On this page
In 2020, a team at the Buck Institute for Research on Aging published a mouse study that quietly reshaped how the longevity supplement market treated one specific compound. That paper is why calcium alpha-ketoglutarate now appears on the ingredient list of nearly every serious longevity stack. It is also more nuanced — and more sex-dependent — than most retailer summaries admit.
This article walks through what the researchers actually did, what they actually found, and where the paper fits in the wider Ca-AKG evidence base. For the full evidence overview across every domain, see our pillar: calcium alpha-ketoglutarate: the complete evidence-based guide. The full paper page with bibliography and PubMed link lives in the Ca-AKG Data Center.
What is the Shahmirzadi 2020 study?
The paper's full title is "Alpha-ketoglutarate, an endogenous metabolite, extends lifespan and compresses morbidity in aging mice." It was published in Cell Metabolism in September 2020 [1].
The research team tested whether feeding middle-aged mice calcium alpha-ketoglutarate would extend their lifespan and, more importantly, keep them healthier for longer. The answer was: yes to both — but the size of the lifespan effect was smaller than the size of the healthspan effect, and the effect varied by sex.
Who ran it and where was it published?
The lead author was Azar Asadi Shahmirzadi. The senior author was Brian K. Kennedy, one of the most-cited names in the modern biology of ageing. Kennedy was at the Buck Institute at the time of the study and has since moved to the National University of Singapore, where he continues to publish on Ca-AKG — including a 2025 paper on synaptic plasticity in Alzheimer mice [5].
Cell Metabolism is one of the highest-impact journals in metabolic biology. Publication there means the paper cleared serious peer review. It does not, on its own, make the findings apply to humans.
What did the researchers actually do?
The design was straightforward. Two independent cohorts of C57BL/6 mice — the standard laboratory mouse strain — were fed regular chow until they reached 540 days of age, which is roughly middle age for a mouse. At that point half of each cohort was switched to chow containing 2% calcium alpha-ketoglutarate by weight. The other half stayed on unsupplemented chow.
Each cohort was made up of roughly 45 males and 45 females, totalling 182 animals across the two cohorts. Two cohorts, not one, is important — reproducing an effect in an independent second group is a much stronger test than a single-cohort result.
The researchers then tracked two things until each mouse died naturally or reached a humane end-point.
Lifespan. How long each mouse lived from the start of Ca-AKG feeding.
Healthspan. Every eight weeks, each mouse was scored on a 31-item frailty index — a validated tool that measures 31 different signs of age-related decline across the musculoskeletal, sensory, digestive, urogenital, and respiratory systems [1]. Every score was done blind to which group the mouse was in.
This dual measurement is the paper's methodological heart. Most longevity studies report only lifespan. Shahmirzadi's team measured whether the extra weeks of life were healthy weeks or frail ones.
What did the mice given Ca-AKG show?
Lifespan
In female mice, the lifespan effect was clear and reproduced across both cohorts. Median lifespan from the start of Ca-AKG feeding was extended by 16.6% in the first cohort and 10.5% in the second. Survival at the 90th percentile of mortality — the age by which nine out of ten mice have died — was extended by 19.7% and 8% respectively [1].
In male mice, median lifespan trended up by 9.6% and 12.8% across the two cohorts — but neither survival effect reached statistical significance [1].
Frailty compression
This is where the paper gets more interesting. Ca-AKG-fed mice spent a much smaller proportion of their lives being frail. When the researchers calculated the area under the frailty curve — a summary measure of total frail time — they found a 46% reduction in females and a 41% reduction in males on Ca-AKG compared to controls [1].
Read that again. The healthspan improvement was roughly two to four times larger than the lifespan improvement. The mice on Ca-AKG did not just live longer. They spent a much smaller proportion of that longer life sick, weak, and declining. In the paper's own words: "This improvement in healthy days of life is larger than the increase in lifespan."
This is what "compressed morbidity" means. It is arguably a more clinically important finding than a modest lifespan extension.
Why did the effect look different in male and female mice?
The retailer summaries that describe this study almost universally leave out the sex-dimorphic pattern. The pattern is worth understanding.
Female mice showed statistically significant lifespan extension in both cohorts. Male mice did not — the survival curves shifted in the same direction, but not enough to clear the significance threshold. Frailty compression, however, showed up strongly in both sexes.
Sex-dimorphic responses to longevity interventions are common in mouse studies. Rapamycin, metformin, 17α-oestradiol and several other geroprotectors show sex-specific patterns in the Interventions Testing Program run by the US National Institute on Aging. Whether the Shahmirzadi effect pattern would repeat in humans — and whether the sex asymmetry would look the same — is one of the questions the ongoing human trials are designed to answer.
The honest read is: Ca-AKG produced clear lifespan extension in female mice, softer and non-significant lifespan trends in male mice, and robust healthspan improvements in both. Do not let anyone tell you the paper showed "mice lived 20% longer." That is a compressed version of what the female data showed in one cohort.
How does Ca-AKG make mice healthier?
The paper offers a mechanism, and it is worth reading in detail because it clarifies what Ca-AKG does and does not do at the cellular level.
Ca-AKG increased T-cell secretion of interleukin-10 (IL-10) — an anti-inflammatory cytokine that plays a central role in dampening chronic inflammation [1]. Chronic low-grade inflammation is one of the well-established drivers of age-related decline, sometimes called inflammaging. The IL-10 finding gives a plausible immune-mediated mechanism for the healthspan effect.
