How Semaglutide Extended Lifespan 12% in Old Female Mice

The Berkeley/NIH paper published in Nature on 2 September 2026 is the first properly designed lifespan trial of any GLP-1 receptor agonist. Old female mice, dosed daily until they died, lived about 12% longer than saline controls. The endurance, memory, senescence and stem-cell gains that came with it are far bigger than anyone clicking "Ozempic" for weight loss would think to expect.

Every outlet ran the same headline. Almost none of them said what a lifter on TRT should actually do with it. That is what this article is for.

What the Berkeley study actually found

The paper is Feng et al., Nature, 2 September 2026 (DOI 10.1038/s41586-026-10940-7; PubMed 42686906). Senior author is Danica Chen at UC Berkeley, funded by the NIA.

The design was clean. Forty 20-month-old female C57BL/6 mice (roughly a 60-year-old woman in human terms) received 10 nmol/kg per day of subcutaneous semaglutide. Thirty-nine got saline. A third arm was pair-fed to match the semaglutide group's 24% drop in food intake, isolating what weight loss alone could and could not do. The physiology cohort ran three months. The lifespan cohort was dosed until natural death, which is close to a year of daily injections in old mice.

The 12% headline and the numbers behind it

Median lifespan went from 742 days on saline to 834 days on semaglutide, a 12.4% extension. That is the number every outlet led with, and it holds up. It is a first-of-its-kind result for the GLP-1 class.

The physiology numbers reported alongside it are large. Endurance on the treadmill roughly tripled, grip strength on the hanging bar more than doubled, rotarod motor coordination and open-field exploration both climbed. Glucose tolerance improved, spatial memory improved, systemic inflammation dropped, cellular senescence markers dropped, and hippocampal neurogenesis lifted several-fold. Bone marrow colony-forming units, one of the better readouts of hematopoietic stem-cell function in an aged animal, rose clearly.

Direction on every readout is confirmed by press coverage and the abstract. The exact effect sizes are behind Nature's paywall, so treat any specific percentage as reported rather than personally verified until the paper is fully out from behind the paywall.

Where semaglutide beat calorie restriction

The pair-fed arm is the interesting part of this paper, and it is what separates it from the influencer coverage.

The CR arm matched semaglutide on endurance, motor coordination and lifespan. That is a strong hint those gains are downstream of weight loss and reduced metabolic burden, not something the drug is doing on top. If you are already lean and already in a deficit, do not assume you would layer more on by adding semaglutide.

The CR arm did not match semaglutide on exploratory drive, spatial memory or glucose control. That is where a genuinely drug-specific mechanism lives. Chen put it plainly in the Berkeley release: "These differences point to the possibility that GLP-1 drugs tap into a biological pathway independent of calorie restriction." The paper's abstract flags nutrient sensors, sirtuins and insulin/IGF-1 signalling as the shared axis, plus something extra centrally. The central GLP-1 receptor is densely expressed in the hippocampal dentate gyrus and subventricular zone, both of which are where the neurogenesis signal lit up.

Metabolic rate is the other interesting difference. CR animals dropped their metabolic rate, which is a downside of any long-term restriction. Semaglutide animals did not. That is another mark against the "just eat less" reading, and one of the reasons to take the study seriously even if you are skeptical of the mouse-to-human jump.

Senescence, stem cells and neurogenesis

The senolytic-adjacent story is the finding most likely to matter for enhanced athletes long term. p16^INK4a^ and p21, the two most-cited markers of cellular senescence, dropped substantially. So did the SASP inflammatory cytokines those cells secrete. Bone marrow stem-cell function lifted. Hippocampal neurogenesis, roughly tripled by press descriptions, lifted enough to be visible under a microscope.

Chronic AAS use elevates systemic inflammation and drives haematocrit up through a mix of erythropoiesis stimulation and reduced iron sequestration. Long-term androgen exposure also engages the senescence machinery in some tissues, which is a separate problem the mouse data does not directly speak to. But an intervention that pulls SASP down and supports HSC function is worth taking seriously for lifters carrying a decade or two of pharmacology, and it is the same mechanism our MOTS-C and SS-31 coverage is built around.

The mouse dose is not a microdose

This is the piece most posts, on Reddit and elsewhere, are getting wrong.

