Effects of Spermidine Supplementation on Cognition and Biomarkers in Older Adults With Subjective Cognitive Decline: A Randomized Clinical Trial

Abstract

Importance: Developing interventions against age-related memory decline and for older adults experiencing neurodegenerative disease is one of the greatest challenges of our generation. Spermidine supplementation has shown beneficial effects on brain and cognitive health in animal models, and there has been preliminary evidence of memory improvement in individuals with subjective cognitive decline.

Objective: To determine the effect of longer-term spermidine supplementation on memory performance and biomarkers in this at-risk group.

Design, setting, and participants: This 12-month randomized, double-masked, placebo-controlled phase 2b trial (the SmartAge trial) was conducted between January 2017 and May 2020. The study was a monocenter trial carried out at an academic clinical research center in Germany. Eligible individuals were aged 60 to 90 years with subjective cognitive decline who were recruited from health care facilities as well as through advertisements in the general population. Data analysis was conducted between January and March 2021.

Interventions: One hundred participants were randomly assigned (1:1 ratio) to 12 months of dietary supplementation with either a spermidine-rich dietary supplement extracted from wheat germ (0.9 mg spermidine/d) or placebo (microcrystalline cellulose). Eighty-nine participants (89%) successfully completed the trial intervention.

Main outcomes and measures: Primary outcome was change in memory performance from baseline to 12-month postintervention assessment (intention-to-treat analysis), operationalized by mnemonic discrimination performance assessed by the Mnemonic Similarity Task. Secondary outcomes included additional neuropsychological, behavioral, and physiological parameters. Safety was assessed in all participants and exploratory per-protocol, as well as subgroup, analyses were performed.

Results: A total of 100 participants (51 in the spermidine group and 49 in the placebo group) were included in the analysis (mean [SD] age, 69 [5] years; 49 female participants [49%]). Over 12 months, no significant changes were observed in mnemonic discrimination performance (between-group difference, -0.03; 95% CI, -0.11 to 0.05; P = .47) and secondary outcomes. Exploratory analyses indicated possible beneficial effects of the intervention on inflammation and verbal memory. Adverse events were balanced between groups.

Conclusions and relevance: In this randomized clinical trial, longer-term spermidine supplementation in participants with subjective cognitive decline did not modify memory and biomarkers compared with placebo. Exploratory analyses indicated possible beneficial effects on verbal memory and inflammation that need to be validated in future studies at higher dosage.

Trial registration: ClinicalTrials.gov Identifier: NCT03094546.

Conflict of interest statement

Conflict of Interest Disclosures: Drs Madeo, Sigrist, and Eisenberg reported having equity interests in and advisory roles for The Longevity Labs GmbH (TLL), and reported receiving support for travel and conference fees. Dr Stekovic reported employment as chief executive officer of TLL during planning, initiation, and the time the trial was conducted and had equity interest at TLL until December 2019. Dr Sigrist reported receiving personal fees from Ludwig Boltzmann Gesellschaft, i2c incubator (TUV), Swiss Milk Association, and ÄGHE outside the submitted work. Dr Sigrist was part of the advisory board of NovoArc GmbH, Global Thinkers Hub GmbH, European Forum Alpbach AIM, and CellEctric Bioscience GmbH, all without payments. Dr Flöel reported receiving personal fees from Bayer, Roche, Biogen Idec, and Novartis outside the submitted work. Dr Eisenberg reported receiving personal fees from Infectopharm and work-related lectures; he reported a US patent pending for a technique in enhancing mitochondrial respiration. Dr Madeo reported advisor fees for Samsara Therapeutics and personal fees from work-related lectures; he reported having a US patent pending for technology related to spermidine. No other conflicts were reported.

Figures

Figure 1.. Study Flowchart

AE indicates adverse event; SAE, serious adverse event; SCD, subjective cognitive decline. Per-protocol set included all participants completing the 12-month intervention (89 participants). Multiple imputation was performed for intention-to-treat analyses of full analysis set and per-protocol set.

Figure 2.. Effect of Spermidine Supplementation on Mnemonic Discrimination (MD) Performance

Notched box plots display within-group effects in MD performance in the spermidine group and the placebo group at baseline and postintervention assessments. Box plots include data of all participants who completed postintervention assessment. Notches in the shaded regions indicate 95% CI of the median, and shaded boxes the IQRs with lower (25th) and upper (75th) percentiles. Dots represent individual data points.

Figure 3.. Effect of Spermidine Supplementation on Primary and Selected Secondary Outcomes in Intention-to-Treat Analysis

ALT indicates alanine aminotransferase; AST, aspartate aminotransferase; CRP, C-reactive protein; DBP, diastolic blood pressure; ECog-39, Everyday Cognition Scales (39 items); eGFR, estimated glomerular filtration rate; HDL, high-density lipoprotein; IFN-gamma, interferon gamma; INR, international normalized ratio of blood clotting; LDL, low-density lipoprotein; MST, Mnemonic Similarity Task; PSQI, Pittsburgh Sleep Quality Index; PSWQ, Penn State Worry Questionnaire; SBP, systolic blood pressure; SF-12, Short Form Health Survey; sICAM-1, soluble intercellular adhesion molecule-1; sVCAM-1, soluble vascular cell adhesion molecule-1; TMT, Trail Making Test; TNF-alpha, tumor necrosis factor alpha; VLMT, Verbaler Lern-und Merkfähigkeitstest, German version of the Rey Auditory Verbal Learning Test (AVLT); WHOQOL-BREF, World Health Organization Quality of Life. Missing data were imputed using multivariate imputation by chained equations (mice) based on 30 imputed data sets and predictive mean matching. Mean changes of each parameter from baseline to postintervention assessment with 95% CIs are presented for both intervention groups separately. Group differences and P values result from analysis of covariance models for change in outcome from baseline to postintervention visit, with the intervention group as the examined factor and adjusted for age, sex, and the particular baseline measure. Mean group differences in the forest plots were standardized by being converted into z scores. Forest plots and group differences were transformed, if necessary, to yield the same direction of effect.