← The Signal Report Work with me

Report 248 · Supplement Reality Check

What the NOVOS Core vascular trial found

A randomized trial published on 4 November 2026 in European Heart Journal Open tested NOVOS Core, a 12-ingredient "longevity" powder, against a placebo powder for six months in healthy adults over 40. It reported better blood-vessel function, less arterial stiffness and lower systolic blood pressure. The trial is real and the design has genuine strengths. It is also small, it analyzed 43 of the 61 people it randomized, and every outcome it measured is a stand-in for the thing people actually care about.

Disclosure up front: I run a functional-mushroom supplement company, so "longevity supplements" is a market I have a commercial stake in. I have no connection to this trial, its authors, NOVOS or the product. The full disclosure is at the bottom.

The trial

The study is called STAMINA. Christian Heiss and colleagues at the University of Surrey and Royal Surrey NHS Foundation Trust randomized 61 apparently healthy, non-smoking adults aged 40 and over: 33 to NOVOS Core and 28 to a control powder. Recruitment ran from July 2023 to January 2025. Each person mixed one 15 g scoop into water or juice once a day for six months.

Per the paper's ingredient table, each scoop holds 100 mg vitamin C, 304 mg magnesium (as magnesium malate), 231 mg calcium (as calcium alpha-ketoglutarate), 2,000 mg glycine, 1,000 mg glucosamine sulphate, 300 mg Rhodiola rosea extract, 150 mg L-theanine, 100 mg hyaluronic acid, 100 mg fisetin, 100 mg ginger extract, 50 mg pterostilbene and 20 mg lithium aspartate (1 mg of lithium). The control powder had the same sweeteners and flavorings (erythritol, malic acid, stevia and so on) with none of the actives.

The pre-specified primary outcome was the change in flow-mediated dilation (FMD) at six months. FMD is an ultrasound test: you image the brachial artery in the upper arm, squeeze off blood flow with a cuff for five minutes, release it, and measure how much the artery widens. It's reported as a percentage of the artery's resting diameter, and it's a standard research measure of how well the artery lining (the endothelium) is working. The registry entry, NCT06145087, lists the same primary outcome, which is a point in the trial's favor.

What it found

Adjusted for baseline, the supplement group's FMD ended 2.9 points higher than the control group's (95% confidence interval 2.1 to 3.8). Pulse wave velocity, a measure of arterial stiffness, ended 1.18 m/s lower (−2.00 to −0.36). Peripheral systolic blood pressure ended 6.1 mmHg lower (−10.9 to −1.3). Diastolic pressure did not differ. Cholesterol and triglycerides did not change. FMD also rose an hour after the very first dose.

The trial also reported a small drop in estimated 10-year cardiovascular risk (SCORE2, −0.45 percentage points). The authors are clear about where that comes from: SCORE2 uses age, sex, smoking, systolic pressure and non-HDL cholesterol, and only the blood pressure moved. It isn't an independent finding. It is the blood-pressure result run through a risk calculator.

Six things to know before you believe the headline

1. "47%" and "2.9%" are different units. Coverage reported that FMD "increased by 47% from baseline" in the supplement group, with "an adjusted improvement versus placebo of 2.9%." Both numbers come from the paper, but FMD is already a percentage. The supplement group started at an FMD of 5.5% and rose by 2.6 points; 2.6 divided by 5.5 is about 47% (my arithmetic). The 2.9 is the difference between groups, in percentage points of FMD. Neither is wrong. They just aren't the same kind of number, and the 47% sounds a lot bigger than "about three points on a test where this trial's participants started at around 5%."

2. 61 were randomized; 43 were analyzed. Nine people in the supplement group and five in the control group didn't come to the six-month visit "for unknown reasons." Three more control participants dropped out (two with mild reflux, one with an unrelated illness after travel), and one was excluded after being diagnosed with atrial fibrillation. The primary analysis used only the 43 people with both FMD measurements (24 versus 19), with no imputation for the missing. The authors note that dropout rates were similar between groups. Still, when nearly 30% of a small trial is missing for mostly unknown reasons, the people who stayed may not be a random slice of the people who started.

