Gray market peptides are injectable compounds sold outside a prescription and licensed-pharmacy relationship. The new Johns Hopkins report doesn't say every peptide is fake. It says a buyer often can't verify what is in the vial, whether it was made to be sterile, or whether the promise on the website rests on good human research. A prescription changes that chain of responsibility. It doesn't turn thin evidence into strong evidence.
What the gray market peptides report found
On September 16, 2026, Johns Hopkins published its report after speaking with physician-scientists Christopher Robinson and Anita Gupta. Their plain concern was that compounds promoted for recovery, sleep, energy and aging have often been studied mainly in mice or rats, while people are already injecting them.
As of September 18, 2026, the day this article was written, the fresh picture was this: the Hopkins report was two days old, a UCLA-led evidence review had been online since August 11, and the FDA had posted materials from its July peptide meeting but no final rule adding the reviewed compounds to the 503A bulks list.
The growth number needs a careful label. Chainalysis estimated that cryptocurrency flowing to the gray-market peptide sellers it tracked had moved from roughly $1 million per quarter in 2024 to a run rate above $100 million a year in 2026. That is the firm's estimate of visible crypto payments, not an official count of every peptide sale.
Johns Hopkins also described the broader U.S. compounding business as roughly $6.8 billion. The market release it linked was more specific: it estimated $6.98 billion in 2025 and projected $7.42 billion for 2026. That number covers compounding generally, not gray-market peptides. Putting the two markets in one sentence can make them sound interchangeable. They aren't.
The human evidence has not caught up
A Partnership for Safe Medicines digest labeled September 8 pointed readers to the same evidence review and the gap between the marketing and the medicine.
The newest review puts a number on that gap. UCLA's September 2 summary says the team examined 565 studies involving BPC-157, TB-500, CJC-1295, MK-677, ipamorelin and GHK-Cu. The peer-reviewed paper in The American Journal of Sports Medicine reports that 67% of the publications used preclinical animal models, meaning studies done before dependable testing in people. Human studies were limited to a handful and usually lacked strong controls.
That does not prove the compounds can never help. It means a rat tendon result cannot tell a person how likely a benefit or side effect is. BPC-157, for example, has a large animal literature but no robust human trial establishing recovery benefits. For TB-500, the review found no peer-reviewed human trial for a musculoskeletal condition.
The distinction matters because the sales pitch often skips the species. "Studied" may mean cells in a dish, a small group of rats, or a controlled human trial. Those are three very different levels of confidence.
A label is not the same as verification
A vial can look tidy and still leave basic questions unanswered. A certificate of analysis, a lab report on one tested sample, may show identity or purity. It does not automatically prove that the vial in hand came from that batch, was filled to the stated strength, or remained sterile through shipping.
One test-purchase study gives this problem real numbers. Researchers ordered semaglutide from six online sellers that did not require a prescription. Only three orders arrived. In the 2024 Journal of Medical Internet Research paper, all three delivered samples contained endotoxin, a residue from bacteria, and measured semaglutide content was about 29% to 39% above the label. Purity ranged from 7.7% to 14.4% despite a 99% claim.
That was a tiny sample, and it tested semaglutide rather than every peptide sold online. It cannot supply a failure rate for the whole market. It can answer the narrower question: a professional-looking site and a printed purity claim are not independent verification.
What a prescription route actually changes
The useful contrast is not gray market versus perfect safety. It is anonymity versus accountability.
A prescribed compounded medication starts with a clinician who can see medical history, current medicines and the reason for the request. A licensed U.S. compounding pharmacy prepares the prescribed formulation and supplies a pharmacy label stating the patient, prescriber, compound, strength and directions. The pharmacy keeps records tied to the preparation, and there is a pharmacist to call.
If something goes wrong, there are places for the report to land: the prescriber, the pharmacy, the state board and FDA MedWatch. Our guide to reporting peptide side effects explains that chain. A vial bought from a research-labeled seller may have none of it.
This route does not repair a weak evidence base. It adds screening, a documented dose, a pharmacy of record and follow-up. The molecule still has whatever human evidence it has.
At Promise, a licensed provider reviews every request and not everyone qualifies. Where the applicable rules permit compounding, a licensed provider may still prescribe a compounded formulation; that decision is between you and your doctor.
What the FDA committee vote did not do
The FDA's Pharmacy Compounding Advisory Committee met on July 23 and 24, 2026 to discuss seven bulk substances. Its votes recommended six, including BPC-157 and TB-500, for the 503A bulks list. The FDA meeting page shows the uses reviewed and the agency's briefing documents.
A committee recommendation is advice, not a final rule and not drug approval. It did not settle the evidence for recovery claims. It also did not make an anonymous online vial equivalent to a medication prepared by a licensed pharmacy for a named patient. Promise's FDA peptide decision explainer follows the rulemaking question without turning it into a verdict on any one person's care.
What to watch next
The next meaningful document is a proposed or final FDA rule, not a social post saying the July vote changed everything. Better human trials would matter too: enough participants, a comparison group and outcomes chosen before the results are known.
The market is moving faster. A separate telehealth peptide expansion explainer follows one business thread raised by the Johns Hopkins story. Whatever company name is attached, the same questions remain: Who reviewed the person? Which licensed pharmacy prepared the medication? What exactly does the label say? Who answers if the vial or the response is wrong?
That is the practical meaning of the Johns Hopkins report. The problem is not that peptides are one bad category. It is that gray-market buying removes the people and records needed to tell a medical decision from a gamble.