What Passing System Suitability Actually Confirms
Consider an illustrative commercial quality control (QC) scenario. A high-performance liquid chromatography (HPLC) sequence passes its opening system suitability injections, and its closing standard checks meet the approved method's limits. Yet a finished-product assay result is outside specification, duplicate sample preparations disagree beyond their variability criterion, or an unexpected shoulder appears on the analyte peak. Does the passing SST settle whether that sample result is usable or whether the batch can be released?
No. Passing SST provides evidence about the system against the checks performed; it does not establish that each sample was prepared correctly or that the batch conforms. Section IV.H of FDA's July 2015 Analytical Procedures and Methods Validation guidance describes SST as checks of the integrated equipment, electronics, analytical operations, and controls at the time of use. WHO Good Chromatography Practices (2020), section 11.10, recommends including SST in the sample set and meeting predefined criteria before sample injection and throughout analysis.
The practical inference is that demonstrating system suitability is necessary for accepting affected chromatographic data, but it is not sufficient for accepting a sample result. SST also does not replace instrument qualification or method validation: WHO sections 4 and 12 address those activities separately, while WHO pharmaceutical QC laboratory guidance (2024), section 6.27, describes suitability checks before and during analysis. A pass answers only the questions covered by the applicable checks and acceptance criteria.
Three Separate Gates: Sequence, Sample, and Batch
A sequence pass and a sample result answer different questions. Standard injections assess defined aspects of system performance. A sample assay also depends on extraction, dilution, solution identity, chromatographic selectivity, integration, calculations, and the method's replicate rules. Satisfactory reference-standard responses do not demonstrate that every one of those sample-specific steps was correct. Conversely, a sample OOS result can be analytically valid and still indicate a quality concern requiring investigation.
The following evidence map synthesizes FDA OOS recommendations, WHO chromatography guidance, and CGMP laboratory-record requirements. The gate names are working labels for reviewers, not an agency-prescribed release algorithm. Read each row as a separate question supported by its own records.
| Decision and authority | Question to resolve | Evidence to examine | Meaning of a pass or failure |
|---|---|---|---|
| Sequence acceptance — approved procedure and applicable compendial method; FDA 2015 guidance and WHO 2020 guidance. | Were system-performance and interference criteria met before samples and across the relevant analysis period? | Standard preparations and SST results; required blanks; method-defined bracketing checks; injection sequence; instrument events and relevant audit trails. | Passing checks support use of data for the covered interval, subject to sample review. If standards indicate malfunction, identify suspect-period data and withhold them from use while investigating. A pass does not establish correct sample preparation. |
| Sample result validity — written approved method, FDA 2022 OOS guidance, and WHO chromatography guidance. | Was the method executed correctly, with acceptable sample-level variability and chromatographic processing? | Preparation records, weights, dilutions, individual replicate results, method variability limits, chromatograms, integration history, and calculations. | An acceptable method execution can yield a valid OOS result. Failed variability criteria prevent using the set as the method-defined result; they do not permit erasing individual values or skipping a causative investigation. |
| Batch disposition — quality control unit review under 21 CFR 211.192, complete laboratory records under 211.194, and FDA OOS guidance. | Does the complete production and control record, including investigations, support the quality unit's release or rejection decision? | Original and reprocessed data, worksheets, specification comparisons, second-person review, deviations, OOS investigations, and quality unit assessment. | An SST pass or a single passing assay cannot authorize release. Unexplained discrepancies and specification failures require investigation. Confirmed batch nonconformance and an inconclusive initial OOS require the distinct disposition assessments described by FDA. |
Sequence Acceptance: What the Run-Level Evidence Must Show
Gate 1 answers whether the chromatograph, column, mobile phase, detector, and integration environment remained fit for purpose across the entire time span in which samples were analyzed. Under Section IV.G of FDA's 2015 Analytical Procedures and Methods Validation guidance, the approved analytical procedure itself should define the scan or injection sequence—specifying the placement of blanks, placebos, samples, controls, sensitivity solutions, and standards—so that system suitability validity is maintained across the span of analysis.
For sequence design, the reviewer needs the controlled method and applicable compendial requirements in force at testing. WHO chromatography guidance, sections 11.10–12.1, ties SST to the relevant monograph or validated in-house procedure and identifies parameters such as selectivity, sensitivity, symmetry, repeatability, and integration conditions. Those parameter names are not universal numeric limits. Use them to locate the actual acceptance criteria and confirm which apply to the assay or impurity procedure being executed.
Identify the controlling requirements. Record the procedure version and applicable monograph or specification, rather than copying a generic SST checklist into the sequence. WHO section 11.10 links execution to those controlled documents.
