Affiliate disclosure: Some links in this article are affiliate links. We may earn a commission at no additional cost to you.

Medical disclaimer: This article reviews clinical research on anti-aging interventions including NMN, NAD+, and metformin. It is not medical advice, diagnosis, or treatment. Consult a qualified healthcare professional before making changes to your supplement regimen or considering any interventional approach.

TL;DR

• Japan has produced several of the most-cited human NMN trials, primarily from Keio University researchers and their US collaborators. Results are promising on biomarker endpoints but remain preliminary on clinical outcomes.
• Japanese institutions are running cohort studies connecting genomic, dietary, and lifestyle data from centenarian-dense regions — particularly Okinawa — with cellular aging markers including NAD+ decline.
• Metformin has been examined in Japan as a potential geroprotective agent in observational and smaller intervention settings; no Japanese trial equivalent in scale to the US TAME study has been registered as of 2025.
• Trial stage, peer-review status, and endpoint differences matter significantly when reading Japan anti-aging research. Most interventional work is Phase I/II evaluating biomarker endpoints, not hard clinical outcomes.

Japan’s position in NAD+ and NMN research

Japan is positioned differently from most Western countries in the anti-aging research landscape for two reasons that compound each other. It has pharmaceutical-scale NMN producers — Mitsubishi Gas Chemical and Shinkoso, among others — with both the capacity and commercial interest to fund domestic clinical work. And it has a population with the world’s highest proportion of centenarians (approximately 92,000 as of 2023 per Japan Statistics Bureau data), which generates research cohort opportunities that are genuinely unavailable in most other countries.

That combination means Japanese institutional NMN research often draws on both industrial-grade supply chains and unusually relevant study populations. It does not mean the field is more conclusive in Japan than elsewhere. Trial sizes remain modest. Endpoints remain largely biomarker-level rather than clinical outcomes. And several of the most-cited Japan-origin NMN papers are international collaborations rather than purely domestic work.

Keio University NMN program

Keio University’s research, particularly work from its Department of Physiology and collaborating US institutions, represents the best-documented Japanese institutional contribution to NMN human trials.

Yoshino et al. 2021 (Science, PubMed 34407019): A randomized, double-blind, placebo-controlled trial enrolling 25 postmenopausal women with prediabetes. Participants received 250 mg/day oral NMN for 10 weeks. The primary finding: NMN supplementation was associated with improved muscle insulin sensitivity, with skeletal muscle gene expression changes consistent with increased glucose utilization. This is a Washington University / Keio collaboration. What it establishes: a specific biomarker endpoint in a specific population. What it does not establish: general anti-aging effects in healthy adults, benefit at doses above 250 mg/day, or effects in men or younger populations.

Igarashi et al. 2022 (NPJ Aging): A Keio-led trial in older adults at 250 mg/day examining functional endpoints including grip strength and walking speed. Modest changes were observed relative to placebo, with within-group variation large enough to warrant replication before strong conclusions. Peer-reviewed and published; effect sizes are not large enough to constitute clinical proof of concept on their own.

Both trials contribute to what researchers call “proof of target engagement” — NMN reaches tissue, raises NAD+, and produces downstream metabolic changes consistent with the NAD+ thesis. Neither trial answers whether those metabolic changes translate to population-level health outcomes. Additional 2024-2025 Japanese and international trials at doses of 250–1000 mg/day in adults aged 40–80 have broadly replicated the NAD+ rise and modest biomarker signal findings, without yet producing a clinical endpoint result.

Tokyo University and NAD+ aging research

University of Tokyo researchers have contributed primarily to the mechanistic and cohort side of NAD+ aging research rather than large interventional NMN trials. Several labs have examined NAD+ metabolism in the context of aging biology, including relationships between NAD+ decline, sirtuin function, and epigenetic aging markers measured in Japanese adult cohorts.

One relevant Tokyo University contribution involves NAD+ measurement in population cohort data. Researchers have documented that blood NAD+ levels in Japanese adults decline substantially between ages 40 and 80, following a trajectory broadly consistent with findings in other populations. What this means for trial interpretation: the reference population has different longevity baseline characteristics than most Western cohorts. Whether this means supplementation effect sizes will differ remains speculative — baseline NAD+ and decline trajectory are only two of many confounders.

As of 2025, no large interventional NMN trial with a registered clinical endpoint has been published primarily from University of Tokyo labs. Most Tokyo University aging work is in genomics, cellular mechanism, and cohort characterization — the scaffolding for future trials rather than the trials themselves. For the genomic side of this work, the centenarian FOXO3 research is covered in Centenarian genome research: Okinawa and FOXO3 variants.

Okinawa centenarian research and regional longevity cohorts

The Okinawa Centenarian Study (OCS) is the longest-running systematic study of centenarians worldwide, initiated in 1975 through the University of the Ryukyus and subsequently involving collaborating institutions across Japan and internationally. Its contribution to the NMN/NAD+ clinical question is indirect but structurally important.

