On 9 September 2026, Crossref recorded Vantini and colleagues in iScience 29(10):117471, "Effects of Pep19 on adipose tissue distribution and metabolic parameters in adults with obesity: A randomized, double-blind, placebo-controlled trial" (DOI 10.1016/j.isci.2026.117471). Print issue October 2026. Cell and ScienceDirect HTML returned HTTP 403 here. Elsevier metadata confirmed the same DOI, PII S2589-0042(26)02850-6, and CC BY 4.0, with no abstract.
Scienmag recapped the paper on 12 September 2026. The numeric cells below for that trial come from that recap. They are not fetched manuscript tables.
The recap's primary claim is a 90-day android-to-gynoid fat-ratio shift on 10 mg oral Pep19 versus placebo, without meaningful weight loss. That is a fat-distribution signal in 60 adults (56 completers). It is not FDA approval for obesity. It is not a GLP-1, GIP, or amylin result.
Pep19 is DIIADDEPLT, a short cyclophilin-A-derived intracellular peptide studied as a peripheral CB1 inverse agonist. Heimann 2025 (PMC12126752, DOI 10.1002/dmrr.70056) is the earlier human trial. Reckziegel 2017 is the discovery paper.
This is the 10:07 Research Roundup for 14 September 2026. It is not the 8:07 Industry News slot. Yesterday's GZC8072 first-dose note is an oral peptide GLP-1 story. Pep19 is not an incretin. The vosoritide CANOPY-HCH-3 note stays as written.
What happened
Scienmag describes a randomized, double-blind, placebo-controlled trial at FMABC University Center in Santo André, São Paulo. Sixty adults with BMI 30 to 39.9 kg/m² were assigned to placebo, 5 mg Pep19, or 10 mg Pep19 daily at bedtime for 90 days. Fifty-six completed.
The prespecified primary endpoint was change in the android-to-gynoid fat ratio by DXA. After 90 days, the 10 mg arm showed a reduction versus placebo. Scienmag reports an observed difference of 3.19 percentage points in favor of 10 mg, and a mixed-effects adjusted treatment effect of −3.58 percentage points (95% CI −6.96 to −0.19; p=0.037). A log-scale analysis showed a 3.4% relative reduction. The absolute-ratio contrast of −0.037 units was p=0.058. The 5 mg arm was not significant on the primary.
Body weight, BMI, total fat mass, and percent body fat were statistically unchanged. Lean mass was preserved or slightly increased. The 10 mg arm had significant reductions in total and abdominal skinfold thickness. The 5 mg arm had a significant reduction in abdominal circumference.
On metabolic markers, Scienmag says 10 mg produced a significant HbA1c reduction versus placebo with fasting glucose stable, and that HOMA-IR fell in both treated arms (already visible at day 60 on 10 mg). This article does not invent HbA1c or HOMA-IR deltas the recap did not print.
An unsupervised k-means analysis split the cohort into 45 low-change participants and 11 high responders with coordinated drops in total fat mass, skinfolds, HOMA-IR, and TNF-alpha. All 11 were Pep19-treated (six on 5 mg, five on 10 mg). No placebo participant entered that cluster. A cluster is a post-hoc phenotype. It is not a registration primary.
Safety, as recapped: renal, hepatic, hematological, and endocrine labs stayed within normal ranges, with no treatment-related toxicology signals. Systolic blood pressure trended down on 10 mg and did not reach significance. Lipids, CRP, and IL-6 did not change significantly.
Why it matters
Most obesity coverage on this site is incretin or amylin work. Pep19 is a different class. Reckziegel 2017 identified DIIADDEPLT as a CB1 inverse agonist from an intracellular-peptide screen. Oral dosing in diet-induced obese rats improved metabolic parameters. Those assays did not produce a CNS tetrad, brain c-Fos induction, or anxiety-depression signals. Silvério 2022 reported oral Pep19 at 1 mg/kg in high-fat-diet Swiss mice attenuated weight gain, improved insulin-sensitivity relationships, improved liver phenotype, and showed inguinal UCP1 signals.
