instrumental assessment of the Neurotrophic peptide mixture effect on motor functions, the Motor tempo test showed higher sensitivity. As a result, the significant decrease in the average, minimum and maximum time of motor reactions was recorded (Fig. 3).\nThose changes are more characteristic of the reactions of the right hand, while for the left hand only the average value of the motor tempo decreased. It can be assumed that this asymmetry of effects is associated with the sidedness of the pathological process, since a significant part of patients (76% of the study group) demonstrated predominantly right-sided symptoms of PD. These findings show a valuable positive impact of Neurotrophic peptide mixture on neurophysiological parameters.\nAt the same time, the latent period of a simple sensorimotor reaction did not change during a course of the Neurotrophic peptide mixture administration (Fig. 3).\nChanges of the H-reflex and M-response after administration of a regimen of Сerebrolysin\nIn the context of administering the Neurotrophic peptide mixture, there was an elevation observed in the threshold for triggering the H-reflex of the gastrocnemius muscle, increasing from 5 [3; 9] mA to 6 [5.8; 18] mA (Fig. 4). Notably, there were no alterations noted in the latency period or power attributes of the H-reflex, nor in the parameters of the M-response, compared to their baseline values.\nPotentials P300\nAnalysis of endogenous components revealed that in the studied group of patients with PD, there was a delay in the formation of the P3 wave (median 434 ms, interquartile range 386–512 ms in the temporal area of the neocortex) compared to the corresponding age norm of 390–410 ms. After the course of the Neurotrophic peptide mixture, a statistically significant increase in the latency period of the P3 component in the central-parietal lead was recorded, up to 394 [304; 446] ms (Fig. 5). There were no significant changes in the latency period of the N2 component and the amplitudes of the N2 and P3 components.\nThe delay in the formation of the P3 wave is a characteristic feature of several neurodegenerative diseases, particularly Alzheimer’s and Parkinson’s diseases. For patients with borderline PD and moderate cognitive impairments, an increase in the latency period of P3 has been described, with a reverse correlation between this parameter and the volume of gray matter in the globus pallidus, dysfunction of which leads to symptoms such as bradykinesia and rigidity25. Improvement in this parameter following the course of the Neurotrophic peptide mixture suggests that a more pronounced effect on the state of higher cortical functions can be expected in patients with increased risk of developing or progressing cognitive impairments.\nCorrelation analysis of clinical and neurophysiological parameters after a course of treatment with the Neurotrophic peptide mixture\nThe degree of improvement in several non-motor characteristics (cognitive functions assessed by the FAB scale, amplitude of cognitive evoked potential P3, and signs of depression measured by the Beck scale) was greater in individuals who initially received a higher daily dose of levodopa (Figs. 6–8).\nFor indicators of cognitive functions (MMSE, P3 cognitive evoked potential amplitude), a similar relationship with the time of onset of the disease was also found: more pronounced positive effects of the Neurotrophic peptide mixture treatment were observed in patients with an earlier onset of PD and a longer duration of the disease (Figs. 6, 8).\nAt the same time, a more pronounced reduction in reaction time was characteristic of individuals with lower initial ratings of PD severity on the UPDRS scale and lower levels of anxiety. Such correlation relationships are found both when directly assessing the time of motor reactions in the Motor tempo and Sensorimotor reaction tests, and when determining the latent periods of the H-reflex and M-response (Figs. 9, 10).\nThe use of the Neurotrophic peptide mixture in the treatment course improves the general condition of patients with Parkinson’s disease, as evidenced by a significant decrease in the quantitative indicators of all UPDRS subscales, attenuation of affective disorders such as anxiety and depression, and improvement in cognitive function. Against the background of the Neurotrophic peptide mixture administration, dynamic characteristics of simple motor reactions improve, and the threshold for the emergence of the H-reflex increases, indicating facilitation of supraspinal control processes over the activity of motor neurons. In general, the results show a strong tendency to improvement in different parameters.\nMitochondrial dysfunction before and after treatment with the Neurotrophic peptide mixture in patients with Parkinson’s disease\nThe conducted studies revealed significant differences in the structural organization of platelets in individuals with PD examined before the start of the standard treatment process. Electron microscopic and morphometric study of platelets in patients with PD revealed that platelets were mainly represented by senescent cells. They had a reduced number of dense δ granules, which was 5.1 ± 0.4 per one cell (in healthy individuals - 8.4 ± 0.5 units per one cell, p < 0.05). However, the number of transparent α-granules turned out to be increased from 5.3 ± 0.2 up to 10.3 ± 0.6 per cell. In addition, the latter often had large sizes (>100 nm) and contained a significant amount of protein debris (Fig. 11).\nThe single mitochondria that were found in the cells (2–4 per cell) were structurally quite preserved and had an average diameter of 0.48 ± 0.02 µm, not significantly different from those determined in the control group (0.41 ± 0.04 μm).\nSignificant abnormalities were found in the ultrastructure of mitochondria in platelets of patients. In addition to increased mitochondrial membrane permeability (outer and/or inner, crystas-forming), the formation of septated organelles was observed. The mechanisms and causes of their formation are still unexplored. It is assumed that septated mitochondria can accumulate Са2+26. Since it is known that the accumulation of Са2+ and the synthesis of ATP are alternative processes associated with the transfer of electrons, an increase in the Ca-accumulating ability of mitochondria may indicate a decrease in the synthesis of ATP in them. In addition, the appearance of septated mitochondria may indicate partial cell necrosis. These ultrastructural changes are evidence of the formation of mitochondrial dysfunction in platelets.\nThe use of the Neurotrophic peptide mixture in the complex therapy of patients with CP was accompaniedby reliable dynamics of the morphofunctional state of platelets (Table 1). The use of the Neurotrophic peptide mixture led to a significant (by 45%) increase in the number of δ-granules. No significant changes in the number of α granules were detected; only a tendency to its decrease was observed. An increase in the number of δ-granules indicates an increase in serotonin reserves. Probably, one of the possible mechanisms of the neuroprotective effect of the Neurotrophic peptide mixture is realized in this way. In addition, the Ca2+ -accumulating capacity in thrombocytes is relatively reduced, which should contribute to the improvement of ATP synthesis.\nDespite the small number of mitochondria in platelets, a slight increase in their number was observed - from 2 to 4 organelles before treatment to 5–7 organelles after treatment with the Neurotrophic peptide mixture, which, in combination with the absence of detection of septated organelles, may indicate a decrease in the severity of mitochondrial dysfunction.\nThus, treatment with the Neurotrophic peptide mixture improves the morphofunctional state of platelets in patients with PD, it increases serotonin reserves, reduces Ca2+ -accumulating capacity, reduces the manifestations of mitochondrial dysfunction, and exhibits mild antithrombotic properties.\nEffect of the Neurotrophic peptide mixture treatment on PARKIN gene expression in patients with Parkinson’s disease\nFor the whole group of tested individuals, taken together, no correlation of the expression of the measured genes was shown using the Spearman test. At the same time, when dividing by sex, multidirectional differences in gene expression before and after therapy with the drug Neurotrophic peptide mixture were found (Table 2).\nIn particular, the group of men was characterized by a decrease in the expression of the BDNF gene, and in women the expression increased after the use of the drug (Fig. 12).\nIt is important to note that when comparing the results of gene expression with cognitive parameters, an increase in BDNF in women (but not in men) was also accompanied by an increase in the MMSE score (Fig. 13).\nNo statistically significant difference was observed in expression of other three genes (DJ-1, PINK1, TNFalfa) before and after treatment with the Neurotrophic peptide mixture (Figs. 14–16).\nThe correlation analysis shows higher values between MMSE and PINK1 expression, FAB a and DJ1, FAB a and PINK1, MOCA a and DJ1, MOCA a and PINK1, MOCA and PINK1 for men (Table 3).\nFor women, the correlation analysis shows strong values of clinically central endpoints UPDRS III and UPDRS S with almost all studied genes expression rates (Table 4).\nEffect of the Neurotrophic peptide mixture treatment on the state of the prooxidant-antioxidant system in patients with Parkinson’s disease\nThe intensification of prooxidant processes in blood plasma and erythrocytes of patients with moderate PD stages was registered. It was established a significant rise (by 40%) in the content of the secondary products of lipid peroxidation in blood plasma as well as a marked increase (by 50%) of the H2O2 formation in erythrocytes of patients with PD compared to healthy participants (Tables 5 and 6). Our assessment of the antioxidant protective effectiveness was largely based on the changes in activity of antioxidant enzymes – SOD and catalase, which represent the first line of such defense, providing detoxification of reactive oxygen metabolites27. In patients with PD, we have found an increase in the activities of SOD and catalase by 21% and 69%, respectively in comparison with healthy subjects. In addition, we analyzed the glutathione cycle, which is one of the main intracellular mechanisms for maintaining redox homeostasis28. In our study, the concentration of GSH was significantly decreased (by 58%) in erythrocytes of PD patients as well as GPx activity though this change was not statistically significant (by 15%) compared to healthy controls. We found that CBL administration has moderate antioxidant effects in PD patients by decreasing the TBARS accumulation (by 17%) in blood plasma as well as H2O2 content in erythrocytes (by 12%) compared with the CBL-untreated patients. These effects of CBL were accompanied by a decrease in the activities of SOD (by 11%) and catalase (by 20%) in blood of patients with PD compared with the CBL-untreated ones. In CBL-treated patients, there was registered an increase in the GSH concentration (by 30%) and GPx activity (by 15%) in erythrocytes compared with PD patients though this change was not statistically significant compared to the CBL-untreated patients (Tables 5 and 6).\nUPDRS score change prediction using multiple machine learning models\nAmong the models, GBMs demonstrated the highest predictive accuracy, achieving an RMSE of 18.59, MSE of 345.82, and MAE of 15.39. The top-performing RF model exhibited slightly higher RMSE (21.99) and MAE (17.78), underscoring the superior capability of GBMs to capture complex patterns in the data.\nThe boxplot of feature importance across the top 10 models provided valuable insights into which factors most significantly influenced the prediction of UPDRS score changes. Below are the most impactful features, sorted by median scaled importance: Fig. 17\n- \n                    BDNF Change (Median Importance: 0.98): Changes in Brain-Derived Neurotrophic Factor (BDNF) were the most influential predictor of UPDRS score change. BDNF’s high importance suggests that neurotrophic support may play a central role in the Neurotrophic peptide mixture’s therapeutic effects, potentially enhancing neuroplasticity and recovery. \n- \n                    MOCA (Median Importance: 0.62): The Montreal Cognitive Assessment (MOCA), a measure of cognitive function, was ranked as a highly predictive feature, highlighting the relevance of cognitive improvement as a response indicator for the Neurotrophic peptide mixture therapy. \n- \n                    BDNF (after treatment) (Median Importance: 0.61): Post-treatment levels of BDNF were also significant, indicating that maintaining or boosting neurotrophic factors is associated with positive therapeutic outcomes in PD patients. \n- \n                    MMSE, S (Median Importance: 0.58): The Mini-Mental State Examination (MMSE) score also emerged as a key predictor, reinforcing the importance of cognitive assessments in evaluating the Neurotrophic peptide mixture’s efficiency. \n- \n                    PINK1 (after treatment) (Median Importance: 0.55): PINK1, a mitochondrial health marker, was another significant predictor, suggesting that the Neurotrophic peptide mixture’s potential benefits may extend to cellular and mitochondrial health. \nThe boxplot analysis revealed that features related to neurotrophic support (BDNF) and cognitive function (MOCA, MMSE) showed the most consistent importance across models, as evidenced by their high median importance values and narrow interquartile ranges. In contrast, features with wider variability, such as TBA-active products and Age, had broader interquartile ranges, indicating that their contribution to predictive accuracy may depend on model-specific interactions or particular patient subgroups.