A QC laboratory analyst completes an HPLC assay, and the reported result meets specification.
During data review, however, a supervisor notices that the analytical sequence was interrupted and restarted. The instrument event history shows an unexpected fault, and the sequence information does not fully match the analyst’s documentation.
The final result may be within specification, but questions remain about how the analysis was performed and whether the data are reliable.
This illustrates an important feature of pharmaceutical laboratory controls, equipment, samples, and supporting data to understand what occurred and determine whether the results can be relied upon.
The investigator must reconstruct the event, evaluate the scientific evidence, establish what caused the problem where possible, and assess any wider quality implications.
What Belongs in a QC-Lab Investigation?
QC laboratory investigations cover a broad range of events, including:
sample identification, handling, storage, or preparation errors
instrument malfunctions or unexpected system behavior
deviations from approved analytical procedures
incorrect standards, reagents, dilutions, or calculations
unexpected chromatographic or analytical observations
missing, incomplete, or inconsistent laboratory records
data-processing discrepancies or audit-trail concerns
out-of-specification (OOS), out-of-trend (OOT), or atypical results
The terms laboratory investigation and analytical investigation are sometimes used interchangeably, but they can serve different purposes.
A laboratory investigation examines the broader laboratory event and its associated controls. An analytical investigation focuses more specifically on the measurement process, analytical method, data, and scientific explanation for a questionable result.
An OOS investigation follows a more defined regulatory pathway, potentially extending beyond the laboratory into manufacturing, product impact, and batch disposition.
Within the broader Pharmaceutical Investigations & CAPA lifecycle, QC investigations provide the evidence needed to determine whether laboratory-related events affect the reliability of quality decisions.
Establish the Event and Preserve the Evidence
The first step is to establish what happened and what may be affected.
An initial laboratory investigation should identify the analytical procedure, sample or batch, instrument, analyst, date and time, observed discrepancy, and applicable requirement.
It should also document immediate actions.
Depending on the event, these may include restricting equipment use, protecting affected samples or preparations, securing electronic records, notifying laboratory supervision and QA, or holding affected analytical results pending assessment.
Evidence preservation is especially important because some laboratory information may be time-sensitive.
Sample solutions may have limited stability. Instrument conditions may change. Temporary files, analytical sequences, or system settings may be affected by subsequent activity.
The investigation should preserve relevant original data and available preparations in accordance with approved procedures so that potential causes can be examined.
An initial event description should distinguish the observation from its proposed explanation.
For example:
“An unexpected autosampler fault was recorded during the analytical sequence.”
This establishes an event.
“The autosampler fault caused an invalid assay result” is a causal conclusion requiring further evidence.
For practical guidance on documenting these distinctions, see Deviation Writing: Examples & Templates.
Examine the Complete Analytical Record
The original analytical record is central to a QC laboratory investigation.
A reported result or certificate of analysis may summarize the outcome, but it does not necessarily show every condition under which the data were generated.
Depending on the analytical method and event, the investigation may need to examine:
Raw analytical data: Chromatograms, spectra, instrument files, electronic data, worksheets, and original observations.
Analytical sequences: Injection order, sample identification, standards, blanks, system suitability injections, and sequence interruptions.
Calculations and data processing: Integration parameters, processing methods, dilution factors, calculations, and any documented changes.
Sample preparation: Weighing, extraction, dilution, filtration, transfer, storage, and preparation stability.
Standards and reagents: Identity, preparation records, qualification, suitability, and expiration or use periods.
Instrument records: Calibration, qualification, maintenance, system errors, operating conditions, and relevant equipment history.
Method requirements: Approved procedure, system suitability, validated conditions, and applicable acceptance criteria.
The investigator should establish what each evidence source demonstrates and how it relates to the observed problem.
For example, a passing system suitability test provides relevant evidence about system performance. It does not automatically resolve every possible concern about sample preparation, identification, or subsequent instrument behavior.
Evaluate Laboratory Execution and Actual Conditions
An analytical procedure describes how testing should be performed. The investigation must determine how the procedure was actually executed.
This may involve reviewing sample preparation activities, instrument configuration, analyst actions, laboratory conditions, or the sequence of operations.
A method may be scientifically appropriate but difficult to execute consistently under certain conditions.
For example, a preparation step may involve several dilutions, time-sensitive activities, or transfers that create opportunities for error.
Likewise, an instrument may operate within its qualification requirements while exhibiting an intermittent problem that affects a particular analysis.
The investigator should examine whether the method, equipment, procedure, workflow, or operating conditions could explain the event.
This assessment should remain evidence-based.