Heimann 2025 moved the same sequence into a 60-day human capsule study. Vantini 2026, as recapped, is the larger DXA-ratio trial. Together they argue for fat-depot remodeling rather than scale weight. That argument is only as strong as two modest trials plus rodent work.
GRAS or supplement marketing is not a drug approval. Heimann 2025 states Pep19 received U.S. GRAS status under 21 C.F.R. § 170.30 and became commercially available as 5 mg capsules from Nutroslim LLC. NCT06359327 is a dietary-supplement record and, in the sponsor brief, claims GRAS at up to 3.8 grams per day. That 3.8 g line is registry language. It is not an FDA obesity or diabetes indication. It is not the 2 mg, 5 mg, or 10 mg trial assignment.
A catalog vial labeled Pep19, WATACAP, or DIIADDEPLT is not the trial capsule. A COA check can test whether a certificate names a real lab. It cannot turn a listing into an approved drug.
Evidence and design
Vantini 2026 (iScience), as recapped
| Item | Figure | Source |
|---|---|---|
| IDs | iScience 29(10):117471 | Crossref 9 Sept 2026 |
| N / assignment | 60 enrolled, 56 completers; placebo vs 5 mg vs 10 mg bedtime, 90 days | Scienmag |
| Primary | Android-to-gynoid fat ratio (DXA) | Scienmag |
| 10 mg primary | Adjusted −3.58 pp vs placebo (95% CI −6.96 to −0.19; p=0.037) | Scienmag |
| 5 mg primary | Not significant | Scienmag |
| Weight / total fat | Unchanged; lean preserved or slightly increased | Scienmag |
| Glycemia | 10 mg HbA1c down vs placebo; HOMA-IR down in both treated arms; fasting glucose stable | Scienmag |
| Cluster | 11 high responders, all Pep19-treated | Scienmag |
This article does not invent arm-level kilogram tables or a target-engagement assay. The iScience HTML was not fetched.
Heimann 2025 (Diabetes Metab Res Rev)
The full PMC text was fetched. This is a 60-day, placebo-controlled, triple-blind capsule trial at Precision Clinical Research Centre, Sunrise, Florida (May-August 2024). Advarra IRB Pro00075623. Registered as NCT06359327. Convenience sample. Twenty-four adults (12 male, 12 female), ages 46 to 59, weight 91-106 kg, BMI 30-35 kg/m², assigned 1:1:1 to placebo, 2 mg, or 5 mg oral capsules once daily at bedtime. One 2 mg participant dropped out for administrative reasons, not an adverse-effect report.
The paper's primary was quality of life (SF-12) plus sleep (PSQI). Key secondaries were visceral fat by DXA, weight, and waist, hip, and chest circumferences.
| Endpoint | Published figure | Source |
|---|---|---|
| 5 mg visceral fat | −17 ± 4.7% vs placebo (p<0.05); no lean-mass loss | Abstract; PMC |
| Sleep (PSQI) | 2 mg 35 ± 10%; 5 mg 25 ± 16% (p<0.05) | Abstract; PMC |
| Poor-sleep share | 43% and 63% reductions on 2 mg and 5 mg vs 13% placebo | PMC |
| Weight and waist | Significant reductions on 5 mg (p<0.05) | Abstract; PMC |
| SBP / fat-mass index | p=0.07 vs placebo | PMC |
| Adverse effects | None reported | Abstract |
The authors state the limits: convenience sampling, small N, short duration, restricted generalization. The 5 mg arm was imbalanced (6 male vs 2 female; heavier and taller). Diet and activity were not collected. A 90-day larger trial is described as underway. That 2025 sentence matches the later Vantini design as recapped. It is not proof the datasets are one protocol.
