How peptides are prescribed in the United States

A peptide is just a short chain of amino acids. The regulatory category it lands in depends on whether the FDA has reviewed it as a drug, whether it's been classified as a biologic, and whether a compounding pharmacy is permitted to prepare it. Those three questions produce four practical buckets.

The first bucket: FDA-approved peptide drugs. Semaglutide (sold as Ozempic and Wegovy), tirzepatide (Mounjaro and Zepbound), liraglutide, tesamorelin for HIV-associated lipodystrophy, and bremelanotide (PT-141, branded Vyleesi) for hypoactive sexual desire disorder. These cleared Phase 1 through 3 trials, carry approved labeling, and are dispensed through normal retail and specialty pharmacy channels.

The second bucket: peptides legally compoundable by licensed pharmacies under FDA's interim Category 1 list. Sermorelin is the cleanest example. A licensed prescriber writes a patient-specific prescription, a 503A pharmacy prepares it, the patient picks it up or has it shipped.

The third bucket: peptides that were widely compounded until late 2023, then reclassified as Category 2 - meaning the FDA flagged enough concern about evidence, immunogenicity, or quality that compounding was prohibited ^[1]. BPC-157. TB-500. CJC-1295. Ipamorelin. Thymosin Alpha-1. GHK-Cu. AOD-9604. Seventeen substances in total moved into this category.

This bucket has been draining since, without anything moving into the legal one. Five - AOD-9604, CJC-1295, Ipamorelin acetate, Thymosin Alpha-1, and Selank - were pulled out in September 2024 and referred to the Pharmacy Compounding Advisory Committee; that review is still pending, and none of the five was on the July 2026 agenda. Twelve more came out of Category 2 in April 2026, and seven of those went before the committee on July 23-24, 2026, which recommended six of them for the permitted list and rejected only emideltide (DSIP). See our report on the vote.

The result is a third bucket that is now mostly a fourth one: substances neither prohibited nor permitted. Coming off Category 2 removes a stated prohibition; it does not put a substance on the Category 1 list, and only that placement — reached through rulemaking FDA has not begun — makes a bulk substance legally compoundable. For a prescriber and a pharmacy, the practical answer for all of these is still no.

The fourth bucket: peptides sold online as "research use only." Not legal for human consumption, not subject to FDA manufacturing or purity standards, and not what this article is about - though they account for a large share of what's actually being injected in the United States.

Short answer: with one important exception, peptides aren't on the DEA's schedules. They aren't Schedule I, II, III, IV, or V. A licensed prescriber doesn't need a DEA registration number to prescribe sermorelin or semaglutide, in the way they would for oxycodone or alprazolam.

The exception is human growth hormone. HGH distribution without a prescription carries specific federal criminal penalties under the Anti-Drug Abuse Act of 1988 - a carve-out written into law decades before the current peptide market existed, aimed at gym-and-locker-room diversion ^[2]. Certain growth hormone secretagogues came under DEA controlled substance review during 2025 because of sports-medicine misuse patterns, though no formal scheduling action has been finalized.

What this means in practice: the regulatory pressure on peptides comes almost entirely from the FDA, not the DEA. The constraints are about manufacturing, compounding, and approval status - not about controlled substance handling.

All peptides intended for human use require a prescription. There's no over-the-counter peptide injection sold legally at a US pharmacy.

What varies is the path the prescription travels. A semaglutide prescription written for weight management goes to a retail or specialty pharmacy and gets filled with manufactured Wegovy. A sermorelin prescription written for adult growth hormone deficiency goes to a 503A compounding pharmacy and gets filled with a custom preparation. A BPC-157 prescription written today still can't be filled by a US compounding pharmacy - not because the substance is prohibited, which it no longer formally is, but because it isn't on the permitted list either, and a pharmacy needs the permission, not the absence of a ban. That is why patients seeking it end up on overseas research-chemical sites or in clinics operating in regulatory gray zones.

The peptides that don't require a prescription, strictly speaking, are the ones being sold with "not for human consumption" labeling. That labeling isn't a loophole. It's a statement that the seller isn't claiming the product is a drug, and a buyer who injects it's doing so outside any regulated framework.

Most peptide prescriptions written today move through telehealth, not brick-and-mortar clinics. The workflow goes like this.

