Rapamycin, Not Metformin, Mimics Calorie Restriction's Lifespan Benefits: What a 167-Study Meta-Analysis Reveals

Table of Contents

• Key Points
• Why This Research Matters
• Understanding the Three Treatments
• How the Study Was Conducted
• Key Findings: What the Data Reveals
• Sex Differences and Dietary Restriction Methods
• What This Means for Patients
• Study Limitations
• Recommendations for Patients and Researchers
• Frequently Asked Questions
• Source Information

Key Points

• A meta-analysis of 167 studies found rapamycin significantly extends lifespan in vertebrates, similar to dietary restriction.
• Metformin did not show a statistically significant lifespan benefit in vertebrates in the same meta-analysis.
• Rapamycin produced about a 24% average lifespan increase; dietary restriction about a 19% increase; the difference was not statistically significant.
• No consistent sex differences were found for dietary restriction or rapamycin lifespan effects in animal studies.
• Neither drug is approved for anti-aging; patients should consult a healthcare professional before any off-label use.

Why This Research Matters: The Search for a "Calorie Restriction in a Pill"

For over 100 years, scientists have known that dietary restriction (DR)—reducing food intake without entering a state of malnutrition—can dramatically extend lifespan. This effect was first documented in the early 20th century (Osborne et al. 1917; McCay et al. 1935) and has since been confirmed across an astonishing range of species, from microscopic roundworms (Caenorhabditis elegans) and fruit flies (Drosophila melanogaster) to mice, rats, and primates (Bodkin et al. 2003; Anderson et al. 2009; Fontana et al. 2010; Nakagawa et al. 2012).

But there is a major catch: studies show that humans find long-term calorie restriction very difficult to maintain. Adherence to such diets is consistently low (Scheen 2008; Barte et al. 2010; Selman 2014; Di Francesco et al. 2024). This reality has fueled intense interest in so-called "DR mimetics"—compounds that could trigger the same biological responses as calorie restriction without requiring people to actually eat less (Mattson et al. 2001; Ingram et al. 2006; Mouchiroud et al. 2010).

Two drugs have dominated this field: rapamycin and metformin. Both are already approved for use in humans, but for entirely different purposes. The question of whether they genuinely extend lifespan—and how their effects compare to actual dietary restriction—has remained unresolved, especially in vertebrates. This meta-analysis was designed to finally answer that question with rigorous statistical methods.

Understanding the Three Treatments

Dietary Restriction (DR)

Dietary restriction involves reducing caloric or food intake while avoiding malnutrition. In this study, the researchers examined two main forms of DR: a percentage reduction in calories or food intake, and fasting (going without food for a defined period). They explicitly excluded studies that tested isocaloric protein reduction or other macronutrient manipulations. Notably, DR is not universally beneficial in every species—some studies have found negative or neutral effects (Harper et al. 2006; Sohal et al. 2009)—which is why a comprehensive meta-analysis was needed.

Rapamycin (Sirolimus)

Rapamycin was discovered in 1975, isolated from soil bacteria found on Easter Island (Vézina et al. 1975). It is an FDA-approved immunosuppressant used for kidney transplant patients and in cardiac stents (Kaeberlein et al. 2023). The drug works by inhibiting the mechanistic target of rapamycin (mTOR) pathway, a master regulator of cell growth, metabolism, and aging. Previous studies have shown that rapamycin extends lifespan in mice (Harrison et al. 2009; Miller et al. 2011), reduces epigenetic aging (Horvath et al. 2019), and may reduce age-related diseases in humans (Lee et al. 2024). However, not all findings have been positive—rapamycin did not reduce epigenetic aging in common marmosets (Horvath et al. 2021) and had mixed effects on aging rates in mice (Neff et al. 2013).

Metformin (Dimethylbiguanide)

Metformin is one of the most widely prescribed drugs in the world, used primarily to treat type II diabetes. It works by lowering circulating glucose and improving insulin sensitivity (Bailey and Turner 1996), and it activates an enzyme called AMPK (adenosine monophosphate-activated protein kinase), a key energy sensor in cells. Studies have reported lifespan extension in nematodes (Onken and Driscoll 2010) and mice (Anisimov et al. 2005), and a 2024 study found it decelerated aging in male cynomolgus monkeys (Yang et al. 2024). However, the overall evidence for metformin's lifespan benefits in vertebrates has remained inconclusive (Selman 2014; Mohammed et al. 2021).

How the Study Was Conducted: A Rigorous Meta-Analysis

The research team—from the University of Glasgow and the University of East Anglia in the UK—followed the gold-standard PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines for their systematic review and meta-analysis. They also adhered to additional specialized checklists, including the PRISMA-EcoEvo checklist (O'Dea et al. 2021) and the MATES (Meta-analysis Appraisal Tool for Environmental Sciences) checklist (Morrison et al. 2025).

