Spermidine for cognitive ageing: insights into observational and interventional studies

Introduction

With the expansion of the ageing population, cognitive decline has become an increasingly pressing challenge.¹ As life expectancy increases, its socioeconomic burden is also increasing, highlighting the urgent need for effective interventions. Numerous studies suggest that specific dietary patterns and nutritional interventions may help mitigate cognitive decline associated with ageing.² Among these, caloric restriction (CR), which aims to extend lifespan through reduced calorie intake, has gained significant research interest in recent years.³ However, the optimal CR regimen remains difficult to define due to the risk of malnutrition and other potential adverse effects. Moreover, maintaining long-term adherence to CR can be difficult for most individuals. Consequently, the use of caloric restriction mimetics (CRMs), compounds that replicate the physiological benefits of CR without requiring stringent dietary modifications, has emerged as a promising and potentially practical alternative. Among these, increasing evidence suggests that CR and intermittent fasting may elevate endogenous levels of spermidine (SPD), a polyamine compound now being investigated as a natural CRM candidate.⁴

SPD is a ubiquitous natural polyamine present across all living organisms, from bacteria to humans, and is naturally found in various dietary sources. Beyond its endogenous role in cellular metabolism, SPD can be obtained from dietary intake and synthesised by commensal gut microbiota. SPD is involved in several critical biological processes, including cell growth, differentiation and autophagy, a fundamental mechanism for cellular maintenance and repair. Recognised as a natural inducer of autophagy, SPD is considered an antiageing compound with properties resembling those of CR, positioning it as a potential CRM.^{4 5} Previous reviews have underscored SPD’s potential to promote cellular autophagy, regulate the cell cycle, inhibit inflammatory responses and maintain gene expression stability.⁵ These mechanisms collectively suggest that SPD may play a significant and multifaceted role in supporting cognitive health during ageing.

This forum article provides a comprehensive synthesis of current evidence on the impact of SPD on cognitive ageing, drawing from both observational and interventional studies. A systematic search of major electronic databases identified 22 relevant studies, comprising 4 interventional trials and 18 observational studies. The quality of the included studies was evaluated using established tools, including the Cochrane risk-of-bias tool, the Risk of Bias in Non-randomised Studies of Interventions (ROBINS-I) and the Newcastle-Ottawa Scale (NOS) (online supplemental table S1–S3). While the majority of studies were of moderate to high quality, limitations such as small sample sizes and heterogeneity in outcome measures were prevalent, warranting cautious interpretation of the findings. Through this synthesis, we aim to elucidate the potential role of SPD in cognitive ageing and identify critical gaps to guide future research directions.

Observational studies: potential sources of inconsistencies

Observational studies have investigated the association between SPD levels and cognitive ageing-related conditions, yielding mixed findings. Several studies suggest a potential link between higher SPD levels and improved cognitive performance. For example, dietary SPD intake has been associated with a reduced risk of cognitive impairment,⁶ larger hippocampal volume and greater cortical thickness.⁷ Additionally, higher blood SPD levels have been correlated with better Mini-Mental State Examination scores and other cognitive outcomes in one study.⁸

However, these findings are inconsistent across studies. Some research has indicated that higher SPD levels may be associated with adverse outcomes, such as reduced hippocampal volume, elevated Alzheimer’s disease (AD) scores⁹ and markers of poor sleep quality¹⁰ in older adults. Further discrepancies have been noted in studies measuring SPD levels in patients with cognitive disorders, including mild cognitive impairment (MCI) and AD. While certain studies report lower SPD levels in these patients compared with healthy controls, others have reported elevated SPD levels in similar populations. Detailed findings are provided in online supplemental table S4.

One potential factor contributing to these inconsistencies is the variation in the sources of SPD measurements. SPD can be quantified from diverse biological and dietary sources, including brain tissue, blood (serum or plasma), red blood cells (RBCs) and dietary intake. Each source introduces distinct challenges in accurately representing SPD bioavailability and its relevance to cognitive function. These variations may complicate the interpretation of findings and contribute to the observed discrepancies across studies.

