USP Revision Watch — Sterility and Endotoxin Testing Updates

Published 2026-07-21 · Research Chem Today Editorial

USP <71> and <85> Revisions Tighten Documentation for Research Chemical Suppliers

The comment period for USP <71> (sterility testing) closed March 2025. USP <85> (endotoxin testing) remains under review. Both revisions shift documentation requirements for suppliers shipping sterile or low-endotoxin products labeled for research use only. Expect tighter expectations on method suitability, sample size, and endotoxin limit calculations.

USP <71> Sterility Testing Changes

Current USP <71> requires testing a statistically defined sample from each sterilized batch via membrane filtration or direct inoculation. The proposed revision does two things: it tightens sample size requirements for aseptically filled products, and it adds explicit language for products labeled "research use only" that claim sterility.

Key change: suppliers must document the rationale for sample size selection based on sterilization method and container configuration. The revision also requires a validated neutralization step for bacteriostatic or fungistatic products — a common failure point in FDA warning letters. Suppliers relying on parametric release without concurrent sterility testing must now demonstrate equivalency.

USP <85> Endotoxin Testing Documentation

USP <85> specifies the Limulus Amebocyte Lysate (LAL) test. The pending revision mandates an endotoxin limit calculation based on the maximum human dose, even for research-use-only products. Many suppliers previously used a default 0.5 EU/mL without justification. That stops.

The revision requires a documented formula: Endotoxin Limit = K / M, where K is 5.0 EU/kg/hour for parenterals and M is the maximum human dose per hour. Research chemical suppliers must either declare a maximum human dose for limit calculation or state "no established limit" — but then cannot claim "low endotoxin" on the Certificate of Analysis (COA). The revision also adds spike recovery verification at the product's final dilution, not just the lysate reconstitution volume.

COA Accuracy Impact

| Documentation Element | Current Practice | Post-Revision Requirement | |----------------------|------------------|---------------------------| | Sample size rationale | Often omitted | Must state sterilization method and justify n | | Neutralization validation | Optional | Required if product is bacteriostatic/fungistatic | | Endotoxin limit source | Default 0.5 EU/mL | Must show calculation or state "no limit" | | Spike recovery verification | At lysate reconstitution | At final product dilution | | Parametric release | Allowed without sterility test | Requires equivalency data |

A COA that simply lists "Sterile" and "<0.5 EU/mL" no longer meets USP standards. Each claim must trace to a specific test method, sample size, and limit calculation. FDA cited at least three research chemical suppliers in 2024 warning letters for COAs lacking this traceability — letters referencing 21 CFR 211.165 and 211.194 for incomplete testing records.

Research Use Only Labeling

The proposed <71> revision explicitly states that products labeled "research use only" but claiming sterility or low endotoxin must comply with the test method and documentation requirements. This closes a loophole: some suppliers used the RUO label to avoid full documentation while marketing to researchers requiring sterile products.

The revision does not require clinical-grade testing for RUO products. But any sterility or endotoxin claim on the label or COA must be supported by documented testing per <71> and <85>. Suppliers stating "Not for human use" without testing data are unaffected. Those claiming "Sterile filtered" or "Low endotoxin" must now provide the same documentation as clinical suppliers.

Common Failure Modes in Sterility Testing

Three failure modes dominate FDA warning letters for research chemical suppliers.

First, inadequate sample size. Many test only one unit per batch. The revision explicitly rejects this for aseptically filled products.

Second, failure to validate neutralization for bacteriostatic products. Benzyl alcohol, phenol, and parabens inhibit microbial growth and require validated neutralization.

Third, parametric release without concurrent sterility testing. The revision requires at least one sterility test per production run for products claiming sterility. It also requires environmental monitoring data during aseptic filling — data many research chemical suppliers lack.

Endotoxin Testing Pitfalls

The revised <85> chapter highlights three specific pitfalls.

Product interference. Many research chemicals — particularly peptides and proteins — can inhibit or enhance the LAL reaction. The revision requires interference testing at the product's final dilution, not just the lysate reconstitution volume.

Endotoxin limit miscalculation. Using 0.5 EU/mL as a default without calculating K/M is no longer acceptable.

Spike recovery at wrong dilution. The revision requires spike recovery at the product's final dilution, which for many research chemicals is higher than the lysate dilution. Suppliers using a single dilution for both product and lysate must revise their protocols.

Quality System Updates

Suppliers should update three documents: the sterility testing SOP, the endotoxin testing SOP, and the COA template.

The sterility testing SOP must include sample size rationale based on sterilization method and container configuration. The endotoxin testing SOP must include the K/M calculation or a statement that no limit is established. The COA template must include test method reference, sample size, and limit calculation. Suppliers outsourcing testing should request a revised Certificate of Testing that includes these elements.

BWD, a supplier of sterile diluents, publishes per-lot COAs with sample size and endotoxin limit calculation — aligning with the revised chapters. Alpha Amino USA, a domestic peptide supplier, provides COAs with documented interference testing and spike recovery at final dilution. Bachem and Genscript have similarly updated documentation.

The revision takes effect 12 months after publication in the USP–NF. Early adoption reduces compliance risk.

The bottom line: USP <71> and <85> revisions require research chemical suppliers to document sterility and endotoxin claims with the same rigor as clinical products. COAs lacking sample size rationale, neutralization validation, or endotoxin limit calculation will not survive regulatory scrutiny. Suppliers updating quality systems now — including testing protocols, COA templates, and vendor qualification — will avoid the warning letters that inevitably follow publication.