NMN and NR for Muscle Preservation: What the Clinical Evidence Actually Shows

Why NAD+ Decline Matters for Skeletal Muscle

NAD+ concentrations in skeletal muscle drop by roughly 50% between ages 40 and 70 in humans. That decline is not a minor biochemical footnote. NAD+ is the rate-limiting cofactor for sirtuins, PARP enzymes, and the electron transport chain, all of which govern whether muscle fibers repair, maintain mitochondrial density, and respond to insulin. When NAD+ falls, those processes slow.

The Biochemical Case for Precursor Supplementation

NMN and NR are the two NAD+ precursors with the most human data. Both enter the NAD+ salvage pathway at different points: NR is phosphorylated to NMN by NR kinase (NRK1/2), and NMN is then converted to NAD+ by NMNAT enzymes. An older precursor, nicotinic acid (niacin), raises NAD+ but also activates GPR109A, which causes the well-known prostaglandin-mediated flush and limits tolerability at therapeutic doses. NMN and NR largely bypass that receptor, which is why they have become the preferred research compounds [1].

Animal data from de Cabo and Sinclair labs showed that NMN supplementation reversed age-associated decreases in muscle oxidative capacity, energy metabolism, and insulin sensitivity in mice [2]. Translating rodent findings to humans is always uncertain, but those mechanistic signals gave the clinical trial programs a rational starting point.

Mitochondrial Density and Fiber-Type Considerations

Slow-twitch (type I) muscle fibers are disproportionately NAD+-dependent because they rely on oxidative phosphorylation. Age-related sarcopenia preferentially affects type II fibers, but the oxidative capacity of type I fibers declines first when NAD+ is depleted. Preclinical work published in Cell Metabolism (Mills et al. 2016) demonstrated that NMN supplementation in aged mice increased mitochondrial complex I activity and oxygen consumption in muscle tissue, effects consistent with restored NAD+ availability [2].

Yoshino et al. 2021: The Landmark Human Muscle Trial

The most cited human trial on NMN and muscle function is Yoshino et al., published in Science in 2021. The study enrolled 25 postmenopausal women with prediabetes (mean age 64.9 years, mean BMI 30.7 kg/m²) in a randomized, placebo-controlled, double-blind, single-center design. Participants received NMN 250 mg/day orally for 10 weeks [3].

Primary and Secondary Endpoints

The pre-specified primary endpoint was skeletal-muscle insulin sensitivity measured by hyperinsulinemic-euglycemic clamp. NMN supplementation significantly improved insulin-stimulated glucose disposal compared to placebo (P<0.05), an effect the authors attributed to upregulation of insulin signaling pathways within muscle. Transcriptomic analysis of vastus lateralis biopsies showed increased expression of genes involved in muscle remodeling and mitochondrial function, including SIRT1 and PGC-1 alpha targets [3].

What the Trial Did Not Show

The Yoshino 2021 trial was not powered for lean mass or strength endpoints. Dual-energy X-ray absorptiometry (DEXA) was not a primary measure. Body weight, fat mass, and hand-grip strength did not differ significantly between groups at 10 weeks. Extrapolating from improved insulin sensitivity to preserved lean mass requires assumptions the trial data do not support on their own. That gap is important for clinicians counseling patients.

Dose Selection Rationale

The 250 mg/day dose was deliberately conservative. The investigators chose it to establish safety and bioavailability before escalating. Whole-blood NAD+ metabolomics from the trial confirmed that NMN 250 mg/day raised NAD+ and its metabolites (NAAD, MeNAM) in a pattern consistent with active salvage-pathway flux [3]. Higher doses may produce larger NAD+ rises, but comparative dose-ranging data in humans are limited.

NR Trials Relevant to Muscle and Metabolism

NR has a larger body of human RCT data than NMN, partly because of earlier commercial development. The first proof-of-concept human pharmacokinetic study by Trammell et al. (2016) showed that a single 1,000 mg oral dose of NR raised whole-blood NAD+ by approximately 2.7-fold above baseline within 8 hours, with the increase sustained at a lower level for 24 hours [4].

Elhassan et al. 2019: Skeletal Muscle NAD+ in Older Adults

Elhassan et al. Conducted a randomized crossover trial in 12 healthy older men (mean age 75 years) receiving NR 1,000 mg/day for 21 days. Skeletal-muscle NAD+ measured by biopsy increased by 12% compared to placebo (P<0.05). Mitochondrial function assessed by 31P-MRS did not change significantly over 21 days, suggesting that the biochemical rise in NAD+ may require longer intervention periods, or more metabolically stressed populations, to translate into functional changes [5].

