FDA scientists recommended against all seven peptides. The panel voting July 23 was rebuilt to say yes anyway.

Written July 2, 2026, twenty-one days before the vote. This is the follow-up we promised in our April preview, and it corrects that preview against the public record that has since landed: the Federal Register notice, the FDA staff reviews, and the reconstituted committee roster. We will update it again the day the vote happens, and ninety days later with market-impact data.

The one-paragraph version

Three weeks before the vote, the case for it fell apart in public. On June 30, 2026, the FDA released its staff reviews of the seven peptides going before the Pharmacy Compounding Advisory Committee, and the agency’s own scientists recommended against all seven.¹²³ The reviews found the human evidence insufficient across the board, and for two of the compounds, TB-500 and KPV, the reviewers reported that they could not find any study in which the substance had been given to a human being.² This is the same conclusion FDA staff reached the last time these questions came up, in 2024, when the committee voted against every peptide it reviewed.⁴ What has changed is not the evidence. It is the committee. HHS reconstituted the panel with members who prescribe, produce, and promote peptides, a roster the Washington Post and PBS reported includes at least seven people with ties to peptide-related businesses.⁵¹ So the July 23-24 meeting is now a collision: a hand-picked advisory panel, under direct pressure from a Health Secretary who has called himself a “big fan” of peptides,⁶ against the agency’s own career reviewers who say the data is not there. Wall Street is still betting the panel says yes.⁷ We think the community has the base rate right and the journalists have the near term right, and that both are less important than the thing almost nobody is saying out loud: even a yes vote changes nothing you can buy for a year or more, and the grey market’s price advantage survives either way.

What we got wrong in April, corrected against the record

Our April preview was written before the Federal Register notice and the staff reviews existed. Three things in it need correcting, and the corrections make the story sharper, not softer.

It is 7 peptides, not 14. The Federal Register notice (FR Doc 2026-07361, published April 16, 2026, docket FDA-2025-N-6895) lists seven peptides across two days.⁸ July 23: BPC-157, KPV, TB-500, MOTS-C. July 24: Emideltide (the FDA’s lead term for DSIP), Semax, Epitalon. The “14” figure that floated through the spring came from two different places and got conflated. Each peptide is being reviewed as two distinct bulk drug substances, a free base and an acetate salt, so the notice enumerates fourteen substances. Separately, “14” was also the press’s rough count of how many peptides RFK Jr. said might become more accessible, out of a starting set he called nineteen.⁹ Those are not the same fourteen. The compounds our April draft listed as candidates, CJC-1295, ipamorelin, GHK-Cu, Selank, thymosin alpha-1, are not on this agenda. Some were already rejected in 2024. GHK-Cu and three others are deferred to a second PCAC meeting scheduled before the end of February 2027.¹⁰

Nothing is being “restored.” Our preview, along with most of the coverage, described this as restoring Category 1 status. That framing is wrong on the law. These peptides were never in Category 1 and were never legally compoundable.¹¹ “Category 1” and “Category 2” are interim enforcement buckets for substances still under review, not the final list that authorizes compounding. In April, the FDA removed twelve peptides from Category 2, the “significant safety risk” bucket, and it did so for a procedural reason: the original nominators had withdrawn their nominations, so there was no active safety nomination to maintain.¹¹¹² Removal from Category 2 does not mean the FDA found anything safe. It means the substance is now in limbo, no longer flagged as dangerous, not yet eligible to compound. The only way a peptide becomes legally compoundable is to be affirmatively added to the 503A Bulks List by final rulemaking, and as of January 2025 the FDA stopped placing newly nominated substances into the interim “enforcement discretion” Category 1 at all.¹³ So there is no Category 1 slot to restore anyone to. The July vote is one early step toward addition, not a reinstatement.

The FDA is reviewing each peptide against a specific medical claim, and the claims are ambitious. This is the detail that changes how you should read the odds. The FDA is not asking “is BPC-157 a healing peptide.” It is asking whether BPC-157 should be compoundable for ulcerative colitis. The agenda pairs each substance with an indication: KPV for wound healing and inflammatory conditions, TB-500 for wound healing, MOTS-C for obesity and osteoporosis, DSIP for opioid withdrawal, chronic insomnia, and narcolepsy, Semax for cerebral ischemia, migraine, and trigeminal neuralgia, Epitalon for insomnia.⁸ These are drug claims. The bar for a drug claim is human efficacy data, and that is exactly the bar the staff reviews say these compounds do not clear.

