Section 7 of 10 — Testing and labeling

"Research Use Only" Labeling and Release Testing Expectations

What "Research Use Only" (RUO) actually means as a legal matter, and what CBER guidance and USP standards actually require before an investigational regenerative product reaches a human being — with a clear correction of the mycoplasma/mycotoxin confusion.

Educational information only — not medical or legal advice. This page is for general public education about FDA regulation of regenerative medicine. It does not diagnose, treat, or recommend treatment for any condition, and it is not a substitute for consulting a licensed physician about a medical question or a licensed attorney about a legal question. See the full disclaimer.

"Research Use Only" is a label, not a legal safe harbor

FDA's own enforcement record makes this explicit. In a Warning Letter to Kimera Labs, Inc. (MARCS-CMS 649343, issued September 1, 2023), FDA documented that components labeled "For research use only" were nonetheless "incorporated into products that are injected into/used by humans" — cited as a violation of 21 CFR § 211.84(a) (failure to release components after adequate testing). The letter also confirms the broader legal point: because Kimera's exosome and amniotic-fluid products are intended to treat disease in humans, they meet the definitions of a drug (21 U.S.C. § 321(g)) and a biological product (42 U.S.C. § 262(i)), and require either a valid biologics license or an effective IND before any clinical distribution — and even with an IND in place, 21 CFR § 312.7(b) separately prohibits commercial distribution or test-marketing of an investigational drug. FDA had previously issued an Untitled Letter to the same company in April 2020 on the same underlying issue.

Bottom line: whether a product is legally an investigational drug/biologic turns on its intended use — including promotional claims — not on the label affixed to the container. An "RUO" label does not exempt a product from drug/biologic regulation once it is marketed or used for human clinical treatment.

What CBER requires before any investigational biologic reaches a human

A biological product without an approved BLA may be administered to humans only under an effective IND (21 U.S.C. § 355(i); 42 U.S.C. § 262(a)(3); 21 CFR Part 312). Under 21 CFR § 312.23, an IND must include nonclinical pharmacology/toxicology data (generally under Good Laboratory Practice, 21 CFR Part 58) sufficient to support the safety of the proposed initial human trial, Chemistry, Manufacturing, and Controls (CMC) information appropriate to the phase of investigation, and a clinical protocol. Even with an active IND, 21 CFR § 312.7(b) bars commercial distribution — an IND authorizes investigation, not sale.

Release testing standards actually referenced by FDA/USP (verified figures only)

Test	Standard	What it establishes
Sterility	21 CFR § 610.12; USP <71>	Validated method to detect viable contaminating microorganisms in each lot.
Bacterial endotoxin	USP <85>	Endotoxin quantity in USP Endotoxin Units. No universal pass/fail number exists — the limit is calculated per the USP <85> formula based on maximum human dose, body weight, and route of administration, and must be set per product.
Mycoplasma	USP <63>; governed today under 21 CFR § 610.9 (equivalent methods)	Detects mycoplasma contamination in cell-culture-derived products. Standard 28-day agar/broth culture, or a validated nucleic-acid-based alternative accepted by FDA.
Identity / purity / potency	21 CFR §§ 610.10, 610.13, 610.14	Product-specific potency assay; freedom from extraneous material; a test specific enough to distinguish the product from any other product made in the same facility.

Correction to a commonly cited regulation. 21 CFR § 610.30 ("Test for Mycoplasma") was revoked by FDA effective August 21, 2020 (85 FR 51383). Mycoplasma testing today is handled under the general "equivalent methods" provision at 21 CFR § 610.9, referencing USP <63>. Do not cite § 610.30 as current law.

Mycoplasma vs. mycotoxin — two different things, only one is a standard release test

Mycoplasma is a genus of cell-wall-deficient bacteria and a well-recognized cell-culture contamination risk. Mycoplasma testing (USP <63>) is a standard, expected release/in-process test for cell-derived biologics like MSC and exosome products.

Mycotoxin is an entirely different thing — a toxic metabolite produced by fungi (e.g., aflatoxins). Mycotoxin testing is addressed in USP General Chapter <561> and applies to plant/botanical-origin raw materials and dietary-supplement ingredients — it is not a standard CBER release specification for cell- or exosome-derived biologic drug products, absent a specific fungal-derived or botanical raw material (such as certain culture-media components) that would independently trigger a mycotoxin/aflatoxin test for that raw material. This is a genuinely common point of confusion, worth stating plainly to the public: "batch-by-batch mycotoxin testing" is not, by itself, an accurate description of what CBER requires for these products in general.

"Predicate testing" / comparability — a plain-language note

The phrase "predicate testing" is FDA medical-device terminology (510(k) predicate device comparison) — it is not the standard CBER term for biologics. For biologics, the relevant concept is comparability: when a manufacturing process changes (raw material, site, scale-up), the manufacturer must demonstrate the changed product remains comparable in quality, safety, and efficacy, per the internationally harmonized standard ICH Q5E (adopted by FDA, June 2005). This is standard regulatory practice applied by extension to cell- and exosome-derived biologics; no CBER guidance was found that names exosome products specifically in connection with ICH Q5E.