\nDiscussion\nThe findings of this exploratory study suggest that the Neurotrophic peptide mixture has a multifaceted impact on various aspects of PD pathology, particularly regarding platelet ultrastructure, gene expression, neurophysiological parameters, and oxidative stress markers. These findings align with existing literature, which underscores the neuroprotective and neurorestorative properties of the Neurotrophic peptide mixture.\nImprovement in motor functions, particularly in the Motor tempo test, was more pronounced in patients with predominantly right-sided symptoms of PD. This asymmetry in response could be related to the lateralization of the pathological process. Additionally, the elevation in the threshold for triggering the H-reflex suggests modifications in spinal cord excitability, which could translate to improvements in motor function. The study reported statistically significant enhancements in cognitive abilities, as measured by the MMSE, FAB, and MoCA scales. Additionally, there was a notable reduction in depression scores (BDI) and reactive anxiety (Spielberger-Hanin scale), although personal anxiety levels remained unchanged. These findings highlight the potential of the Neurotrophic peptide mixture in ameliorating both motor and non-motor symptoms of PD. The improvements in motor function, cognitive abilities, and emotional well-being observed in this study suggest that the Neurotrophic peptide mixture has a broad therapeutic potential.\nThe study also observed significant neurophysiological improvements post-Neurotrophic peptide mixture treatment. The latency period of the P3 component, a marker of cognitive processing, was significantly reduced. This improvement in P3 latency indicates enhanced cortical function and suggests that the Neurotrophic peptide mixture may have beneficial effects on cognitive processing speed and efficiency in PD patients. These neurophysiological benefits align with the observed cognitive improvements, further supporting the potential of the Neurotrophic peptide mixture in mitigating cognitive decline in PD.\nIn PD patients, platelets exhibited several signs of senescence, including a reduction in dense δ-granules and an increase in α-granules29. The number of δ-granules per cell was significantly lower in PD patients (5.1 ± 0.4) compared to healthy individuals (8.4 ± 0.5). The administration of the Neurotrophic peptide mixture significantly increased the number of δ-granules in platelets by 45%, indicating enhanced serotonin reserves, which could contribute to the neuroprotective effects observed clinically. This increase in δ-granules likely reflects a normalization of platelet storage function, which is crucial given the role of serotonin in mood and neuroprotection. Despite no significant change in the number of α-granules, the trend towards a decrease might indicate an alteration in platelet activation status.\nAdditionally, the Neurotrophic peptide mixture reduced the calcium-accumulating capacity of platelets, potentially improving ATP synthesis and overall platelet function. This reduction in calcium accumulation is crucial as it indicates a shift away from the pathological state observed in untreated PD platelets. The observed improvements in platelet ultrastructure and function may translate to better overall cellular health, reflecting the systemic benefits of the Neurotrophic peptide mixture treatment beyond its direct neural effects. The enhancement in serotonin reserves and mitochondrial function underscores the multifaceted role of the Neurotrophic peptide mixture in addressing both neurological and peripheral manifestations of PD.\nThe Neurotrophic peptide mixture treatment appears to modulate BDNF gene expression in a sex-specific manner in patients with Parkinson’s disease, with women showing an increase and men showing a decrease in expression levels. This modulation is accompanied by improved cognitive function in women, highlighting the potential of the Neurotrophic peptide mixture as a therapeutic option that may benefit female PD patients more significantly. The stable expression of DJ-1, PINK1, and TNFα genes suggests that the primary neuroprotective mechanisms of the Neurotrophic peptide mixture in PD may involve pathways other than those mediated by these genes. These findings emphasize the need for further research to explore the underlying mechanisms of sex differences in PD treatment responses and to optimize therapeutic strategies accordingly.\nThe excessive production and release of ROS in PD occurs as a result of the combined action of mitochondrial dysfunction, dopamine autooxidation, and neuroinflammation30. In present study, the Neurotrophic peptide mixture demonstrated a moderate antioxidant effect, reducing the levels of lipid peroxidation products and H2O2 in erythrocytes. This reduction was accompanied by a decrease in SOD and catalase activities, and an increase in GSH concentration and GPx activity. These changes indicate a restoration of redox balance and a reduction in oxidative stress, which are crucial in the pathology of PD. One of the most promising effects of CBL on OS developing in blood and tissues in PD could be its ability to decrease intracellular α-synuclein aggregation in neurons, mimicking the action of classical NTFs and growth factors as well as of NTFs belonged to a novel family CDNF/MANF31. The ability of the Neurotrophic peptide mixture to modulate the antioxidant system suggests its potential role in protecting neuronal cells from oxidative damage and subsequent apoptosis.\nThe findings underscore the central role of cognitive and neurotrophic markers in predicting UPDRS score changes as an indicator of the Neurotrophic peptide mixture efficacy. The high importance of BDNF aligns with previous research on neuroprotective therapies, where neurotrophic support has been linked to improved neuronal survival and plasticity. The inclusion of mitochondrial markers, such as PINK1, suggests that the Neurotrophic peptide mixture may exert beneficial effects at the cellular level, addressing mitochondrial dysfunction—a known factor in PD pathology.\nThe boxplot analysis provided further clarity on the stability and reliability of each feature’s importance. Features with high median importance and narrow interquartile ranges, like BDNF and MOCA, indicate consistent predictive power across models, making them robust indicators of therapeutic response. The broader variability observed in other features, such as oxidative stress markers and demographic factors, suggests that while they may contribute to specific aspects of disease progression, they are less central to predicting therapeutic efficacy. The superior performance of GBMs in capturing non-linear relationships reinforces the value of using advanced machine learning techniques to analyze complex clinical datasets. These findings support the potential of the Neurotrophic peptide mixture as a multifaceted therapeutic option, with the capacity to improve cognitive and neurotrophic parameters in PD patients. The results of the present study indicate an antioxidant action of the Neurotrophic peptide mixture in patients with moderate stages of Parkinson’s disease. This action is mainly based on the decrease of the prooxidant events in peripheral blood of PD patients as well as on the prevention of thiol redox defects and reducing an imbalance in oxidant–antioxidant status.\nOverall, statistically significant improvements and consistent correlations were observed in the key clinical endpoints (UPDRS Part III and UPDRS S), while secondary and exploratory outcomes demonstrated patterns suggestive of mechanistic links that warrant confirmation in larger studies.\nInterpretation of its results is limited by several factors, including the small sample size, lack of a control group receiving placebo in addition to standard therapy, and a relatively short duration of the observation period. Since the study is preliminary, exploratory in nature and aimed at collecting primary data on the response to the drug, only one group of patients was formed, which allows identifying primary trends and assessing the feasibility of future larger studies. Future research should focus on larger cohorts and longer treatment durations to confirm these findings and to explore the long-term effects of neurotrophic peptide mixture on gene expression and cognitive function in PD patients. Particular attention should be paid to investigating sex differences in response to the drug in future studies.\nThe results of this exploratory study provide a preliminary rationale for investigating the Neurotrophic peptide mixture as an adjunct therapy in Parkinson’s disease. During the treatment period, changes were observed in platelet ultrastructure, gene expression, neurophysiological parameters, and oxidative stress markers. These findings, along with shifts in cognitive and neurotrophic markers, may suggest a multidimensional biological response.\nHowever, these observations must be interpreted with extreme caution. Given the limited sample size and the absence of a control group, no causal inferences can be drawn. Therefore, larger, placebo-controlled clinical trials are warranted to determine if these initial observations can be substantiated, to investigate potential sex-specific and long-term effects, and to explore biomarkers that may help elucidate the mechanisms of the Neurotrophic peptide mixture in PD.\nMethods\nPatient information\nThe exploratory study included 17 patients with Parkinson’s disease with an established diagnosis according to the inclusion-exclusion criteria of the British Brain Bank32, aged 45–74 years with disease stage 2.0–3.0 (2.9 ± 0.3, according to Hoehn and Yahr scale). Written informed consent was obtained from all participants.\nThe neurotrophic peptide mixture was prescribed against the background of basic stable doses of antiparkinsonian therapy (Levodopa-containing drugs), which were used during 28–30 days before starting the course of neurotrophic peptide mixture and during the entire study period. No dose adjustments were allowed during the study, and medication dose was not included as a covariate in the statistical analysis. The median daily Levodopa dose was about 500 mg, it was individually optimized depending on disease stage and the patient’s condition. The Neurotrophic peptide mixture dosage regimen was 20 ml (215.2 mg/ml) intravenously, once a day for 10 days. Infusions were carried out in the first half of the day.\nEthics statement and study design\nThe Ethics Committee of the Clinical Department of the Institute of Gerontology named after D.F. Chebotariov of the National Academy of Medical Sciences of Ukraine approved the protocol of current research (Protocol No. 4, 09/06/2022). Written informed consent was obtained from all participants prior to enrollment.\nThe study was conducted in accordance with the principles of the Declaration of Helsinki (2000) and the European Council Convention on Human Rights and Biomedicine (2005), and relevant national standards for the participation in the research were taken into account.\nClinical assessment\nCharacterization of the stages of PD was carried out using the Hoehn and Yahr scale. To assess the dynamics of motor disorders, the standardized international Unified Parkinson’s Disease Rating Scale (UPDRS) was used. For a comprehensive assessment of the state of non-motor functions, the following clinical scales were used:\n- \n                    Mini-Mental State Examination (MMSE); \n- \n                    Test battery for assessing frontal dysfunction (Frontal Assessment Battery, FAB); \n- \n                    Montreal Cognitive Assessment (MoCA); \n- \n                    Beck Depression Inventory (BDI); \n- \n                    Spielberger-Hanin Anxiety Scale. \nSensorimotor reaction and motor tempo\nMotor tempo was determined using a computer program as the time between consecutive presses of the same finger on two keys spaced 20 cm apart on the keyboard. The patient was instructed to press the keys at the fastest possible tempo. The duration of each test was 20 s. Testing was conducted for each hand separately, and then the data were averaged.\nThe latent period of simple sensorimotor reaction was determined as the interval between the moment of presentation of a visual signal on the computer monitor and the moment the patient pressed the key in response to the signal. Geometric shapes (circle, square, oval) of red and white colors were used as signals, with 20 signals presented at intervals of 2–4 s. Testing was conducted for each hand separately, and the data were averaged.\nElectroneuromyography\nThe technique of registering the H-reflex, as a reflex response of muscles to the stimulation of sensory fibers of a mixed nerve followed by monosynaptic activation of spinal motoneurons and nerve motor fibers, was performed using disposable surface recording electrodes placed in the projection of the posterior group of muscles of the calf, with stimulation of the tibial nerve in the popliteal fossa. The dynamics of the amplitudes of H- and M-responses (mV) with increasing strength of nerve electrical stimulation (mA), the threshold for the occurrence of the H-reflex (mA), the latency period of the H-reflex (ms), and the ratio of the maximum amplitude of H max/M max (%) were evaluated.