ON SPEC

Testing intelligence for supplement, peptide, and compounding teams. A monthly briefing from Prodigy Labs.

Archive · Issue 2 · September 2026

Last month the story was a committee voting to let six peptides into compounding. This month the story is what the agency actually did instead. Sources are linked so you can check my work.

The Big Read: the committee voted yes. The agency is still saying no.

Seven weeks after the Pharmacy Compounding Advisory Committee recommended six of seven nominated peptides for the 503A bulk substances list, nothing has been published. I searched the Federal Register for compounding bulk drug substances and for peptides across August 1 to September 13. Zero documents touch the 503A list or the July nominations. The compounding items that did publish were an animal drug compounding draft guidance and an outsourcing facility registration notice.

What did happen: on August 24 the FDA issued unapproved new drug and misbranding letters to at least five peptide sellers, posted publicly on September 1.

Recipient Issued Cited as
Peptide Partners LLC Aug 24 Unapproved new drugs
Royal Peptides LLC Aug 24 Unapproved new drugs
Peak Performance Peptides Aug 24 Unapproved new drugs
NuScience Peptides LLC Aug 24 Unapproved new drugs
TXP Innovations LLC dba Tex Peptides Aug 24 Unapproved new drugs

Products cited across the letters include semaglutide, tirzepatide, retatrutide (sold under coded names such as GLP-3 Reta), SS-31, tesamorelin and PT-141. The agency held that research use only disclaimers were defeated by therapeutic marketing claims and by selling administration supplies, bacteriostatic water and dosing calculators, alongside the vials.

Here is the part worth reading twice. Not one of those letters cites a purity, assay, sterility or endotoxin failure. The theory is intended use, start to finish. A spotless certificate of analysis would not have changed a single outcome. That is worth saying plainly, because the opposite belief is common and expensive.

It does not follow that characterization is optional. Counsel writing on the vote note there is no statutory deadline for FDA to act, that the closest precedent took more than two years to reach a final rule, and that the enforcement discretion bucket closed to substances nominated after January 2025, so there is no interim safe harbor. Their advice to industry is to propose the guardrails first: enhanced testing requirements, sourcing standards, certificates of analysis, labeling, physician oversight. FDA's stated objections were physical and chemical characterization, purity, safety, immunogenicity and effectiveness. Three of those five are lab questions. For the next year or so, the analytical file is not only your compliance position, it is your lobbying position. Holland and Knight put the current state bluntly: as of this writing, these peptides still cannot be lawfully compounded.

Testing Watch

The GLP-1 import exemption is supplier level, not batch level. FDA revised Import Alert 66-80 on August 21, adding orforglipron API for three manufacturers and a drug product intermediate for a fourth, plus new product codes. The exempted supplier list now runs past twenty firms across eight countries, covering semaglutide, tirzepatide, liraglutide, exenatide and dulaglutide. Read the mechanism carefully: the exemption attaches to the manufacturer, not to your lot. Material from a listed firm still arrives with no agency statement about the drum in front of you, which leaves incoming identity and assay verification where it has always been, with the buyer.

NSF put numbers on the gap between testing and certification. At the inaugural Future Nutra Innovation Summit on September 9, NSF's David Trosin reported roughly a 10 percent failure rate for products entering NSF's certification program, about 30 percent for products tested outside it, and near 50 percent in sexual health, muscle building and weight loss. The Certified for Sport banned substances panel has grown from about 100 items to 329. His framing: third party testing is not certification, it is a single test versus an ongoing one.

Self-affirmed GRAS would end under a proposed rule. FDA proposed on August 11 to require GRAS notices for substances used in human or animal food, closing the self-affirmation pathway. Comments close December 9. If it lands, ingredient suppliers running on internal dossiers will need notice grade identity, composition and specification data, which is analytical work that mostly has not been done yet.

On the Bench

565 studies on six popular peptides, and the human evidence is not there. A UCLA scoping review of BPC-157, TB-500, CJC-1295, MK-677, ipamorelin and GHK-Cu found claimed musculoskeletal and performance benefits unsubstantiated by current human trials, with more than two thirds of the literature animal only and dosing highly heterogeneous. MK-677 was linked to congestive heart failure in about 6.5 percent of trial participants. American Journal of Sports Medicine, August 11.

Where the plus 56 peak comes from. A study in the Journal of Pharmaceutical and Biomedical Analysis identifies the likely origin of plus 56 Da impurities formed during solid phase peptide synthesis, and how to suppress them. If you have ever had a mass shift on a purity profile read as contamination, this is the paper. September 3.

Confirming identity when the peptide is cyclic. A review of mass spectrometry strategies for confirming cyclic peptide identity and profiling impurities, where cyclization defeats the usual sequencing logic. Journal of the American Society for Mass Spectrometry, August 1.

A possible alternative chemistry for endotoxin. A peptide based digital ELISA for endotoxin detection in clinical plasma and pharmaceutical products, reported as an ultrasensitive option alongside existing LAL and rFC methods. Early stage, but worth tracking if you run injectables. Analyst, August 28.

New Developments

Per batch reporting is becoming the vendor pitch. On September 1 a research peptide vendor announced a system linking each product batch to its own independent analytical report, with a declared panel of identity, purity, measured quantity, heavy metals, endotoxin and sterility. Treat it as an announcement rather than verified practice, but the direction is the signal: one marketing certificate for a product line is being replaced by documentation per batch.

Capital is arriving in the peptide supply chain. System, a peptide telehealth and compounding platform, raised 20 million dollars on September 9 and says it plans to add raw ingredient manufacturing to a stack that already includes registered compounding pharmacies. Vertical integration changes who owns the quality question, and it usually lands on whoever is closest to the API.

Testing capacity keeps consolidating. Eurofins is acquiring Element Materials Technology's North American life sciences testing business, 27 laboratories at a 400 million dollar enterprise value, with closing expected this quarter. Eurofins is also one of the three verifiers Amazon will accept for supplements, which is a lot of the market's testing capacity and a lot of its retail gatekeeping in one place.

From the Lab: a report nobody can check is not documentation

If the market is moving to per batch reporting, the obvious failure mode is a certificate that cannot be verified by the person relying on it. Screenshots circulate. Numbers get retyped. Every certificate we issue carries a verification link, so a customer, or a customer's customer, can confirm the report came from us and matches the batch in hand. If you are asking a lab for per batch documentation this quarter, ask them how a third party checks it.

Send a sample: https://app.prodigy-labs.com/join/testing Peptide identity, purity and content pricing: https://prodigy-labs.com/peptide-testing-pricing


Sources

Get the next issue

One email a month. What actually changed in testing and regulation, with sources.

You are on the list. The next issue comes the first Tuesday of the month.