In detail
Legal basis
Chapter 6 of the EU GMP Guide requires a written procedure for investigating out-of-specification and out-of-trend results. Any OOS or OOT result must be addressed and investigated. For stability testing, Chapter 6.35 states that confirmed OOS results or significant negative trends affecting batches already released on the market must be reported to the competent authority. The impact on those batches must be assessed in line with Chapter 8 and in consultation with the authority.
How an investigation runs
Detailed guidance on the approach is given in the FDA guidance on OOS results (Revision 1, May 2022), which is also widely used as a reference in Europe. It distinguishes two phases:
- Phase I, laboratory investigation: an initial assessment of whether the laboratory data are accurate and a laboratory error has occurred, ideally before the test preparations are discarded.
- Phase II, full-scale investigation: if no laboratory error can be demonstrated, production and sampling are reviewed, including the impact on batches already distributed.
Retesting and averaging
A retest is performed on the same homogeneous sample, following a predefined plan and often by a second analyst who is at least as experienced and qualified. The number of retests must not depend on the results obtained. The FDA describes testing until a passing result is obtained, known as testing into compliance, as unscientific and objectionable under GMP. An original OOS result may only be replaced by the retest result where a laboratory error has been clearly identified. Averaging is critical because it can hide the variability of individual results.
Consequences
A confirmed OOS result usually leads to rejection of the batch. Even then, the investigation remains necessary to establish whether other batches are affected. Corrective and preventive actions are derived from the root cause analysis. Batches that failed to meet specification, together with their investigation, are also part of the annual product quality review under EU GMP Chapter 1.10.
In practice at A&O Pharma
Our GMP-certified laboratory in Itzehoe performs release testing, EU retesting and stability testing, which are exactly the tests in which OOS results can occur. On request, we handle the subsequent processing of deviations and CAPA as part of our operational QM/QA support.
Analytical ServicesFrequently asked questions
What is the difference between OOS and OOT?
OOS means a result lies outside the specification. OOT (out of trend) means a result is within specification but deviates noticeably from previous data. Both must be investigated under EU GMP Chapter 6.
Can you simply retest after an OOS result?
Not at will. Retests must be planned and justified in advance, and their number must not depend on the results. The original result may only be invalidated where a laboratory error has been demonstrated.
Does an OOS result have to be reported to the authority?
Under EU GMP Chapter 6.35, yes, if it is confirmed and affects batches already on the market. The same applies to significant negative trends.
