What is the difference between OOS, OOT and OOE results?

1What each term means.

FDA’s guidance on out-of-specification results, Revision 1 of May 2022, defines OOS results as “all test results that fall outside the specifications or acceptance criteria established in drug applications, drug master files (DMFs), official compendia, or by the manufacturer”. The term also covers in-process laboratory tests outside established specifications.¹ The guidance does not define out of trend, but says much of it “may be useful for examining results that are out of trend”.¹

MHRA’s guidance, Out of Specification and Out of Trend Investigations (October 2017), names three kinds of result.²

• An OOS result does not comply with predetermined acceptance criteria, such as those in filed applications, drug master files, approved marketing submissions, official compendia or internal acceptance criteria.
• An OOT result “is generally a stability result that does not follow the expected trend”, compared with other stability batches or with earlier results in the same study. Starting materials and in-process samples can be out of trend too, and MHRA says the idea should be considered for environmental trend analysis of viable and non-viable data. The result “is not necessarily OOS but does not look like a typical data point”.
• An atypical, aberrant or anomalous result is “still within specification” but “unexpected, questionable, irregular, deviant or abnormal”, such as a chromatogram with unexpected peaks or an unexpected stability result.

Neither FDA’s nor MHRA’s guidance uses the term out of expectation (OOE). The result it describes, one that meets specification but is not what was expected, is MHRA’s atypical result.²

2How they compare.

The assay example is FDA’s.¹ The detection methods for OOT are MHRA’s: it says it is advantageous to use “objective (often statistical) methods” and notes that “historical data are needed to identify OOT alerts”.²

3What the rules require for each.

In the US, 21 CFR 211.192 requires that “the failure of a batch or any of its components to meet any of its specifications shall be thoroughly investigated, whether or not the batch has already been distributed”. The investigation extends to other batches and products that may be associated with the failure, and its written record includes the conclusions and follow-up.³ Products failing their specifications are rejected under 211.165(f).⁴ FDA’s guidance adds that “batch rejection does not negate the need to perform the investigation”.¹

21 CFR Part 211 does not use the terms out of trend or out of expectation.⁵ FDA’s guidance does warn that “an assay result that is low, but within specifications, should also raise a concern”, because the batch may not have been formulated properly, and that caution should be used in the release decision.¹

In the EU, Chapter 6 of the GMP guide, in operation since 1 October 2014, lists “a procedure for the investigation of Out of Specification and Out Of Trend results” among the documents the quality control department should have readily available (6.7). Test results, yields and environmental controls should be recorded so they can be trended, and “any out of trend or out of specification data should be addressed and subject to investigation” (6.9). Results for critical parameters should be trended and checked for consistency (6.16).⁶

For the on-going stability programme, section 6.35 says “out of specification or significant atypical trends should be investigated”. A confirmed OOS result or significant negative trend affecting batches already on the market should be reported to the relevant competent authorities.⁶

MHRA expects OOS, OOT and atypical results to be investigated in batch release testing and starting-material testing, in in-process control testing if the data are used for batch calculations or decisions and appear in a dossier and on certificates of analysis, in stability studies on marketed batches, and for clinical trial batches. Its OOS process does not apply to in-process tests run to reach a manufacturing end point, such as adjusting pH or viscosity.² FDA’s guidance likewise leaves out in-process tests done solely to adjust equipment in real time.¹

4How a trend signal escalates.

MHRA sorts OOT stability alerts into three levels: analytical, process control and compliance. The depth of investigation should increase from one to the next.²

• An analytical alert is a single result that is aberrant but within specification, outside normal analytical or sampling variation and normal change over time.
• A compliance alert is an OOT result that suggests OOS results are likely before the expiry date, in the same study or others on the product.

If a result at any stability interval predicts an OOS before the next interval, MHRA says to schedule extra testing before that next interval. Stability OOS and OOT situations should be escalated as soon as the suspect result is found, and a stability OOS should link to the product recall procedure.²

For monitoring against warning (alert) levels, MHRA says that when two or more consecutive samples exceed the alert level, or counts rise or the same organisms recur over a short period, consideration should be given to treating the results as action level excursions.²

5How it runs in Seal.

In Seal, OOS and OOT investigations record OOS, OOT, atypical and invalid as different states, each with its triggering rule and the specification and method versions in effect, and a later reclassification keeps its rationale. Trend rules are versioned with their product, test, population and effective dates, so a signal does not come from mixing incompatible cohorts. The first reportable value stays permanent with its raw data, and no later passing result replaces it.

Neil can compare an out-of-trend assay with its history: it sets an in-specification but unusual result against a defined historical population and drafts the trend investigation plan, and the analytical owner decides whether the difference is significant. The pharmaceutical QC blueprint shows how an investigation opens with the sample, method and instrument context already linked.