\nEvoked potentials P300\nEEG was recorded monopolarly from 21 leads (Fp1, Fp2, Fpz, F3, F4, Fz, C3, C4, Cz, P3, P4, Pz, T3, T4, F7, F8, T5, T6, O1, O2, Oz). The combined ear electrode was used as a reference electrode. The interelectrode resistance was less than 5 kOhm, the signal sampling frequency was 500 Hz, and the high- and low-pass filters were 0.3 Hz and 35 Hz, respectively. Endogenous cognitive evoked potentials N2 and P3 were recorded according to the oddball paradigm using sound stimuli with a frequency of 1000 and 2000 Hz and a duration of 10 ms.\nPlatelet and mitochondrial structure with Electron microscopy\nBlood plasma enriched with platelets containing leukocytes was obtained by centrifugation of whole blood obtained from the ulnar vein of the subjects at room temperature for 15 min at 120 g in a T-30 centrifuge. The plasma was carefully separated from the sediment and centrifuged at 2000 g for 20 min using a Vortecs Combispin FVL-2400N minicentrifuge33.\nElectron microscopic studies of platelets and mitochondria were carried out according to the generally accepted method with double fixation with glutaraldehyde and OSO4, dehydration with alcohols of increasing concentrations and subsequent embedding in Epon. Ultrathin sections with a thickness of 40–60 nm, contrasted with uranyl acetate and lead citrate, were examined using an electron microscope PEM-125K (Ukraine).\nMorphometric calculations of platelets’ structure were performed using the Image Tool (USA) on 130–150 fields for each group of subjects. The number of optically dense small σ-granules and optically transparent large α-granules was determined.\nPBMC isolation and quantitative PCR\nBlood collection, PBMC isolation, RNA extraction and cDNA synthesis were performed by standard methods. The collection of blood for further cell isolation and qPCR was made on 0 and 11th day (after a 10-day course of neurotrophic peptide mixture treatment). Briefly, patients’ blood was collected in 4 ml vacutainers with EDTA. Density gradient centrifugation of diluted whole blood layered over a density gradient medium was used for PBMC isolation. PBMC were gradually frozen at a temperature of −20 °C, and then transferred to the temperature of −80 °C for long-term storage. RNA was isolated from cells using the Direct-zol RNA Miniprep Kit (Zymo Research, USA) following the manufacturer’s protocols. The isolated RNA was treated with DNase to remove DNA residues that could give a false positive signal for the expression of target genes. After this, reverse transcription was carried out using a commercial kit ZymoScript RT PreMix (Zymo Research, USA), guided by the manufacturer’s recommendations.\nSamples of this cDNA were used for further qPCR-RT (quantitative real-time polymerase chain reaction), which was used to measure the expression of the DJ-1, PINK1, BDNF and TNFα genes. To identify a suitable internal control for RT-qPCR data normalization, we empirically tested the expression stability of three candidate reference genes (rps18, gusb, and beta-actin). The sequences of all used primers are listed in the Supplementary Information. Analysis of the raw Cqt values revealed that both beta-actin and gusb were the most stably expressed genes, showing minimal variation and a strong positive correlation with each other. In contrast, rps18 expression was found to be significantly more variable across our samples and was therefore excluded. Although both beta-actin and gusb were suitable candidates, we selected beta-actin for normalization due to its slightly higher expression stability.\nThe sequences of all used primers are listed in the Supplementary Information. The PCR reaction mix was prepared using a commercial kit of reagents for RT-PCR (HOT FIREPol® EvaGreen® qPCR Mix Plus, Solis BioDyne, Estonia) - reverse transcription and polymerase chain reaction with the addition of betaine (Sigma, USA) at a final concentration of 1 M. To determine differences in mRNA expression of Parkin genes, Repeated Measures Anova was used.\nAccumulation curves were generated using Opticon Monitor 3 software. After amplification was completed, raw gene product amplification data were used to calculate Cy0.\nMeasurement of oxidative stress biomarkers\nSecondary products of lipid peroxidation, the activity of antioxidant enzymes superoxide dismutase (SOD) and catalase (CAT) were measured in blood plasma. In red blood cells – the content of hydrogen peroxide (H2O2); the state of the glutathione system was assessed by the content of reduced glutathione (GSH) and the activity of the enzyme glutathione peroxidase (GP).\nFor erythrocyte isolation, samples were centrifuged at 8000 rpm for 15 min to separate plasma and red blood cells. Using centrifugation, erythrocytes were washed twice of plasma with cold sterile physiological sodium chloride solution. Erythrocyte hemolysate was obtained by adding distilled water to washed erythrocytes in a 1:1 ratio. The hemoglobin (Hb) content was also measured (Hemoglobin Assay kit MAK115, Sigma-Aldrich, St. Louis, MO, USA).\nDetermination of thiobarbituric acid reactive substances (TBARS) levels: TBARS were isolated by boiling plasma with the thiobarbituric acid reagent (0.5% 2-thiobarbituric acid/10% trichloroacetic acid/0.63 mM hydrochloric acid) and measuring the absorbance at 532 nm. The results are expressed as µM/ml using a molar extinction coefficient 1.56 * 105 mm−1 cm−1.\nCatalase activity measurement: The method is based on the ability of molybdenum salts to form a colored complex with hydrogen peroxide, which was determined at a wavelength of 410 nm. Enzyme activity (μM/min/L) was calculated using a molar extinction coefficient of 22.2 * 103 mm−1cm−1.\nSuperoxide dismutase (SOD) activity was determined spectrophotometrically. The method is based on the ability of SOD to inhibit the autoxidation of adrenaline at pH 10.2. The results were expressed as specific activity of the enzyme in units per ml of plasma. One unit of SOD activity is defined as the amount of protein causing 50% inhibition of the conversion rate of adrenaline to adrenochrome under specified conditions.\nThe activity of selenium-dependent glutathione peroxidase (GPx) was determined by recording the rate of NADPH oxidation in the presence of reduced glutathione at a wavelength of 340 nm. Enzyme activity was expressed in nM NADPH/min/ml.\nThe amount of reduced glutathione (μM/ml) was determined using Elman’s reagent (5.5’-dithio-bis-nitrobenzoic acid in absolute ethanol). The change in extinction was recorded at a wavelength of 412 nm every minute for 5 minutes.\nHydrogen peroxide content was studied using the FOX method, which is based on the peroxide-mediated oxidation of Fe2 + , followed by the reaction of Fe3+ with xylenol orange. Absorbance of the Fe3 + -xylenol orange complex was determined at a wavelength of 560 nm. The amount of H2O2 (μM/mgHb) was calculated using the H2O2 standard curve for each independent experiment.\nMachine learning prediction of UPDRS score change\nTo model UPDRS score changes, H2O’s AutoML framework was employed34, allowing for the training, tuning, and evaluation of multiple Machine Learning models, including Gradient Boosting Machines (GBMs) and Random Forests (RFs). AutoML facilitated hyperparameter optimization, ensuring robust and accurate predictions.\n- \n                    Gradient Boosting Machines (GBMs): GBMs iteratively build a series of decision trees, where each subsequent tree focuses on correcting errors made by its predecessors. This sequential approach enables GBMs to capture intricate, non-linear relationships within the data, making them particularly suitable for analyzing how clinical and biochemical factors impact UPDRS score changes and, consequently, therapeutic response35. \n- \n                    Random Forests (RFs): RFs are ensemble models that combine multiple decision trees, each trained on a random subset of the data. Known for their robustness and reduced susceptibility to overfitting, RFs provide valuable insights into feature importance by averaging across numerous trees36. This model serves as a stable baseline to complement the more flexible GBMs. \nThe models were evaluated using the following metrics to quantify prediction accuracy:\n- \n                    Root Mean Squared Error (RMSE): Captures the average deviation between predicted and actual values. \n- \n                    Mean Squared Error (MSE): Measures the variance of predictions relative to actual outcomes. \n- \n                    Mean Absolute Error (MAE): Reflects the average absolute deviation between predicted and observed values. \nThese metrics allowed for a thorough assessment of each model’s ability to predict UPDRS score changes, highlighting the Neurotrophic peptide mixture’s potential impact on disease progression.\nFeature importance was evaluated using the top 10-performing models from a set of 50 trained with H2O AutoML. Each feature’s importance was measured by its “scaled importance,” which represents its relative contribution to prediction accuracy. The models were trained using two-fold cross-validation, and feature importance was extracted during the validation stage.\nData from the top 10 models were aggregated, and features were sorted by their median feature importance. It is worth noting that while building a generalized model on such a small dataset is not feasible, the features identified with the highest median importance could be considered potential markers in future studies. The results were visualized using a boxplot (Fig. 16) to highlight the median importance and variability of each feature.\nStatistical analysis\nHypotheses about the effects of the drug were tested using the Wilcoxon (for related samples) and Mann-Whitney (for independent groups) tests. The relationship between the investigated indicators was determined in the form of Spearman’s rank correlation coefficients.\nStatistical processing of gene expression data was performed using Statistica 8.0 software (StatSoft Inc., USA). For statistical analysis, Spearman’s non-parametric test was used to investigate the correlation between parameters. The difference in expression of the studied genes before and after treatment was determined using the Wilcoxon t-test (Wilcoxon signed-rank test).\nThe results of the study of the state of the prooxidant-antioxidant system were processed statistically using the Origin program. The significance of differences between comparison groups was determined by the method of analysis of variance (ANOVA) followed by the Bonferroni test (post-hoc test).\nOutcomes and hierarchy\nFor this exploratory study, outcomes were grouped into three hierarchical levels.\n- \n                    Clinically central endpoints were prespecified as the UPDRS III (motor score) and the UPDRS Total score, as the primary efficacy measures, as these are the most widely accepted, clinically relevant, and objective indicators of motor impairment and overall disease severity in Parkinson’s disease. \n- \n                    Secondary clinical endpoints included UPDRS I (non-motor experiences of daily living) and UPDRS II (motor experiences of daily living), and validated scales of non-motor function: MMSE, MoCA, FAB (cognitive measures), and BDI and Spielberger–Hanin Anxiety Scale (mood and affective measures). These outcomes were considered supportive of the clinically central endpoints. \n- \n                    Exploratory outcomes included additional non-motor measures, neurophysiological parameters (P300 latency and amplitude), and molecular/structural biomarkers (gene expression, platelet ultrastructure, mitochondrial morphology, and oxidative stress markers). These outcomes were included to generate mechanistic insights into the potential modes of action of the Neurotrophic peptide mixture. \nFormal hypothesis testing was restricted to the clinically central endpoints. Secondary and exploratory outcomes are presented with descriptive statistics and effect sizes. To minimize Type I error, no multiple-testing correction was applied outside of the clinically central endpoints.\nCorrelation analyses were performed using Spearman’s rank correlation coefficient (ρ). For correlations involving clinically central endpoints, p-values are reported. Correlations among secondary and exploratory outcomes are reported without formal inferential testing, to avoid Type I error inflation.\nData availability\nThe datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.\nReferences\n- Alexoudi, A., Alexoudi, I. & Gatzonis, S. Parkinson’s disease pathogenesis, evolution and alternative pathways: A review. Rev. Neurol. (Paris) 174, 699–704 (2018). \n- Chung, S. J. et al. The Pattern of Striatal Dopamine Depletion as a Prognostic Marker in De Novo Parkinson Disease. Clin. Nucl. Med 201843, 787–792 (2018). \n- Moon, H. E. & Paek, S. H. Mitochondrial Dysfunction in Parkinson’s Disease. Exp. Neurobiol. 24, 103–116 (2015). \n- Mansur, A. et al. Characterization of 3 PET tracers for quantification of mitochondrial andsynaptic function in healthy human brain:18F-BCPP-EF,11C-SA-4503, and11C-UCB-J. J. Nucl. Med 61, 96–103 (2020). \n- Simonenkov, A. P. & Kluyev, V. M. Syndrome serotonin deficiency. M.: BINOM, 96-97 (2013). \n- Mauricio, R. et al. Tackling gaps in developing life-changing treatments for dementia. Alzheimers Dement 5, 241–253 (2019). \n- Gonchar, O. O., Klymenko, O. O., Drevytska, T. I., Bratus, L. V. & Mankovska, I. M. Oxidative stress in rat heart mitochondria under a rotenone model of Parkinson’ disease: a corrective effect of capicor treatment. Ukr. Biochem J. 93, 21–30 (2021). \n- Schneider-Garces, N. J. et al. Span, CRUNCH, and beyond: working memory capacity and the aging brain. J. Cogn. Neurosci. 22, 655–669 (2010). \n- Fiani, B. et al. Cerebrolysin for stroke, neurodegeneration, and traumatic brain injury: review of the literature and outcomes. Neurol. Sci. 42, 1345–1353 (2021). \n- Brainin, M. Cerebrolysin: a multi-target drug for recovery after stroke. Expert Rev. Neurother. 