General information about U.S. regulatory definitions and published chemistry literature — not legal, medical, or manufacturing advice.

"Peptide synthesis company" covers everything from a cGMP contract manufacturer supplying an approved drug's active ingredient to a website shipping vials behind a "research use only" disclaimer. U.S. law puts those in very different categories, and every fact that separates them is checkable against a primary source.

What does a peptide synthesis company do?

It builds peptides chemically, chain by chain, then purifies and identifies what it built. The dominant method is solid-phase peptide synthesis (SPPS), originating with Merrifield's 1963 tetrapeptide "assembled under heterogeneous conditions from the C- to the N-terminus on a polymeric solid 'resin'" through repeated "alternating steps of Nα-deprotection, activation and coupling," per *Beilstein Journal of Organic Chemistry* (2014). Boc chemistry dominated SPPS's first years; Fmoc chemistry opened "a novel, more variable synthesis concept."

The work does not end at the resin. After cleavage, the same article reports, the crude peptide is "purified by standard analytical methods such as the diverse chromatographic techniques," with identity confirmed by soft-ionization mass spectrometry such as MALDI-TOF.

Is the output a drug or a biologic?

The dividing line is 40 amino acids. FDA's regulations define a protein as "any alpha amino acid polymer with a specific, defined sequence that is greater than 40 amino acids in size" under 21 CFR § 600.3(h)(6), so a polymer of 40 residues or fewer is a peptide regulated as a drug, and anything longer is a biological product that may not enter interstate commerce unless a license "is in effect," under 42 U.S.C. § 262(a)(1).

Chemical synthesis is no escape hatch: the Further Consolidated Appropriations Act, 2020 (Pub. L. 116-94), enacted December 20, 2019, struck the former exclusion "(except any chemically synthesized polypeptide)" from the definition of "biological product" in 42 U.S.C. § 262(i)(1).

Does "FDA registered" mean the products are FDA approved?

No — and saying so is itself a violation. 21 CFR § 207.77(a) states that "registration of an establishment or listing of a drug does not denote approval of the establishment, the drug, or other drugs of the establishment." Subsection (b) goes further: any representation that a drug is approved or legally marketable because the establishment holds a registration number, a Unique Facility Identifier, or an NDC "is misleading and constitutes misbranding."

Registration is clerical: a domestic establishment registers within 5 calendar days of beginning to manufacture a drug, per 21 CFR § 207.21. Nor is a Drug Master File a credential — FDA states DMFs "are neither approved nor disapproved," and reviews their technical contents only in connection with an application that references them, per FDA's DMF page.

Does a "research use only" label make selling peptides legal?

It does not, because intended use — not the disclaimer — controls. A "drug" includes articles "(other than food) intended to affect the structure or any function of the body of man or other animals" under 21 U.S.C. § 321(g)(1), and 21 U.S.C. § 355(a) bars any new drug from interstate commerce unless "an approval of an application… is effective with respect to such drug."

FDA has said so repeatedly and recently. In an August 24, 2026 warning letter covering eight peptide products plus bacteriostatic water, FDA wrote: "Despite statements on your product labeling marketing your products 'for laboratory, research, and analytical use'… evidence obtained from your website establishes that your products are intended to be drugs," in violation of FD&C Act sections 505(a) and 301(d) — noting that injectable products "bypass some of the body's key defenses against toxins and microorganisms." A June 17, 2026 letter reached the same conclusion where "RESEARCH USE ONLY" labeling was contradicted by the seller's own website claims.

What testing is a legitimate manufacturer required to perform?

Identity, strength, and — for sterile products — sterility, all before release. A drug is deemed adulterated if its manufacturing methods, facilities, or controls "do not conform to… current good manufacturing practice" under 21 U.S.C. § 351(a)(2)(B), and 21 CFR § 210.1 makes parts 210, 211, 213, 225, and 226 "the minimum" cGMP. Three requirements are worth checking against a supplier's paperwork:

- Components. "At least one test shall be conducted to verify the identity of each component," and a supplier's certificate of analysis substitutes for full testing only if the manufacturer still runs its own specific identity test and validates the supplier's results at appropriate intervals — 21 CFR § 211.84. - Finished batches. "For each batch… there shall be appropriate laboratory determination of satisfactory conformance to final specifications… including the identity and strength of each active ingredient, prior to release," and each method's accuracy, sensitivity, specificity, and reproducibility "shall be established and documented" — 21 CFR § 211.165(a), (e). - Sterility claims. Any batch purporting to be sterile or pyrogen-free requires "appropriate laboratory testing to determine conformance to such requirements" — 21 CFR § 211.167(a).

For active ingredients, FDA's ICH Q7 guidance (September 2016) governs API GMP, so that APIs "meet the quality and purity characteristics that they purport, or are represented, to possess." And clinical study has its own gate: a sponsor "shall not begin a clinical investigation… until the investigation is subject to an IND which is in effect," under 21 CFR § 312.20.