First, whether dietary SPD intake reliably reflects blood SPD levels remains uncertain. Several studies employed food frequency questionnaires,^{6 7} which are susceptible to recall bias and may not accurately quantify SPD bioavailability. This discrepancy raises concerns about the validity of dietary SPD as a proxy for its physiological impact. Notably, in both healthy adult animals and humans, SPD absorption depends on the intestinal environment, where it is synthesised. Given the significant interindividual variability in gut microbiota composition and function, studies examining dietary SPD may produce highly heterogeneous findings.¹¹ Spermine (SPM), a downstream metabolite of SPD, plays a crucial role in cellular homeostasis, including oxidative stress regulation, ion channel modulation and autophagy. A recent investigation of high-dose SPD supplementation revealed increased blood SPM levels, whereas blood SPD levels remained unchanged. This phenomenon may be explained by enterocytic and hepatic conversion of SPD to SPM during absorption and metabolism, resulting in elevated SPM concentrations despite stable SPD levels.¹² Similarly, two studies assessing long-term adherence to a polyamine-rich diet, such as traditional Japanese foods like natto, observed significant alterations in SPM levels but no substantial changes in blood SPD levels. These findings underscore the ambiguity in attributing observed clinical improvements in cognitive function following a high-SPD diet to SPD itself, as opposed to other concurrent metabolic or physiological reactions.¹³

Second, discrepancies in SPD measurement methods, such as differences between serum/plasma and RBC SPD concentrations, may lead to varied clinical outcomes. Most circulating SPD (>90%) is associated with RBCs, while plasma SPD accounts for only approximately 1.2% of whole-blood SPD. The low concentration of SPD in plasma makes accurate detection challenging. Moreover, even minor haemolysis during sample collection or processing can result in SPD leakage from RBCs, potentially skewing measured concentrations.¹⁴ These factors may partly explain the inconsistencies observed in studies relying on serum SPD measurements. To enhance reliability, measuring SPD levels in whole blood is recommended, as this approach more accurately reflects its true concentration in the circulatory system.

Third, the relationship between peripheral SPD levels and brain SPD concentrations remains poorly understood, introducing further complexity in interpreting observational data. Under normal physiological conditions, SPD is generally unable to cross the blood–brain barrier (BBB). However, studies suggest that SPD may cross the BBB in extreme scenarios, such as in animal models subjected to lethal stress or in patients with traumatic brain injuries where BBB dysfunction occurs. It has been hypothesised that individuals with cognitive impairment may exhibit compromised BBB integrity, potentially allowing SPD to accumulate in the brain, which could explain higher SPD levels observed in both brain and blood in such patients compared with healthy controls.¹⁴ For instance, Graham et al reported that participants with MCI who subsequently progressed to AD had higher plasma SPD levels than those with stable MCI. In addition to the reasons mentioned above, this observation has been attributed to the conversion of putrescine to SPD and SPM in individuals progressing to AD, as opposed to alternative pathways for putrescine metabolism (eg, conversion to N-acetyl putrescine or 4-amino butanal) in those with stable MCI.¹¹

On the other hand, some evidence indicates that SPD may cross the BBB even in non-pathological conditions, as observed in studies involving healthy elderly mice.⁶ This raises questions about whether blood SPD levels reliably reflect brain SPD concentrations in both healthy and pathological states. Given the ongoing debate regarding SPD’s ability to traverse the BBB in non-pathological conditions, caution is warranted when interpreting blood SPD levels as proxies for brain SPD levels.

Despite these limitations, 18 studies were rated as good quality based on the NOS, lending a degree of credibility to the findings (online supplemental table S1). However, the observed variations in SPD measurements underscore the critical need for more standardised methodologies in future observational studies to enable a more accurate assessment of the relationship between SPD levels and cognitive ageing. Future research efforts should prioritise the integration of diverse SPD measurement sources while accounting for their unique challenges and limitations. Additionally, longitudinal studies investigating the temporal dynamics of SPD levels about cognitive trajectories would be invaluable in addressing current inconsistencies and offering a more comprehensive understanding of SPD’s potential impact on cognitive health.

Interventional studies: dosage, tools and safety considerations

Based on the findings from interventional studies (see table 1), SPD supplementation appears to have potential as a strategy to mitigate memory decline in older adults. Among the four interventional trials reviewed, three studies reported positive effects of SPD supplementation on cognitive function, including improvements in memory performance and cognitive assessments.15,17 However, one study did not observe significant improvement in cognitive outcomes following 12 months of supplementation.¹⁸ These inconsistencies may reflect the influence of several interacting factors, such as variations in study design, dosage and duration of supplementation as well as variations in participants’ baseline SPD levels and overall health status. Further investigation is warranted to elucidate these factors and optimise supplementation protocols for cognitive health.

Table 1. Summary of included interventional studies.

