HHS temporarily opens a peptide compounding channel
Changes what gets built, funded, approved or priced: 2 yes · 0 not yet · 0 no
◆◆◆◇◇ Every study sits on a five-step ladder, from lab to a person's actual outcome. How to read it
Not a study
HHS temporarily allowed compounding pharmacies to distribute unapproved peptide injections
| Who | US Department of Health and Human Services |
|---|---|
| What | Policy temporarily allowing compounded manufacture and wide distribution; duration unreported |
| How much | Volume not stated |
| Evidence | Not a study: a policy announcement, so stage not assignable |
| Product, price | None announced; the products are unapproved peptide injections |
HHS said compounding pharmacies may temporarily manufacture and widely distribute unapproved peptide injection products, the kind sold by online wellness influencers. The FDA had previously advised against that.
Regulatory action here had been narrowing supply. This widens it: clinics and telehealth prescribers who relied on grey-market import now have a legal domestic route, and the policy creates no new mechanism to collect safety or efficacy data.
A commentary the same week argued sermorelin marketing had outrun its clinical evidence. 1,2
HARMSReported: the FDA warned consumers off two products for undisclosed ingredients the same week
WHAT WOULD MOVE THISA randomised trial of a named peptide against a clinical endpoint, or an adverse-event registry for compounded supply.
1 NYT Health · Not a study · 2 Not a study
◆◆◆◇◇ People, watched
Biograph and Prenuvo published prevalence findings from their own paying customers
| Who | Biograph with Caristo, Prenuvo, Fitnescity with Superpower |
|---|---|
| What | Two company-reported studies and a distribution launch |
| How much | 14,031 adults; more than 1,000 imaging locations |
| Evidence | People, watched: cross-sectional, self-referred, nobody assigned to anything (People, watched) for both studies; the distribution launch is stage not assignable |
| Product, price | DEXA and calcium scans resold through apps; no price disclosed |
◆ Funding is not validationBiograph's finding was company-conducted and company-reported. Prenuvo reported its own with academic collaborators. Both samples are customers who paid for the scan.
Biograph, using Caristo's CaRi-Heart analysis of coronary CT angiograms, reported elevated perivascular inflammation in 42.9% of asymptomatic, conventionally low-risk adults (16298, People, watched). Of the adults with elevated inflammation, 47% had no detectable plaque.
Prenuvo reported across 14,031 adults that higher sleep efficiency tracked with lower visceral fat and more parietal grey matter, and sleep beyond eight hours with smaller hippocampal volume (16234, People, watched). Both samples are self-referred people who paid for imaging, the group least likely to represent anyone else.
Fitnescity opened more than 1,000 US imaging locations to third-party platforms through one API (stage not assignable). Distribution is solved ahead of validity. 3–8
HARMSNot measured: neither Biograph nor Prenuvo reports overdiagnosis, downstream procedures or cost. The 30-year calcium cohort behind early scoring was assembled from referred people (15563, People, watched)
WHAT WOULD MOVE THISA randomised comparison of acting versus not acting on an early calcium score, with procedures and costs reported.
3 Longevity.Technology · company-reported observational analysis of AI-derived coronary inflammation on routine CT in low-risk adults · 4 Longevity.Technology · cross-sectional screening cohort, asymptomatic adults, AI-enabled CCTA with cardiometabolic profiling · 5 Longevity.Technology · observational analysis of whole-body MRI and self-reported sleep data, company-reported · n = 14,031 · 6 Longevity.Technology · cross-sectional whole-body MRI observational study · n = 14,031 · 7 Longevity.Technology · Not a study · 8 JAMA Cardiology · retrospective cohort, long-term follow-up
Can a clock rank mortality risk across a cohort and still tell one person nothing?