18, 681–687 (2018). \n- Requeejoe, C. et al. Nanodelivery of Cerebrolysin and rearing enriched environment induce neuroprotective effects in a preclinical rat model of Parkinson’s disease. Mol Neurobiol. Springer Science+Business Media, LLC (2017). \n- Neveen, A. et al. A promising therapeutic potential of Cerebrolysin in the 6-OHDA rat model of Parkinson's disease. Life Sci 155, 174–179 (2016). \n- Luchanina, E., Karaban, I. & Berezetskaya, N. Diagnosis of Parkinsonas disease by electrophysiological methods. Diagn. Rehabil. Parkinson. Dis. 2, 27–58 (2011). \n- Eder, P., Reinprecht, I., Schreiner, E., Skofitsch, G. & Windisch, M. Increased density of glutamate receptor subunit 1 due to Cerebrolysin treatment: an immunohistochemical study on aged rats. Histochem J. 33, 605–612 (2001). \n- Stepanichev, M. et al. Effects of cerebrolysin on nerve growth factor system in the aging rat brain. Restor. Neurol. Neurosci. 35, 571–581 (2017). \n- Veinbergs, I., Mante, M., Mallory, M. & Masliah, E. Neurotrophic effects of Cerebrolysin in animal models of excitotoxicity. J. Neural Transm. Suppl. 59, 273–280 (2000). \n- Alvarez, X. A. et al. Neuroimmunotrophic effects of cerebrolysin in an animal model of hippocampal degeneration induced by β-amyloid. In: Korczyn, A. D., ed. International Congress on Vascular Dementia 2017. Bologna, Italy: Monduzzi Editore, 233–237 (1999). \n- Gavrilova, S. I. & Alvarez, A. Cerebrolysin in the therapy of mild cognitive impairment and dementia due to Alzheimeras disease: 30 years of clinical use. Med Res Rev. 41, 1–29 (2020). \n- DeBoer, S. R. et al. Enhanced spontaneous motor recovery after stroke in mice treated with cerebrolysin. Neurorehabil Neural Repair 35, 525–533 (2021). \n- Muresanu, D. F. et al. Efficacy and safety of cerebrolysin in neurorecovery after moderate-severe traumatic brain injury: results from the CAPTAIN II trial. Neurol. Sci. 41, 1171–1181 (2020). \n- Lu, W. et al. Cerebrolysin alleviates early brain injury after traumatic brain injury by inhibiting neuroinflammation and apoptosis via TLR signaling pathway. Acta Cir. Bras. 37, e370605 (2022). \n- Chen, C. C., Wei, S. T., Tsaia, S. C., Chen, X. X. & Cho, D. Y. Cerebrolysin enhances cognitive recovery of mild traumatic brain injury patients: double-blind, placebo-controlled, randomized study. Br. J. Neurosurg. 27, 803–807 (2013). \n- Cui, S. Cerebrolysin for vascular dementia. Cochrane Database Syst Rev. 11, 19–20 (2019). \n- Rochenstein, E. et al. Neuropeptide treatment with cerebrolysin enhances the survival of grafted neural stem cells in a synuclein transgenic model of Parkinson’s disease. J. Exp. Neurosci. 9, 131–140 (2015). \n- Hünerli, D. et al. Mild cognitive impairment in Parkinson’s disease is associated with decreased P300 amplitude and reduced putamen volume. Clin. Neurophysiol. 130, 1208–1217 (2019). \n- Alvarez, X. A., Lombardi, V. R., Fernández-Novoa, L., García, M. & Sampedro, C. Cerebrolysin reduces microglial activation in vivo and in vitro: a potential mechanism of neuroprotection. J. Neural Transm. 59, 281–292 (2000). \n- Kulinskiĭ, V. I. & Kolesnichenko, L. S. Glutathione system. I. Synthesis, transport, glutathione transferases, glutathione peroxidases. Biomed. Khim 55, 255–277 (2009). \n- Zenkov, N. K., Lankin, V. Z. & Menshikova, E. B. Oxidative stress: The biochemical and pathophysiological aspects. M.: MAIK. Science / Interperiodica 4, 352–361 (2001). \n- Park, S. M. et al. Evidence that alpha-synuclein functions as a negative regulator of Ca(++)-dependent alpha-granule release from human platelets. Blood 100, 2506–2514 (2002). \n- Luo, Y., Hoffer, A., Hoffer, B. & Qi, X. Mitochondria: A therapeutic target for Parkinson’s disease?. Int J. Mol. Sci. 16, 20704–20730 (2015). \n- Voutilainen, M. H., Arumae, U., Airavaara, M. & Saarma, M. Therapeutic potential of the endoplasmic reticulum located and secreted CDNF/MANF family of neurotrophic factors in Parkinson’s disease. FEBS Lett. 589, 3739–3748 (2015). \n- Hughes, A. J., Daniel, S. E., Kilford, L. & Lees, A. J. Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: A clinico-pathological study of 100 cases. J. Neurol. Neurosurg. Psychiatry 55, 181–184 (1992). \n- Filippova, O. I., Koloskov, A. V. & Stolitsa, A. A. Methods for studying the functional activity of platelets (literature review). Transfusiology 13, 493–514 (2012). \n- LeDell, E. & Posse, C. H2O AutoML: Scalable Automatic Machine Learning. J. Open Source Softw. 5, 1–10 (2020). \n- Chen, J. & Guestrin, C. XGBoost: A Scalable Tree Boosting System. Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery Data Mining, 785–794 (2016). \n- Liaw, A. & Wiener, M. Classification and regression by random Forest. R. N. 2, 18–22 (2002). \nAcknowledgements\nThis work was supported by the National Academy of Medical Sciences, Ukraine (Grant # 0123U100771).\nAuthor information\nAuthors and Affiliations\nContributions\nD.K. took part in writing of the main text of manuscript, coordinated the work of scientific laboratories in their experiments and results, contributed to the conceptualization of the project, I.K. was the lead scientist of part connected to clinical assessment, data analysis, project conception and organization, N.K., N.M. and S.K. performed and analyzed electroencephalograms and elecroneuromyograms of patients, collected patient data regarding sensorimotor reactions and motor tempo, K.R. participated in the preparation of the literature review, implemented the methodology of morphological studies, processing the data of the section about the morphological parameters. O.G. examined markers of oxidative stress and made substantial contributions to the analysis of data as well as participated in a critical review of the manuscript, I.M. helped with the preparation of the manuscript and analysis of data. O.Z. wrote an abstract, implemented the main methodology for gene expression measurement, analysis and interpretation of data, O.B. and K.M. were involved to gene expression experiments and analysis of qPCR-RT data, as well as collection of biological material, S.S. provided the results of deep analysis of obtained data using multiple machine learning models, performed feature importance analysis, T.P. contributed to the administration of project, wrote and edited the main manuscript text. All authors reviewed the manuscript.\nCorresponding author\nEthics declarations\nCompeting interests\nThe authors declare no competing interests.\nAdditional information\nPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.\nRights and permissions\nOpen Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.\nAbout this article\nCite this article\nKrasnienkov, D., Karaban, I., Karasevych, N. et al. Evaluation of the neurotrophic peptide mixture in pathogenetic therapy of patients with Parkinson’s disease. npj Parkinsons Dis. 12, 55 (2026). https://doi.org/10.1038/s41531-026-01270-6\n- Received: \n- Accepted: \n- Published: \n- Version of record: \n- DOI: https://doi.org/10.1038/s41531-026-01270-6","reading_time_min":42,"extraction":{"state":"ok","confidence":0.9,"error":null,"explanation":"High confidence: full text extraction produced 58549 characters.","diagnostics_url":"/api/diagnose?url=https%3A//www.nature.com/articles/s41531-026-01270-6","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 58549 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":58549,"summary_length":313,"usable_text_length":58549,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":58549,"summary_length":313}}},"quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 58549 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":58549,"summary_length":313,"usable_text_length":58549,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":58549,"summary_length":313}},"actions":{"read":"/item/34350","export_markdown":"/api/items/34350/export?format=markdown","export_json":"/api/items/34350/export?format=json","diagnose":"/api/diagnose?url=https%3A//www.nature.com/articles/s41531-026-01270-6"}},"digest":{"id":34350,"title":"Evaluation of the neurotrophic peptide mixture in pathogenetic therapy of patients with Parkinson’s disease - Nature","url":"https://www.nature.com/articles/s41531-026-01270-6","source":"Nature","topic":"peptides","published_at":"2026-01-23T08:00:00+00:00","excerpt":"Abstract This exploratory, single-group, open-label study investigated 17 patients with Parkinson’s disease (PD) using a pre-post design. Motor and non-motor outcomes were assessed through clinical scales, biochemical and genetic analyses, and machine learning models (Gradient…","quality_bucket":"high","quality_reason":"High confidence: full text extraction produced 58549 characters.","reading_time_min":42,"cluster_id":null},"card":{"display_title":"Evaluation of the neurotrophic peptide mixture in pathogenetic therapy of patients with Parkinson’s disease - Nature","subtitle":"Nature · 2026-01-23","summary":"Abstract This exploratory, single-group, open-label study investigated 17 patients with Parkinson’s disease (PD) using a pre-post design. Motor and non-motor outcomes were assessed through clinical scales, biochemical…","badges":["quality:high"],"links":{"read":"/item/34350","original":"https://www.nature.com/articles/s41531-026-01270-6","diagnose":"/api/diagnose?url=https%3A//www.nature.com/articles/s41531-026-01270-6"},"quality_warning":null},"export":{"title":"Evaluation of the neurotrophic peptide mixture in pathogenetic therapy of patients with Parkinson’s disease - Nature","url":"https://www.nature.com/articles/s41531-026-01270-6","summary":"Abstract\nThis exploratory, single-group, open-label study investigated 17 patients with Parkinson’s disease (PD) using a pre-post design. Motor and non-motor outcomes were assessed through clinical scales, biochemical and genetic analyses, and machine learning models (Gradient Boosting Machines, Random Forests).","source":"Nature","date":"2026-01-23T08:00:00+00:00","content":"Abstract\nThis exploratory, single-group, open-label study investigated 17 patients with Parkinson’s disease (PD) using a pre-post design. Motor and non-motor outcomes were assessed through clinical scales, biochemical and genetic analyses, and machine learning models (Gradient Boosting Machines, Random Forests). After treatment with a neurotrophic peptide mixture, improvements were observed in daily activity (16%), cognition (11%), depression (10% reduction), and reactive anxiety (23% reduction). Biological changes included a 45% increase in platelet δ-granules, higher mitochondrial counts, elevated gene expression (notably BDNF in women, p = 0.046), and modulation of oxidative stress markers (17% reduction in TBARS, 30% increase in GSH). Machine learning identified BDNF and PINK1 expression, along with MOCA and MMSE scores, as key predictors of UPDRS improvement. These findings suggest that neurotrophic peptide therapy may influence clinical, structural, and molecular domains in PD. Larger, controlled trials are warranted to confirm therapeutic potential and clarify associations with cognitive and neurotrophic parameters.\nSimilar content being viewed by others\nIntroduction\nParkinson’s disease is a neurodegenerative condition typically observed in older individuals, characterized by a gradual onset and progressive nature. The initial manifestation often includes tremors, followed by symptoms such as slowed movement (bradykinesia) and muscle stiffness (rigidity)1. The disease is linked to the degeneration of dopaminergic neurons in the substantia nigra, the presence of alpha-synuclein aggregates within neuronal cells (Lewy bodies and Lewy neurites), and subsequent pathology in regions like the basal forebrain, amygdala, and medial temporal lobe structures, progressing to cortical areas in advanced stages2.\nMitochondrial abnormalities and dysfunction are recognized as key contributors to Parkinson’s disease pathogenesis in a subset of patients3. Some data was obtained on the availability correlation of the functional activity of mitochondria in lymphocytes, which displays multisystem mitochondrial failure and indicates of energy dysfunction in general4. The spectrum of mitochondrial dysfunctions in PD includes dysfunction of the respiratory chain (inhibition of the NADH-oxidase oxidation pathway), hyperproduction of ATP, disruption of Ca-regulatory systems, opening of the mitochondrial pore, perturbations of mitochondrial dynamics, and dysregulation of mitochondrial clearance5.\nAlthough oxidative stress plays a significant role in the pathogenesis of PD, therapy based on the use of exogenous antioxidants has not been shown to be highly effective. An alternative strategy that can counteract the damaging effects of free radicals and restore cellular redox balance is the activation of endogenous antioxidant enzymes present in the central neural system, such as superoxide dismutase, catalase, and glutathione peroxidase. The transcription of these cytoprotective proteins is regulated by the transcription factor Nrf2, which plays a leading role in the regulation of cellular redox status and is the main regulator of the cellular response to oxidative stress through the induction of antioxidant and detoxifying enzymes and proteins6. This is consistent with our previous data, which showed that lipid peroxidation was significantly increased while reduced glutathione (GSH) and glutathione-dependent antioxidant enzymes activity markedly decreased in blood of patients with PD as well as in rat brain under modeling of PD7. Various activators of the Nrf2/ARE signaling pathway, including peptides, have shown their effectiveness in both in vitro and in vivo models of neurological disorders8.