A complicated procedure does not establish that the analyst made a preparation error. An instrument alarm does not necessarily mean that every result generated during the sequence is invalid.
The investigation must determine the significance of the observed condition.
Interview Analysts and Technical SMEs
Analysts and laboratory SMEs can provide important information about events that are not fully explained by records.
An analyst may describe the sequence of actions taken after an instrument fault. An instrument specialist may identify the meaning of an error code. A method SME may explain whether a particular preparation or operating condition could affect the analytical result.
Interviews should begin with neutral questions.
For example:
“Can you describe what happened when the instrument stopped?”
“Which actions did you take before restarting the sequence?”
“Were any preparations, settings, or analytical conditions changed?”
The investigator should distinguish firsthand observations from recollections, opinions, and proposed explanations.
Relevant statements should be corroborated through original records, electronic data, technical assessments, or other available evidence. The approach is discussed further in Interviewing SMEs for GMP Investigations.
Investigative Testing, Orthogonal Methods, and Scientific Hypotheses
When the original evidence does not fully explain a laboratory event, additional scientific investigation may be useful.
However, investigative testing should answer a defined question.
Suppose an unexpected impurity peak appears during HPLC analysis.
The investigator suspects that the peak may represent interference from another sample component rather than the impurity being measured.
A useful investigation begins with a specific hypothesis:
“An interfering compound may be contributing to the response measured at the impurity retention time.”
The proposed experiment should then be designed to distinguish between the possible explanations.
Investigative approaches may include reviewing the original chromatographic data, examining retained preparations, conducting controlled experiments, or using an appropriate alternative analytical technique.
Where Orthogonal Testing Can Help
Orthogonal testing uses an analytical procedure with different selectivity or, ideally, a different measurement principle to provide additional evidence about a scientific question.
For example, a suitably characterized LC-MS method may help investigate whether an unexpected HPLC peak represents the intended analyte or another component.
The alternative method must be scientifically suitable for the question, with relevant performance characteristics and limitations understood.
Agreement between methods may strengthen confidence in an interpretation. Disagreement may identify a possible analytical limitation requiring further investigation.
Neither outcome automatically establishes the cause of the original event.
An orthogonal result does not, by itself, justify discarding or invalidating an original analytical result.
ICH Q2(R2), Validation of Analytical Procedures, recognizes orthogonal procedure comparisons as an approach to evaluating analytical specificity or selectivity.
The application of orthogonal testing during a GMP investigation must still be scientifically justified, documented, and evaluated within the applicable investigation and result-validity requirements.
Distinguish Investigative Testing From Retesting
Reinjection, data reprocessing, hypothesis testing, retesting, and resampling are different activities.
Reinjection may help investigate instrument behavior using a retained preparation where scientifically appropriate.
Data processing may help evaluate an integration or calculation issue, provided original records and processing changes remain traceable.
Retesting involves additional analytical determinations and may be subject to specific procedural and regulatory requirements, particularly during OOS investigations.
Investigators should not repeatedly test samples until an acceptable result appears.
The detailed distinctions, stopping rules, and limitations of these practices are addressed in OOS Investigations (Phase I and II).
Distinguish Supported Causes from Suspected Causes
One of the most important laboratory investigation decisions is whether the evidence actually establishes a laboratory-related root cause.
A laboratory error may be suspected because a result is unexpected, an instrument fault occurred, or subsequent testing produced different results.
These observations may support further investigation, but they do not automatically establish causation.
For example, an investigator may suspect incomplete sample transfer.
A controlled experiment may demonstrate that incomplete transfer can produce a lower assay result.
That establishes a possible mechanism.
To conclude that incomplete transfer caused the original discrepancy, the investigation needs evidence connecting that mechanism to the actual test.
This distinction becomes particularly important when the laboratory cannot identify a clear cause.
An inconclusive laboratory investigation should document the hypotheses evaluated, evidence obtained, alternatives considered, and remaining uncertainty.
If a formal OOS result is involved and no clearly causative laboratory error is established, the investigation must proceed through the applicable broader OOS process.
The absence of an identified manufacturing cause cannot be used as proof that the original problem was caused by the laboratory.
Assess Data Reliability and Wider Impact
The laboratory investigation must determine what the event means for the reliability of the affected analytical results.
This requires more than identifying why an instrument stopped or why an analyst performed a particular action.
The investigation should establish whether the observed condition could have affected the analytical measurement, reporting, or interpretation.
For example, an instrument malfunction may affect one injection, an entire sequence, or no reported analytical result, depending on the mechanism and supporting evidence.
A preparation error may affect one sample, multiple samples prepared using the same solution, or other testing performed under comparable conditions.