| Reading A | The literature is large. A meta-analysis pooled 48 methylation clocks against mortality across 380,705 participants in 53 studies (15301, People, watched). Its pooled effect estimates are not in the available reporting, so the association's size is unstated. |
|---|---|
| Reading B | Association is not utility. A meta-analysis of four such clocks against cardiovascular outcomes found clinical-utility evidence limited (15464, People, watched). In 178 adults with matched samples, only 50 muscle proteins and 17 ovarian proteins tracked their own tissue (15773, People, watched); muscle RNA clocks could not separate sarcopenia status (15424, Lab). |
| What would settle it | A prospective trial where clinicians see a clock reading and those patients do better than the control arm. 9–13 |
Regulation
- The FDA cleared NEXT-Cue, combining brain MRI with cognitive scores to assess dementia risk after mild cognitive impairment (stage not assignable). Clearance is a marketing authorisation, not evidence that using it changes an outcome. 14
Money
- NIH committed $29M over five years to the Dog Aging Project, covering nearly 55,000 dogs (stage not assignable). 15
- Seragy raised $13.4M to expand protein-aggregation measurement for Alzheimer's, Parkinson's and ALS (stage not assignable). 16
Pipeline
- Artiva Biotherapeutics is winding down after two patient deaths in its cell-therapy trial for autoimmune disease (stage not assignable). 17
- VG-3927, an oral TREM2 agonist, reported CNS penetration, target engagement and tolerability in healthy volunteers (People, tested on a stand-in). No efficacy outcome. Company-reported. 18
- Beacon Biosignals acquired Snap Diagnostics, a home sleep testing company in all 50 US states (stage not assignable). 19
- Samsung Health and Lark Health will deliver AI-led prevention to adults 65+ through the Medicare Diabetes Prevention Program (stage not assignable). 20
- Stenting or surgery for severely narrowed carotid arteries gave no cognitive benefit over medication alone at one year (People, tested on a stand-in). 21
India Watch
- Cent, founded by Practo's founder, raised $15M and acquired Medsee for AI-led preventive scanning, with no clinical data (stage not assignable). Overdiagnosis, procedures and cost go unquantified. 22
- ICMR funded a four-year nationwide study of the genetic roots of rising autoimmune disease; AIIMS is asking lupus patients to enrol, and no results exist yet (stage not assignable). 23
- Tooth loss tracked worse daily function in adults 60 and over across England, the US and India (People, watched); LASI's sample was largest, 31,816. Whether restoring dentition restores function is untested here. 24
- A geriatrics framework names sarcopenia and frailty as GLP-1 risks, landing where generic semaglutide is widely available and FSSAI challenged a muscle-preservation claim (stage not assignable). 25,26
| Trend | Activity / evidence | Where it stands |
|---|---|---|
| Age clocks, commercialised | ↑ rising Measurement | Large mortality-association literature (People, watched); no clock in this week's evidence was tested as a decision tool. |
| Biomarker standardisation | ↑ rising Review | One consensus framework, a validation argument and a methods guide; one lab published its own pipeline's confounds. |
| Senolytics in the clinic | → flat◆◆◇◇◇ Animals | Widening mouse targets; a review restates that no drug alters calcific aortic valve disease. |
| Partial epigenetic reprogramming | ↑ rising◆◆◇◇◇ Animals | Mouse lifespan gains only; no delivery route, dose or human data. |
| GLP-1 into healthspan medicine | ↑ rising◆◆◆◇◇ People, watched | Indications sprawling, all observational; a framework now names sarcopenia and frailty. |
| GLP-1 muscle-preservation adjuncts | ↑ rising◆◆◆◆◇ People, tested on a stand-in | Androgens, SARMs and myostatin inhibitors pooled on mass and strength, not disability. |
| India preventive executive health | ↑ rising Not a study | Scan distribution consolidating; no clinical or research data accompanies the Cent raise or Medsee acquisition. |
| Longevity clinics globalising | ↑ rising◆◆◆◇◇ People, watched | Imaging sold as an API; its calcium-score evidence is a referred cohort. |
| Dogs as a regulatory wedge | ↑ rising◆◆◇◇◇ Animals | NIH's $29M went to the cohort, not a drug; no pathway moved. |
| Ovarian aging as a target | ↑ rising◆◇◇◇◇ Lab | A named cohesion protein and a postmenopausal atlas; no delivery route. |
| Gut microbiome geroscience | ↑ rising◆◆◇◇◇ Animals | Yanomami donor microbiota protected germ-free mice against diet-induced metabolic dysfunction (Animals); E. ramulus is the candidate strain. |
| NAD precursor supplementation | → flat◆◆◇◇◇ Animals | Depletion makes mice frail; a second registration adds no outcome. |
| South Asian metabolic phenotype | → flat◆◆◆◇◇ People, watched | BMI misses sarcopenic obesity here; adiposity definitions still differ across studies. |