A patient fills out an intake questionnaire - medical history, current medications, weight, goals, sometimes lab results. The platform matches the patient to a licensed clinician in the patient's state. The clinician reviews the intake and conducts either an asynchronous evaluation (reviewing the file, sometimes exchanging messages) or a synchronous evaluation (video visit). If the clinician determines the peptide is appropriate, they write a prescription and route it to either a 503A compounding pharmacy partnered with the platform or, for approved drugs, a retail or specialty pharmacy.

Asynchronous models are faster and cheaper. The clinician can review dozens of cases in the time a single video visit would take, and many states permit asynchronous evaluation as long as the clinician has the information needed to make a reasonable clinical judgment. Critics - including some state medical boards - argue that asynchronous prescribing of injectables, particularly for off-label indications, doesn't meet the standard of care. The argument hasn't been resolved and varies by state.

Synchronous models look more like traditional telemedicine. A video visit, a more thorough history, sometimes labs ordered before the prescription is written. Slower, more expensive, and - at least in the view of regulators - more defensible.

The clinician writing the prescription bears the legal responsibility. The platform that connects them to the patient generally doesn't, though that division of liability has been tested in litigation since 2023 and continues to evolve.

A patient seeking peptides through a conventional clinic is generally going to one of three places: a primary care physician, an endocrinologist, or an anti-aging or functional medicine clinic.

Primary care is where most semaglutide and tirzepatide prescriptions originate, written for diabetes management or, increasingly, weight management with off-label use becoming on-label use as the FDA expanded indications. The workflow is conventional: visit, evaluation, prescription to a retail pharmacy, insurance adjudication.

Endocrinology becomes relevant for growth hormone deficiency, HIV-associated lipodystrophy, and the more complex metabolic cases. A specialist evaluation, labs, sometimes imaging. Tesamorelin prescriptions almost always run through endocrinology.

Anti-aging and functional medicine clinics are where the compounded-peptide market lives. These clinics - which proliferated between roughly 2019 and 2023 - built their practices around sermorelin, ipamorelin, CJC-1295, BPC-157, and similar compounds, paired with hormone optimization and a wellness aesthetic. The 2023 Category 2 reclassification hit this business model directly. Clinics that had built protocols around BPC-157 had to either substitute, send patients overseas, or pivot to whatever remained legally compoundable. Many pivoted to sermorelin and the GLP-1s. Some closed.

The difference between 503A and 503B is the central piece of infrastructure most patients never learn about, and it determines almost everything about how a non-approved peptide reaches a syringe.

Section 503A of the Food, Drug, and Cosmetic Act covers traditional compounding pharmacies ^[1]. They prepare medications in response to individual prescriptions - patient-specific, written by a licensed prescriber for a specific person. A 503A pharmacy can only compound from bulk substances on the FDA's Category 1 list. Category 2 substances are off-limits. State boards of pharmacy regulate 503A facilities, and the standards vary; some states permit "office stock" compounding without patient-specific prescriptions, which the FDA considers a violation of federal law but hasn't aggressively prosecuted.

Section 503B was created by the Drug Quality and Security Act in 2013, after the New England Compounding Center meningitis outbreak killed 64 people. 503B "outsourcing facilities" can prepare medications in larger batches without patient-specific prescriptions, selling to healthcare facilities. In exchange, they must comply with Current Good Manufacturing Practice standards, register annually with the FDA, submit biannual reports, and accept federal inspection. Higher bar, broader latitude.

For peptides, the 503A pathway is where most patient-specific compounding happens. The 503B pathway became briefly important during the GLP-1 shortage - when the FDA declared semaglutide and tirzepatide officially in shortage, 503B facilities could compound them at scale. That window closed when the shortages were declared resolved (tirzepatide in December 2024, semaglutide in February 2025), with compounding deadlines enforced through April and May 2025.

The result is a narrower compounding landscape than existed two years ago. Fewer peptides on the legal list, more enforcement, and ongoing litigation - including the Evexias v. FDA settlement in 2024, which forced the FDA to route reclassification decisions through the PCAC process rather than unilateral action.

A patient walks into a CVS or Walgreens with a peptide prescription. What can they actually get?

The FDA-approved peptides: Ozempic, Wegovy, Mounjaro, Zepbound, Saxenda, Vyleesi, and a small number of others. These are manufactured drugs with NDC numbers, distributed through normal pharmaceutical wholesale channels, dispensed like any other prescription. Tesamorelin (Egrifta) generally requires a specialty pharmacy rather than retail, given its narrow indication and cost.