\nThe Neurotrophic peptide mixture is peptidergic mixture with neurotrophic-like properties, made from a mixture of low molecular weight, porcine-derived peptides and free amino acids. The pharmaceutical preparation includes the peptide fragments: nerve growth factor (NGF), brain-derived neurotrophic factor (BDNF), ciliary neurotrophic factor (CNTF), enkephalins, orexin, and P219.\nThe use of the Neurotrophic peptide mixture significantly reduces neurological deficits, improves cognitive abilities, and reduces the volume of the infarction zone during stroke10,11. The Neurotrophic peptide mixture has an organ-specific multimodal effect on the brain, i.e., it provides metabolic regulation, neuroprotection, functional neuromodulation, and neurotrophic activity12. The Neurotrophic peptide mixture protects neurons from the damaging effects of lactic acidosis, prevents the formation of free radicals13, increases survival and prevents neuronal death under conditions of hypoxia and ischemia, and reduces the damaging neurotoxic effect of glutamate14.\nOver the past decade, compelling evidence has emerged neuroprotective properties of the Neurotrophic peptide mixture, explaining its multi-level impact on both the pathological cascade of damage and on subsequent recovery processes15. The ability of the Neurotrophic peptide mixture to influence the processes of neuroprotection and neuroplasticity was proven in a series of studies15,16,17 on a model of Alzheimer’s disease in transgenic mice for the amyloid precursor protein (mThy-1-hAPP751 tg). Using a model of Alzheimer’s disease, it was shown that the Neurotrophic peptide mixture, through the regulation of the synthesis and transport of amyloid precursor protein and the deposition of β-amyloid in synapses, has a synaptotrophic effect, which plays a primary role in the processes of neuroplasticity, which are extremely important in the acquisition of new skills and the maintenance of existing ones18.\nIncreasing studies have demonstrated that the Neurotrophic peptide mixture administration can promote recovery of motor function, improve early brain injury, and decrease hippocampal neuronal death19. In recent clinical studies, the Neurotrophic peptide mixture improved overall outcomes after moderate to severe traumatic brain injury and was also safe and better for early rehabilitation patients after ischemic stroke20. Lu et al. reported that the Neurotrophic peptide mixture can improve neurological dysfunction and brain damage, relieve neuroinflammation and prevent apoptosis after brain injury21.\nBased on the multifaceted mechanism of action of the Neurotrophic peptide mixture, including its effect on the processes of neurogenesis, neuroplasticity and neuroprotection, the drug currently has three main indications for use - stroke, dementia and traumatic brain injury10,22,23.\nLuchanina et al. demonstrated for the first time the effect of Neurotrophic peptide mixture on motor activity and muscle tone in patients with PD treated with levodopa13. It was established that the Neurotrophic peptide mixture improved the engraftment of a DOPA transplant from stem cells in the substantia nigra24 and reduced the severity of symptoms in model animals12.\nThe current exploratory study dataset comprised clinical, cognitive, biochemical, and demographic information from PD patients undergoing the Neurotrophic peptide mixture therapy. Key features included cognitive assessments (MMSE, MOCA), neurotrophic markers (BDNF, PINK1), oxidative stress indicators (SOD, DJ1), and demographic factors (age, sex). The primary outcome was the change in UPDRS scores, serving as an objective measure of therapeutic response to the Neurotrophic peptide mixture. Also, in this study, machine learning models were employed to predict UPDRS score changes as a key indicator of the Neurotrophic peptide mixture’s therapeutic effectiveness, offering a data-driven evaluation of treatment response.\nResults\nClinical and neurophysiological assessment of the effectiveness of course treatment\nA course of the Neurotrophic peptide mixture administration against the background of basic antiparkinsonian therapy had a beneficial effect on the condition of the majority of patients, primarily on their activity in everyday life and the ability to self-care. Objectively, this effect was expressed in a significant decrease in scores of all UPDRS subscales (~16% on average) (Fig. 1).\nBoth clinically central endpoints improved significantly (p < 0.01). Secondary measures (UPDRS I and UPDRS II) generally trend in the same direction; effect sizes are provided without formal inference.\nA beneficial impact was observed following administration of a regimen of the Neurotrophic peptide mixture on a spectrum of non-motor dysfunctions concomitant with the motor manifestations of Parkinson’s disease (PD). Participants exhibited statistically significant enhancements in cognitive abilities as assessed by specialized instruments including MMSE, FAB, and MoCA (~11% on average). Moreover, there was a notable reduction in scores indicative of depression measured via the Beck Depression Inventory scale (~10% on average), as well as in reactive anxiety assessed by the Spielberger-Hanin scale (~23% on average). Notably, levels of personal anxiety remained consistent relative to baseline measurements. As a result, all observed Secondary outcomes are trended in the direction of improvement (Fig. 2).\nFor instrumental assessment of the Neurotrophic peptide mixture effect on motor functions, the Motor tempo test showed higher sensitivity. As a result, the significant decrease in the average, minimum and maximum time of motor reactions was recorded (Fig. 3).\nThose changes are more characteristic of the reactions of the right hand, while for the left hand only the average value of the motor tempo decreased. It can be assumed that this asymmetry of effects is associated with the sidedness of the pathological process, since a significant part of patients (76% of the study group) demonstrated predominantly right-sided symptoms of PD. These findings show a valuable positive impact of Neurotrophic peptide mixture on neurophysiological parameters.\nAt the same time, the latent period of a simple sensorimotor reaction did not change during a course of the Neurotrophic peptide mixture administration (Fig. 3).\nChanges of the H-reflex and M-response after administration of a regimen of Сerebrolysin\nIn the context of administering the Neurotrophic peptide mixture, there was an elevation observed in the threshold for triggering the H-reflex of the gastrocnemius muscle, increasing from 5 [3; 9] mA to 6 [5.8; 18] mA (Fig. 4). Notably, there were no alterations noted in the latency period or power attributes of the H-reflex, nor in the parameters of the M-response, compared to their baseline values.\nPotentials P300\nAnalysis of endogenous components revealed that in the studied group of patients with PD, there was a delay in the formation of the P3 wave (median 434 ms, interquartile range 386–512 ms in the temporal area of the neocortex) compared to the corresponding age norm of 390–410 ms. After the course of the Neurotrophic peptide mixture, a statistically significant increase in the latency period of the P3 component in the central-parietal lead was recorded, up to 394 [304; 446] ms (Fig. 5). There were no significant changes in the latency period of the N2 component and the amplitudes of the N2 and P3 components.\nThe delay in the formation of the P3 wave is a characteristic feature of several neurodegenerative diseases, particularly Alzheimer’s and Parkinson’s diseases. For patients with borderline PD and moderate cognitive impairments, an increase in the latency period of P3 has been described, with a reverse correlation between this parameter and the volume of gray matter in the globus pallidus, dysfunction of which leads to symptoms such as bradykinesia and rigidity25. Improvement in this parameter following the course of the Neurotrophic peptide mixture suggests that a more pronounced effect on the state of higher cortical functions can be expected in patients with increased risk of developing or progressing cognitive impairments.\nCorrelation analysis of clinical and neurophysiological parameters after a course of treatment with the Neurotrophic peptide mixture\nThe degree of improvement in several non-motor characteristics (cognitive functions assessed by the FAB scale, amplitude of cognitive evoked potential P3, and signs of depression measured by the Beck scale) was greater in individuals who initially received a higher daily dose of levodopa (Figs. 6–8).\nFor indicators of cognitive functions (MMSE, P3 cognitive evoked potential amplitude), a similar relationship with the time of onset of the disease was also found: more pronounced positive effects of the Neurotrophic peptide mixture treatment were observed in patients with an earlier onset of PD and a longer duration of the disease (Figs. 6, 8).\nAt the same time, a more pronounced reduction in reaction time was characteristic of individuals with lower initial ratings of PD severity on the UPDRS scale and lower levels of anxiety. Such correlation relationships are found both when directly assessing the time of motor reactions in the Motor tempo and Sensorimotor reaction tests, and when determining the latent periods of the H-reflex and M-response (Figs. 9, 10).\nThe use of the Neurotrophic peptide mixture in the treatment course improves the general condition of patients with Parkinson’s disease, as evidenced by a significant decrease in the quantitative indicators of all UPDRS subscales, attenuation of affective disorders such as anxiety and depression, and improvement in cognitive function. Against the background of the Neurotrophic peptide mixture administration, dynamic characteristics of simple motor reactions improve, and the threshold for the emergence of the H-reflex increases, indicating facilitation of supraspinal control processes over the activity of motor neurons. In general, the results show a strong tendency to improvement in different parameters.\nMitochondrial dysfunction before and after treatment with the Neurotrophic peptide mixture in patients with Parkinson’s disease\nThe conducted studies revealed significant differences in the structural organization of platelets in individuals with PD examined before the start of the standard treatment process. Electron microscopic and morphometric study of platelets in patients with PD revealed that platelets were mainly represented by senescent cells. They had a reduced number of dense δ granules, which was 5.1 ± 0.4 per one cell (in healthy individuals - 8.4 ± 0.5 units per one cell, p < 0.05). However, the number of transparent α-granules turned out to be increased from 5.3 ± 0.2 up to 10.3 ± 0.6 per cell. In addition, the latter often had large sizes (>100 nm) and contained a significant amount of protein debris (Fig. 11).\nThe single mitochondria that were found in the cells (2–4 per cell) were structurally quite preserved and had an average diameter of 0.48 ± 0.02 µm, not significantly different from those determined in the control group (0.41 ± 0.04 μm).\nSignificant abnormalities were found in the ultrastructure of mitochondria in platelets of patients. In addition to increased mitochondrial membrane permeability (outer and/or inner, crystas-forming), the formation of septated organelles was observed. The mechanisms and causes of their formation are still unexplored. It is assumed that septated mitochondria can accumulate Са2+26. Since it is known that the accumulation of Са2+ and the synthesis of ATP are alternative processes associated with the transfer of electrons, an increase in the Ca-accumulating ability of mitochondria may indicate a decrease in the synthesis of ATP in them. In addition, the appearance of septated mitochondria may indicate partial cell necrosis. These ultrastructural changes are evidence of the formation of mitochondrial dysfunction in platelets.\nThe use of the Neurotrophic peptide mixture in the complex therapy of patients with CP was accompaniedby reliable dynamics of the morphofunctional state of platelets (Table 1). The use of the Neurotrophic peptide mixture led to a significant (by 45%) increase in the number of δ-granules. No significant changes in the number of α granules were detected; only a tendency to its decrease was observed. An increase in the number of δ-granules indicates an increase in serotonin reserves. Probably, one of the possible mechanisms of the neuroprotective effect of the Neurotrophic peptide mixture is realized in this way. In addition, the Ca2+ -accumulating capacity in thrombocytes is relatively reduced, which should contribute to the improvement of ATP synthesis.\nDespite the small number of mitochondria in platelets, a slight increase in their number was observed - from 2 to 4 organelles before treatment to 5–7 organelles after treatment with the Neurotrophic peptide mixture, which, in combination with the absence of detection of septated organelles, may indicate a decrease in the severity of mitochondrial dysfunction.\nThus, treatment with the Neurotrophic peptide mixture improves the morphofunctional state of platelets in patients with PD, it increases serotonin reserves, reduces Ca2+ -accumulating capacity, reduces the manifestations of mitochondrial dysfunction, and exhibits mild antithrombotic properties.\nEffect of the Neurotrophic peptide mixture treatment on PARKIN gene expression in patients with Parkinson’s disease\nFor the whole group of tested individuals, taken together, no correlation of the expression of the measured genes was shown using the Spearman test. At the same time, when dividing by sex, multidirectional differences in gene expression before and after therapy with the drug Neurotrophic peptide mixture were found (Table 2).