Author (publication year) country	Study design	Sample size and subject characteristics	Intervention	Duration of follow-up	Cohort study	Cognitive function assessments	Findings
Wirth et al (2018), Germany15	RCT A pilot study	n=28 (I: 14; P: 14) Age: I: 70.4 (5.2), P: 69.4 (5.6), Male=35.70% SCD	I: 750 mg/day of SPD-rich plant extract (with a daily SPD dose of 1.2 mg) and 510 mg cellulose, dispersed into three capsules. P: three capsules/day, contained in a total of 750 mg potato starch and 510 mg cellulose.	3 months	SmartAge	Memory performance: MST (MD, RM); GDS, MMSE, LMS delayed recall, TMT A; conventional neuropsychological tests: the German version of the AVLT, total immediate recall, DSST.	Memory performance was moderately enhanced in the SPD group compared with placebo at the end of the intervention (contrast mean=0.17, 95% CI –0.01 to 0.35; Cohen’s d=0.77, 95% CI 0 to 1.53). MD ability improved in the SPD-treated group with a medium effect size (MD=–0.11, 95% CI –0.19 to –0.03; Cohen’s d=0.79, 95% CI 0.01 to 1.55). A similar effect was not found in the placebo-treated group (MD=0.07, 95% CI –0.13 to 0.27; Cohen’s d=–0.20, 95% CI 0.94 to 0.54).
Schwarz et al (2022), Germany18	RCT	n=100 (I: 51; P: 49) Age: 69.0 (5.0), Male=51.00% SCD	I: the daily dose of administered plant extract was 750 mg (corresponding to 0.9 mg SPD, 0.5 mg SPM, 0.2 mg PUTR) administered in six capsules of 125 mg each. P: 750 mg of microcrystalline cellulose.	12 months	SmartAge	Memory performance: MST (MD, RM); GDS, MMSE, LMS delayed recall, TMT A; conventional neuropsychological tests: the German version of the AVLT total immediate recall, DSST.	SPD supplementation over 12 months did not result in a significant beneficial effect on MD performance (between-group difference, –0.03; 95% CI –0.11 to 0.05; p=0.470) as compared with placebo. Exploratory analyses indicated possible beneficial effects of the intervention on inflammation and verbal memory.
Pekar et al (2021), Austria16	RCT	n=85 (A group: 43; B group: 42) Age: 83.1 Male: A group: 23.3%, B group: 19.0% Older adults with dementia	A group: a grain roll with wheat germ (1075 mg/kg SPD) for breakfast six times a week (roll A). Each roll A contained 3.3 mg of SPD after baking. B group: rolls baked with wheat bran (115 mg/kg SPD) instead of wheat germ (roll B). Each finished roll B contained 1.9 mg of SPD.	3 months	NI	CERAD-Plus test: MMSE; verbal fluency; Boston naming test; learn, recall and recognise a word list; sign and recall figures; TMT-A and B; phonemic fluency.	A clear correlation between the intake of SPD and the improvement in cognitive performance in subjects with mild and moderate dementia was observed in the group treated with the higher SPD dosage. The most substantial improvement in test performance was found in the group of subjects with mild dementia, with an increase of 2.23 points (p=0.026) in MMSE and 1.99 (p=0.470) in phonematic fluidity. By comparison, the group that had a lower SPD intake showed consistent or declining cognitive performance.
Pekar et al (2023), Austria17	Single-arm trial	n=45 Age=83.0 (9.5), Male=N/A Older adults with dementia	A daily dose of 3.3 mg SPD in their diet.	12 months	NI	CERAD-Plus test: MMSE; verbal fluency; Boston naming test; learn, recall and recognise a word list; sign and recall figures; TMT-A and B; phonemic fluency.	The comparison of the MMSE test results at baseline and after 1 year demonstrated a significant difference (p<0.001). The mean improvement is five points.

AVLT, the German version of the Rey Auditory Verbal Learning Test; CERAD, Consortium to Establish a Registry for Alzheimer’s Disease; CI, confidence interval; DSST, digit symbol substitution test; GDS, Geriatric Depression Scale; I, intervention; LMS, Logical Memory Subscale; MD, mnemonic discrimination; MMSE, Mini-Mental State Examination; MST, mnemonic similarity task; N/A, not applicable; NI, no information; P, placebo; PUTR, putrescine; RCT, randomised controlled trial; RM, recognition memory; SCD, subjective cognitive decline; SPD, spermidine; SPM, spermine; TMT, trail making test.

First, the impact of varying dosages of SPD intake warrants careful consideration. Senekowitsch et al conducted the first pharmacokinetic study on oral SPD supplements, suggesting that the insufficient dosage of SPD might have contributed to the lack of significant effects reported in the SmartAge study.¹⁸ In this study, the dosage of SPD supplementation, selected for safety and tolerability, was notably lower than the levels associated with adverse reactions in animal models, which could potentially explain the absence of pronounced effects with low-dose SPD.¹⁶ While it is plausible that long-term supplementation with low-dose SPD may exert effects through mechanisms such as epigenetic regulation or metabolic cascades, these hypotheses require further investigation.¹² Therefore, future experimental designs should aim to optimise SPD dosages to achieve a balance between maximising cognitive benefits and ensuring safety, highlighting the importance of further research in this area.