Separate standard repeatability from sample variability. Standard-injection agreement addresses a system-performance question. Sample preparations or injections may have a different number of replicates and different variability limits under FDA's reportable-result discussion. Satisfying one criterion does not satisfy the other.
Record the solutions and checks actually required. FDA sections IV.F–H describe standard/control preparation, the injection sequence, and SST. Where appropriate, include sensitivity solutions, blanks, placebos, and standards as specified by the procedure.
Keep thresholds attached to their method and purpose. Sensitivity, resolution, and symmetry checks should use the approved criteria for the application. Record the numerical limit and governing reference alongside the result so that the reviewer can assess the correct criterion; a parameter name by itself is insufficient.
Ongoing controls matter as well as opening injections. WHO sections 11.7–11.13 recommend verifying vial identity and sequence setup before starting, establishing SST and peak-shape criteria, and including bracketing standards at defined intervals where appropriate. Detected blank interference should meet predefined limits. Section 6 calls for documented audit-trail review, including sample sequences or sets. Neither this guidance nor the evidence map supplies a universal closing-bracket frequency; the method and justified sequence design determine the checks required.
Two practices deserve particular scrutiny because a final SST summary can conceal their history:
Exceptional injections outside the defined sample set: WHO section 11.9 discourages trial or system-check injections outside the sample set. Where exceptionally used, they should follow authorized procedures with the approach justified and electronic records retained alongside the analytical sample set. Do not treat an injection as disposable because it preceded the routine sequence.
Repeated adjustments to obtain a passing SST: WHO section 12.6 allows adjustments within specified limits and written procedures. FDA 2015 section VIII recommends reevaluation, revalidation, or amendment, as appropriate, when an analytical procedure meets SST only through repeated operating-condition adjustments. The fact that the final adjustment produced a pass does not remove the need to review the procedure's performance.
| Sequence control | Evidence question | Retention and failure response |
|---|---|---|
| Opening SST | Do the required checks meet the applicable procedure and compendial criteria before samples are injected? | Retain individual standard injections, calculations, acceptance criteria, and their link to the worksheet. Do not substitute the summary pass flag for the underlying records. |
| Blank and placebo checks, where specified | Are relevant interferences controlled within predefined limits? | Record blank values and maintain traceable graphical data, as described in WHO 2024 sections 6.37–6.38. Assess any interference under the approved procedure. |
| Bracketing standards, where appropriate | Do the scheduled standard checks support system performance across the analysis interval? | Retain all checks and evaluate any failed response. A previous passing bracket is evidence for an investigation, not automatic proof that every earlier sample is usable. |
| Trial or conditioning injections | Were exceptional injections authorized and justified, with the solution identity documented? | WHO section 11.9 discourages injections outside the defined sample set and calls for retained records when exceptionally used. FDA Q&A 13 addresses actual-sample misuse and the conditions for a characterized secondary standard. |
Sample Validity: The Result Has Its Own Acceptance Gate
Even when every opening SST parameter and every closing bracketing standard passes without adjustment, an individual sample result inside that sequence is not automatically valid or reportable. Gate 2 evaluates whether the execution of the test method on the sample itself satisfied the method's internal validity requirements.
In Investigating Out-of-Specification (OOS) Test Results for Pharmaceutical Production (Revision 1, May 2022), FDA defines a reportable result as a final analytical result that is defined in the written, approved test method and derived from one full execution of that method, starting from the original sample. Critically, the agency pairs that definition with an explicit acceptance rule for sample replicates:
Method-defined replicate count and variability limits: FDA's OOS guidance discusses methods in which one reportable result is an average of specified replicates. The method should define their number and acceptable variability. A set that fails those variability limits should not be used as that reportable result. Retain the individual values and investigate the failure instead of selecting the most favorable replicate.
Sample chromatogram and integration criteria: WHO sections 12.1 and 13 discuss method-specific acceptance criteria, scientifically sound integration, and investigation of atypical peak shapes. A passing standard does not resolve an unexpected shoulder, co-elution concern, or inappropriate integration in the sample. Investigate the observation against the procedure and supporting data; neither the visual anomaly nor the SST pass alone proves that a particular value is invalid.
Audit-trail preservation for reintegration: If an analyst adjusts baseline drop points or integration events on a sample chromatogram, If an analyst adjusts baseline drop points or integration events on a sample chromatogram, FDA's 2018 Data Integrity Q&A guidance recommends established written procedures for reprocessing chromatography and retaining each generated result for review. (For the specific electronic record controls required when chromatograms are reprocessed, see our analysis of what the CGMP audit trail must preserve during chromatogram reintegration; for setting method-level acceptance criteria before routine use, see our guide to analytical method transfer acceptance criteria for a release-testing lab.)
