This clinical trial investigates the effects of spermidine supplementation on sleep quality and memory function in older adults with Mild Cognitive Impairment (MCI), a condition associated with an increased risk of developing dementia, particularly in patients with MCI due to Alzheimer's disease. Impaired sleep has been identified as a modifiable factor contributing to cognitive decline, and interventions targeting sleep architecture could offer therapeutic potential to prevent or slow down this decline. Spermidine is a naturally occurring polyamine found in foods such as wheat germ and soybeans. It induces autophagy, a cellular degradation and recycling process essential for neuronal maintenance and function. In animal studies, spermidine has been shown to improve memory performance, reduce neuroinflammation, and support mitochondrial health. Preliminary findings from human trials in individuals with subjective cognitive decline or MCI suggest potential cognitive benefits of spermidine, but results are not unequivocal, and the impact on sleep has not been systematically evaluated. In this randomized, double-blind, placebo-controlled trial, 76 participants aged 55 to 70 years with MCI will receive either spermidine (6 mg/day) or a placebo for 12 weeks. Sleep will be evaluated using overnight EEG in a controlled laboratory setting, focusing on measures such as slow-wave sleep and sleep spindle activity. Memory performance will be assessed before and after the intervention using standardized neuropsychological testing. Numerical skills will be tested at baseline only to compare MCI patients with healthy controls. Blood samples will be collected to quantify metabolic indicators, neurodegeneration-related biomarkers, and autophagy-associated proteins. A control group of 38 cognitively healthy individuals will undergo comparable sleep and cognitive assessments without receiving any supplementation. The primary objective of the study is to characterize the impact of spermidine on sleep-dependent memory consolidation and to identify associated biological changes relevant to aging and neurodegeneration. The results may inform the development of non-pharmacological strategies aimed at preserving cognitive function in individuals at risk for dementia.
Mild Cognitive Impairment (MCI) is a clinical condition characterized by measurable cognitive decline that exceeds age-related norms but does not meet criteria for dementia, often due to Alzheimer's Disease (AD). It is associated with an increased likelihood of progressing to dementia due to AD or other forms of dementia. Disruption of sleep architecture is increasingly implicated in the pathophysiology of neurodegenerative disorders, and early interventions targeting sleep-related mechanisms could help delay further cognitive decline. This clinical trial evaluates the effects of spermidine supplementation on sleep quality and sleep-dependent memory consolidation in older adults diagnosed with MCI. Following an adaptation night and an initial baseline sleep assessment with overnight electroencephalography (EEG), participants will be randomly assigned to either the spermidine or placebo group. The intervention consists of a daily oral dose of 6 mg spermidine (administered as three 2 mg sachets), continued over a 12-week period. The trial employs a randomized, double-blind, placebo-controlled design. Placebo sachets, identical in appearance and taste, contain only microcrystalline cellulose. After the 12-week supplementation period, participants will return for a second overnight EEG assessment. A healthy control group (n=38), matched for age and sex, will undergo comparable baseline assessments but will not receive any intervention. These data will provide normative reference values for sleep and cognitive parameters. The primary objective of the study is to assess the impact of spermidine on sleep architecture, measured via overnight polysomnography, with a specific focus on slow-wave sleep and sleep spindle activity, EEG markers associated with sleep-dependent memory consolidation and known to decline with age and neurodegeneration. Secondary outcomes include changes in memory consolidation (assessed using a battery of cognitive tasks that target declarative, procedural, and visuospatial memory domains), as well as numerical skills (tested via e.g., digit-letter-decision task, Berlin Numeracy Test). Testing occurs before and after each EEG night to evaluate overnight consolidation effects. All participants will wear actigraphs prior to both EEG nights to monitor sleep-wake cycles and physical activity. Additional biological endpoints will examine changes in circulating neuropeptides, insulin-glucose homeostasis, and autophagy-related biomarkers. Blood samples are collected at each EEG session and two weeks after the start of supplementation. Physiological assessments include oral glucose tolerance tests, measurements of inflammatory markers (e.g., interleukin (IL)-6, tumor necrosis factor (TNF-α), neuroprotective factors (e.g., Neuropeptide Y, eukaryotic translation initiation factor 5A (eIF5A) hypusination), and metabolic indicators (e.g., fasting insulin, glucose, and lipid profiles). Participants will undergo structural brain imaging using a 3-Tesla MRI scanner to exclude individuals with comorbid neurological conditions (e.g., prior stroke, other neurodegenerative diseases) and to assess brain morphology. Eligibility is determined through standardized pre-screening procedures, including clinical interviews and neuropsychological assessments. The study includes 76 participants with MCI, aged 55-70 years, randomly allocated to either the spermidine or placebo group (n=38 per group). The sample size provides adequate statistical power to detect medium-to-large effect sizes, based on previous pilot findings and anticipated attrition rates. This study aims to investigate the effects of spermidine supplementation on neurophysiological and cognitive outcomes in individuals with early-stage cognitive decline. The results are expected to contribute to the growing body of evidence evaluating spermidine as a safe, nutrition-based strategy for maintaining brain health and cognitive function in the context of aging.
