Regulations and compliance · Practical guide

Aborted GMP analyses: what needs to be resolved before a repeat?

A sequence stops, an injection is interrupted or preparation starts again. Which data already exist? Why was the run interrupted, what else was affected and can the repeat legitimately replace the original attempt? Retain the context needed to assess the results, including what happened before the last successful run.

Table of contents

At a glance

How should aborted analyses be handled?

Preserve and assess the data already generated during an aborted GMP analysis, including the relevant circumstances of the interruption. An abort is neither automatically OOS nor automatically insignificant. A repeat needs a documented purpose and a decision consistent with the method. For chromatographic injections within drug CGMP scope, FDA explicitly addresses recording incomplete and aborted runs.

  • Planned test scope, initiated runs, and generated data must be traceable and consistent.
  • Termination reason, data usability, and repeatability are separate decisions.
  • A technically successful restart does not automatically conclude a professional assessment.

1. Not treat every termination the same way

This operational guide addresses laboratory testing of human medicinal products under GMP. A stoppage may occur prior to measurement, during data entry, or only upon data processing. Accordingly, the data set and potential implications differ. A device malfunction must be evaluated differently from a deliberate stoppage following the identification of an adverse signal.

EU GMP Chapter 6 requires traceable test results and controlled investigations for out-of-specification (OOS) and out-of-trend (OOT) results. The FDA Guidance clarifies chromatographic processes within the US pharmaceutical context. It does not establish a blanket requirement for the retention of all research studies.

Sources: [1] [2] [3]

2. The first decision depends on the timing.

First, ensure that the actually achieved processing stage is secured. An empty results report does not prove that no data were generated. Raw signals, sequence entries, device alerts, or processing histories may already exist and could be required for investigation. The information needed depends on the process and the reason for termination.

The matrix is a standalone reference tool. The procedures are not a universal menu sequence for a specific system. A function labeled "Cancel" does not define the professional meaning of the action.

Self-assessment based on the cutoff time point
SituationFirst, secure and clarifyTechnical Implication
Start prior to data entry rejected Task, Notification, Actual Start Status Verify that relevant records have been generated
Injection data capture terminated Partial raw data, time, user and reason Assess usability and additional impacted areas
Sequence stopped after multiple samples Complete sequence of all previous runs Not only investigate the most recent batch.
Full run evaluation completed Raw data and processing status Breakdown and processing breakdown separate
Restart after correction Correction and reference to the original process Rationale to be justified and jointly reviewed

Sources: [2] [4] [5]

3. Do not lose sight of the beginning because of a successful end

Assign unique identifiers to initiated relevant activities. The linkage between order, interrupted run, corrective action, and reexecution must be preserved. A newly created order must not inadvertently split the history into two apparently independent test runs.

For electronic systems, verify when data are actually captured and stored. The FDA states that chromatographic records do not emerge only at the end of a complete sequence. Therefore, it is essential to determine whether original data remain if a sequence is interrupted prior to the final report and whether such data are accessible during review.

Sources: [2] [5] [4]

4. Example: Six planned, seven initiated activities

A sequence includes two standard and four sample injections, resulting in six planned activities. All six activities are initiated. One sample injection is interrupted due to a documented technical disruption; five activities are completed. Upon evaluation, this sample injection is repeated once and then completed.

Seven initiated processes are in place: six completed and one discontinued. A results report may relate to six used injections; however, the traceability chain must still account for the seventh process. Whether standards and other samples remain methodologically usable after the interruption is an additional determination. Counting alone does not answer this question.

Example: Sequence case comparison matrix
GroupPlannedStartedCompletedAborted
Standards 2 2 2 0
Original samples 4 4 3 1
Additional repetition 0 1 1 0
Total 6 7 6 1

Sources: [1] [2]

5. Investigate root cause and extent

A run may be affected by a local event. However, a shared component such as an autosampler, power supply, or method can impact multiple runs. Review the documentation to determine the actual scope of impact. Neither the blanket rejection of all results nor full confidence in completed runs should be inferred solely from a run termination.

