Posted on January 16, 2023 at 1:51 PM by Marie Matthews

Blog on FDA updates

*This article was originally published on July 21, 2022. Last May, the FDA published updated guidance on “Investigating Out-of-Specification (OOS) Test Results for Pharmaceutical Production.” The prior guidance is now 16 years old and was issued after the groundbreaking case of US versus Barr Laboratories, Inc., which focused heavily on this issue. In the Barr case, the FDA clarified its thinking in court on the right and wrong way to handle out of specification (OOS) results, and the policy statements that emerged became the first guide.

This publication has all the latest guidance disclaimers, including the caveat that “the contents of this document do not have the force of law…,” which may lead some readers to question, “if it is only guidance, can the FDA legally enforce it?” The answer is found in this portion of the text:  “FDA guidance documents, including this guidance, should be viewed as recommendations, unless specific regulatory or statutory requirements are cited.” This document has numerous mentions of statutory requirements, and as the Barr case adjudicated way back in 1993, the FDA can and will legally enforce them. For example, the guidance mentions Code of Federal Regulations (CFR) /(CGMP regulations) such as:

  • 21 CFR § § 211.160 and 211.165 - Laboratory testing is necessary to confirm that components, containers and closures, in-process materials, and finished products conform to specifications, including stability specifications.
  • 21 CFR 211, subparts I (Laboratory Controls) and J (Records and Reports):  Testing also supports analytical and process validation efforts. These regulations provide for the establishment of scientifically sound and appropriate specifications, standards, and test procedures that are designed to ensure that components, containers and closures, in-process materials, and finished drug products conform to the established standards.
  • 211.165(f) of the CGMP regulations specifies that finished drug products that fail to meet established standards, specifications, or other relevant quality control criteria must be rejected.
  • Active pharmaceutical ingredients (API) regulatory requirements, found in Q7 Good Manufacturing Practice Guidance for Active Pharmaceutical Ingredients (September 2016) aka (ICH Q7), which the FDA has adopted as a regulatory requirement, as reflected in a 2018  FDA Compliance Policy Guide.
  • ICH Q1E Evaluation of Stability Data (2004). 

The new OOS guidance also updates a few terms, such as substituting the term “quality unit” for the old “quality control unit,” and clarifies some important issues, such as scope. For example, the guidance now says that while it still does not cover biological assays, it does apply to “chemistry-based laboratory testing of biotechnology products” that are covered by CDER, which includes monoclonal antibodies, insulin, enzymes, growth factors and many more products.

The new guide helpfully points to other guidance and statutory references like the US Pharmacopeia and ICH, so the footnotes are a "must read." But the most helpful changes are the addition of some new topics, as well as examples and clarifications for others, which are summarized here:

Outlier Tests: Section  IV.C.2.  now reads, “Occasionally, an outlier test may be of some value in understanding how discordant from a data set a result is, but can be used solely in an informational capacity in the course of an investigation to determine the distance of a result from the mean." The guide previously said something similar but suggested the data might be of use in an “auxiliary fashion,” which apparently had been subject to misinterpretation. The new guide now goes into greater detail and provides example data sets of when an outlier examination might be of use (for information only).

In Section V.B., “Concluding the Investigation/ Cautions”  has new subheadings for:

  • “Averaging results from multiple sample preparations” The new guide counsels that firms “err on the side of caution” and treat the average of OOS and acceptable results as OOS overall, and references USP General Notices Section 7.10.
  • “Averaging results from the same final sample preparation” This section points out that these instances should be predefined in the test method and gives the example of an HPLC test method that may “specify both acceptance criteria for variability and that a single reportable result be determined by averaging the peak response from a number of consecutive, replicate injections from the same test vial.”
  • “Borderline results that are within specifications” This section contains language from the previous guide but is now set off with a heading for greater emphasis. The guidance still cautions that a low result could indicate an improperly formulated batch, and that a firm should be careful when making a release or reject decision. The rationale is included only in the footnote – which is that a lot with an unusually low result might drop over time and end up with a failing stability result. The footnote also refers readers to ICH guide Q1E Evaluation of Stability Data (2004).  

While the FDA uses the standard language that this is guidance only and is not binding, it should be noted that alternate approaches should be approached with great care. FDA’s Office of Compliance (OC) has mentioned the concepts in this guidance in many prior warning letters issued by CDER and the FDA field organization (ORA).  The guidance does not represent new requirements, but rather clarification of the FDA’s interpretation of the CFRs and current Good Manufacturing Practices as applied to the laboratory.

While the updated guidance does touch on some subjects badly in need of elucidation, it is important to remember the main themes for which FDA repeatedly sends warning letters, which remain fairly unchanged in this guide:

  • If this initial assessment indicates that errors were made in the analytical method used to arrive at the data, a full-scale OOS investigation should be conducted.
  • OOS test results should not be attributed to analytical error without completing an investigation that clearly establishes a laboratory root cause. 
  • A full-scale investigation (phase 2 investigation) should include a review of production and sampling procedures and will often include additional laboratory testing. Such investigations should be given the highest priority.
  • For contract laboratories, the laboratory should convey its data, findings, and supporting documentation to the manufacturing firm’s quality unit (QU). The manufacturing firm’s QU should then initiate a Phase 2 (full-scale) OOS investigation, whenever no clearly causative laboratory was identified.
  • Among the elements of this phase is evaluation of the impact of OOS result(s) on already distributed batches.

The guidance has additional information on how to document all phases of the investigation and the proper rationales to employ in decision-making. This document, like its predecessor, is a "must read" for anyone working in any way with analytical data and laboratory operations for products regulated by FDA’s Center for Drugs.