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
BASIC_SCIENCE
Masking
DOUBLE
Enrollment
76
supplementation of 6 mg Spermidine per day across 3 doses
supplementation of 6 mg Placebo per day across 3 doses (placebo consists of maltodextrin, rice extract microcrystalline cellulose mixture, citric acid (anhydrous), silicon oxide (precipitated, E551))
Department of Neurology, University Medicine Greifswald
Greifswald, Mecklenburg-Vorpommern, Germany
RECRUITINGEffects of spermidine supplementation on sleep quality as measured by electroencephalography (EEG) mean power spectra
Individual mean power spectra are calculated. Subsequently, the mean power is calculated in the following EEG bands: slow oscillations (0.5-1 Hz), low delta (1-1.5 Hz), delta (1-4 Hz), theta (4-8 Hz), and beta (15-25 Hz). For the spermidine effect, baseline and follow-up EEG data are compared with each other.
Time frame: baseline, 12-week follow-up
Effects of spermidine supplementation on sleep quality as measured by sleep spindle count assessed from EEG
Individual mean power spectra are calculated. Subsequently, the mean power is calculated in the following EEG bands: slow frontal spindle (8-12 Hz), fast parietal spindle (12-15 Hz), and the number of slow frontal and fast parietal discrete spindles are counted. Spindle detection is performed based on an algorithm adopted from previous studies. For the spermidine effect, baseline and follow-up EEG data are compared with each other.
Time frame: baseline, 12-week follow-up
Effects of spermidine supplementation on sleep quality as measured by sleep spindle power assessed from EEG
Individual mean power spectra are calculated. Subsequently, the mean power is calculated in the following EEG bands: slow frontal spindle (8-12 Hz), fast parietal spindle (12-15 Hz). Spindle power is analyzed as a further measure of spindle activity. Spindle detection is performed based on an algorithm adopted from previous studies. For the spermidine effect, baseline and follow-up EEG data are compared with each other.
Time frame: baseline, 12-week follow-up
Effects of spermidine supplementation on sleep-related alertness as measured by computer-based task
The alertness task is trained at adaptation night and assessed at baseline and follow-up before and after sleep. It comprises two runs without and two runs with a prior warning tone. For the spermidine effect, baseline and follow-up task performances of MCI patients are compared with each other. Task performance is measured by speed (in ms) and hit/false or miss.
Time frame: baseline, 12-week follow-up
Effects of spermidine supplementation on sleep-related visual-spatial memory as measured by computer-based task
The visual-spatial memory task is trained at adaptation night and assessed at baseline and follow-up before and after sleep. After an encoding phase consisting of 60 objects (e.g., photo of a band-aid) placed in front of a background (e.g., photo of a treatment room in a doctor's office), a retrieval phase follows directly after encoding in the evening before bedtime, in which 30 randomly selected images from the total of 60 are tested. Participants should memorize the position of the object placed in front of the background and select the correct position by pressing a button. In the next morning, the remaining 30 images are tested in a recall phase. Task performance is measured by speed (in ms) and hit/false or miss. For the spermidine effect, baseline and follow-up task performances of MCI patients are compared with each other.
Time frame: baseline, 12-week follow-up
Effects of spermidine supplementation on sleep-related verbal memory as measured by computer-based task
The verbal memory task is trained at adaptation night and assessed at baseline and follow-up before and after sleep. After an encoding phase consisting of 44 word pairs, a retrieval phase follows directly after encoding in the evening before bedtime, in which all word pairs are tested in randomized order. The next morning, all word pairs are tested in a recall phase. Task performance is measured by hit/false or miss. For the spermidine effect, baseline and follow-up task performances of MCI patients are compared with each other.
Time frame: baseline, 12-week follow-up
Differences in autophagy-related blood markers pre and post spermidine intervention in MCI as measured by polyamine concentration
concentration in nmol / mg Protein assessed via Peripheral Blood Mononuclear Cells (PBMCs)
Time frame: baseline, 2 weeks after baseline, 12-week follow-up
Differences in sleep quality between MCI and healthy controls (HC) as measured by electroencephalography (EEG) mean power spectra
Individual mean power spectra are calculated. Subsequently, the mean power is calculated in the following EEG bands: slow oscillations (0.5-1 Hz), low delta (1-1.5 Hz), delta (1-4 Hz), theta (4-8 Hz), and beta (15-25 Hz). The differences between MCI and healthy controls are assessed from baseline EEG data.