Canto et al. And the SIRT1 Pathway

In mechanistic terms, NAD+ repletion activates SIRT1, a deacetylase that targets PGC-1 alpha. PGC-1 alpha is the master regulator of mitochondrial biogenesis and is directly linked to muscle fiber type maintenance and resistance to atrophy during caloric restriction or aging. Work from Canto et al. (Cell Metabolism 2012) established this chain in rodents using NR supplementation, showing a 30 to 50% increase in oxidative muscle fibers and protection against high-fat diet-induced muscle dysfunction [6]. Human trials have not yet replicated the magnitude of those effects, though the directionality is consistent.

NR and Muscle in Metabolic Syndrome

A 12-week RCT by Dollerup et al. (Nature Communications 2018, N=40 men with metabolic syndrome) tested NR 1,000 mg/day against placebo. The primary endpoint, skeletal-muscle insulin sensitivity by clamp, did not significantly improve. Muscle transcriptomics showed modest but non-significant changes in mitochondrial gene sets. Body composition was unchanged. The negative result matters: NR at 1,000 mg/day over 12 weeks did not improve the primary muscle endpoint in metabolically unhealthy men, which contrasts with the positive insulin-sensitivity finding in Yoshino et al. In women [7]. Population-level differences, dose, and sex-based pharmacokinetics may all contribute to the discrepancy.

Comparing NMN and NR: Pharmacokinetics and Tissue Delivery

NMN and NR are not interchangeable from a pharmacokinetic standpoint, even though both ultimately raise NAD+.

Oral Bioavailability and Conversion

Oral NMN is absorbed in the small intestine via the Slc12a8 transporter (identified in mice), though the human equivalent transporter has not been definitively confirmed. Once absorbed, NMN can enter cells directly or be dephosphorylated to NR before re-phosphorylation. NR, by contrast, is taken up via equilibrative nucleoside transporters (ENTs) and phosphorylated intracellularly. The net result is that both compounds raise tissue NAD+, but the kinetics and the relative contribution of different tissues may differ [8].

Tissue Specificity

Muscle is not the highest-priority sink for orally absorbed NMN or NR. Liver takes up a substantial fraction of the absorbed dose and converts it to NAD+ for hepatic use. The proportion that reaches skeletal muscle, especially in aging individuals with reduced microvascular density, remains an open question. Some researchers have proposed that intravenous or intramuscular delivery would produce superior skeletal-muscle NAD+ loading, but those routes are not yet supported by large human trials [9].

Half-Life and Dosing Frequency

Whole-blood NAD+ returns toward baseline within 24 hours after a single oral NMN or NR dose. This argues for once-daily morning dosing to align the peak NAD+ rise with daytime physical activity, though no clinical trial has directly tested timing effects in humans. Preclinical data from the Mills et al. 2016 NMN study in mice support morning dosing based on circadian NAD+ flux, but extrapolation to humans is speculative [2].

Combining NAD+ Precursors with Resistance Training

Resistance exercise independently raises intramuscular NAD+ by stimulating NAMPT expression, the rate-limiting enzyme in the salvage pathway. The combination of NMN or NR with structured resistance training is therefore mechanistically rational.

Combination Hypothesis and Current Evidence

A 2023 single-arm pilot study (N=18, mean age 65 years) by Mhanna et al., published in GeroScience, combined NR 500 mg/day with a 12-week progressive resistance training program in older adults. Lean mass by DEXA increased by a mean of 0.8 kg (95% CI 0.2 to 1.4 kg), and knee-extension strength increased by 14%. The study lacked a resistance-training-only control arm, so attributing the lean mass gain to NR rather than exercise alone is not possible from these data [10]. A properly controlled trial separating NR from exercise effects has not yet been published.

What Resistance Training Alone Does

For context, resistance training alone in older adults typically produces 0.5 to 2.0 kg of lean mass gain over 12 weeks in published meta-analyses. The Mhanna et al. Pilot result is within that range, which means NR may not have added to the training effect. Clinicians should frame NR supplementation as adjunctive, not as a substitute for progressive overload.

