Safety & evidence

How we grade
the evidence

Our compound guides grade the evidence for each use they cover. Here is what each grade means, how it is assigned, and when it changes.

The 6AM mark rendered in glass
Short answerSafety & evidence · 7 min read · Updated 18 Nov 2026

Research guides on 6AM grade evidence by the strongest study design available for each use: Preclinical (cell or animal studies only), Very limited / early human (small uncontrolled human reports), Controlled human trials, or Approved (that use only). Approval is tracked separately. A grade shows how far research has gone, not whether a compound works.

Key takeaways

  • The grade is set by study design: cell, animal, small human report, controlled human trial, and, on a separate axis, regulatory approval.
  • Grades attach to one compound, one use and one route. Topical GHK-Cu and injected GHK-Cu are graded separately.
  • A controlled trial earns its grade even when its result is negative or mixed. The grade says how well a question was tested, not what the answer was.
  • Animal results are a reason to study something in people, not evidence that it works in people: a systematic review found treatments that helped in animal models and then showed no benefit, or worse outcomes, in patients.
  • Grades change when new studies publish, when we correct a classification, and when we re-read a page against current evidence.

01The scale

The scale runs Cell → Animal → Human pilot → Controlled trials, with regulatory approval on its own line. Some guides show it as a research bar; others give the internal tier from the table below (T4 to T1). Each area of research gets the highest step it has reached; a guide’s headline shows the area most of its research addresses.

StepWhat countsGrade shownInternal tierExample from our register
CellCells or tissue studied outside a living body (in vitro, explants), including cells taken from animalsPreclinicalT4BPC-157 on rat tendon explants and cultured tendon cells (Chang et al., 2011)
AnimalStudies in live animalsPreclinicalT3BPC-157 after Achilles tendon transection in rats (Staresinic et al., 2003); KPV in mouse colitis models (Dalmasso et al., 2008)
Human pilotSmall human studies without a proper comparison group, case series, case reports, early pharmacology studiesVery limited / early humanT2BPC-157: three small uncontrolled pilot studies (McGuire et al., 2025); ipamorelin: a pharmacology study in 40 healthy men (Gobburu et al., 1999)
Controlled trialsRandomised human trials with a placebo or comparison groupControlled human trialsT1CJC-1295, in the long-acting form tested in 2006: two placebo-controlled studies that measured GH and IGF-1 (Teichman et al., 2006); thymosin β4 eye drops, phase 2, n=72 (Sosne and Ousler, 2015)
Regulatory approvalA regulator approved a specific product for a specific useApproved (that use only)factSemaglutide and tirzepatide, prescription medicines approved after large phase 3 trials
Not gradedAnecdotes, forums, influencer claims, vendor pagesNever usedT0—

How sources are ranked before they reach this scale is set out in source standards. To compare compounds side by side, use the evidence grade explorer in tools.

02Six rules for assigning a grade

1
One use, one route

Each area studied gets the grade of its strongest source. GHK-Cu has small human studies for topical use and none for injection, so the two are graded apart. Thymosin β4 eye-drop trials used the full 43-amino-acid protein; they do not grade injected TB-500, which usually means a 7-amino-acid fragment. A guide’s headline grade reflects the area most of its research addresses; the evidence map grades every area separately.

2
One compound at a time

Evidence attaches to the compound studied. A blend or a pair gets no grade from its parts. We found, for example, no published or registered human study of CJC-1295 and ipamorelin together (PubMed and ClinicalTrials.gov, November 2026).

3
Design, not direction

A controlled trial earns the controlled-trial grade whatever it found. Thymosin β4 eye drops have controlled trials, and the phase 2 dry-eye trial missed both of its primary endpoints (Sosne and Ousler, 2015). The grade stays; the guide says the results are mixed.

4
What was measured counts

The 2006 CJC-1295 trials measured hormone levels, not muscle, fat or performance. The grade carries that qualifier: controlled trials, biomarkers only.

5
Live animal means live animal

A study on cells or tissue taken from an animal is a cell study. The animal grade needs a living animal.

6
Registered is not reported

Several BPC-157 studies are registered on ClinicalTrials.gov; as of 18 November 2026 none has posted results. Until results are published, they do not change the grade.

03Why animal results don't transfer automatically

Animal studies are how almost every medicine starts. They are also a weak predictor of what happens in people.

  • Animal benefit can become human harm. A systematic review (Perel et al., BMJ, 2007) found that the stroke drug tirilazad improved outcomes in animal models, yet was associated with worse outcomes in patients. Corticosteroids helped in animal models of head injury and showed no benefit in clinical trials.
  • It cuts both ways. Semaglutide causes thyroid C-cell tumours in rodents; its US label says whether it does so in humans is unknown. Animal findings, good or bad, are questions for human research, not answers.

The reasons are ordinary: different species, disease models that do not mimic the human condition, and small studies prone to bias.

04How grades change

1
A new study publishes

A controlled trial published in full moves a compound up the bar for that use. A press release, a conference talk or a conference abstract does not.

2
We correct a classification

Grades can go down. Chang et al. (2011) was first filed in our register as an animal study of BPC-157; on re-reading, it was tendon explants and cultured cells, so it was re-tiered to cell evidence.