3. The randomization came out lopsided on age. In the full randomized group, the control participants were older (60 versus 53 on average, p = 0.017). Among the 43 analyzed the gap shrank to 59 versus 54 and was no longer statistically significant. Age matters a lot for every outcome here, and the authors list this as a limitation. The analysis adjusted each outcome for its own baseline value, which helps, but a trial this size can't fully adjust its way out of a group difference like that.

4. The effect was about three times what the trial was planned to find. The sample-size calculation assumed a 1-point FMD change. The result was 2.9. That's good news for the product if it holds up. My own view, not the paper's: a small first trial that finds an effect much bigger than planned is exactly the kind of result that needs an independent replication before anyone treats the size as settled, because early estimates from small trials often come down when larger ones follow.

5. The stiffness number came from a cuff device, not the reference test. One trade article described the stiffness measure as carotid-femoral pulse wave velocity, which is the reference method (sensors over the neck and groin arteries). The paper's methods say something different: PWV was measured with an Arteriograph, a cuff-based oscillometric device on the upper arm that derives PWV and central blood pressure from the arm's pressure waveform using the manufacturer's algorithm. That's a recognized research tool, but it's not the same measurement, and it's worth knowing which one produced the number. The device-derived central systolic pressure, by the way, did not differ significantly between groups (p = 0.134).

6. It tested a blend, so it can't name a cause. The authors say plainly that the trial cannot identify the active ingredient or ingredients, cannot tell whether they act together, and cannot establish dose-response. Several components already have their own blood-pressure or vascular literature (the paper cites, for example, an umbrella meta-analysis on magnesium and blood pressure). This trial answers one question, "does this exact powder beat this control powder on these measures," and only that one.

The surrogate problem

None of the outcomes in this trial is a heart attack, a stroke or a death. FMD, pulse wave velocity and blood pressure are intermediate markers, and the authors say the trial "did not assess clinical endpoints." The case for FMD as a meaningful marker is real: a 2015 meta-analysis of 35 FMD studies (17,280 participants) in the Journal of the American Heart Association found each 1-point-higher FMD was associated with about 12% lower risk of cardiovascular events. But that is an association between people who have better FMD and people who have fewer events. The same meta-analysis says more research is needed on whether improving endothelial function as a target actually improves outcomes. A powder that moves the marker has not yet been shown to move the outcome.

Two more measurement details belong in the fine print. The six-month test happened roughly 24 hours after each person's last dose, without a formal washout, so the authors say leftover short-term effects "cannot be excluded." And adherence was tracked by self-reported diary only, which the authors note tends to overestimate compliance.

Who paid, and what the company said

The paper calls this an investigator-initiated study. NOVOS Labs supplied the product and the matching control and made an unrestricted research contribution to the University of Surrey. The funding statement says the company had no role in the design, data collection, primary analysis, interpretation or the decision to publish, and that the academic authors kept control of the data. One co-author, Diogo Barardo, is a NOVOS employee; the author-contributions section says he "independently reproduced the statistical analyses." The lead investigator also discloses research funding from NOVOS. None of that makes the result wrong. It does mean this is a company-supported first trial of a company's product, and an independent replication would carry more weight.

In the trade coverage, NOVOS's director of R&D also said the product "extended lifespan in aged mice by more than 18%." That claim isn't part of this trial. The paper's own reference for the mouse work is a preprint that the reference list labels "not peer reviewed," and the authors write that those preclinical findings "cannot be directly extrapolated" to humans. To the company's credit, its CEO also called STAMINA "early human evidence" and said direct measures of biological aging "come next." That's the right frame.

One small registry note: the ClinicalTrials.gov record lists daily walking distance (measured by smartphone) as a secondary outcome. I didn't find it reported in the main paper. It may be in the supplementary material, which I did not read.