The math most posts get wrong

Semaglutide has a molecular weight of 4113 g/mol. The mouse dose was 10 nmol/kg per day, which works out to 0.041 mg/kg per day. If you scale that naively to a 70 kg human by mg/kg body weight, you get about 2.9 mg per day, or 20 mg per week. That is roughly eight times a Wegovy maintenance dose, which is why the naive calculation is obviously wrong and worth doing anyway. It is what non-technical readers land on when they try to translate the paper into a personal protocol.

The FDA's 2005 guidance on translating animal doses uses body surface area (BSA) rather than body mass, because clearance scales with metabolic rate and metabolic rate scales with BSA rather than weight. The Km values are 3 for mouse and 37 for human, giving a divisor of about 12.3. The human equivalent of the mouse dose is therefore closer to 0.23 mg per day, or about 1.6 mg per week.

For reference:

• Ozempic maintenance for type 2 diabetes: 2.0 mg per week
• Wegovy maintenance for obesity: 2.4 mg per week
• BSA-scaled mouse dose: ~1.6 mg per week
• Ozempic starting titration: 0.25 mg per week
• Bodybuilder cruise microdose: 0.25 mg per week

The mouse dose lands at roughly two-thirds of a Wegovy maintenance dose, and about 6.5 times the cruise microdose people are actually running. That is the number to walk away with. The 2026 netnography of older bodybuilders' semaglutide use (Turnock et al., Performance Enhancement & Health) captured users running 1.25 mg per week for anti-inflammatory and joint-pain reasons, and 5 mg per week of tirzepatide, both of which are closer to the BSA-scaled mouse dose than to a cruise dose.

What 1.6 mg per week actually looks like

Standard-dose semaglutide, not micro. The only randomised human data with a longevity signal, Sehl et al. at UCLA (medRxiv, 2025) showing GrimAge decelerated by around three years and DunedinPACE slowed roughly 9%, ran participants on the standard 1.0 to 2.4 mg per week titration, not a microdose.

There is no controlled human evidence that 0.25 mg per week delivers the SELECT-scale cardiovascular benefits or the epigenetic-age deceleration seen at label doses. The AgelessRx microdosing RCT is the trial that will actually answer the microdose question. Until it reads out, "Chen's mice validate microdosing" is a marketing story rather than an evidence-based one.

Half-life and why humans dose weekly

Mouse semaglutide half-life is short (hours). Human half-life is 165 to 184 hours, about a week (Hall et al., 2018). The gap is driven by albumin binding of the C18 fatty-diacid side chain, which is affinity-tuned to human albumin turnover. This is why mice were dosed daily and you are dosed weekly, and why the useful comparison across species is total weekly area under the curve, not per-injection amount. The BSA calculation above already gives a weekly figure, so it is the right lens.

What this study does not tell men on TRT

The paper is explicit about running female C57BL/6 mice only. That single design choice does more damage to the "should I add semaglutide to my TRT cruise for longevity" question than any other caveat in the study.

Every prior longevity drug has shown sex-differential effects

The NIA Interventions Testing Program has spent two decades documenting sex-specific responses. Metformin at 1000 ppm gave a non-significant 7% median lifespan gain in male mice and nothing in females (Strong et al., 2016). 17-alpha-estradiol extended male lifespan 19% and did nothing in females, and the effect vanished in castrated males (Garratt et al., 2018). Even rapamycin, the most reproducible geroprotector in the ITP catalogue, extends female lifespan more than male at matched doses (Miller et al., 2014; Strong et al., 2020). Calorie restriction itself shows the opposite pattern in young mice, with CR improving glucose homeostasis in young males but being blunted or absent in young females (Suchacki et al., 2023).

Direction, magnitude, and sometimes sign are not predictable across sex. Feng et al. did not test males. Assuming a 12% median lifespan gain will show up in male mice, let alone male humans on chronic androgen replacement, is not warranted by the published record.

Androgens engage the senescence machinery

Testosterone is not a neutral background for a senomorphic drug. Androgen receptor activation induces p21 and drives senescence in prostate-derived cells (Mirochnik et al., 2012), and supraphysiological androgen engages a Src-Akt senescence axis (Roediger et al., 2014). Female mice on semaglutide showed reduced hallmark-of-aging markers. Male athletes on TRT are running a chronic androgenic signal that plausibly engages the opposite arm of the same machinery in at least some tissues. The interaction is unstudied.