The paper also tested whether Ca-AKG cleared senescent cells — the misbehaving cells that accumulate with age and secrete a pro-inflammatory cocktail called the senescence-associated secretory phenotype (SASP). The answer was no. Ca-AKG did not reduce senescent cell numbers in mouse tissues, and it did not prevent cells from becoming senescent in culture [1].
But Ca-AKG did alter what the senescent cells secreted. In primary fibroblasts, Ca-AKG reduced the levels of the SASP factors IL-1β, IL-6, CCL2, and MMP3, without affecting classical senescence markers β-galactosidase or p21 [1]. This makes Ca-AKG a candidate senomorphic — a compound that quiets the harmful output of senescent cells without killing them — rather than a senolytic like fisetin or dasatinib, which physically clear the cells.
The pathway sits alongside earlier mechanistic findings from invertebrate work. Ca-AKG extends lifespan in worms by inhibiting ATP synthase and TOR [2], and in flies through TOR inhibition combined with AMPK activation [3]. Shahmirzadi 2020 extended the biology into a mammal and added the immune-modulation angle. For a broader look at what this all means at the cellular level, see how Ca-AKG works. The specific SASP and inflammation implications are covered in Ca-AKG and inflammation.
What the study does not show
Five things Shahmirzadi 2020 is not evidence for. All of them get claimed anyway by supplement marketing.
It does not show that Ca-AKG extends human lifespan. It is a mouse study. Mouse-to-human translation for longevity interventions has a poor track record. Every longevity intervention that has ever moved from mouse to man has done so with reduced effect size, and many have failed to translate at all.
It does not show that Ca-AKG reverses ageing. "Reverse ageing" is a marketing phrase. The paper showed compression of frailty and modest lifespan extension in one sex — neither of which is the same thing as reversing anything.
It does not show that Ca-AKG helps young or middle-aged mice. Ca-AKG was started at 18 months of age — late middle age for a mouse. The study cannot say what would happen with earlier or shorter intervention windows.
It does not show a dose–response curve. Only one dose was tested: 2% by weight in food. Whether a lower dose would work as well, whether a higher dose would work better, or whether there is a hormetic sweet spot — none of these questions was addressed.
It does not show which cellular mechanism dominates. The IL-10 finding is one plausible mechanism. The altered SASP is another. Epigenetic effects via dioxygenase cofactor function are a third. The paper points at all three; it does not prove which one drives the phenotype.
From mouse to human: what does a 2% food dose mean for a person?
This is one of the questions people search for most and the SERP most consistently gets wrong.
The Shahmirzadi mice ate roughly 4 to 5 grams of chow per day. At 2% Ca-AKG by weight, that works out to about 80 to 100 mg of Ca-AKG per day per mouse. Scaled to a 70 kg human by the body-surface-area conversion that regulators use for allometric dosing, that becomes roughly 400 to 500 mg per day — actually somewhat lower than the 1,000 mg doses used in the human observational studies and the ongoing ABLE trial.
Two caveats matter. First, body-surface-area conversion is a rough approximation, not a validated pharmacokinetic tool for Ca-AKG specifically. Second, the mice received Ca-AKG continuously in food, spread across every meal and every hour. Human capsule dosing is discrete. Whether the pharmacokinetic profile matters is unknown.
The rough translation puts human doses in the same order of magnitude as the mouse dose. It does not prove they will produce the same biological effect. For a full treatment of the trial-vs-supplement dose question, see Ca-AKG dosage: what the trial data supports.
Where does this paper fit in the broader Ca-AKG evidence?
Shahmirzadi 2020 is the anchor preclinical mammalian paper in the entire modern Ca-AKG longevity literature. Almost every subsequent claim about Ca-AKG and ageing traces back to it, either directly or through work it inspired.
The Rejuvant retrospective in 2021, which reported an average 8-year DNA methylation age reduction in supplement users, was published one year after Shahmirzadi and is often paired with it in supplement marketing — even though the two studies used entirely different populations, formulations, and endpoints [6]. A 2022 review in Trends in Endocrinology and Metabolism synthesised Shahmirzadi with the broader AKG literature and concluded the compound warranted human study across multiple domains [7]. Kennedy's own subsequent work on Ca-AKG in Alzheimer mice built directly on the Shahmirzadi biology [5].
The 2017 Archivum Immunologiae et Therapiae Experimentalis review on the pleiotropic activity of AKG situates Shahmirzadi's findings in the decades-longer clinical history of the compound — bone health, wound healing, hemodialysis, protein metabolism — showing how the ageing biology fits alongside a body of pre-existing clinical work [4].
The specific downstream questions Shahmirzadi opened up — biological age, grey hair, muscle and frailty — have each become dedicated evidence articles in this cluster: Ca-AKG and biological age, Ca-AKG and grey hair, and Ca-AKG for muscle, frailty and sarcopenia.
Bottom line
The Shahmirzadi 2020 paper is a well-designed, well-powered, dual-cohort mouse study that demonstrated modest lifespan extension in female mice and substantial frailty compression in both sexes on a Ca-AKG-supplemented diet. The healthspan finding is arguably more important than the lifespan finding, and both are more nuanced than the retailer summaries usually admit.
It is a Tier 3 preclinical study — foundational biology, not human evidence. It is the paper that legitimately opened up Ca-AKG as a serious longevity supplement candidate. It is not, on its own, evidence that any human who takes Ca-AKG will live longer or be less frail. For the full evidence picture — including the human observational data, the trials that will report in 2026, and the honest bottom line on whether to supplement — see our pillar: calcium alpha-ketoglutarate: the complete evidence-based guide.