How I'd read it

This is a properly randomized, placebo-controlled, pre-registered trial whose primary outcome moved in the right direction with a tight confidence interval, alongside a blood-pressure drop that would matter if it held up over years. That's more than most longevity products can show, and it's worth saying. It's also one single-center trial, 43 analyzed people, a sponsor-supplied product, surrogate markers only, a stiffness number from a cuff algorithm, and a 12-ingredient blend that can't tell you which part did anything. The honest summary: this powder improved blood-vessel test scores and systolic pressure over six months in one small, company-supported study. It has not been shown to slow aging, and it has not been shown to prevent a single cardiovascular event.

If you're judging any "clinically studied" longevity claim, the questions are the same ones I use in what "clinically proven" really means: what was measured, in how many people, against what, for how long, and who paid.

What would change my read: an independent replication, a larger trial using carotid-femoral PWV and ambulatory blood pressure, or longer follow-up. If one lands, I'll update this post in place.

Not medical advice. This is educational analysis, not a recommendation — a study is not a prescription. Talk to a qualified clinician before acting on anything you read here. Full disclaimer →

Sources

  1. Charles Piercy, Jenny Harris, Kenza Sackho, Paola Campagnolo, Diogo Barardo, Ben Creagh-Brown and Christian Heiss, "Acute and 6-month vascular effects of a multi-component nutritional supplement: a randomized controlled trial," European Heart Journal Open 6(6), oeag157 (published 4 November 2026), DOI 10.1093/ehjopen/oeag157; full text on PMC13730552. (Primary source, open access, CC BY 4.0. Full text read via Europe PMC; the supplementary material was not read. Source of the design, ingredient table, outcomes, Arteriograph method, sample-size calculation, participant flow, baseline tables, results, limitations, funding, conflict-of-interest and author-contribution statements, the mouse-preprint reference and the quoted phrases.)
  2. ClinicalTrials.gov, NCT06145087, "Effects of a Supplement to Target Ageing Mechanisms on Vascular Function (STAMINA)" (first posted 22 November 2023, last updated 22 January 2026). (Registry record read via the ClinicalTrials.gov API. Source of the registered primary and secondary outcomes, including daily walking distance, and the enrollment of 61. The registry lists the University of Surrey as lead sponsor.)
  3. Stephen Daniells, "RCT supports NOVOS Core's benefits for cardiovascular health in healthy adults," NutraIngredients, 5 November 2026. (Read. Source of the "47%" and "2.9%" phrasing, the carotid-femoral description, and the quoted company statements about mice, "early human evidence" and what comes next.)
  4. Yasushi Matsuzawa, Taek-Geun Kwon, Ryan J. Lennon, Lilach O. Lerman and Amir Lerman, "Prognostic Value of Flow-Mediated Vasodilation in Brachial Artery and Fingertip Artery for Cardiovascular Events: A Systematic Review and Meta-Analysis," Journal of the American Heart Association 4:e002270 (2015). (Abstract read via Europe PMC. Source of the 35 studies, 17,280 participants, the relative risk of 0.88 per 1% FMD, and the call for research on endothelial-function-guided strategies. Cited by the trial as its reference 36.)

A Consensus literature search was used to check how FMD change over time relates to outcomes; the papers it surfaced were in people with existing coronary disease, read in abstract only, and are not cited above. "About 47%" (2.6 ÷ 5.5) is my arithmetic from the paper's Table 3. The point that small first trials often overestimate effects is my general view, not a claim the paper makes.

Disclosure, plainly: I founded and run Shroombiosis (a company I run), which formulates and sells functional-mushroom supplements, so I have a commercial stake in how people think about supplements and healthy aging. I have no relationship with this trial, its authors, NOVOS Labs or the product. Nothing here is sponsored and no link earns a commission; here's the full policy. A recommendation with no stake at all: for performance nutrition, Die Tryin Co. is a fellow combat-veteran-owned brand I'm glad to point people to. I don't own it and earn nothing from the link.

Onur Oncer
Onur Oncer

U.S. Army combat veteran (Counter-IED / Electronic Warfare), peer-reviewed researcher in microwave spectroscopy, and founder & CEO of Shroombiosis. Consults on laboratory operations, AI, and supplement formulation.

← All reports