GLP-1 receptor biology is estrogen-modulated

The 2024 Börchers and Skibicka review in Endocrinology documents differential GLP-1R expression between sexes in adult brain regions, with estrogen up-regulating responsiveness of hindbrain GLP-1R-positive neurons. Human clinical data mirrors this: women lose more weight than men on semaglutide in heart failure (STEP-HFpEF, 2024) and in peripheral artery disease (Bonaca et al., 2025, STRIDE trial).

A TRT user on an aromatase inhibitor with suppressed estradiol has removed the exact upstream signal the female mice were riding on. That does not mean semaglutide will not work in that user. It means the mouse paper does not tell you it will.

Sedentary mice, trained lifter

The mice were sedentary and singly housed. Every reader is neither. Rapamycin blunts exercise-induced hypertrophy in older adults in some settings (Stanfield et al., 2026, RAPA-EX-01) and preserves adaptations in other settings (Elliehausen et al., 2025). The direction is not automatic for longevity drugs interacting with resistance training. Semaglutide's well-documented lean-mass loss on monotherapy, roughly a third of weight lost, is not addressed at all by a sedentary lifespan model, and is one of the reasons the existing coverage on retatrutide muscle loss matters.

The honest read is not "semaglutide extends male TRT-user lifespan." It is "semaglutide plus TRT plus training is a body-composition play, and the longevity claim from this paper does not port over."

Bloodwork to track if you use semaglutide for longevity

The mouse readouts, p16 in tissue, SASP cytokines in tissue, do not have clean human blood equivalents. But the human trial record on semaglutide is now large enough to give a workable proxy panel. If you are running semaglutide with a longevity motive rather than pure fat loss, this is the order to test in.

hs-CRP: the closest human proxy

hs-CRP is the cleanest blood signal you can track. In the SELECT trial (n=17,604, no diabetes, prior CVD, 2.4 mg per week), semaglutide dropped hs-CRP by roughly 38% versus placebo at 104 weeks, with the effect visible by weeks 4 to 8 (Plutzky et al., 2026). Pooled STEP 1/2/3 analyses showed treatment differences of 39% to 48% (Verma et al., 2023). This is one of the largest anti-inflammatory effects for any commonly-used drug in this population, and it reproduces in participants who did not lose weight.

Target: under 1 mg/L. A move from 2 mg/L at baseline to around 1.2 mg/L by month six is a realistic on-treatment endpoint.

HOMA-IR, glucose and HbA1c

HOMA-IR tracks the mechanism the mouse paper implicated. Semaglutide 1.0 mg per week in a 24-week HIV/MASLD pilot cut HOMA-IR by about 1.5 units and fasting glucose by roughly 0.5 mmol/L (Lake et al., 2024, SLIM LIVER). HbA1c falls 0.4 to 0.5% in non-diabetic obesity trials over 68 weeks. The full HbA1c effect needs about 12 weeks of steady dosing because of red cell turnover. Fasting glucose moves faster, usually within 4 to 8 weeks.

Compute HOMA-IR from a fasting insulin and fasting glucose drawn together: HOMA-IR = insulin (μU/mL) × glucose (mg/dL) / 405, or insulin (μU/mL) × glucose (mmol/L) / 22.5 in SI units. A drop from 3.5 to 2.0 is a real signal. See the existing GLP-1 bloodwork guide for the full monitoring panel.

IGF-1 and the growth-vs-longevity tradeoff

Low IGF-1 is one of the best-replicated biomarkers of exceptional longevity. Nonagenarian women in the lowest IGF-1 tertile survived significantly longer and had roughly half the cognitive impairment (Milman et al., 2014). Centenarians are enriched for IGF-1 receptor loss-of-function variants.

Semaglutide does not meaningfully lower total IGF-1 in humans. Published human data on GLP-1 agonists show total IGF-1 stays essentially flat with a modest rise in IGFBP-1, which nudges bioavailable IGF-1 down slightly without collapsing GH signalling. That is arguably the ideal profile if you care about the growth-versus-longevity tradeoff, because it moves you toward the low-IGF-1 phenotype associated with longevity without crashing the axis you need for tissue repair. For contrast, look at what MK-677 does to the same axis in the MK-677 blood sugar monitoring article. The two compounds sit on opposite ends of the tradeoff.

Cystatin-C eGFR, not creatinine

Kidney monitoring on a drug that changes muscle mass is where creatinine-based eGFR starts to lie. Lose lean mass, lose the creatinine substrate the equation is calibrated on, and eGFR looks better than it is.