Literature Search

The researchers searched two major scientific databases, Scopus and Web of Science, in July 2023, and then updated the search in December 2024 to capture the most recent research. Both published and unpublished studies were included. After removing duplicates using the synthesisr software package, they screened thousands of papers using specialized tools (Rayyan and metRscreen). They also added references that appeared in five key papers found during the search but were not captured by the initial database queries.

Data Extraction

To be included in the analysis, a study had to meet strict criteria:

• Included an experimental group receiving a lifespan intervention (dietary restriction, rapamycin, or metformin) and a corresponding control group
• Involved vertebrate animals only
• Provided a measure of lifespan (mean, median, or survival curve data)
• Reported sample sizes and either standard deviation or standard error (or had survival curves that crossed the 50% survival threshold)

From each study, the team extracted mean and median lifespans. When data was presented only in survival curves, they used specialized digitizing software (WebPlotDigitizer and metaDigitise) to extract the numbers. They also recorded, when applicable, data separately for male and female animals—never combining them into a "mixed" group unless that was how the original study reported it.

Statistical Approach

The analysis used a statistical metric called the log-response ratio, which measures the proportional difference in lifespan between treated and control animals. This metric was adjusted for sample size bias following established methodology (Lajeunesse 2015). The team used multi-level modeling—a sophisticated statistical technique that accounts for the fact that multiple measurements from the same study or species are not truly independent of one another.

Two different approaches were used to handle missing standard deviations, and because both produced qualitatively identical results, the authors presented the "all cases" method. They also tested for publication bias using two standard methods: small-study bias (whether smaller studies produce different results than larger ones) and time-lag bias (whether the field's findings have shifted over time).

In total, the analysis included data from 167 papers, yielding 911 effect sizes (effect size is a standardized measure of how large a treatment's effect is). Not every treatment group was represented equally: dietary restriction contributed the most effect sizes (k = 677 from 115 papers), followed by rapamycin (k = 188 from 38 papers), and metformin (k = 46 from 17 papers).

Key Findings: What the Data Reveals

Dietary Restriction Works—Robustly

Dietary restriction produced a significant lifespan extension in vertebrates. The average effect size was 0.172 (95% confidence interval: 0.132 to 0.213; p < 0.001) when adjusting for publication bias, and 0.177 (0.143 to 0.210; p < 0.001) without adjustment. In plain terms, this translates to roughly a 19% average increase in lifespan (since e^0.172 ≈ 1.19). This result was consistent whether the analysis used mean lifespans, median lifespans, or both combined.

Rapamycin Mirrors Dietary Restriction

Rapamycin also produced a significant, robust lifespan extension. Its average effect size was 0.216 (95% CI: 0.152 to 0.279; p < 0.001) after publication bias adjustment, and 0.204 (0.147 to 0.261; p < 0.001) without adjustment. This corresponds to approximately a 24% average increase in lifespan (e^0.216 ≈ 1.24).

Importantly, when the researchers directly compared rapamycin to dietary restriction, the two treatments were not statistically different from each other—whether adjusting for publication bias (p = 0.221; difference = 0.044, 95% CI: −0.026 to 0.114) or not (p = 0.406; difference = 0.028, 95% CI: −0.038 to 0.093). This means that, at the population level, rapamycin produces a degree of lifespan extension comparable to dietary restriction itself, even though rapamycin's average effect size was consistently slightly higher.

Metformin Falls Short

In striking contrast, metformin failed to demonstrate a statistically significant lifespan extension in vertebrates. Its average effect size was 0.086 (95% CI: −0.007 to 0.178; p = 0.069) with publication bias adjustment, and 0.078 (95% CI: −0.012 to 0.168; p = 0.088) without adjustment. Because the confidence intervals in both models crossed zero, the effect could not be distinguished from "no effect."

The difference between metformin and rapamycin was statistically significant (adjusted p = 0.017; difference = 0.130, 95% CI: 0.023 to 0.237), and metformin was also significantly worse than dietary restriction in the unadjusted model (p = 0.044; difference = 0.098, 95% CI: 0.003 to 0.194)—though this difference did not reach significance in the publication-bias-adjusted model (p = 0.081; difference = 0.086, 95% CI: −0.011 to 0.184).

Species Breakdown: Mice Dominated the Data

The eight vertebrate species represented in the meta-analysis were heavily weighted toward rodents:

• Mice: 787 effect sizes from 127 papers (by far the most studied)
• Rats: 83 effect sizes from 32 papers
• Rhesus macaques: 23 effect sizes from 4 papers
• Dogs: 6 effect sizes from 2 papers
• Redtail killifish: 5 effect sizes from 2 papers
• Turquoise killifish: 4 effect sizes from 1 paper
• Stickleback fish: 2 effect sizes from 1 paper
• Mouse lemurs: 1 effect size from 1 paper

Reassuringly, the key patterns held up even when the researchers limited the analysis to mice alone, the most represented species. For dietary restriction, the lifespan-extending effect also remained when they applied the "900-day rule" (a quality-control criterion proposed by Pabis et al. 2024 that excludes potentially problematic long-lived mouse cohorts), although the number of available metformin effect sizes became too small for reliable analysis in these sensitivity checks.