A data-processing deficiency may require assessment of previously reported results generated using the same processing method.
The scope should follow the identified or plausible failure mechanism.
Where electronic data are involved, the review should consider relevant audit trails, original files, processing changes, and records necessary to reconstruct the analytical activities.
A data discrepancy does not automatically establish intentional manipulation. The investigation should evaluate the nature, cause, extent, and significance of the discrepancy.
Practical Example: HPLC Sequence Interruption
Consider a routine QC assay in which an autosampler fault interrupts an HPLC sequence.
The analyst restarts the analysis and obtains results within specification.
During review, the supervisor identifies differences between the reported sequence and the original instrument event history.
The investigation examines the following evidence:
| Evidence Source | Investigation Finding | Significance |
|---|---|---|
| Original chromatograms | Data generated before and after the interruption are retained | Allows evaluation of the affected injections |
| Instrument event history | Autosampler fault recorded during the sequence | Confirms the equipment event |
| Sequence and audit trail | Sequence restart and associated changes are recorded | Supports reconstruction of the analytical activity |
| Analyst interview | Analyst describes restarting the sequence after the fault | Provides context requiring corroboration |
| System suitability | Applicable criteria were met for the relevant analytical runs | Supports assessment of method/system performance |
| Maintenance history | Similar autosampler faults were previously recorded | Raises a possible recurring equipment concern |
Note: All findings are illustrative examples
The investigation must determine which injections were affected, whether the sequence was restarted appropriately, and whether the applicable analytical requirements remained satisfied.
The instrument fault may explain the interruption, but it does not automatically establish that all prior data were invalid.
Similarly, acceptable results after restarting the instrument do not independently establish that the original analytical event was adequately resolved.
If the records support continued use of particular results, the rationale should explain why they remain reliable.
If uncertainty remains, the investigation should identify the affected results and determine what additional assessment or action is required.
CAPA and Laboratory Control Improvement
Once the failure mechanism is sufficiently understood, corrective actions should address the conditions established by the investigation.
Depending on the findings, appropriate actions may involve instrument repair, improved maintenance controls, revised sample-handling procedures, analytical-method improvements, stronger data review, or more reliable system restrictions.
Training may be appropriate when the investigation establishes a specific knowledge or competency deficiency.
An equipment or system-design weakness should not automatically be addressed through analyst retraining.
The investigation should also determine whether the same weakness may affect other instruments, methods, analytical sequences, or previously reported results.
Repeated instrument interventions, unexplained sequence interruptions, or recurring data corrections may indicate a broader control issue requiring trend review.
Regulatory Perspective
Several regulatory expectations are directly relevant to QC laboratory investigations.
21 CFR 211.160 requires scientifically sound laboratory controls and that deviations from established laboratory procedures and controls be recorded and justified.
21 CFR 211.194 requires complete laboratory records, including relevant original analytical data, calculations, methods, results, and review.
21 CFR 211.192 establishes investigation requirements for applicable unexplained discrepancies and specification failures.
FDAs Data Integrity and Compliance with Drug cGMP guidance emphasizes reliable, accurate, and complete data supported by appropriate controls.
The current PIC/S GMP Guide PE 009-18, Chapter 6, addresses QC laboratory practices, analytical records, and investigation of OOS and OOT results.
ICH Q2(R2) and ICH Q14 provide relevant scientific principles for analytical procedure validation and development.
The particular investigation pathway must reflect the event, applicable regulations, approved analytical procedures, and the organization’s quality system.
QA Review Perspective
A QA reviewer evaluating a QC laboratory investigation should be able to answer:
Is the laboratory event clearly described and classified?
Were original analytical data and relevant preparations appropriately preserved?
Were the analytical method, instrument, sample preparation, and actual execution evaluated?
Were analyst and SME statements corroborated where possible?
Were investigative tests scientifically justified and adequately documented?
Does the evidence support the proposed laboratory cause?
Are unresolved uncertainties clearly identified?
Has the reliability of affected results and possible wider impact been assessed?
Are corrective actions aligned with the supported failure mechanism?
The strength of a QC laboratory investigation lies in its ability to reconstruct the analytical event and explain what the available evidence establishes.
A laboratory investigation may identify a clear technical cause, reveal a broader system weakness, or remain inconclusive despite appropriate scientific evaluation.
Each outcome requires a defensible explanation.
The final quality decision should reflect the strength of the evidence, the reliability of the analytical results, and any remaining uncertainty that may affect product or process quality.
Explore more on Investigations & CAPA Excellence
Browse VerethiQ resources on deviation handling, root cause analysis, investigation quality, CAPA design, effectiveness checks, recurrence prevention, and investigation governance.