Cystatin-C-based eGFR is the cleaner marker. The FLOW trial (n=3,533, T2D + CKD on 1.0 mg per week) was stopped early for benefit: kidney events cut 24%, eGFR slope preserved by 1.16 mL/min per year, MACE cut 18%, all-cause mortality cut 20% (Perkovic et al., 2024). The eGFR preservation held up under cystatin-C, so it is not an artefact of muscle-mass-driven creatinine changes. This is a reason to insist your kidney panel includes cystatin-C on any GLP-1 protocol.

ApoB, triglycerides and the lipid panel

Semaglutide drops triglycerides by 15 to 22% and ApoB by roughly 5 to 10% across the STEP program, with a smaller 3 to 5% LDL-C effect (Wilding et al., 2025). ApoB is the better marker to track because it moves more, and it captures the atherogenic particle count rather than the cholesterol content per particle. Full lipid interpretation on cycle is in the cholesterol on steroids article.

ALT falls too, driven by hepatic fat clearance. The existing semaglutide + ALT interaction page covers this in depth.

Epigenetic clocks: the UCLA signal

If you have the budget, the UCLA 2025 randomised trial in HIV lipohypertrophy is the strongest human epigenetic-clock signal on semaglutide to date. Adjusted PCGrimAge fell 3.1 years, PhenoAge fell 4.9 years, DunedinPACE slowed by roughly 9% (Sehl et al., 2025). The earlier SLIM LIVER pilot from the same group showed noise-limited effects, so the RCT is the paper to lean on.

Clocks are expensive and noisy. Run them at baseline and annually rather than every three months, and interpret them cautiously.

What if you're already cutting on a GLP-1?

The most useful reframing of the Feng et al. paper for an athlete comes from noting exactly what CR could and could not do.

Your cut is the CR arm. Semaglutide is the CR-plus arm. If you are actively dieting on 0.25 mg per week for appetite control, the endurance and motor gains in the mouse paper are already on your side, because they came out of the shared weight loss. What semaglutide is bringing on top, per the mouse paper, is exploratory drive, spatial memory, glucose control and (per the human trials) hs-CRP suppression and cardiovascular event reduction. That is the "beyond weight loss" argument, and it is a real one at the standard 1.0 to 2.4 mg per week dose. It is not established at 0.25 mg per week.

When to cruise instead of come off. For a lifter who has already used semaglutide to strip fat, coming off the drug returns hs-CRP toward baseline within weeks, fasting insulin bounces back over similar timescales, and HbA1c drifts back over three to six months. The kidney-slope benefit and any cardiovascular reduction is bought only while you are dosing. The mouse paper does not resolve the question of intermittent versus continuous dosing, because those animals were dosed until death.

If your goal is longevity rather than fat loss, and your hs-CRP is elevated at baseline, and you have the budget and the tolerance for GI side effects, the case for staying on a low maintenance dose during a lean off-season is stronger after this paper than it was before. It is still not the case Feng et al. actually proved.

Does this read across to tirzepatide, retatrutide and eloralintide?

One molecule, one strain, one sex, tested for lifespan. The urge to fan the finding out across the whole GLP-1 class is understandable and largely unsupported.

Tirzepatide has no lifespan data. Zero. The closest signal is a 2025 bioRxiv preprint showing tirzepatide reduced diet-induced senescence and NETosis-mediated liver fibrosis in obese mice, which is a hepatic-senescence-in-a-diseased-organ result rather than a lifespan or brain result. The Moody Longevity Trial (NCT07220473) is running tirzepatide 2.5 mg per week against no drug for 24 weeks with DNA methylation clocks in 55 to 70 year olds. NCT07293325 is directly comparing semaglutide against tirzepatide on genetic aging endpoints. Neither has read out. "Similar to semaglutide" is a reasonable prior for tirzepatide, not a finding.

Retatrutide is the more interesting extrapolation, because its glucagon arm drives hepatic FGF21 secretion (Cyphert et al., 2014), and FGF21 transgenic mice live 30 to 35% longer without caloric restriction via blunted hepatic GH/IGF-1 signalling (Zhang et al., 2012). That is a real longevity axis, replicated in an independent transgenic model. But nobody has run retatrutide through a lifespan protocol, so the mechanistic story is a story, not evidence. The retatrutide bloodwork guide is the place to look for the practical monitoring side.