Time frame: baseline data
Differences in sleep quality between MCI and healthy controls (HC) as measured by sleep spindle count assessed from EEG
Individual mean power spectra are calculated. Subsequently, the mean power is calculated in the following EEG bands: slow frontal spindle (8-12 Hz), fast parietal spindle (12-15 Hz), and the number of slow frontal and fast parietal discrete spindles are counted. Spindle detection is performed based on an algorithm adopted from previous studies. The differences between MCI and healthy controls are assessed from baseline EEG data.
Time frame: baseline data
Differences in sleep quality between MCI and healthy controls (HC) as measured by sleep spindle power assessed from EEG
Individual mean power spectra are calculated. Subsequently, the mean power is calculated in the following EEG bands: slow frontal spindle (8-12 Hz), fast parietal spindle (12-15 Hz). Spindle power is analyzed as a further measure of spindle activity. Spindle detection is performed based on an algorithm adopted from previous studies. The differences between MCI and healthy controls are assessed from baseline EEG data.
Time frame: baseline data
Differences in autophagy-related blood markers pre and post spermidine intervention in MCI as measured by eIF5A hypusination
assessed via Hypusinated eIF5A levels (Western Blot) from Peripheral Blood Mononuclear Cells (PBMCs)
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.
Time frame: baseline, 2 weeks after baseline, 12-week follow-up
Differences in neuropeptide Y blood levels pre and post spermidine intervention in MCI
Neurppeptide Y blood levels in pg/ml analyzed from serum blood sample
Time frame: baseline, 12-week follow-up
Differences in sleep-related alertness between MCI patients and healthy controls as measured by computer-based task
The alertness task is trained at adaptation night and assessed at baseline and follow-up before and after sleep. It comprises two runs without and two runs with a prior warning tone. The differences between MCI and healthy controls are assessed from baseline task data. Task performance is measured by speed (in ms) and hit/false or miss.
Time frame: baseline
Differences in sleep-related visual-spatial memory between MCI patients and healthy controls as measured by computer-based task
The visual-spatial memory task is trained at adaptation night and assessed at baseline and follow-up before and after sleep. After an encoding phase consisting of 60 objects (e.g., photo of a band-aid) placed in front of a background (e.g., photo of a treatment room in a doctor's office), a retrieval phase follows directly after encoding in the evening before bedtime, in which 30 randomly selected images from the total of 60 are tested. Participants should memorize the position of the object placed in front of the background and select the correct position by pressing a button. In the next morning, the remaining 30 images are tested in a recall phase. Task performance is measured by speed (in ms) and hit/false or miss. The differences between MCI and healthy controls are assessed from baseline task data.
Time frame: baseline
Differences in sleep-related verbal memory between MCI patients and healthy controls as measured by computer-based task
The verbal memory task is trained at adaptation night and assessed at baseline and follow-up before and after sleep. After an encoding phase consisting of 44 word pairs, a retrieval phase follows directly after encoding in the evening before bedtime, in which all word pairs are tested in randomized order. The next morning, all word pairs are tested in a recall phase. Task performance is measured by hit/false or miss. The differences between MCI and healthy controls are assessed from baseline task data.
Time frame: baseline
Differences in neuropeptide Y blood levels between MCI and healthy controls
blood levels in pg/ml analyzed from serum blood sample
Time frame: baseline data
Differences in autophagy-related blood markers between MCI and healthy controls as measured by polyamine concentration
concentration in nmol / mg Protein assessed via Peripheral Blood Mononuclear Cells (PBMCs)
Time frame: baseline data
Differences in autophagy-related blood markers between MCI and healthy controls as measured by eIF5A hypusination
assessed via Hypusinated eIF5A levels (Western Blot) from Peripheral Blood Mononuclear Cells (PBMCs)
Time frame: baseline data
Differences in numerical skills between MCI patients and healthy controls as measured by lexical decision task
Participants complete a computer-based lexical decision task, in which a row of six symbols is presented after a fixation cross. Depending on the condition, the test subjects must decide by pressing a key on the keyboard whether the row presented contains a letter (e.g., condition 1) or a digit (e.g., condition 2). Each condition consists of two blocks. After the first block, there is a short pause before the second block starts. The condition (digit or letter) with which the trial starts is randomized for each test subject.
Time frame: assessed at baseline only
Differences in numerical skills between MCI patients and healthy controls as measured by Berlin Numeracy Test
The Berlin Numeracy Test contains 4 short, text-based questions testing statistical numeracy and risk literacy, requiring Bayesian inference for problems like understanding probabilities in health concerns (e.g., poisonous mushrooms) or lotteries, often presenting data in relative/absolute numbers and asking for probability estimates. Correct answers scored for a total of 0-4 points.
Time frame: assessed at baseline only