SOP: Result Invalidation for Cell and Gene Therapy Potency and Release Assays

This is a ready-to-use SOP for the highest-risk decision in a cell and gene therapy laboratory: whether a failing or atypical potency, viability, or identity result may be invalidated. Biology-driven assays are variable, so the temptation to explain away a result is constant; this procedure removes the judgment from the moment of the result and puts it into a pre-approved framework. Replace every <<FILL: ...>> placeholder with your own specifics, set your document numbers and dates, and route it through your normal document control, review, and approval. A worked filled specimen follows the template. Verify each cited regulation against the current source before you rely on it. This is general guidance to adapt, not legal or regulatory advice.

Document control header

1. Purpose

This procedure defines when and how a result from a <<FILL: COMPANY NAME>> cell or gene therapy potency, viability, identity, or other release-relevant assay may be invalidated. Its object is to ensure that a result is set aside only for a documented, pre-approved, and independently evidenced reason, never because the assay is variable or the number is inconvenient, and that the original result and its raw data always survive the decision.

2. Scope

This procedure applies to biology-driven analytical results supporting lot disposition of cell and gene therapy products, including cell-based potency assays, flow cytometry phenotype and viability, quantitative PCR / droplet digital PCR for vector genome copies and transduction, and functional or reporter-gene assays. It applies to in-process results used in a potency assurance strategy as well as final release results. It does not replace the OOS investigation SOP; invalidation is a possible outcome of, not an alternative to, an OOS investigation.

3. Responsibilities

4. Definitions

• Assignable cause: a documented, independently evidenced reason the result does not reflect the true quality of the material (for example a confirmed instrument fault, a reagent lot that failed its own qualification, a confirmed incubator excursion). High assay variability is not an assignable cause.
• Invalidation: a formally recorded decision that a result will not be used for disposition, made against a named assignable cause. The result and its raw data are retained and remain visible; invalidation is not deletion.
• Original result: the first result the assay produced, with its complete raw data (acquisition files, gating templates, amplification curves, instrument metadata), preserved unchanged.
• System suitability: the pre-set acceptance gates (reference-standard recovery, replicate precision, control populations) that qualify a run as valid before its sample results are read.

5. Procedure

5.1 Before anything else, protect the original

1. On any failing, atypical, or suspect result, stop. Do not re-gate, re-integrate, re-run, or reprocess.
2. Confirm the original result and all raw data are saved and cannot be overwritten, with the acquisition time and analyst captured in the audit trail.
3. Notify the supervisor and open the record required by section 8.

5.2 Determine whether an assignable cause exists

1. Check the result against the enumerated assignable-cause list in Attachment A (section 9). Only causes on that list may support invalidation.
2. For the claimed cause, obtain the independent evidence the list requires. Evidence must exist outside the analyst’s account: an instrument maintenance record, an alarm or excursion log, a calibration certificate, a reagent-lot qualification failure, or a second analyst’s confirmation.
3. If no cause on the list applies, or the required evidence does not exist, the result stands and drives an OOS investigation. Record that outcome. Do not invalidate.

5.3 Invalidate correctly, if justified

1. The method owner or SME confirms the cause is scientifically valid for this assay.
2. The second reviewer independently confirms the original data is retained, the cause is on the list, and the evidence is real and independent.
3. Record the invalidation contemporaneously in the audit trail and on the record: who invalidated, when, why, and against which enumerated criterion, with the evidence referenced.
4. Retain the original result and raw data for the full retention period. Both the invalidated original and any valid repeat remain visible to the reviewer.
5. QA approves the invalidation before any repeat result is used for disposition.

5.4 Repeat testing and disposition

1. A repeat may only follow a valid invalidation or a pre-approved OOS retest rule, never a bare “the number looked wrong.”
2. The reviewer sees both the original and the repeat. Disposition follows the pre-approved data-review rule, not selection of the more favourable number.
3. Do not release or use affected product until the invalidation and any resulting investigation are closed.

5.5 Trend the invalidation rate

1. QC trends the invalidation rate per assay against the rate established during method validation.
2. An invalidation rate above the defined trigger routes to the method owner as a method-robustness signal, not to a quiet repeat-until-pass loop.

6. Acceptance criteria

An invalidation is acceptable only when all of the following are true:

• The original result and its raw data are retained and remain visible.
• The reason is a cause enumerated in Attachment A, supported by an independent record that existed before the result was known.
• The invalidation is recorded contemporaneously with who, when, why, and against which criterion, and is independently reviewed and QA-approved.
• No affected product was released before the decision closed.
• The invalidation is counted toward the trended rate, and the rate remains inside the validated range.

7. References

21 CFR 211.192 (production record review and investigations), 211.194 (completeness of laboratory records). 21 CFR Part 11 (electronic records; audit trails and original records). EU GMP Annex 11 (computerised systems) and Annex 1 for sterile advanced therapies. FDA guidance, Investigating Out-of-Specification Test Results (2006, rev. 2022). FDA guidance, Data Integrity and Compliance With Drug CGMP (2018). FDA guidance, Potency Tests for Cellular and Gene Therapy Products (2011), and the draft Potency Assurance for Cellular and Gene Therapy Products (2023) when finalized. PIC/S PI 041, Good Practices for Data Management and Integrity.

Confirm the current version and clause numbers of each reference before issue.

8. Record generated: result invalidation record

9. Attachment A: enumerated assignable-cause list (example, adapt to your assays)

10. Revision history

11. Approvals

Filled specimen

The following shows the record completed for a flow cytometry viability result on an autologous lot. The company, IDs, and numbers are illustrative; replace them with your own.

In this specimen the analyst believed a debris cluster skewed the gate, but no assignable cause on the list applied and no independent evidence supported one, so the result was not invalidated. The original file was retained, the lot was held, and an OOS investigation followed. That is the outcome the procedure is built to force: a variable-assay result you dislike is not, by itself, a reason to make it disappear.

Common inspection findings this SOP prevents

• A failing potency or viability result re-run until it passes, with no retained original (repeat-to-pass).
• Flow cytometry or PCR data re-gated or reprocessed with the original overwritten.
• An invalidation justified only by “high assay variability” with no assignable cause and no evidence.
• The analyst who ran the assay being the only person who decided to invalidate it.
• Invalidation rates never trended, so a systemic method-robustness problem hides inside individual “one-off” invalidations.

How to adapt this SOP

1. Set your document number, owner, and effective date in the header.
2. Rewrite Attachment A for your specific assays: the real, enumerated assignable causes and the exact independent evidence each one requires.
3. Point the cross-references to your real OOS, deviation, and data-review procedures.
4. Define your invalidation-rate trigger from your own validation data, and name where the trend is reviewed.
5. Confirm every regulation in section 7 against the current published version before issue.