\nIn particular, the group of men was characterized by a decrease in the expression of the BDNF gene, and in women the expression increased after the use of the drug (Fig. 12).\nIt is important to note that when comparing the results of gene expression with cognitive parameters, an increase in BDNF in women (but not in men) was also accompanied by an increase in the MMSE score (Fig. 13).\nNo statistically significant difference was observed in expression of other three genes (DJ-1, PINK1, TNFalfa) before and after treatment with the Neurotrophic peptide mixture (Figs. 14–16).\nThe correlation analysis shows higher values between MMSE and PINK1 expression, FAB a and DJ1, FAB a and PINK1, MOCA a and DJ1, MOCA a and PINK1, MOCA and PINK1 for men (Table 3).\nFor women, the correlation analysis shows strong values of clinically central endpoints UPDRS III and UPDRS S with almost all studied genes expression rates (Table 4).\nEffect of the Neurotrophic peptide mixture treatment on the state of the prooxidant-antioxidant system in patients with Parkinson’s disease\nThe intensification of prooxidant processes in blood plasma and erythrocytes of patients with moderate PD stages was registered. It was established a significant rise (by 40%) in the content of the secondary products of lipid peroxidation in blood plasma as well as a marked increase (by 50%) of the H2O2 formation in erythrocytes of patients with PD compared to healthy participants (Tables 5 and 6). Our assessment of the antioxidant protective effectiveness was largely based on the changes in activity of antioxidant enzymes – SOD and catalase, which represent the first line of such defense, providing detoxification of reactive oxygen metabolites27. In patients with PD, we have found an increase in the activities of SOD and catalase by 21% and 69%, respectively in comparison with healthy subjects. In addition, we analyzed the glutathione cycle, which is one of the main intracellular mechanisms for maintaining redox homeostasis28. In our study, the concentration of GSH was significantly decreased (by 58%) in erythrocytes of PD patients as well as GPx activity though this change was not statistically significant (by 15%) compared to healthy controls. We found that CBL administration has moderate antioxidant effects in PD patients by decreasing the TBARS accumulation (by 17%) in blood plasma as well as H2O2 content in erythrocytes (by 12%) compared with the CBL-untreated patients. These effects of CBL were accompanied by a decrease in the activities of SOD (by 11%) and catalase (by 20%) in blood of patients with PD compared with the CBL-untreated ones. In CBL-treated patients, there was registered an increase in the GSH concentration (by 30%) and GPx activity (by 15%) in erythrocytes compared with PD patients though this change was not statistically significant compared to the CBL-untreated patients (Tables 5 and 6).\nUPDRS score change prediction using multiple machine learning models\nAmong the models, GBMs demonstrated the highest predictive accuracy, achieving an RMSE of 18.59, MSE of 345.82, and MAE of 15.39. The top-performing RF model exhibited slightly higher RMSE (21.99) and MAE (17.78), underscoring the superior capability of GBMs to capture complex patterns in the data.\nThe boxplot of feature importance across the top 10 models provided valuable insights into which factors most significantly influenced the prediction of UPDRS score changes. Below are the most impactful features, sorted by median scaled importance: Fig. 17\n- \n                    BDNF Change (Median Importance: 0.98): Changes in Brain-Derived Neurotrophic Factor (BDNF) were the most influential predictor of UPDRS score change. BDNF’s high importance suggests that neurotrophic support may play a central role in the Neurotrophic peptide mixture’s therapeutic effects, potentially enhancing neuroplasticity and recovery. \n- \n                    MOCA (Median Importance: 0.62): The Montreal Cognitive Assessment (MOCA), a measure of cognitive function, was ranked as a highly predictive feature, highlighting the relevance of cognitive improvement as a response indicator for the Neurotrophic peptide mixture therapy. \n- \n                    BDNF (after treatment) (Median Importance: 0.61): Post-treatment levels of BDNF were also significant, indicating that maintaining or boosting neurotrophic factors is associated with positive therapeutic outcomes in PD patients. \n- \n                    MMSE, S (Median Importance: 0.58): The Mini-Mental State Examination (MMSE) score also emerged as a key predictor, reinforcing the importance of cognitive assessments in evaluating the Neurotrophic peptide mixture’s efficiency. \n- \n                    PINK1 (after treatment) (Median Importance: 0.55): PINK1, a mitochondrial health marker, was another significant predictor, suggesting that the Neurotrophic peptide mixture’s potential benefits may extend to cellular and mitochondrial health. \nThe boxplot analysis revealed that features related to neurotrophic support (BDNF) and cognitive function (MOCA, MMSE) showed the most consistent importance across models, as evidenced by their high median importance values and narrow interquartile ranges. In contrast, features with wider variability, such as TBA-active products and Age, had broader interquartile ranges, indicating that their contribution to predictive accuracy may depend on model-specific interactions or particular patient subgroups.\nDiscussion\nThe findings of this exploratory study suggest that the Neurotrophic peptide mixture has a multifaceted impact on various aspects of PD pathology, particularly regarding platelet ultrastructure, gene expression, neurophysiological parameters, and oxidative stress markers. These findings align with existing literature, which underscores the neuroprotective and neurorestorative properties of the Neurotrophic peptide mixture.\nImprovement in motor functions, particularly in the Motor tempo test, was more pronounced in patients with predominantly right-sided symptoms of PD. This asymmetry in response could be related to the lateralization of the pathological process. Additionally, the elevation in the threshold for triggering the H-reflex suggests modifications in spinal cord excitability, which could translate to improvements in motor function. The study reported statistically significant enhancements in cognitive abilities, as measured by the MMSE, FAB, and MoCA scales. Additionally, there was a notable reduction in depression scores (BDI) and reactive anxiety (Spielberger-Hanin scale), although personal anxiety levels remained unchanged. These findings highlight the potential of the Neurotrophic peptide mixture in ameliorating both motor and non-motor symptoms of PD. The improvements in motor function, cognitive abilities, and emotional well-being observed in this study suggest that the Neurotrophic peptide mixture has a broad therapeutic potential.\nThe study also observed significant neurophysiological improvements post-Neurotrophic peptide mixture treatment. The latency period of the P3 component, a marker of cognitive processing, was significantly reduced. This improvement in P3 latency indicates enhanced cortical function and suggests that the Neurotrophic peptide mixture may have beneficial effects on cognitive processing speed and efficiency in PD patients. These neurophysiological benefits align with the observed cognitive improvements, further supporting the potential of the Neurotrophic peptide mixture in mitigating cognitive decline in PD.\nIn PD patients, platelets exhibited several signs of senescence, including a reduction in dense δ-granules and an increase in α-granules29. The number of δ-granules per cell was significantly lower in PD patients (5.1 ± 0.4) compared to healthy individuals (8.4 ± 0.5). The administration of the Neurotrophic peptide mixture significantly increased the number of δ-granules in platelets by 45%, indicating enhanced serotonin reserves, which could contribute to the neuroprotective effects observed clinically. This increase in δ-granules likely reflects a normalization of platelet storage function, which is crucial given the role of serotonin in mood and neuroprotection. Despite no significant change in the number of α-granules, the trend towards a decrease might indicate an alteration in platelet activation status.\nAdditionally, the Neurotrophic peptide mixture reduced the calcium-accumulating capacity of platelets, potentially improving ATP synthesis and overall platelet function. This reduction in calcium accumulation is crucial as it indicates a shift away from the pathological state observed in untreated PD platelets. The observed improvements in platelet ultrastructure and function may translate to better overall cellular health, reflecting the systemic benefits of the Neurotrophic peptide mixture treatment beyond its direct neural effects. The enhancement in serotonin reserves and mitochondrial function underscores the multifaceted role of the Neurotrophic peptide mixture in addressing both neurological and peripheral manifestations of PD.\nThe Neurotrophic peptide mixture treatment appears to modulate BDNF gene expression in a sex-specific manner in patients with Parkinson’s disease, with women showing an increase and men showing a decrease in expression levels. This modulation is accompanied by improved cognitive function in women, highlighting the potential of the Neurotrophic peptide mixture as a therapeutic option that may benefit female PD patients more significantly. The stable expression of DJ-1, PINK1, and TNFα genes suggests that the primary neuroprotective mechanisms of the Neurotrophic peptide mixture in PD may involve pathways other than those mediated by these genes. These findings emphasize the need for further research to explore the underlying mechanisms of sex differences in PD treatment responses and to optimize therapeutic strategies accordingly.\nThe excessive production and release of ROS in PD occurs as a result of the combined action of mitochondrial dysfunction, dopamine autooxidation, and neuroinflammation30. In present study, the Neurotrophic peptide mixture demonstrated a moderate antioxidant effect, reducing the levels of lipid peroxidation products and H2O2 in erythrocytes. This reduction was accompanied by a decrease in SOD and catalase activities, and an increase in GSH concentration and GPx activity. These changes indicate a restoration of redox balance and a reduction in oxidative stress, which are crucial in the pathology of PD. One of the most promising effects of CBL on OS developing in blood and tissues in PD could be its ability to decrease intracellular α-synuclein aggregation in neurons, mimicking the action of classical NTFs and growth factors as well as of NTFs belonged to a novel family CDNF/MANF31. The ability of the Neurotrophic peptide mixture to modulate the antioxidant system suggests its potential role in protecting neuronal cells from oxidative damage and subsequent apoptosis.\nThe findings underscore the central role of cognitive and neurotrophic markers in predicting UPDRS score changes as an indicator of the Neurotrophic peptide mixture efficacy. The high importance of BDNF aligns with previous research on neuroprotective therapies, where neurotrophic support has been linked to improved neuronal survival and plasticity. The inclusion of mitochondrial markers, such as PINK1, suggests that the Neurotrophic peptide mixture may exert beneficial effects at the cellular level, addressing mitochondrial dysfunction—a known factor in PD pathology.\nThe boxplot analysis provided further clarity on the stability and reliability of each feature’s importance. Features with high median importance and narrow interquartile ranges, like BDNF and MOCA, indicate consistent predictive power across models, making them robust indicators of therapeutic response. The broader variability observed in other features, such as oxidative stress markers and demographic factors, suggests that while they may contribute to specific aspects of disease progression, they are less central to predicting therapeutic efficacy. The superior performance of GBMs in capturing non-linear relationships reinforces the value of using advanced machine learning techniques to analyze complex clinical datasets. These findings support the potential of the Neurotrophic peptide mixture as a multifaceted therapeutic option, with the capacity to improve cognitive and neurotrophic parameters in PD patients. The results of the present study indicate an antioxidant action of the Neurotrophic peptide mixture in patients with moderate stages of Parkinson’s disease. This action is mainly based on the decrease of the prooxidant events in peripheral blood of PD patients as well as on the prevention of thiol redox defects and reducing an imbalance in oxidant–antioxidant status.\nOverall, statistically significant improvements and consistent correlations were observed in the key clinical endpoints (UPDRS Part III and UPDRS S), while secondary and exploratory outcomes demonstrated patterns suggestive of mechanistic links that warrant confirmation in larger studies.