3
A regulator decides

Approvals update the separate approval line, dated to the decision. Advisory votes and enforcement policies do not change a grade or an approval status.

4
A re-read

Published pages are re-read against current evidence, and legal-status pages are rechecked on a fixed schedule and after every relevant regulatory event, as set out in our editorial policy and how we check every page.

05What a grade can't tell you

  • Whether something works. Controlled trials can be negative. Preclinical findings can be striking and still fail in people.
  • How big an effect is, or whether it matters to a person. Read the guide for what was measured.
  • Whether something is safe. See are peptides safe and peptide side effects.
  • Study quality within a step. Formal systems such as GRADE (Guyatt et al., BMJ, 2008) also rate risk of bias, consistency and precision. Our scale is simpler and based on design; we do not present it as a GRADE rating.

This content is educational. It is not medical advice and does not describe use in humans. 6AM Wellness products are sold for research use only and are not for human consumption. If you have a health condition, consult a licensed clinician who knows you.

RefSources

  1. Perel P, Roberts I, Sena E, et al. Comparison of treatment effects between animal experiments and clinical trials: systematic review. BMJ, 2007;334(7586):197. PMID 17175568. pubmed.ncbi.nlm.nih.gov/17175568/
  2. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations. BMJ, 2008;336(7650):924–926. PMID 18436948. pubmed.ncbi.nlm.nih.gov/18436948/
  3. Chang CH, Tsai WC, Lin MS, et al. The promoting effect of pentadecapeptide BPC 157 on tendon healing involves tendon outgrowth, cell survival, and cell migration. J Appl Physiol, 2011;110(3):774–780. PMID 21030672. pubmed.ncbi.nlm.nih.gov/21030672/
  4. Staresinic M, Sebecic B, Patrlj L, et al. Gastric pentadecapeptide BPC 157 accelerates healing of transected rat Achilles tendon and in vitro stimulates tendocytes growth. J Orthop Res, 2003;21(6):976–983. PMID 14554208. pubmed.ncbi.nlm.nih.gov/14554208/
  5. McGuire FP, Martinez R, Lenz A, et al. Regeneration or risk? A narrative review of BPC-157 for musculoskeletal healing. Curr Rev Musculoskelet Med, 2025;18(12):611–619. PMID 40789979. pubmed.ncbi.nlm.nih.gov/40789979/
  6. Teichman SL, Neale A, Lawrence B, et al. Prolonged stimulation of growth hormone (GH) and insulin-like growth factor I secretion by CJC-1295, a long-acting analog of GH-releasing hormone, in healthy adults. J Clin Endocrinol Metab, 2006;91(3):799–805. PMID 16352683. pubmed.ncbi.nlm.nih.gov/16352683/
  7. Sosne G, Ousler GW. Thymosin beta 4 ophthalmic solution for dry eye: a randomized, placebo-controlled, phase II clinical trial conducted using the controlled adverse environment (CAE™) model. Clin Ophthalmol, 2015;9:877–884. PMID 26056426. pubmed.ncbi.nlm.nih.gov/26056426/
  8. Novo Nordisk. WEGOVY (semaglutide) prescribing information. FDA-approved label via DailyMed, revised June 2026. dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=ee06186f-2a…
  9. ClinicalTrials.gov. NCT07437547: BPC 157 for acute hamstring muscle strain repair (phase 2). Trial registry record, 2026. clinicaltrials.gov/study/NCT07437547
  10. Dalmasso G, Charrier-Hisamuddin L, Nguyen HT, et al. PepT1-mediated tripeptide KPV uptake reduces intestinal inflammation. Gastroenterology, 2008;134(1):166–178. PMID 18061177. pubmed.ncbi.nlm.nih.gov/18061177/
  11. Eli Lilly and Company. ZEPBOUND (tirzepatide) prescribing information. FDA-approved label via DailyMed, revised August 2026. dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=487cd7e7-43…

Plain-English summary of public sources, not legal or medical advice. Last checked 18 Nov 2026.

Questions

Why is BPC-157 graded Preclinical when people have used it in studies?

Its headline grade reflects where most of its research sits: tendon, gut and muscle studies in animals. Its human data, six small studies, are graded separately, as Very limited / early human, in the guide’s evidence map: the only randomised, placebo-controlled data are in conference abstracts from 2002 to 2005, and no peer-reviewed randomised trial has been published.

Does a Controlled human trials grade mean a compound works?

No. It means the question was tested in a controlled trial. The result can be positive, negative or mixed, and each guide says which.

Why don't you count forum reports or influencer results?

They cannot separate an effect from placebo, chance, other substances or a mislabeled vial. They are never used as evidence on this site.

What does Approved (that use only) mean?

A regulator approved a specific product for a specific use and population. The approval does not extend to other uses, other routes or unapproved copies of the molecule.

How often are grades reviewed?

We re-read published pages against current evidence and correct what has changed. Every published guide shows the date it was last updated.

Evidence before hype.

One email a week: new research explained and regulatory changes. No prices, no hype.

Get the 6:00 brief

Free. Unsubscribe any time.

Written by 6AM Research. Drafted from primary sources; every claim links to its source, and the page is dated. Editorial policy · How we check every page