Second, the limitations of assessment tools must be acknowledged, as research outcomes can be influenced by their sensitivity. In interventional studies, some tools, such as the Memory Screening Tool, may exhibit greater sensitivity to specific cognitive domains, such as memory-related functions.^{15 16} This variability in sensitivity may contribute to discrepancies in observed outcomes across different cognitive assessment measures.

Another critical consideration is the ‘ceiling effect’, wherein participants’ performances reach the upper measurement limit of the tools or experimental design, masking their actual ability levels.¹⁹ As a result, variations in cognitive function among participants may not be accurately captured, even if such differences exist. This limitation underscores the importance of utilising comprehensive and sensitive assessment instruments in future studies to better understand the cognitive effects of SPD supplementation.

While SPD supplementation has demonstrated potential cognitive benefits in ageing populations, its safety profile remains a key area for further investigation. Among the four interventional studies reviewed, one study reported adverse events ranging from mild to severe; however, these events were deemed unrelated to SPD supplementation. Furthermore, no significant differences were observed between the intervention and control groups in the incidence of life-threatening events, hospitalisations or malignancies, indicating that SPD supplementation is generally safe.

These findings underscore the necessity of rigorous safety monitoring in future studies. Comprehensive safety evaluations, including long-term follow-up, are essential to confirm SPD’s safety and feasibility as a potential strategy for mitigating cognitive decline in ageing populations.

Synthesis of findings: bridging evidence gaps

The findings from both observational and interventional studies offer a nuanced perspective on the role of SPD in cognitive ageing. Observational evidence suggests a potential association between SPD levels and cognitive function, with indications of a protective effect against cognitive decline. However, the variability in results, driven by inconsistencies in SPD measurement methods (eg, brain tissue, blood serum/plasma, RBC or dietary intake), poses challenges to drawing definitive conclusions.

To address these gaps, future research should prioritise standardising SPD measurement techniques and exploring whether SPD levels in the blood or dietary intake reliably reflect brain concentrations. Such efforts would provide a more robust foundation for understanding SPD’s role in cognitive ageing.

Interventional studies offer preliminary evidence suggesting that SPD supplementation may serve as a potential strategy to mitigate age-related cognitive decline. Some studies have indicated positive cognitive effects of SPD supplementation on cognitive function, such as improvements in memory performance and cognitive assessments. However, inconsistencies remain. The observed differences may be potentially due to variations in SPD dosage, the sensitivity of cognitive assessment tools and other methodological differences. Furthermore, while existing studies reported no direct association between SPD supplementation and severe adverse events, comprehensive safety monitoring remains essential in future research.

The mechanisms through which SPD may exert its cognitive benefits are still under investigation. Current evidence suggests that SPD could potentially support brain health through cellular processes, particularly via the enhancement of autophagy, a critical mechanism for clearing damaged proteins and maintaining neuronal homeostasis.^{5 11 20} SPD has also been identified as a key mediator of fasting- or CR-induced autophagy, acting through the polyamine-hypusination axis to regulate translation via eukaryotic translation initiation factor 5A.⁴ However, further research is needed to verify these mechanisms and their clinical implications in cognitive ageing.

Variations in SPD measurement methods across studies, along with individual differences in dietary intake and genetic predispositions, may partly explain the observed inconsistencies in findings from observational and interventional studies. Nevertheless, these hypotheses require further high-quality evidence to be substantiated. Future research should focus on exploring the relationship between blood, brain and dietary SPD levels while accounting for individual differences in cognitive outcomes. Such efforts will be critical for achieving a more comprehensive understanding of SPD’s role in cognitive ageing.

Conclusions

In conclusion, while SPD demonstrates potential as an intervention for cognitive ageing, the inconsistencies in existing evidence highlight the need for further standardised and long-term studies to better explore its role and efficacy.

Biography

Lirong Yu, Master of Medicine, graduated with a Bachelor of Medicine degree from Shandong Second Medical University, China, in 1999 and obtained a master's degree from West China Medical Center of Sichuan University, China, in 2006. Since 1999, she has been working as a lecturer at the School of Nursing, Shandong Second Medical University, China. From July 2023 to July 2024, she was a visiting scholar at the Department of Psychological Medicine, National University of Singapore. Her main research interests include healthy ageing, epidemiology of cognitive diseases in older adults, and prevention.