Sex Differences and Dietary Restriction Methods

Sex Does Not Consistently Matter

The study included data from male-only (k = 428 from 114 papers), female-only (k = 380 from 77 papers), and mixed-sex (k = 103 from 35 papers) groups. Across nearly all statistical models, there were no consistent sex-specific differences in how treatments affected lifespan. In only one model involving metformin did publication bias adjustment reveal a sex-related difference. This suggests that the lifespan benefits of dietary restriction and rapamycin apply roughly equally to both sexes—good news for translational potential.

Method of Dietary Restriction Didn't Change the Outcome

Dietary restriction was implemented in three ways across the included studies: percentage reduction in caloric intake (k = 610 from 103 papers), fasting (k = 63 from 18 papers), or a combination (k = 4 from 1 paper). The effect of DR was robust to these methodological differences—it did not matter whether animals were fed fewer calories daily or fasted intermittently; the lifespan benefit appeared either way. This finding is clinically relevant because fasting and calorie reduction may have different practical implications for translation to human use.

What This Means for Patients

This study offers the strongest evidence to date that rapamycin—not metformin—holds the greater promise as a true "calorie restriction mimetic" for humans. It is important to be clear that neither drug is currently approved for anti-aging purposes, and no one should take them off-label for longevity without medical supervision.

However, the findings do have tangible implications:

• For people already taking metformin for diabetes: This study does not suggest metformin is harmful—it simply found no consistent evidence that it extends lifespan in vertebrates. Its established benefits for blood sugar control and insulin sensitivity remain valid.
• For aging research: Rapamycin's consistent, significant lifespan extension across species—comparable to actual dietary restriction—makes it the leading candidate for future clinical trials in humans.
• For the "anti-aging" supplement market: Consumers should be skeptical of metformin-based longevity products, as the scientific evidence for lifespan extension in vertebrates is weak and statistically nonsignificant.
• For males and females alike: The absence of consistent sex differences suggests that benefits, if they translate to humans, would likely apply to both men and women.

It is also worth emphasizing that dietary restriction itself remains the gold standard. The fact that rapamycin's effect was statistically indistinguishable from DR means the drug might eventually serve as a substitute for people who cannot maintain severe calorie restriction—but it does not mean rapamycin is proven safe or effective for longevity in humans yet.

Study Limitations: What This Study Couldn't Prove

Meta-analyses are powerful, but they inherit the weaknesses of the studies they combine. The authors were transparent about several important limitations:

• Very high heterogeneity (I² = 96.5%): This means the individual study results varied enormously, far beyond what would be expected from random chance alone. Only a tiny portion of this variability was explained by species differences (0%), with the rest split between within-study variation (58.0%) and between-study variation (38.5%). In plain terms, this means that individual studies' results were all over the map, even though the overall pattern was consistent.
• Publication bias: When analyzing mean lifespans separately, the researchers found significant small-study bias (p < 0.001) and time-lag bias (p = 0.011). Both biases were underestimating the true effects—meaning the published literature likely skews results. When analyzing all measures combined, no significant bias was detected, but the sensitivity of results to how lifespan is reported means conclusions should be interpreted with caution.
• Data reporting inconsistencies: Results differed depending on whether studies reported mean lifespan, median lifespan, or survival curves, though some patterns (like rapamycin's benefit) were consistent across all reporting types.
• Species imbalance: With 787 of 911 effect sizes (86%) coming from mice, the ability to draw conclusions about any single non-rodent species—including primates—is limited.
• Normality violations: Only 5 of 911 effect sizes (0.5%) failed the Geary normality test, and the authors included them in the analysis, noting that excluding them made no qualitative difference.
• What it doesn't measure: This meta-analysis focused on lifespan (quantity of life), not healthspan (quality of life). A treatment that extends lifespan could still leave patients with significant age-related disease burden, and the analysis could not assess this crucial dimension.