The buried lede: PCAC has already rejected these, and it went 0-for-7

The single most important number for anyone handicapping this vote is not in the MAHA press releases. It is in the FDA’s own recent history.

The lawsuit that forced all of this, brought by the compounders Evexias and Farmakeio, argued that the FDA had reclassified peptides to Category 2 in 2023 without following its own procedure.¹⁴ The FDA settled in September 2024. It did not concede that peptides were safe. It conceded the process: it agreed to run proper notice-and-comment rulemaking, including PCAC review, before categorizing these substances.¹⁵ That settlement produced two PCAC meetings almost immediately, and they are the closest precedent that exists for what happens on July 23.

On October 29, 2024, PCAC reviewed ipamorelin, ibutamoren, L-theanine, and kisspeptin-10. It voted against including all four.⁴ On December 4, 2024, it reviewed CJC-1295, thymosin alpha-1, and AOD-9604. It voted against those too.¹⁶ That is seven peptide-class substances reviewed by this committee, out of the same legal machinery, in the same regulatory era, with the same staff scientists writing the reviews, and the committee rejected every single one.

It is worth being precise about what a PCAC vote even is, because the coverage tends to inflate it. The vote is advisory. It is non-binding.¹¹¹⁴ Historically the committee is small; the compounding trade association itself has noted the panel can seat as few as three voting members against a slate authorized for up to twelve.¹⁷ The path from a favorable PCAC recommendation to a substance you can actually get filled at a pharmacy runs through a proposed rule, a public comment period, and a final rule. On the optimistic timeline that is more than a year. On the real timeline it is longer: the first 503A Bulks List final rule did not publish until 2019, roughly six years after the underlying law passed, and it added six substances, all small molecules, no peptides.¹⁸ To this day, no therapeutic peptide of the class under review has ever completed that path.

The defensible reading of the precedent is not “PCAC always says no.” It is narrower and more useful than that: the FDA reliably follows the committee’s negative votes, and a positive vote is necessary but nowhere near sufficient. The 80-percent-follow-rate figure that circulates in the vendor blogs has no FDA source and should not be repeated.¹⁶ What is documented is that the substances PCAC voted down stayed down.

So why does everyone expect it to pass?

Because the committee is not the same committee.

In the spring, HHS reconstituted the peptide panel. The prior panels were built from academics and researchers. This one, according to reporting from PBS, the Washington Post, and STAT, was seated with health professionals who prescribe, produce, or promote peptides.⁵¹¹⁹ PBS named a physician who runs peptide clinics, another who charges around five hundred dollars for peptide consultations and promotes them on social media, and a pharmacist from a family compounding pharmacy.⁵ STAT’s headline asked whether the Secretary had “stacked the deck.”¹⁹ Three former FDA officials told ProPublica that Kennedy had mischaracterized their prior safety work.²⁰

This is why the sophisticated read splits from the community read. The peptide forums, working off the 0-for-7 base rate, mostly expect another no or a deferral. The reporters who covered the roster expect this particular panel to lean yes, because it was assembled to. And then, three days ago, the FDA’s own staff put out reviews recommending against all seven anyway, setting up the actual drama of July 23: a stacked panel voting in one direction while the agency’s scientists point in the other. A committee can recommend yes. The staff review, the proposed rule, and the final rule are all still downstream, and all still run through the career reviewers who just went on record.

There is one shortcut worth naming, because it is the thing to watch. The FDA Law Blog notes that the Secretary could, in theory, invoke the public-health provision of Section 503A to place a peptide into enforcement-discretion status directly, bypassing the committee entirely.¹⁴ That is the move that would actually change access quickly. It is also revocable by the next administration with a stroke, which makes it a weak foundation for a compounding pharmacy to build a product line on. Watch for it. It would tell you the administration decided the science argument was unwinnable and went around it.

What our data says about the seven, and the question it cannot answer

This is where TitrateLab has something no one else in this debate has: the receipts. We hold 22,762 third-party Certificates of Analysis across 1,091 distinct manufacturers. We pulled every cert we have on the seven compounds the FDA is about to vote on. Here is what the grey market’s own lab record looks like, as of July 2, 2026.