\nInterpretation of its results is limited by several factors, including the small sample size, lack of a control group receiving placebo in addition to standard therapy, and a relatively short duration of the observation period. Since the study is preliminary, exploratory in nature and aimed at collecting primary data on the response to the drug, only one group of patients was formed, which allows identifying primary trends and assessing the feasibility of future larger studies. Future research should focus on larger cohorts and longer treatment durations to confirm these findings and to explore the long-term effects of neurotrophic peptide mixture on gene expression and cognitive function in PD patients. Particular attention should be paid to investigating sex differences in response to the drug in future studies.\nThe results of this exploratory study provide a preliminary rationale for investigating the Neurotrophic peptide mixture as an adjunct therapy in Parkinson’s disease. During the treatment period, changes were observed in platelet ultrastructure, gene expression, neurophysiological parameters, and oxidative stress markers. These findings, along with shifts in cognitive and neurotrophic markers, may suggest a multidimensional biological response.\nHowever, these observations must be interpreted with extreme caution. Given the limited sample size and the absence of a control group, no causal inferences can be drawn. Therefore, larger, placebo-controlled clinical trials are warranted to determine if these initial observations can be substantiated, to investigate potential sex-specific and long-term effects, and to explore biomarkers that may help elucidate the mechanisms of the Neurotrophic peptide mixture in PD.\nMethods\nPatient information\nThe exploratory study included 17 patients with Parkinson’s disease with an established diagnosis according to the inclusion-exclusion criteria of the British Brain Bank32, aged 45–74 years with disease stage 2.0–3.0 (2.9 ± 0.3, according to Hoehn and Yahr scale). Written informed consent was obtained from all participants.\nThe neurotrophic peptide mixture was prescribed against the background of basic stable doses of antiparkinsonian therapy (Levodopa-containing drugs), which were used during 28–30 days before starting the course of neurotrophic peptide mixture and during the entire study period. No dose adjustments were allowed during the study, and medication dose was not included as a covariate in the statistical analysis. The median daily Levodopa dose was about 500 mg, it was individually optimized depending on disease stage and the patient’s condition. The Neurotrophic peptide mixture dosage regimen was 20 ml (215.2 mg/ml) intravenously, once a day for 10 days. Infusions were carried out in the first half of the day.\nEthics statement and study design\nThe Ethics Committee of the Clinical Department of the Institute of Gerontology named after D.F. Chebotariov of the National Academy of Medical Sciences of Ukraine approved the protocol of current research (Protocol No. 4, 09/06/2022). Written informed consent was obtained from all participants prior to enrollment.\nThe study was conducted in accordance with the principles of the Declaration of Helsinki (2000) and the European Council Convention on Human Rights and Biomedicine (2005), and relevant national standards for the participation in the research were taken into account.\nClinical assessment\nCharacterization of the stages of PD was carried out using the Hoehn and Yahr scale. To assess the dynamics of motor disorders, the standardized international Unified Parkinson’s Disease Rating Scale (UPDRS) was used. For a comprehensive assessment of the state of non-motor functions, the following clinical scales were used:\n- \n                    Mini-Mental State Examination (MMSE); \n- \n                    Test battery for assessing frontal dysfunction (Frontal Assessment Battery, FAB); \n- \n                    Montreal Cognitive Assessment (MoCA); \n- \n                    Beck Depression Inventory (BDI); \n- \n                    Spielberger-Hanin Anxiety Scale. \nSensorimotor reaction and motor tempo\nMotor tempo was determined using a computer program as the time between consecutive presses of the same finger on two keys spaced 20 cm apart on the keyboard. The patient was instructed to press the keys at the fastest possible tempo. The duration of each test was 20 s. Testing was conducted for each hand separately, and then the data were averaged.\nThe latent period of simple sensorimotor reaction was determined as the interval between the moment of presentation of a visual signal on the computer monitor and the moment the patient pressed the key in response to the signal. Geometric shapes (circle, square, oval) of red and white colors were used as signals, with 20 signals presented at intervals of 2–4 s. Testing was conducted for each hand separately, and the data were averaged.\nElectroneuromyography\nThe technique of registering the H-reflex, as a reflex response of muscles to the stimulation of sensory fibers of a mixed nerve followed by monosynaptic activation of spinal motoneurons and nerve motor fibers, was performed using disposable surface recording electrodes placed in the projection of the posterior group of muscles of the calf, with stimulation of the tibial nerve in the popliteal fossa. The dynamics of the amplitudes of H- and M-responses (mV) with increasing strength of nerve electrical stimulation (mA), the threshold for the occurrence of the H-reflex (mA), the latency period of the H-reflex (ms), and the ratio of the maximum amplitude of H max/M max (%) were evaluated.\nEvoked potentials P300\nEEG was recorded monopolarly from 21 leads (Fp1, Fp2, Fpz, F3, F4, Fz, C3, C4, Cz, P3, P4, Pz, T3, T4, F7, F8, T5, T6, O1, O2, Oz). The combined ear electrode was used as a reference electrode. The interelectrode resistance was less than 5 kOhm, the signal sampling frequency was 500 Hz, and the high- and low-pass filters were 0.3 Hz and 35 Hz, respectively. Endogenous cognitive evoked potentials N2 and P3 were recorded according to the oddball paradigm using sound stimuli with a frequency of 1000 and 2000 Hz and a duration of 10 ms.\nPlatelet and mitochondrial structure with Electron microscopy\nBlood plasma enriched with platelets containing leukocytes was obtained by centrifugation of whole blood obtained from the ulnar vein of the subjects at room temperature for 15 min at 120 g in a T-30 centrifuge. The plasma was carefully separated from the sediment and centrifuged at 2000 g for 20 min using a Vortecs Combispin FVL-2400N minicentrifuge33.\nElectron microscopic studies of platelets and mitochondria were carried out according to the generally accepted method with double fixation with glutaraldehyde and OSO4, dehydration with alcohols of increasing concentrations and subsequent embedding in Epon. Ultrathin sections with a thickness of 40–60 nm, contrasted with uranyl acetate and lead citrate, were examined using an electron microscope PEM-125K (Ukraine).\nMorphometric calculations of platelets’ structure were performed using the Image Tool (USA) on 130–150 fields for each group of subjects. The number of optically dense small σ-granules and optically transparent large α-granules was determined.\nPBMC isolation and quantitative PCR\nBlood collection, PBMC isolation, RNA extraction and cDNA synthesis were performed by standard methods. The collection of blood for further cell isolation and qPCR was made on 0 and 11th day (after a 10-day course of neurotrophic peptide mixture treatment). Briefly, patients’ blood was collected in 4 ml vacutainers with EDTA. Density gradient centrifugation of diluted whole blood layered over a density gradient medium was used for PBMC isolation. PBMC were gradually frozen at a temperature of −20 °C, and then transferred to the temperature of −80 °C for long-term storage. RNA was isolated from cells using the Direct-zol RNA Miniprep Kit (Zymo Research, USA) following the manufacturer’s protocols. The isolated RNA was treated with DNase to remove DNA residues that could give a false positive signal for the expression of target genes. After this, reverse transcription was carried out using a commercial kit ZymoScript RT PreMix (Zymo Research, USA), guided by the manufacturer’s recommendations.\nSamples of this cDNA were used for further qPCR-RT (quantitative real-time polymerase chain reaction), which was used to measure the expression of the DJ-1, PINK1, BDNF and TNFα genes. To identify a suitable internal control for RT-qPCR data normalization, we empirically tested the expression stability of three candidate reference genes (rps18, gusb, and beta-actin). The sequences of all used primers are listed in the Supplementary Information. Analysis of the raw Cqt values revealed that both beta-actin and gusb were the most stably expressed genes, showing minimal variation and a strong positive correlation with each other. In contrast, rps18 expression was found to be significantly more variable across our samples and was therefore excluded. Although both beta-actin and gusb were suitable candidates, we selected beta-actin for normalization due to its slightly higher expression stability.\nThe sequences of all used primers are listed in the Supplementary Information. The PCR reaction mix was prepared using a commercial kit of reagents for RT-PCR (HOT FIREPol® EvaGreen® qPCR Mix Plus, Solis BioDyne, Estonia) - reverse transcription and polymerase chain reaction with the addition of betaine (Sigma, USA) at a final concentration of 1 M. To determine differences in mRNA expression of Parkin genes, Repeated Measures Anova was used.\nAccumulation curves were generated using Opticon Monitor 3 software. After amplification was completed, raw gene product amplification data were used to calculate Cy0.\nMeasurement of oxidative stress biomarkers\nSecondary products of lipid peroxidation, the activity of antioxidant enzymes superoxide dismutase (SOD) and catalase (CAT) were measured in blood plasma. In red blood cells – the content of hydrogen peroxide (H2O2); the state of the glutathione system was assessed by the content of reduced glutathione (GSH) and the activity of the enzyme glutathione peroxidase (GP).\nFor erythrocyte isolation, samples were centrifuged at 8000 rpm for 15 min to separate plasma and red blood cells. Using centrifugation, erythrocytes were washed twice of plasma with cold sterile physiological sodium chloride solution. Erythrocyte hemolysate was obtained by adding distilled water to washed erythrocytes in a 1:1 ratio. The hemoglobin (Hb) content was also measured (Hemoglobin Assay kit MAK115, Sigma-Aldrich, St. Louis, MO, USA).\nDetermination of thiobarbituric acid reactive substances (TBARS) levels: TBARS were isolated by boiling plasma with the thiobarbituric acid reagent (0.5% 2-thiobarbituric acid/10% trichloroacetic acid/0.63 mM hydrochloric acid) and measuring the absorbance at 532 nm. The results are expressed as µM/ml using a molar extinction coefficient 1.56 * 105 mm−1 cm−1.\nCatalase activity measurement: The method is based on the ability of molybdenum salts to form a colored complex with hydrogen peroxide, which was determined at a wavelength of 410 nm. Enzyme activity (μM/min/L) was calculated using a molar extinction coefficient of 22.2 * 103 mm−1cm−1.\nSuperoxide dismutase (SOD) activity was determined spectrophotometrically. The method is based on the ability of SOD to inhibit the autoxidation of adrenaline at pH 10.2. The results were expressed as specific activity of the enzyme in units per ml of plasma. One unit of SOD activity is defined as the amount of protein causing 50% inhibition of the conversion rate of adrenaline to adrenochrome under specified conditions.\nThe activity of selenium-dependent glutathione peroxidase (GPx) was determined by recording the rate of NADPH oxidation in the presence of reduced glutathione at a wavelength of 340 nm. Enzyme activity was expressed in nM NADPH/min/ml.\nThe amount of reduced glutathione (μM/ml) was determined using Elman’s reagent (5.5’-dithio-bis-nitrobenzoic acid in absolute ethanol). The change in extinction was recorded at a wavelength of 412 nm every minute for 5 minutes.\nHydrogen peroxide content was studied using the FOX method, which is based on the peroxide-mediated oxidation of Fe2 + , followed by the reaction of Fe3+ with xylenol orange. Absorbance of the Fe3 + -xylenol orange complex was determined at a wavelength of 560 nm. The amount of H2O2 (μM/mgHb) was calculated using the H2O2 standard curve for each independent experiment.\nMachine learning prediction of UPDRS score change\nTo model UPDRS score changes, H2O’s AutoML framework was employed34, allowing for the training, tuning, and evaluation of multiple Machine Learning models, including Gradient Boosting Machines (GBMs) and Random Forests (RFs). AutoML facilitated hyperparameter optimization, ensuring robust and accurate predictions.\n- \n                    Gradient Boosting Machines (GBMs): GBMs iteratively build a series of decision trees, where each subsequent tree focuses on correcting errors made by its predecessors. This sequential approach enables GBMs to capture intricate, non-linear relationships within the data, making them particularly suitable for analyzing how clinical and biochemical factors impact UPDRS score changes and, consequently, therapeutic response35. \n- \n                    Random Forests (RFs): RFs are ensemble models that combine multiple decision trees, each trained on a random subset of the data. Known for their robustness and reduced susceptibility to overfitting, RFs provide valuable insights into feature importance by averaging across numerous trees36. This model serves as a stable baseline to complement the more flexible GBMs. \nThe models were evaluated using the following metrics to quantify prediction accuracy:\n- \n                    Root Mean Squared Error (RMSE): Captures the average deviation between predicted and actual values. \n- \n                    Mean Squared Error (MSE): Measures the variance of predictions relative to actual outcomes. \n- \n                    Mean Absolute Error (MAE): Reflects the average absolute deviation between predicted and observed values. \nThese metrics allowed for a thorough assessment of each model’s ability to predict UPDRS score changes, highlighting the Neurotrophic peptide mixture’s potential impact on disease progression.