Cagrilintide and eloralintide are amylin agonists, a different pathway entirely, and the class extrapolation from GLP-1 longevity data does not extend to them. Eloralintide is worth watching because it appears to preserve lean mass better than semaglutide in early trials (covered here), which matters for the population reading this article. Whether it does anything for hallmarks of aging is completely unknown.

If you are on tirzepatide today and doing well on it, this paper is not a reason to switch to semaglutide. If your metabolic panel is already tight (fasting insulin, HbA1c, hs-CRP, visceral fat, ApoB), the marginal geroscience benefit from any GLP-1 is unproven and probably small. The stronger case is for the metabolically drifting cohort on higher-dose AAS with creeping insulin, ferritin and blood pressure, where a low-dose GLP-1 plausibly walks those markers back.

Practical recommendations

If you're on TRT and 40+

Baseline: hs-CRP, fasting glucose and insulin (compute HOMA-IR), HbA1c, ApoB, cystatin-C eGFR, ALT, IGF-1 with IGFBP-3. This is your comparator for anything semaglutide changes.

If the panel is tight, this paper is not a reason to start. If hs-CRP is above 2 mg/L, HOMA-IR is above 2.5, or you carry visceral fat that resistance training has not touched, the case for a longevity-motivated 1.0 mg per week is stronger than it was a month ago. Rescan the panel at 12 and 24 weeks.

Do not stack semaglutide with harsh compounds (trenbolone, high-dose stimulants, clenbuterol) without a lifter-experienced clinician watching heart rate and blood pressure. The lean-mass loss on GLP-1 monotherapy is real, and TRT plus resistance training is the standard defence.

If you're actively cutting

Keep doing what you are doing. Your deficit is doing most of the work that the mouse study showed CR could reproduce. Do not add pharmacology because a mouse study went viral. Recheck ALT, ApoB and hs-CRP at week 12 to document any drift.

If you are on a cruise microdose (0.25 mg per week), be honest with yourself that the human evidence for geroprotection at that dose is not there yet. The AgelessRx trial is what will actually answer that question.

If you're cruising through a longevity block

If you decide to sit at a maintenance 1.0 mg per week for a longevity block off-season, plan the exit: hs-CRP, HOMA-IR and fasting insulin rebound within weeks of stopping, so anything you gained on hs-CRP will fade unless the weight loss holds. The kidney-slope benefit is bought only during dosing.

Track cystatin-C eGFR rather than creatinine-based eGFR. On a drug that changes muscle mass, creatinine lies. See the cystatin-C article for the full argument.

Key takeaways

• Feng et al. (Nature, 2 Sep 2026) extended aged female mouse median lifespan 12.4% on daily semaglutide, plus large gains in endurance, memory, glucose control and stem-cell function, with reduced systemic inflammation and cellular senescence.
• A pair-fed 24% calorie-restriction arm matched the drug on endurance and motor gains and on lifespan, so those are likely weight-loss effects. CR did not match the drug on exploration, spatial memory or glucose control, which is the "beyond CR" mechanistic wedge.
• The mouse dose (10 nmol/kg per day) is roughly 1.6 mg per week in a 70 kg human by BSA scaling, about two-thirds of a Wegovy maintenance dose. It is not a cruise microdose.
• The study was female-only. Every previously tested longevity drug in ITP data has shown sex-differential effects. Androgen signalling engages the senescence machinery separately, and TRT-suppressed estradiol removes one of the mechanistic pathways the mouse gains rode on.
• The best current human proxy panel is hs-CRP, HOMA-IR, cystatin-C eGFR, ApoB, ALT and IGF-1 with IGFBP-3. The UCLA 2025 randomised trial is the strongest human epigenetic-clock signal to date, at 1.0 to 2.4 mg per week.
• Do not extrapolate the finding to tirzepatide, retatrutide, cagrilintide or eloralintide. One molecule has been tested. The Moody Trial (tirzepatide) will read out first.