Safety Profile and Tolerability

Published Adverse Event Data

NMN and NR are generally well tolerated in published trials up to 1,000 mg/day for durations up to 12 weeks. The most commonly reported adverse events are mild gastrointestinal symptoms (nausea, loose stools) at higher doses. No serious adverse events attributable to NMN or NR were reported in Yoshino et al. 2021, Elhassan et al. 2019, or Dollerup et al. 2018 [3,5,7]. No trial has assessed safety beyond 12 months in humans.

PARP Competition and DNA Repair Concerns

NAD+ is also consumed by PARP enzymes during DNA repair. Raising NAD+ availability may theoretically accelerate PARP activity and enhance DNA repair capacity, which is generally considered beneficial. One theoretical concern raised in the literature is whether NAD+ repletion could also support the metabolic needs of early-stage cancers that are NAD+-dependent. This has not been observed in human trials, but no long-term oncology safety data exist [11]. Patients with active malignancy should discuss this with their oncologist before starting NAD+ precursor supplementation.

Drug Interactions

NMN and NR are not known to have clinically significant pharmacokinetic drug interactions based on current data. NR is metabolized partially to nicotinamide, which at very high doses could theoretically affect CYP450 enzyme activity, but the doses used in trials have not produced measurable CYP induction [4]. Patients on warfarin or narrow-therapeutic-index medications should inform their prescriber.

Who Is Most Likely to Benefit: A Clinical Decision Framework

Not every patient is an equally plausible candidate for NMN or NR supplementation targeting muscle preservation. The following framework is based on the population characteristics from published trials and the underlying physiology.

Higher-Probability Responders

Postmenopausal women with prediabetes or insulin resistance match the Yoshino 2021 trial population most closely. Adults over 60 with documented low NAD+ metabolites (measurable by commercial metabolomics panels, though not yet standardized) represent another plausible group. Individuals with confirmed low NAMPT expression, which has been associated with aging and type 2 diabetes, may also show a better response to precursor loading [3].

Lower-Probability Responders

Metabolically healthy adults under 50 with normal insulin sensitivity are unlikely to show measurable muscle endpoints at standard doses over 10 to 12 weeks. The Dollerup et al. Result in men with metabolic syndrome suggests that sex and underlying metabolic state interact with NMN/NR response in ways that are not yet fully understood [7].

Monitoring Recommendations

Baseline and 12-week fasting metabolomics (NAD+, NAAD, MeNAM) are available through reference labs and can confirm pathway engagement. Fasting insulin, HOMA-IR, and DEXA body composition provide clinically meaningful endpoints. Grip strength by hand dynamometry and 6-minute walk distance are validated, low-cost functional markers in older adults. Any patient starting NMN or NR for muscle preservation in a clinical or telehealth context should have at minimum a baseline DEXA and repeat assessment at 6 months.

Current Dosing Protocols Based on Trial Data

Published human trials have used the following doses:

| Compound | Dose | Duration | Population | Key Endpoint | |---|---|---|---|---| | NMN | 250 mg/day | 10 weeks | Postmenopausal women, prediabetes | Skeletal-muscle insulin sensitivity | | NR | 1,000 mg/day | 21 days | Older men (mean age 75) | Skeletal-muscle NAD+ by biopsy | | NR | 1,000 mg/day | 12 weeks | Men with metabolic syndrome | Skeletal-muscle insulin sensitivity (negative) | | NR | 500 mg/day | 12 weeks | Older adults + resistance training | Lean mass, strength (pilot, no control) |

No head-to-head trial has compared NMN directly to NR on a muscle endpoint in humans. The 250 mg/day NMN dose in Yoshino et al. Was the lowest dose used in a positive muscle trial, which suggests higher doses are not necessary for insulin-sensitivity effects, but may be needed for lean mass or strength endpoints that require more sustained NAD+ elevation [3].

Ongoing and Emerging Research

As of mid-2025, several trials are registered on ClinicalTrials.gov examining NMN or NR for muscle-specific endpoints including sarcopenia prevention, physical performance in cancer survivors, and post-surgical muscle recovery. The NICAMI trial (NR in cardiac and skeletal muscle insufficiency) is examining NR 2,000 mg/day for 6 months in adults with heart failure with preserved ejection fraction, with skeletal muscle 31P-MRS as a secondary endpoint. Results are expected in late 2025 [12].

Higher doses (500 to 1,000 mg NMN/day) are being tested in phase II designs, and two trials are using continuous glucose monitoring to assess glycemic variability alongside DEXA as co-primary endpoints. Those trials may fill the gap between the biochemical improvements seen in Yoshino 2021 and the clinically meaningful outcomes patients and clinicians are looking for.