Three things jump out, and each one matters to the vote.

The FDA is ruling on seven compounds it has wildly unequal evidence for, and so is the market. Independent verification depth spans roughly fifteen to one across these seven. BPC-157 has 1,187 lab reports across 280 vendors. DSIP has 79. Four of the seven, DSIP, KPV, Epitalon, and Semax, each sit on roughly eighty to one hundred thirty tests. When the FDA staff write that they could not find human studies for KPV, they are describing a compound the grey market has also barely characterized: 110 certs, 65 vendors, a niche. The agenda treats all seven as a slate. They are not a slate. They are one heavily-traded compound, one well-traded compound, and five thin ones.

The most-tested compound is also the one with documented fraud. BPC-157’s median purity is 99.6 percent, which sounds like a clean bill of health until you look at the tail. Thirty-four of its certs come in under 90 percent purity, and twenty-three of those assayed at zero: vials sold as BPC-157 that contained no detectable BPC-157 at all. One cert in our set reads plus 729 percent on fill, a vial with more than eight times the labeled dose. So BPC-157 carries, simultaneously, the strongest evidence base and the clearest record of adulteration of any peptide on the agenda. Both facts are true, and both are relevant: the compound the panel is most likely to wave through is the one whose grey-market supply most demonstrably includes counterfeits.

The dosing failure mode is overfill, not shorting. Every one of the seven runs a positive median deviation. Tested contents come in over the labeled dose more often than under it, sometimes far over: KPV plus 16 percent at the median, DSIP plus 13, Semax plus 13. This cuts directly against the FDA’s, and the public’s, mental model of grey-market peptides as underdosed junk. The real dosing risk in this corpus is a fat right tail of overfilled vials, not systematic cheating on quantity. That is a more interesting problem than fraud, and a harder one to regulate, because an overfilled vial passes a “is the drug in there” test and fails a “how much is in there” test.

Now the part that should reframe the entire debate. Every number in that table answers one question: is the compound real, and is it pure? That is a chemistry question, and the grey market, with its Janoshik and Finnrick receipts, has answered it thousands of times over. The FDA’s four-factor test asks a different question. Its second and third factors are safety in humans and evidence of effectiveness in humans.²¹ No Certificate of Analysis speaks to either. A vial can be 99.7 percent pure DSIP and that tells you exactly nothing about whether DSIP treats narcolepsy or is safe to inject for a year. The grey market built an extraordinary apparatus for verifying identity and purity. It built almost nothing for verifying safety and efficacy, because that requires clinical trials, and clinical trials are the one thing a Chinese contract synthesizer and a Telegram vendor cannot produce. This is the structural reason the staff reviews read the way they do, and it is why our own data, as good as it is, does not rebut them. We can tell you what is in the vial. We cannot tell you it works.

Why peptides are hard to compound, mechanically

Even setting the politics aside, these seven face a wall built into the statute.

Compounding law draws a hard line at forty amino acids. Below it, a molecule is a peptide and can, in principle, be compounded. Above it, the molecule is a biologic, and compounding requires a biologics license that 503A pharmacies cannot obtain.²² All seven of these clear the forty-amino-acid threshold, but the threshold is the reason the whole category lives on a knife edge.

Then there is the supply chain, which is the part almost no vendor blog mentions. To compound legally, a pharmacy must source its bulk active ingredient from an FDA-registered establishment, accompanied by a Certificate of Analysis, and it may not use material labeled “research use only.”²² The entire existing grey-market supply, the Chinese-direct API that our 280 BPC-157 vendors are reselling, is RUO material. It is categorically disqualified. A yes vote does not launder it. It means a compounding pharmacy would have to build a compliant, FDA-registered supply line from scratch, which does not currently exist for these compounds at scale, and which, per Foreign Policy’s reporting, would still run back to Chinese synthesis for the foreseeable future because the domestic capacity to chain these amino acids has not been built.²³ That last point is the intra-administration fault line worth watching: peptides are made predominantly in China, so a MAHA deregulation that expands demand largely expands demand for Chinese product, which puts the Health Secretary on a collision course with the China hawks in his own party.²³

And then there is immunogenicity, the safety concern the FDA has led with since 2023.²⁴ Synthetic peptides carry peptide-related impurities, and the FDA’s impurity bar is strict: any new peptide-related impurity above 0.5 percent of the drug substance is a problem, and anything at or above 0.10 percent must be identified.²⁵ The immune response to an injected synthetic peptide can, per the FDA’s own bulletins, range from nothing to anaphylaxis, and the human data that would let anyone estimate where on that range a given compound falls is exactly the data the staff reviews say is missing.²⁰

What the people who actually use these think

We read the forums so you do not have to, and the picture is more interesting than either the hype or the panic.