Everything else: not available at retail. Sermorelin, ipamorelin (during periods when it's compoundable), and the rest of the compounded-peptide universe travels through 503A pharmacies - many of them mail-order, some of them brick-and-mortar - that specialize in this category. A CVS won't have it. A specialty compounding pharmacy will.

This is one of the points patients most often misunderstand. The peptide market that anti-aging clinics and telehealth platforms have built doesn't run through the retail pharmacy system. It runs alongside it, on a parallel infrastructure of compounding pharmacies with different rules and different supply chains.

Insurance coverage tracks the regulatory bucket. FDA-approved peptides with on-label indications are eligible for coverage, subject to the same formulary and prior-authorization games that govern every expensive drug. Semaglutide and tirzepatide for type 2 diabetes are generally covered. For weight management - Wegovy and Zepbound - coverage is patchier, with many employer plans excluding GLP-1s for obesity entirely, and Medicare prohibited from covering them for weight loss until recent rule changes began to shift that.

Compounded peptides are essentially never covered. Insurance pays for FDA-approved drugs in approved indications. A compounded sermorelin prescription is out-of-pocket. A compounded GLP-1, even during the shortage window when it was legally available, was out-of-pocket. This is one reason the compounded GLP-1 market grew so quickly: it was the cheaper option for patients whose insurance wouldn't cover the brand.

IQVIA data from 2025 showed that compounded GLP-1 users had significantly lower three-month continuation rates than branded-drug users ^[3]. The honest read of that data: paying cash for a less-supported product, then stopping, is a different experience than paying a $25 copay for Wegovy.

Off-label prescribing is legal. A licensed physician in the United States may prescribe an FDA-approved drug for any indication they consider clinically appropriate, even if that indication isn't on the label. This is settled law and standard practice across medicine - a large share of pediatric prescribing, oncology prescribing, and psychiatric prescribing is off-label.

What off-label prescribing doesn't permit is prescribing a drug that isn't FDA-approved at all. The legal latitude applies to indication, not to drug status. A clinician can prescribe semaglutide for an off-label use. They can't prescribe an unapproved peptide off-label, because there's no label to be off of.

The liability picture for off-label peptide prescribing - particularly through telehealth, particularly for cosmetic or performance indications - has been tested in malpractice cases and state medical board actions since 2023. The general principle: clinicians remain responsible for meeting the standard of care, and "the patient asked for it" is not a defense. Clinicians who prescribe peptides without adequate evaluation, documentation, or follow-up face the same exposure they would for any other prescribing decision.

The regulatory state of compounded peptides changed more in the last 30 months than in the previous decade. A summary of what's currently in play.

Category 2 list, established late 2023: 17 peptides reclassified as compounding-prohibited. Five of those - AOD-9604, CJC-1295, Ipamorelin acetate, Thymosin Alpha-1, Selank - were removed from Category 2 and referred to PCAC in September 2024 for formal review. As of mid-2026 that review is still pending; these five were not on the FDA's July 23-24, 2026 PCAC agenda.

FDA docket FDA-2025-N-6895: opened to collect public comment on whether to add roughly 12 additional peptides - including Cathelicidin/LL-37, GHK-Cu, Dihexa acetate, Melanotan II, and PEG-MGF - to the bulk compounding list. Comment period open through July 22, 2026.

GLP-1 shortage compounding: ended. Tirzepatide shortage declared resolved December 2024. Semaglutide shortage declared resolved February 2025. Enforcement deadlines for compounded preparations ran through April and May 2025. Some patient-specific 503A compounding may continue under narrow clinical-need exceptions, but the mass-market compounded GLP-1 channel is effectively closed.

Import Alert 66-78: expanded in 2025 to add unapproved peptides, restricting importation.

RFK Jr. announcement, early 2026: stated goal of returning roughly 14 peptides to compounding eligibility ^[4]. As of this writing, no formal FDA rulemaking has been published reflecting that goal. The PCAC review process in July 2026 is the proximate decision point.

The honest summary: the peptide market is contracting in some directions (compounded GLP-1s, BPC-157, the broader anti-aging compound list) and possibly expanding in others (whatever survives PCAC review, plus continued growth of the FDA-approved category). What looks settled today may not look settled in twelve months.

That uncertainty is what makes prescribing in this space genuinely hard. A clinician who built a practice around compounded peptides in 2022 has had to rebuild it twice since. A patient trying to understand what's legal, what's available, and what's worth taking is navigating a landscape where the rules keep moving.