Where recurring patterns are observed, technical and organizational conditions must also be investigated. "User error" is not an adequate explanation without a proper root cause analysis. Immediate corrective actions may be taken prior to full root cause determination; such actions do not replace the appropriate completion of a root cause investigation.

Sources: [6] [1] [4]

6. Start with a clear question when repeating

Before the next run, describe what needs to be clarified or replaced. A trial to verify a technical hypothesis is not automatically equivalent to a new routine test result. If an invalid measurement is to be replaced, it must be consistent with the method, sample condition, and test stage.

Only considering appropriate repetitions is insufficient for a complete evaluation. In the context of Drug CGMP, the FDA cautions against inferring compliance. Also evaluate unfavorable or previously discontinued relevant studies. An interruption must not be used to make previously evident problematic data disappear.

Sources: [3] [2] [1]

7. Operation and technical inspection are interlinked

End-users document observations and immediate corrective actions; equipment or IT personnel investigate technical root causes. Professional assessment determines the impact on the method and results. These responsibilities may overlap in small teams, but must be clearly attributable.

A concise run summary is often more useful than a disorganized collection of error reports. It should display planned and additional activities, status, root cause, and the associated decision. An automated exception summary can only be reliably supported when the data foundation and evaluation logic are appropriately suited to this purpose.

Sources: [5] [4] [7]

8. Checklist for a Closed Breakage History

Another qualified person should be able to review the relevant steps based on the evidence. A new green status alone does not indicate that the original finding was appropriately addressed.

The following internal support document does not require a standalone CAPA for each termination and does not mandate automatic reexecution of all previously completed tests. Adjust the scope and responsibilities in accordance with the meaning in the process.

  • Planning is compared with all the relevant processes started.
  • Original and partial data together with necessary context securely preserved.
  • Record when the run stopped, what was observed, who was involved and the reason.
  • Assessed for potential impacts on other runs and samples.
  • Purpose and scope of the repetition are justified.
  • Original run, correction, new cycle, and decision linked.
  • Remaining uncertainty regarding the usage decision is considered.

Sources: [6] [1] [2]

Project experience and collaboration

Include the aborted run in the scientific review

With laboratory management, QA and the system team, we examine an aborted analysis using existing or specially prepared test records. We trace the raw data, reason, status changes and any repeat. A technical interruption warrants a different investigation from a discarded run that already contains interpretable data. Relevant information must be retained and assessed under the appropriate procedure. We connect laboratory practice with system controls and identify gaps in the historical evidence as such.

Frequently asked questions about Aborted analyses

Is an interrupted injection automatically considered out of specification?

No. An out-of-specification (OOS) result is defined as a finding outside predefined specifications or acceptance criteria. An OOS may occur in the case of a study termination, but it is not necessarily present. Document the actual situation and select the appropriate investigation procedure.

Sources: [3] [1]

Must partial raw data also be retained?

To the extent that they constitute required GMP records and are relevant for reconstruction or evaluation, yes. The FDA explicitly references incomplete and interrupted chromatographic injections. The absence of a final report does not justify treating already generated relevant data as non-existent.

Sources: [2]

Can we immediately restart after a device error?

This depends on the specific situation and procedure. Secure the finding and verify whether the error has been resolved and whether the restart serves a defined purpose. The restart must not override original data or the rationale for repetition.

Sources: [1] [5]

Does every shutdown require its own CAPA?

No. Investigations and corrective actions are determined by significance, root cause, and recurrence. Recurrent or systemic issues may necessitate more extensive corrective actions. A brief, justified technical fix and a comprehensive CAPA process are not interchangeable for all situations.

Sources: [6]

What is particularly important regarding manually stopped runs?

Assignment to the responsible user and justification for the stop. Check which data were previously visible and whether these continue to be evaluated. A manual stop following an adverse signal must not result in the disappearance of the finding.