Recommendations for Patients and Researchers

For patients interested in healthy aging, this study offers several practical takeaways:

1. Don't start taking rapamycin or metformin for anti-aging purposes without a doctor's supervision. Rapamycin is an immunosuppressant with potentially serious side effects. Its long-term effects on healthy humans are unknown.
2. Talk to your doctor about proven interventions. Although hard to maintain, dietary restriction—or its more practical cousin, intermittent fasting—remains the most robust, scientifically supported lifestyle intervention for extending lifespan and healthspan in animals.
3. Check whether you are a candidate for existing clinical trials. Research efforts like the TAME (Targeting Aging with Metformin) trial are actively investigating these drugs. Participating in rigorous trials helps generate the evidence needed. Notably, this meta-analysis suggests metformin's promise may need re-evaluation, while rapamycin seems more deserving of future human testing.
4. Remain cautiously optimistic. The finding that a single drug (rapamycin) can reproduce the lifespan benefits of dietary restriction in animals is genuinely exciting. But the high heterogeneity and publication bias revealed in this analysis should temper expectations—what works in mice doesn't always translate to humans.
5. For researchers: The authors' data and code are freely available (Zenodo DOI: 10.5281/zenodo.15673918), enabling other scientists to build on this work.

In summary, this comprehensive meta-analysis answers a long-standing question in aging research: if a pill could replicate the benefits of eating less, what would it be? Based on 911 effect sizes spanning 167 papers and eight vertebrate species, that pill looks far more like rapamycin than metformin. But the journey from animal studies to human therapies remains long, and the statistically rigorous caution flags raised by this study—high heterogeneity and publication bias—are a reminder that aging science is still a field in adolescence, not maturity.

Frequently Asked Questions

What is the main finding of this meta-analysis about rapamycin and metformin?

In a meta-analysis of 167 studies covering 911 measurements in eight vertebrate species, rapamycin significantly extended lifespan, closely matching the effect of dietary restriction. Metformin did not produce a statistically significant lifespan benefit in vertebrates. This suggests rapamycin is the more promising drug-based alternative to calorie restriction for longevity research.

Should I take rapamycin or metformin to slow aging?

No. Neither drug is approved for anti-aging purposes, and the study does not recommend taking them off-label. Rapamycin is an immunosuppressant with potentially serious side effects, and long-term effects on healthy humans are unknown. Always talk to a qualified healthcare professional before considering any medication or dietary change for longevity.

Does metformin have any benefit for people with diabetes based on this study?

The study found no consistent evidence that metformin extends lifespan in vertebrates, but it also did not suggest metformin is harmful. For people with type II diabetes, metformin's established benefits for blood sugar control and insulin sensitivity remain valid. This meta-analysis does not change those proven uses.

How does dietary restriction compare to rapamycin for lifespan extension?

In the meta-analysis, dietary restriction produced about a 19% average increase in lifespan, while rapamycin produced about a 24% average increase. Statistically, the two treatments were not different from each other. This means rapamycin's effect at the population level is comparable to actual calorie restriction in these animal studies.

Do the lifespan benefits of rapamycin or dietary restriction differ between males and females?

The study found no consistent sex-specific differences in lifespan effects for dietary restriction or rapamycin. In nearly all statistical models, the benefits applied similarly to male, female, and mixed-sex animal groups. Only one model involving metformin showed a sex-related difference after publication bias adjustment, but this was not a consistent finding.

What animal species were included in this meta-analysis?

The eight vertebrate species were mice, rats, rhesus macaques, dogs, redtail killifish, turquoise killifish, stickleback fish, and mouse lemurs. Mice contributed 787 of 911 effect sizes, meaning 86% of the data came from mice. Therefore, conclusions about non-rodent species, including primates, are limited.

What are the main limitations of this meta-analysis?

The authors noted very high heterogeneity (I² = 96.5%), meaning individual study results varied widely. They also found publication bias that underestimated true effects, and data reporting inconsistencies. Species imbalance toward mice limits generalizability. The analysis measured lifespan, not healthspan, so it could not assess quality of life or age-related disease burden.

Source Information

This patient-friendly article is based on the following peer-reviewed research:

Original Article: "Rapamycin, Not Metformin, Mirrors Dietary Restriction-Driven Lifespan Extension in Vertebrates: A Meta-Analysis"

Authors: Edward R. Ivimey-Cook, Zahida Sultanova, and Alexei A. Maklakov

Author Contributions: Edward R. Ivimey-Cook and Zahida Sultanova contributed equally to this work.

Affiliations: School of Biodiversity, One Health, and Veterinary Medicine, University of Glasgow, UK; School of Biological Sciences, University of East Anglia, UK

Journal: Aging Cell, 2025, Volume 0, Article e70131

DOI: https://doi.org/10.1111/acel.70131

Received: 8 May 2025 | Accepted: 22 May 2025

Funding: Leverhulme Trust (ECF-2022-214) and Natural Environment Research Council (NERC NE/W001020/1)

Data Availability: All data and code are available from Zenodo (DOI: 10.5281/zenodo.15673918)

Note: This article is an open-access publication under the Creative Commons Attribution License, which permits reuse with proper attribution. This patient-friendly summary was created to make the research accessible to a broader audience. It does not constitute medical advice. Always consult a qualified healthcare professional before making any decisions about medications or dietary changes.