---

References

1. Feng, Y., Barthez, M., Wang, Y., Chen, Y., Qiu, H., Wang, C.-L., Heydari, K., Delcroix, M., Rasmussen, L. J., Bohr, V. A., & Chen, D. (2026). Late-life semaglutide treatment slows ageing and extends lifespan in female mice. Nature. DOI: 10.1038/s41586-026-10940-7. PubMed
2. Plutzky, J., Lincoff, A. M., Colhoun, H. M., et al. (2026). Effect of semaglutide on the inflammatory biomarker high-sensitivity CRP in patients with established cardiovascular disease and overweight or obesity in SELECT. Circulation. PubMed
3. Verma, S., Ryan, D., Jacob, S., et al. (2023). Effects of once-weekly semaglutide 2.4 mg on C-reactive protein in adults with overweight or obesity (STEP 1, 2, and 3). eClinicalMedicine, 55, 101737. PubMed
4. Perkovic, V., Tuttle, K. R., Rossing, P., et al. (2024). Effects of semaglutide on chronic kidney disease in patients with type 2 diabetes (FLOW). New England Journal of Medicine, 391(2), 109-121. PubMed
5. Milman, S., Atzmon, G., Huffman, D. M., et al. (2014). Low insulin-like growth factor-1 level predicts survival in humans with exceptional longevity. Aging Cell, 13(4), 769-771. PubMed
6. Sehl, M. E., et al. (2025). Semaglutide slows epigenetic aging in people with HIV-associated lipohypertrophy: Evidence from a randomized controlled trial. medRxiv. DOI: 10.1101/2025.07.09.25331038
7. Lake, J. E., et al. (2024). Semaglutide reduces metabolic-associated steatotic liver disease in people with HIV (SLIM LIVER). Annals of Internal Medicine. PMC
8. Hall, S., Isaacs, D., & Clements, J. N. (2018). Pharmacokinetics and clinical implications of semaglutide. Clinical Pharmacokinetics, 57(12), 1529-1538. PubMed
9. Turnock, L. A., Germain, J., Hirst, M., & Hearne, E. (2026). Off-label semaglutide and GLP-1 weight-loss medicine use for anti-ageing: A netnography of older bodybuilders' experiences and understandings. Performance Enhancement & Health, 14, 100355. DOI: 10.1016/j.peh.2025.100355
10. Strong, R., Miller, R. A., Astle, C. M., et al. (2016). Longer lifespan in male mice treated with a weakly estrogenic agonist, an antioxidant, an alpha-glucosidase inhibitor or a Nrf2-inducer. Aging Cell, 15(5), 872-884. PubMed
11. Garratt, M., Bower, B., Garcia, G. G., & Miller, R. A. (2018). Male lifespan extension with 17-alpha-estradiol is linked to a sex-specific metabolomic response. Aging Cell, 17(4), e12786. PMC
12. Miller, R. A., Harrison, D. E., Astle, C. M., et al. (2014). Rapamycin-mediated lifespan increase in mice is dose and sex dependent. Aging Cell, 13(3), 468-477. PMC
13. Suchacki, K. J., et al. (2023). The effects of caloric restriction on adipose tissue and metabolic health are sex- and age-dependent. eLife. PMC
14. Börchers, S., & Skibicka, K. P. (2024). GLP-1 and its analogs: Does sex matter? Endocrinology, 166(2), bqae165. PMC
15. Mirochnik, Y., Veliceasa, D., Williams, L., et al. (2012). Androgen receptor drives cellular senescence. PLoS ONE, 7(1), e31052. PLOS
16. Roediger, J., Hessenkemper, W., Bartsch, S., et al. (2014). Supraphysiological androgen levels induce cellular senescence in human prostate cancer cells through the Src-Akt pathway. Molecular Cancer. PubMed
17. Zhang, Y., Xie, Y., Berglund, E. D., et al. (2012). The starvation hormone, fibroblast growth factor-21, extends lifespan in mice. eLife, 1, e00065. PubMed
18. Cyphert, H. A., Alonge, K. M., Ippagunta, S. M., & Hillgartner, F. B. (2014). Glucagon stimulates hepatic FGF21 secretion through a PKA and EPAC-dependent posttranscriptional mechanism. PLOS ONE, 9(4), e94996. PubMed
19. Wilding, J. P. H., et al. (2025). Effects of semaglutide on lipid metabolism and CRP in obesity: A review of STEP trials. Drugs and Drug Candidates, 5(2), 35. DOI: 10.3390/ddc5020035
20. U.S. Food and Drug Administration. (2005). Guidance for industry: Estimating the maximum safe starting dose in initial clinical trials for therapeutics in adult healthy volunteers. FDA