Sources: [5] [2]

Are all previously completed results invalid?

Not automatically. Assess whether the cause or interruption affects the suitability. Common settings or methodological sequence conditions may be relevant. The specific relationship determines the issue—not merely the position of a run before or after the interruption.

Sources: [1] [3]

How does an interrupted measurement differ from interrupted data processing?

In the event of a data acquisition interruption, the raw data capture may be incomplete. In the case of an analysis interruption, complete raw data may be available, while the processing has not yet been finalized. These initial conditions determine whether a new measurement is required or whether controlled processing is sufficient.

Sources: [2] [4]

Does a fault code constitute sufficient justification?

An error code may serve as a potential original record. However, it does not necessarily indicate the root cause, extent, or data usability. It should be correlated with the relevant process and the associated assessment. An unknown code should not be replaced with the generic note "technical error, recurring."

Sources: [1] [4]

What numbers must our run summary display?

The process must clearly depict: planned, started, completed, discontinued, and any additional relevant activities. Deviations in the numbers may be justified. What is essential is a transparent explanation, not mandatory alignment with the original plan.

Sources: [2] [7]

Must each system retain all error messages indefinitely?

No. The scope and duration shall be determined based on the required data and evidence. Not every technical notification is critical; conversely, a relevant interruption notice must not be omitted merely because of its technical origin. The appropriate scope and retention period shall be clearly justified.

Sources: [4] [5]

Sources and review date

Source date: 7 October 2026. The technical examples relate to human medicinal product GMP. FDA statements are contextualized within the applicable U.S. medicinal product framework. Tables and numerical examples are intended as internal work aids.

  1. European Commission: EU GMP Chapter 6, Quality ControlRevision 2014; effective from 01.10.2014; §6.7 Documentation; §6.9 Out-of-Specification/Out-of-Trend; §§6.15–6.17 Test methods, calculations, and laboratory records
  2. FDA: Data Integrity and Compliance With Drug CGMP, Questions and AnswersFinal Guidance, December 2018; Section III Q1(b)–(d), Q2, Q4–6, Q7–10, Q12–13, and Q18
  3. FDA: Investigating Out-of-Specification Test Results for Pharmaceutical ProductionFinal guidance, Level 2 revision, May 2022. Section I defines OOS. Sections III–IV address investigations, IV.B retesting, IV.C averaging and outlier tests, and V the conclusion.
  4. PIC/S: PI 041-1, Data Management and IntegrityPIC/S PI 041-1, 1 July 2021. See section 3.7 for status, sections 7.5–7.8 and 8.1–8.8, sections 9.1.5 and 9.3 and 9.5–9.10, section 12.1 and the glossary.
  5. European Commission: EU GMP Annex 11, Computerised SystemsRevision January 2011; from 30.06.2011; Principle; §§1,4,5,7,9–13,17; §12.4 Identity, Date and Time; §10 controlled changes
  6. European Commission: EU GMP Chapter 1, Pharmaceutical Quality SystemRevision 2013; effective from 31.01.2013; §1.4(xii)–(xiv): proactive change evaluation, post-evaluation, and root cause analysis/CAPA; §1.5 Management
  7. European Commission: EU GMP Chapter 4, DocumentationJanuary 2011; in effect since 30.06.2011; Principle; §§4.1–4.9, in particular timely recording 4.8 and visible correction 4.9; §§4.10–4.12 Storage

Sources checked: .

Daniel Herrmann, Daniel Herrmann Consulting

Daniel Herrmann Consulting

Investigate and document out-of-specification results and repeat analyses traceably.

We help you integrate run histories, technical findings, and technical decisions into a practical laboratory process.

  • Computer system validation →

    Bring your GxP systems into validated operation, from laboratory software, MES and eQMS to ERP. We define the scope, plan risk-based tests using GAMP 5 and hand the evidence to QA for release.

Send a project enquiry Customer testimonials →