The peptide community, on Reddit, is mostly cynical that the vote passes, anchored on that 0-for-7 base rate. The sharper commenters have the law right, that coming off the “bad list” is not the same as getting onto the “good one,” and that the recommendation is non-binding.²⁶ But the more revealing sentiment is about what they would do if it did pass, and the answer is mostly: nothing. The legal compounded channel is expected to run three to ten times grey-market prices, and the community’s read is that legality does not move the price of a compound people can already get for thirty-five dollars a vial. Their evidence is testosterone and HGH, both legally prescribable and both among the cheapest things on the grey market anyway. Prescription status has never been what sets the price.

The veterans on MESO-Rx go a step further, into a genuinely counterintuitive place: a faction of them does not want legalization, and articulates exactly why. One long-running thread lays out the thesis that transparency is what kills the cheap grey market, not enforcement. “You cannot have cheap peptides and full public transparency and expect the industry to continue existing,” one veteran wrote. “Consumers want cheap research chemical prices, but they also want pharmaceutical grade assurance. And that’s a contradiction.”²⁷ It is the same tension we sit inside every day. The thing that makes a compound trustworthy, a paper trail of lab reports and named manufacturers, is also the thing that hands regulators and pharma a target map. The grey market’s cheapness and its opacity are the same property viewed from two sides.

The money is positioning hard but quietly. The compounding trade association is running a real, organized effort to collect five years of member peptide-dispensing data to submit to the committee, the closest thing to a coordinated evidentiary case for the yes side.²⁸ On the public markets, Hims and Hers stock jumped about 50 percent in a week in April on the peptide optionality, and even after the June 30 staff reviews, a Needham analyst said he was “still operating under the assumption that they will get approved.”²⁹⁷ That is Wall Street pricing the stacked panel over the staff science. It may be right about the committee vote. We think it is wrong about how quickly that vote becomes a product, for all the rulemaking reasons above.

Our read, per peptide

We will state odds, because you asked us to, and because a preview that refuses to predict is just a reading list. These are our estimates of the probability that the committee recommends the substance for inclusion on July 23-24. They are not predictions that you will be able to buy compounded product any time soon; that is a separate, much longer, much lower-probability event gated on rulemaking. We will grade ourselves publicly after the vote.

• BPC-157 (ulcerative colitis) - the best-documented and most-lobbied compound, and the one the stacked panel most wants to advance. But the staff review says the data does not support the UC indication, and BPC-157 carries the immunogenicity flag and the cancer-angiogenesis question the skeptics keep raising, still theoretical in humans but still unanswered.¹⁹³⁰ A favorable committee vote is plausible here, maybe the only one that is; call it a coin flip leaning slightly yes on the panel, near-zero on near-term access.
• KPV and TB-500 (wound healing) - the staff reported no human studies at all. It is hard to see even a stacked panel voting yes into a record that says “we found zero human data,” because that vote is the one that gets quoted in the eventual lawsuit. We lean no on both.
• MOTS-C (obesity and osteoporosis) - clean purity in our data, zero fails across 290 certs, but near-zero controlled human data and two ambitious indications. Lean no.
• Semax and Epitalon and DSIP - the thin four. The human literature is mostly old, mostly ex-Soviet, and mostly not reproduced to Western regulatory standards. These are the compounds most likely to be deferred rather than voted at all. Lean no or deferral.

The realistic aggregate, and it matches the sharpest neutral industry read we found,³¹ is one, possibly two favorable committee votes out of seven, the rest rejected or deferred, and every one of them still a year or more of contested rulemaking away from a pharmacy shelf. That is not “the grey market collapses.” That is “the grey market gets a slow, partial, revocable competitor sometime in 2027 or 2028, for one or two compounds, at three to ten times the price.”