ClinicalTrials.gov, queried 14 September 2026, still shows NCT06359327 COMPLETED, enrollment 24, hasResults false, last update 6 November 2024. Registry masking is SINGLE (participant). The paper says triple-blind. This article keeps both. The 2025 numbers are not on the registry.
Reckziegel 2017 and Silvério 2022
Reckziegel et al., Scientific Reports 7:14781 (DOI 10.1038/s41598-017-13690-9, PMC5665932): Pep19 from a conformational-antibody screen; CB1 inverse agonism; UCP1 induction blocked by AM251. Oral DIO-rat work improved adiposity, weight, glucose, lipids, and blood pressure. PMC legends report oral 600 µg/kg and 1 mg/kg in named assays. Intraperitoneal 0.1 or 1 mg/kg did not produce a cannabinoid tetrad.
Silvério et al. (DOI 10.3390/ijms23084082): oral Pep19 1 mg/kg for 30 days in HFD Swiss mice attenuated weight gain, reduced endocrine-pancreas mass by about 40%, and showed inguinal UCP1 signals. Rodent milligrams per kilogram are not human capsule milligrams.
Trial doses (not approved obesity doses)
The rows below are trial, registry, and paper fields. Pep19 is not an FDA-approved prescription therapy for obesity, fat redistribution, diabetes, or sleep. That does not make 2 mg, 5 mg, or 10 mg an approved dose. Nothing here is a consumer protocol.
| Setting | Reported assignment | Source |
|---|---|---|
| Vantini 2026, as recapped | Placebo, 5 mg, or 10 mg oral daily at bedtime for 90 days | Scienmag 12 Sept 2026 |
| Heimann 2025 | Placebo, 2 mg, or 5 mg vegetable capsules at bedtime for 60 days | PMID 40450548; PMC12126752 |
| NCT06359327 | Dietary-supplement capsules: placebo, 2 mg, or 5 mg for 60 days | ClinicalTrials.gov, 14 Sept 2026 |
| Silvério 2022 mice | Oral 1 mg/kg once daily for 30 days | PMID 35456900 |
| Reckziegel 2017 | Oral 600 µg/kg or 1 mg/kg in named assays; i.p. 0.1 or 1 mg/kg in tetrad assays | PMC5665932 |
Do not treat a 5 mg supplement capsule, a 10 mg trial arm, or a 1 mg/kg rodent line as a home schedule. Research-vendor vials, if they appear, are not the study product. Use the vendor checklist and the COA verification guide before treating any certificate as proof.
Limitations
The iScience full text was not fetched. Ratio, HbA1c, HOMA-IR, cluster, and safety sentences for 2026 are Scienmag's recap. A later open PDF may add intervals, analyzed N by arm, and a conflict statement.
n=60 (56 completers) at one Brazilian center for 90 days is modest. Diet and activity were self-monitored. Scienmag says baseline HOMA-IR was numerically higher in the 10 mg group. There was no target-engagement biomarker. The 11-person high-responder cluster is post-hoc.
Heimann 2025 is n=24, 60 days, convenience sampled, with a sex and size imbalance on 5 mg. Sleep gains sit on PSQI.
Conflicts are material. Heimann 2025: Andrea S. Heimann, Emer S. Ferro, and Arnon Krongrad are co-founders of Proteimax Biotechnology Israel Ltd. Frank Greenway lists multiple advisor and equity disclosures, including Pep19 Inc. Prachi Singh declares none. Vantini 2026 includes Heimann, Ferro, and Krongrad. This article did not fetch the iScience conflict block. NCT06359327 is sponsored by Proteimax Biotechnology Israel LTD.
Reckziegel's tetrad and c-Fos assays are not a human neuropsychiatric program.
What this does not mean
- It is not FDA approval for obesity, diabetes, or fat redistribution.
- It is not meaningful weight-loss efficacy in the 2026 trial as recapped. Weight, BMI, and total fat were essentially unchanged.
- It is not a GLP-1, GIP, dual/triple incretin, or amylin agonist. Do not file this next to GZC8072.