OCS data establishes that Okinawan centenarians tend to have better-preserved metabolic markers than age-matched peers from other populations. This is correlated with dietary patterns, social structures, and genetic variants — but confounding from each of these factors is substantial, and the study design does not support isolating any single cause. The FOXO3 longevity-associated variant appears at higher frequency in Okinawan and Japanese centenarian cohorts than in most other studied populations (Willcox et al., 2008), and the FOXO3 pathway intersects with NAD+/sirtuin signaling — which is why the centenarian cohort data is cited in mechanistic arguments for NAD+ precursor research.

Critically: post-WWII dietary change in Okinawa has substantially weakened the region’s longevity advantage relative to mainland Japan — a natural experiment that argues for lifestyle factors contributing independently of genetics. This cuts both ways for the NMN thesis. It supports the relevance of metabolic interventions. It also illustrates how quickly dietary and behavioral confounders can outweigh any single bioactive compound.

What the Okinawa cohort does not provide is direct evidence that NMN supplementation replicates centenarian metabolic characteristics in people who are not already long-lived. That evidence would require a randomized trial in a relevant population with a clinical endpoint — which does not yet exist.

TMDU and metformin as a geroprotective candidate

Tokyo Medical and Dental University (TMDU) has an active aging research program that includes work on geroprotective pharmacology — the use of existing drugs with established safety profiles to potentially modify biological aging processes.

Metformin is the most prominent candidate in this category globally. It is an inexpensive, off-patent type-2 diabetes medication with decades of safety data in diabetic populations, and observational studies have consistently found lower cancer incidence and modestly better survival in metformin-treated diabetic patients compared to those on other medications. The US TAME trial (Targeting Aging with Metformin, Albert Einstein College of Medicine) is the most ambitious attempt to formalize this into a registered clinical trial using a composite aging endpoint recognized by the FDA as a trial outcome.

Japanese institutional work on metformin and aging — including TMDU-adjacent research — sits at a different stage. As of 2025, the published Japanese work is primarily mechanistic and observational rather than large-scale interventional. AMPK activation, mTOR inhibition, and mitochondrial effects are well-characterized at the cellular level. Observational analyses of metformin use in Japanese diabetic cohorts provide supporting epidemiological data. A large-scale registered Japanese trial equivalent to TAME has not been announced or registered as of 2025.

Framing metformin as a confirmed geroprotective agent in humans — Japanese or otherwise — runs ahead of the available evidence. The TAME results, expected later this decade, will be the more informative data point.

How to read Japan anti-aging trial claims

Several calibration frameworks are useful when evaluating Japanese anti-aging research claims:

Phase I/II versus clinical outcome: The Keio NMN trials are Phase I or II equivalent by design. They establish bioavailability, tolerability, and biomarker engagement. A positive Phase II biomarker result does not establish clinical efficacy. Phase III trials with preregistered clinical endpoints — hospitalization, diagnosed disease, mortality — distinguish “associated with biomarker change” from “associated with reduced risk of disease.”

Peer review versus preprint or press release: Several NMN trial results circulate in conference abstracts, press releases, or preprints before appearing in peer-reviewed journals with full methodology and data. Peer-reviewed publication is the minimum interpretive floor.

Endpoint selection: A trial measuring NAD+ blood levels, grip strength, or subjective wellness scores answers a different question at a different confidence level than a trial measuring hospitalization rates or disease incidence. Both are informative at their respective levels; conflating them produces overclaiming.

Funding source and conflict of interest: NMN trials funded by NMN manufacturers should be read with the same attention to conflict of interest that applies to pharmaceutical-industry-funded drug trials. The results may be valid; endpoint selection and interpretation may reflect optimism that independent replication does not always confirm.

Where the trial landscape stands

Japan’s institutional contributions to anti-aging research — particularly the Keio NMN work, the Okinawa centenarian cohorts, and mechanistic work on sirtuin and NAD+ biology — are substantive and internationally recognized. They are not, however, ready evidence for strong efficacy claims about NMN supplementation.

The honest position in 2025:

• NMN raises NAD+ in humans at studied doses. Established.
• Some biomarker changes consistent with the aging hypothesis occur at studied doses. Preliminary, not consistently replicated across populations.
• No clinical outcome evidence from trials — in Japan or elsewhere — supports claims that NMN supplementation is associated with reduced mortality, cancer incidence, or cardiovascular events in humans.
• Human outcome data is preliminary for NMN and NAD+ precursor research generally; the next generation of trials is running but not yet complete.

For a product-focused view of the NMN market, see NMN supplements: Japan hype vs. evidence and NMN vs NR: what human trials actually compare on dose, cost, and safety. For context on Japan’s longevity statistics and why Japanese cohorts matter to this research, see Japan longevity statistics: WHO and OECD data.

If you are considering NMN supplementation based on the Japanese trial data: iHerb carries NMN from brands including those with current certificates of analysis. The evidence supports a starting dose of 250–500 mg/day rather than the higher marketing-driven doses; going above this range is not supported by the dose-response data from any of the trials above. And the question of whether NMN makes sense for you given your specific health history belongs with a clinician, not a supplement retailer.

For a broader framework on stacking Japanese longevity supplements, see Japanese longevity supplement stack: a beginner’s guide.

---

Part of our research series. See also: Centenarian genome research: Okinawa and FOXO3, Japan longevity statistics: WHO and OECD data.