What this actually means

For the grey market: less than the headlines claim, and the timing is backwards from the fear. The thing hollowing out the grey market is not this vote. It is the enforcement that already happened, the 2025 raids and the December 2025 guilty pleas and the vendor exits, Science.bio in January, Peptide Sciences in March.³² And yet, tellingly, grey-market prices fell through that die-off rather than spiking, because the survivors compete harder. A yes vote on July 23 does not reverse any of that. It introduces, eventually, a legal channel that is more expensive, prescription-gated, and supply-constrained, aimed at a different customer than the one buying at thirty-five dollars a vial. The two markets bifurcate; they do not merge.

For consumers: the honest guidance has not changed, and the vote does not change it. A compounded peptide is not automatically safer than a grey-market one. It is more legally covered, which is a real thing that matters if you value not being part of a federal case. But a compounding pharmacy that mislabels a vial harms a patient the same way a Chinese OEM that mislabels a vial does, with a different lawyer attached. The thing that protects you in either channel is the same thing it has always been: independent verification of what is actually in the vial. Our data says the grey market’s identity-and-purity verification is, for these seven, genuinely good, 99-percent-plus median purity across the board, with a real but minority fraud tail on BPC-157 and TB-500 you can screen for by demanding a recent, matching lab report. What no channel gives you, grey or compounded, is proof that the compound is safe and effective for the indication on the FDA’s agenda. That evidence does not exist yet. That is the actual finding of June 30, and no vote on July 23 creates it.

For TitrateLab: this is the market we were built for, and the vote sharpens rather than threatens the role. The entire debate is a fight over verification: the FDA does not trust the grey market’s paper, the grey market does not trust the FDA’s motives, and the compounders are scrambling to assemble a paper trail credible to a hostile panel. In a world where identical molecules will soon be sold through two channels with two legal profiles and two price points, the only thing that keeps either channel honest is third-party lab data applied equally to both. That is the whole thesis. Whether these seven pass or fail, the demand for an independent answer to “what is actually in this vial” goes up, not down.

What to watch between now and July 23

• The full FDA background package, expected in the days before the meeting, will contain the per-substance reviews in detail. The June 30 staff recommendations were the headline; the packet is the evidence behind them.
• The 503A bypass. If the Secretary moves to place any peptide into enforcement discretion directly, without waiting for the committee, that is the signal that the administration decided the science fight was unwinnable and went around it. It would be fast and it would be fragile.
• The vote split itself. A stacked panel voting yes 7-0 against its own staff’s recommendation is a different story, legally and politically, than a divided panel splitting on the two best-documented compounds. The split is the tell for how the rulemaking fight goes.
• China. Any move by the China hawks in the GOP to tie peptide deregulation to Chinese supply dependence would reframe this from a health story into a trade story, and trade stories move faster in this administration than health ones.

Editorial positioning

TitrateLab takes no position on whether peptide compounding should be more or less restricted. We are not a patient-advocacy group, a compounding trade association, or a grey-market vendor. Our job is to describe what is happening accurately enough that you can make your own sourcing decisions.

Two things we will stand behind. First, the June 30 staff reviews are the most credible document in this entire debate, more credible than the MAHA press releases and more credible than the vendor-blog handicapping, because they are the FDA’s own scientists reaching the same conclusion they reached in 2024 with the same evidence, and because reversing them requires new human data that simply does not exist. Second, a stacked advisory panel overruling its own agency’s science is not a restoration of “the rule of law,” which is how the reclassification was sold. It is regulatory capture wearing the other team’s jersey. The 2023 decision that created this grey market was shaped in part by pharmaceutical interests that benefit from restricted peptide access. A 2026 reversal driven by a panel of people who sell peptides is the same dynamic pointed the other way. Neither produces good policy. What produces good outcomes for the person holding the vial is neither channel winning outright, but both channels being forced to show their work.

We will update this article the day the vote happens with the actual outcome, and again ninety days later with what it did to the market.

TitrateLab is the intelligence layer for the peptide and grey-market underground. The lab-data figures in this article were computed from our Certificate-of-Analysis corpus on July 2, 2026 and are reproducible on request. This article will be updated on July 23-24, 2026 with the committee’s vote and again ninety days later with market-impact data.

Footnotes