- GRAS is not a prescription indication. Heimann's Nutroslim sentence and the NCT 3.8 g sponsor line are not drug approval.
- 5 mg and 10 mg are not approved doses. Pep19 is not approved as a prescription therapy for these indications. That does not make a bedtime capsule an approved dose.
- A high-responder cluster is not a companion diagnostic.
- Rodent UCP1 findings are not the human primary.
- A research-catalog peptide is not the trial product.
- This is not medical advice, not a consumer protocol, and not a rewrite of ARA-290.
No PeptidePrices research page for Pep19 existed on 14 September 2026, and none was created.
What to watch
- The open iScience PDF. Confirm arm-level N, the mixed-effects model, HbA1c and HOMA-IR tables, and the conflict statement.
- A posted NCT06359327 results module. The 2025 paper is not on the registry as of this query.
- A trial that is not Proteimax-authored. Heimann, Ferro, and Krongrad sit on both human papers.
- Durability past 90 days, and diet/activity capture.
- Target engagement. Without a CB1 or browning biomarker, the human mechanism remains inferred.
- How later copy treats "obesity pill." Scienmag used that headline. The recapped 2026 primary is a fat-ratio shift without weight loss.
FAQs
Is Pep19 FDA-approved for obesity or diabetes?
No. It is not an FDA-approved prescription therapy for obesity, fat redistribution, diabetes, or sleep. GRAS or supplement marketing is not that approval.
Is Pep19 a GLP-1 or amylin drug?
No. It is a short intracellular peptide, sequence DIIADDEPLT, studied as a peripheral CB1 inverse agonist. It is not a GLP-1, GIP, or amylin agonist.
What did the 2026 iScience trial show?
As recapped by Scienmag, 10 mg for 90 days reduced the DXA android-to-gynoid fat ratio versus placebo (adjusted −3.58 pp; 95% CI −6.96 to −0.19; p=0.037) without meaningful weight loss. The 5 mg arm missed that primary. The paper HTML was not fetched here.
What did the 2025 Heimann trial show?
In 24 adults over 60 days, 5 mg was associated with a 17 ± 4.7% visceral-fat reduction by DXA (p<0.05) without lean-mass loss, plus significant weight and waist reductions (p<0.05). Sleep improved on 2 mg and 5 mg. The authors flag convenience sampling, small N, and short duration.
What dose was used?
Recapped 2026 arms: placebo, 5 mg, and 10 mg at bedtime for 90 days. Heimann 2025 and NCT06359327: placebo, 2 mg, and 5 mg for 60 days. Pep19 is not approved as a prescription therapy for obesity or metabolic disease. That does not make those capsule strengths approved doses.
This article is for general informational purposes only and is not medical advice. Talk to a licensed clinician about personal medical decisions. Pep19 (DIIADDEPLT) remains investigational as a metabolic or obesity therapy. Nothing here is a dosing protocol, an availability claim for a research vendor, or a recommendation to obtain an unapproved product.
Continue your research
- GZC8072 oral weekly peptide GLP-1 first dose
- Vosoritide CANOPY-HCH-3
- ARA-290 human trial doses
- How to verify a peptide COA
- COA Authenticity Checker
Sources
- Vantini et al. iScience 2026;29(10):117471
- Crossref work record, fetched 14 September 2026
- Scienmag recap, 12 September 2026 (secondary)
- Heimann et al. Diabetes Metab Res Rev. 2025;41(5):e70056. PMID 40450548
- Heimann 2025 PMC12126752
- Heimann 2025 DOI
- ClinicalTrials.gov NCT06359327
- Reckziegel et al. Sci Rep. 2017;7:14781. PMID 29093454
- Reckziegel 2017 DOI
- Reckziegel 2017 PMC5665932
- Silvério et al. Int J Mol Sci. 2022;23(8):4082. PMID 35456900
- Silvério 2022 DOI