\nFeature importance was evaluated using the top 10-performing models from a set of 50 trained with H2O AutoML. Each feature’s importance was measured by its “scaled importance,” which represents its relative contribution to prediction accuracy. The models were trained using two-fold cross-validation, and feature importance was extracted during the validation stage.\nData from the top 10 models were aggregated, and features were sorted by their median feature importance. It is worth noting that while building a generalized model on such a small dataset is not feasible, the features identified with the highest median importance could be considered potential markers in future studies. The results were visualized using a boxplot (Fig. 16) to highlight the median importance and variability of each feature.\nStatistical analysis\nHypotheses about the effects of the drug were tested using the Wilcoxon (for related samples) and Mann-Whitney (for independent groups) tests. The relationship between the investigated indicators was determined in the form of Spearman’s rank correlation coefficients.\nStatistical processing of gene expression data was performed using Statistica 8.0 software (StatSoft Inc., USA). For statistical analysis, Spearman’s non-parametric test was used to investigate the correlation between parameters. The difference in expression of the studied genes before and after treatment was determined using the Wilcoxon t-test (Wilcoxon signed-rank test).\nThe results of the study of the state of the prooxidant-antioxidant system were processed statistically using the Origin program. The significance of differences between comparison groups was determined by the method of analysis of variance (ANOVA) followed by the Bonferroni test (post-hoc test).\nOutcomes and hierarchy\nFor this exploratory study, outcomes were grouped into three hierarchical levels.\n- \n                    Clinically central endpoints were prespecified as the UPDRS III (motor score) and the UPDRS Total score, as the primary efficacy measures, as these are the most widely accepted, clinically relevant, and objective indicators of motor impairment and overall disease severity in Parkinson’s disease. \n- \n                    Secondary clinical endpoints included UPDRS I (non-motor experiences of daily living) and UPDRS II (motor experiences of daily living), and validated scales of non-motor function: MMSE, MoCA, FAB (cognitive measures), and BDI and Spielberger–Hanin Anxiety Scale (mood and affective measures). These outcomes were considered supportive of the clinically central endpoints. \n- \n                    Exploratory outcomes included additional non-motor measures, neurophysiological parameters (P300 latency and amplitude), and molecular/structural biomarkers (gene expression, platelet ultrastructure, mitochondrial morphology, and oxidative stress markers). These outcomes were included to generate mechanistic insights into the potential modes of action of the Neurotrophic peptide mixture. \nFormal hypothesis testing was restricted to the clinically central endpoints. Secondary and exploratory outcomes are presented with descriptive statistics and effect sizes. To minimize Type I error, no multiple-testing correction was applied outside of the clinically central endpoints.\nCorrelation analyses were performed using Spearman’s rank correlation coefficient (ρ). For correlations involving clinically central endpoints, p-values are reported. Correlations among secondary and exploratory outcomes are reported without formal inferential testing, to avoid Type I error inflation.\nData availability\nThe datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.\nReferences\n- Alexoudi, A., Alexoudi, I. & Gatzonis, S. Parkinson’s disease pathogenesis, evolution and alternative pathways: A review. Rev. Neurol. (Paris) 174, 699–704 (2018). \n- Chung, S. J. et al. The Pattern of Striatal Dopamine Depletion as a Prognostic Marker in De Novo Parkinson Disease. Clin. Nucl. Med 201843, 787–792 (2018). \n- Moon, H. E. & Paek, S. H. Mitochondrial Dysfunction in Parkinson’s Disease. Exp. Neurobiol. 24, 103–116 (2015). \n- Mansur, A. et al. Characterization of 3 PET tracers for quantification of mitochondrial andsynaptic function in healthy human brain:18F-BCPP-EF,11C-SA-4503, and11C-UCB-J. J. Nucl. Med 61, 96–103 (2020). \n- Simonenkov, A. P. & Kluyev, V. M. Syndrome serotonin deficiency. M.: BINOM, 96-97 (2013). \n- Mauricio, R. et al. Tackling gaps in developing life-changing treatments for dementia. Alzheimers Dement 5, 241–253 (2019). \n- Gonchar, O. O., Klymenko, O. O., Drevytska, T. I., Bratus, L. V. & Mankovska, I. M. Oxidative stress in rat heart mitochondria under a rotenone model of Parkinson’ disease: a corrective effect of capicor treatment. Ukr. Biochem J. 93, 21–30 (2021). \n- Schneider-Garces, N. J. et al. Span, CRUNCH, and beyond: working memory capacity and the aging brain. J. Cogn. Neurosci. 22, 655–669 (2010). \n- Fiani, B. et al. Cerebrolysin for stroke, neurodegeneration, and traumatic brain injury: review of the literature and outcomes. Neurol. Sci. 42, 1345–1353 (2021). \n- Brainin, M. Cerebrolysin: a multi-target drug for recovery after stroke. Expert Rev. Neurother. 18, 681–687 (2018). \n- Requeejoe, C. et al. Nanodelivery of Cerebrolysin and rearing enriched environment induce neuroprotective effects in a preclinical rat model of Parkinson’s disease. Mol Neurobiol. Springer Science+Business Media, LLC (2017). \n- Neveen, A. et al. A promising therapeutic potential of Cerebrolysin in the 6-OHDA rat model of Parkinson's disease. Life Sci 155, 174–179 (2016). \n- Luchanina, E., Karaban, I. & Berezetskaya, N. Diagnosis of Parkinsonas disease by electrophysiological methods. Diagn. Rehabil. Parkinson. Dis. 2, 27–58 (2011). \n- Eder, P., Reinprecht, I., Schreiner, E., Skofitsch, G. & Windisch, M. Increased density of glutamate receptor subunit 1 due to Cerebrolysin treatment: an immunohistochemical study on aged rats. Histochem J. 33, 605–612 (2001). \n- Stepanichev, M. et al. Effects of cerebrolysin on nerve growth factor system in the aging rat brain. Restor. Neurol. Neurosci. 35, 571–581 (2017). \n- Veinbergs, I., Mante, M., Mallory, M. & Masliah, E. Neurotrophic effects of Cerebrolysin in animal models of excitotoxicity. J. Neural Transm. Suppl. 59, 273–280 (2000). \n- Alvarez, X. A. et al. Neuroimmunotrophic effects of cerebrolysin in an animal model of hippocampal degeneration induced by β-amyloid. In: Korczyn, A. D., ed. International Congress on Vascular Dementia 2017. Bologna, Italy: Monduzzi Editore, 233–237 (1999). \n- Gavrilova, S. I. & Alvarez, A. Cerebrolysin in the therapy of mild cognitive impairment and dementia due to Alzheimeras disease: 30 years of clinical use. Med Res Rev. 41, 1–29 (2020). \n- DeBoer, S. R. et al. Enhanced spontaneous motor recovery after stroke in mice treated with cerebrolysin. Neurorehabil Neural Repair 35, 525–533 (2021). \n- Muresanu, D. F. et al. Efficacy and safety of cerebrolysin in neurorecovery after moderate-severe traumatic brain injury: results from the CAPTAIN II trial. Neurol. Sci. 41, 1171–1181 (2020). \n- Lu, W. et al. Cerebrolysin alleviates early brain injury after traumatic brain injury by inhibiting neuroinflammation and apoptosis via TLR signaling pathway. Acta Cir. Bras. 37, e370605 (2022). \n- Chen, C. C., Wei, S. T., Tsaia, S. C., Chen, X. X. & Cho, D. Y. Cerebrolysin enhances cognitive recovery of mild traumatic brain injury patients: double-blind, placebo-controlled, randomized study. Br. J. Neurosurg. 27, 803–807 (2013). \n- Cui, S. Cerebrolysin for vascular dementia. Cochrane Database Syst Rev. 11, 19–20 (2019). \n- Rochenstein, E. et al. Neuropeptide treatment with cerebrolysin enhances the survival of grafted neural stem cells in a synuclein transgenic model of Parkinson’s disease. J. Exp. Neurosci. 9, 131–140 (2015). \n- Hünerli, D. et al. Mild cognitive impairment in Parkinson’s disease is associated with decreased P300 amplitude and reduced putamen volume. Clin. Neurophysiol. 130, 1208–1217 (2019). \n- Alvarez, X. A., Lombardi, V. R., Fernández-Novoa, L., García, M. & Sampedro, C. Cerebrolysin reduces microglial activation in vivo and in vitro: a potential mechanism of neuroprotection. J. Neural Transm. 59, 281–292 (2000). \n- Kulinskiĭ, V. I. & Kolesnichenko, L. S. Glutathione system. I. Synthesis, transport, glutathione transferases, glutathione peroxidases. Biomed. Khim 55, 255–277 (2009). \n- Zenkov, N. K., Lankin, V. Z. & Menshikova, E. B. Oxidative stress: The biochemical and pathophysiological aspects. M.: MAIK. Science / Interperiodica 4, 352–361 (2001). \n- Park, S. M. et al. Evidence that alpha-synuclein functions as a negative regulator of Ca(++)-dependent alpha-granule release from human platelets. Blood 100, 2506–2514 (2002). \n- Luo, Y., Hoffer, A., Hoffer, B. & Qi, X. Mitochondria: A therapeutic target for Parkinson’s disease?. Int J. Mol. Sci. 16, 20704–20730 (2015). \n- Voutilainen, M. H., Arumae, U., Airavaara, M. & Saarma, M. Therapeutic potential of the endoplasmic reticulum located and secreted CDNF/MANF family of neurotrophic factors in Parkinson’s disease. FEBS Lett. 589, 3739–3748 (2015). \n- Hughes, A. J., Daniel, S. E., Kilford, L. & Lees, A. J. Accuracy of clinical diagnosis of idiopathic Parkinson’s disease: A clinico-pathological study of 100 cases. J. Neurol. Neurosurg. Psychiatry 55, 181–184 (1992). \n- Filippova, O. I., Koloskov, A. V. & Stolitsa, A. A. Methods for studying the functional activity of platelets (literature review). Transfusiology 13, 493–514 (2012). \n- LeDell, E. & Posse, C. H2O AutoML: Scalable Automatic Machine Learning. J. Open Source Softw. 5, 1–10 (2020). \n- Chen, J. & Guestrin, C. XGBoost: A Scalable Tree Boosting System. Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery Data Mining, 785–794 (2016). \n- Liaw, A. & Wiener, M. Classification and regression by random Forest. R. N. 2, 18–22 (2002). \nAcknowledgements\nThis work was supported by the National Academy of Medical Sciences, Ukraine (Grant # 0123U100771).\nAuthor information\nAuthors and Affiliations\nContributions\nD.K. took part in writing of the main text of manuscript, coordinated the work of scientific laboratories in their experiments and results, contributed to the conceptualization of the project, I.K. was the lead scientist of part connected to clinical assessment, data analysis, project conception and organization, N.K., N.M. and S.K. performed and analyzed electroencephalograms and elecroneuromyograms of patients, collected patient data regarding sensorimotor reactions and motor tempo, K.R. participated in the preparation of the literature review, implemented the methodology of morphological studies, processing the data of the section about the morphological parameters. O.G. examined markers of oxidative stress and made substantial contributions to the analysis of data as well as participated in a critical review of the manuscript, I.M. helped with the preparation of the manuscript and analysis of data. O.Z. wrote an abstract, implemented the main methodology for gene expression measurement, analysis and interpretation of data, O.B. and K.M. were involved to gene expression experiments and analysis of qPCR-RT data, as well as collection of biological material, S.S. provided the results of deep analysis of obtained data using multiple machine learning models, performed feature importance analysis, T.P. contributed to the administration of project, wrote and edited the main manuscript text. All authors reviewed the manuscript.\nCorresponding author\nEthics declarations\nCompeting interests\nThe authors declare no competing interests.\nAdditional information\nPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.\nRights and permissions\nOpen Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.\nAbout this article\nCite this article\nKrasnienkov, D., Karaban, I., Karasevych, N. et al. Evaluation of the neurotrophic peptide mixture in pathogenetic therapy of patients with Parkinson’s disease. npj Parkinsons Dis. 12, 55 (2026). https://doi.org/10.1038/s41531-026-01270-6\n- Received: \n- Accepted: \n- Published: \n- Version of record: \n- DOI: https://doi.org/10.1038/s41531-026-01270-6","confidence":0.9,"diagnostics_url":"/api/diagnose?url=https%3A//www.nature.com/articles/s41531-026-01270-6","quality_bucket":"high","failure_kind":"none","retryable":false,"quality_reason":"High confidence: full text extraction produced 58549 characters.","quality_profile":{"profile_version":"extraction_quality.v2","bucket":"high","confidence":0.9,"failure_kind":"none","retryable":false,"retry_after_attempts":0,"reason":"High confidence: full text extraction produced 58549 characters.","operator_guidance":{"severity":"ok","recommended_action":"trust_full_text","next_step":"Use the extracted full text as the primary article source.","operator_label":"Ready","can_retry":false,"can_use_summary":false,"diagnostics_required":false},"content_depth":{"contract_version":"content_depth.v1","category":"full_text","label":"Full text","has_full_text":true,"has_summary":true,"content_length":58549,"summary_length":313,"usable_text_length":58549,"source_field":"content"},"legacy_collapsed":false,"signals":{"extract_state":"ok","extract_error":null,"extract_retries":0,"content_length":58549,"summary_length":313}},"tags":[],"format_contract_version":"news_item_formats.v1"}}}