Can a compounding pharmacy use what a synthesis company sells?

Only if the bulk substance clears one of three gates. Under 21 U.S.C. § 353a, a 503A compounder may use a bulk drug substance only if it complies with an applicable USP or NF monograph, is a component of an FDA-approved drug where no monograph exists, or appears on FDA's 503A bulks list — and must compound pursuant to "a valid prescription order." A 503B outsourcing facility, per 21 U.S.C. § 353b(d)(4), registers with FDA and "is not required to be a licensed pharmacy."

Either way, the output is not approved: "compounded drugs are not FDA-approved… FDA does not verify the safety, effectiveness or quality of compounded drugs before they are marketed," and 503B-compounded drugs are subject to cGMP while drugs compounded under 503A are not — FDA's compounding Q&A. Several research-market peptides sit in FDA's Category 2, "bulk drug substances that may present significant safety risks" — GHRP-2, GHRP-6, ibutamoren mesylate, ipamorelin acetate, and kisspeptin-10 — per FDA's category listing (updated April 22, 2026). Another group — BPC-157, KPV, TB-500, MOTS-c, emideltide (DSIP), semax, and epitalon — was taken up by FDA's Pharmacy Compounding Advisory Committee at its July 23–24, 2026 meeting. That committee only advises, so a discussion or a vote there is not itself a legal status: what controls is still 21 U.S.C. § 353a(b)(1)(A)(i) — monograph, approved-drug component, or actually on the published bulks list. Check FDA's 503A bulks list and its category listing as they stand the day you order, not a meeting agenda.

Why can't a synthesis company make any peptide you ask for?

Chemistry limits the menu. Even with modern Fmoc SPPS, "it is still challenging to routinely prepare peptides > 40 amino acids in length," because of "poor solvation and β-sheet formation of the growing peptide chain on the solid support," according to *Chemistry – A European Journal* (2025). That paper's own controls make the point concretely: an unprotected synthesis of PrP(106–126) reached only ~4% crude purity, and amyloid-β(32–43) only ~40%. Because crude purity is sequence-dependent rather than guaranteed, the analytical record is the only thing that tells you what is in a given vial. Browse the Cornerstone Peptide catalog, where each item is listed with its identity and analytical documentation, so you can check the paperwork rather than the label.

What changed for generic peptides in 2026?

FDA reset its expectations in July. On July 28, 2026, it published revised draft product-specific guidances covering 17 peptide products — among them glucagon, liraglutide, semaglutide, teriparatide, and tirzepatide — updating recommendations on ANDA submission of "recombinantly, synthetically, or semi-synthetically produced peptides," innate immune response testing, impurity thresholds, higher order structure, and biological activity, and it simultaneously withdrew its May 2021 guidance on ANDAs for certain highly purified synthetic peptides, which "no longer reflects FDA's current scientific thinking," per FDA's announcement. Separately, *Clinical Pharmacology Considerations for Peptide Drug Products* (December 2023) remains a draft marked "Not for implementation" — see FDA's guidance page.

Frequently asked questions

What does a peptide synthesis company actually do? It assembles amino-acid chains chemically, most often by solid-phase peptide synthesis — repeating cycles of Nα-deprotection, activation, and coupling on a polymeric resin — then cleaves, purifies by chromatography, and confirms identity by mass spectrometry, as described in *Beilstein Journal of Organic Chemistry* (2014).

Does "FDA registered" mean FDA approved? No. 21 CFR § 207.77(a) says registration "does not denote approval," and § 207.77(b) makes a contrary representation misleading and a misbranding violation.

Is "research use only" a legal shield? No — the disclaimer alone does not control. Intended use is defined by 21 U.S.C. § 321(g)(1), and FDA rejected the disclaimer as recently as its August 24, 2026 warning letter, citing FD&C Act §§ 505(a) and 301(d).

Is a peptide a drug or a biologic? The line is 40 amino acids. 21 CFR § 600.3(h)(6) defines a protein as an alpha amino acid polymer greater than 40 amino acids, so 40 residues or fewer is regulated as a drug and anything longer is a biological product requiring a license under 42 U.S.C. § 262(a)(1).

Why can't a synthesis company make any peptide on request? Chain length and sequence limit what is practical: peer-reviewed work in *Chemistry – A European Journal* (2025) reports it remains challenging to routinely prepare peptides longer than 40 amino acids by Fmoc SPPS, because of poor solvation and β-sheet aggregation of the growing chain on the resin.

Can a compounding pharmacy use any peptide a synthesis company sells? No. Under 21 U.S.C. § 353a a bulk drug substance must meet a USP/NF monograph, be a component of an FDA-approved drug, or appear on FDA's 503A bulks list — and FDA has placed several research-market peptides in a category of substances that may present significant safety risks.

What must be tested before release? Component identity per 21 CFR § 211.84, batch identity and strength per 21 CFR § 211.165, and sterility for any batch claiming it per 21 CFR § 211.167(a).