For an explicitly hypothetical example, assume a method averages two independent preparations only when their relative percent difference is no more than 1.5%, calculated as absolute difference divided by their mean, multiplied by 100. Results of 100.4% and 96.8% have a mean of 98.6% and a relative difference of about 3.65%, so the set fails that assumed variability criterion. If the illustrative assay specification were 98.0%–102.0%, the second preparation would also be outside it. These invented limits illustrate two separate checks; they are not compendial defaults. The laboratory cannot select the passing preparation or treat the average as resolving the OOS. Preserve both results and investigate before deciding whether any result can be invalidated.
When System Suitability Fails Mid-Sequence: Bounding the Suspect Data
Suppose the opening SST passes but an intermediate or closing standard fails its method-defined response or drift criterion. Two shortcuts are unsafe: discarding the whole run without investigating, or accepting all earlier samples solely because they preceded the failed check. The question is which period the evidence makes suspect, and what the investigation establishes about the remainder.
FDA's 2022 OOS guidance, section III.A recommends identifying suspect-period data and withholding them from use. The workflow below translates that recommendation into review questions; it is not a regulator-specified quarantine form or fixed rule for bounding injections:
Identify and withhold suspect-period data. FDA advises against using systems that do not meet SST requirements. If reference-standard responses show malfunction, identify the data generated during the suspect period and withhold them from use. Keep the raw data available for the investigation and subsequent quality review.
Establish the cause and affected span before accepting earlier data. FDA recommends identifying the cause and, if possible, correcting it before deciding whether to use data from before the suspect period. Review standard responses, instrument events, preparation records, and the timing of the observed change. A passing bracket can help bound an investigation, but it cannot exclude intermittent or progressive malfunction by itself. Document why any retained span remains scientifically supportable; an unknown onset warrants a wider impact assessment.
Never continue a failing run expecting to invalidate it later: FDA's 2022 OOS guidance explicitly warns that analysts should not knowingly continue an analysis they expect to invalidate at a later time for an assignable cause (for example, watching a reference standard drift out of limits mid-run but letting the remaining batch samples finish injecting to "see what the numbers look like").
Record and disposition the SST failure as nonconforming work: Under WHO's Good Practices for Pharmaceutical Quality Control Laboratories (TRS 1052, Annex 4, 2024), deviations from system suitability criteria are formally classified as nonconforming work. They cannot simply be deleted or ignored as an aborted run; the laboratory must record the nonconforming work, perform a documented impact assessment (including potential impact on previously reported results run on the same system), and disposition the event under the quality management system with defined authority for resuming testing.
When System Suitability Passes and the Sample Is Still OOS
A second error arises when SST passes but the sample remains OOS. The laboratory may treat the SST pass as the entire laboratory investigation, or invoke a minor chromatogram feature as a reason to dismiss the result without showing a causative error. Neither approach connects the proposed explanation to the actual sample outcome.
In WHO 2024 section 6.76, checking SST is one part of the OOS laboratory review, alongside raw data, calculations, equipment, reagents, and procedures. FDA's 2022 guidance separately sets the threshold for invalidation and explains the different treatment of method-defined replicates and investigation retests:
Treat SST as one investigation checkpoint. Acceptable SST results support the system-performance checks performed, but do not rule out every intermittent instrument problem, preparation error, or sample-specific interference. Review the relevant raw data and conditions without assuming that a passing check identifies or eliminates the cause.
Require a causative basis for invalidation. FDA section V.A links invalidating a discrete result to an observed, documented test event reasonably determined to have caused the OOS. WHO sections 6.77–6.81 similarly distinguish identified error from an inconclusive investigation. If no error is found, retain and report all results as appropriate and give the initial OOS full consideration. It does not become invalid merely because later testing passes.
Keep averaging and retesting rules separate. FDA sections IV.C.1 and V.B distinguish averages permitted by a written method from averaging an original OOS with investigation retests or resamples. Predefined consecutive injections from the same final solution may form one reportable result when their variability criterion is met; an individual injection does not necessarily make that result OOS. By contrast, do not pool mixed preparation or retest results to conceal an OOS. Retesting should follow a justified, predefined procedure with a stopping point. The broader disposition and reporting questions are addressed in our OOS investigation and disposition analysis.
The Invalidation Trap: SST as a Disguise for Testing Into Compliance
System-check labels can also conceal exploratory sample testing. Question 13 of FDA's December 2018 Data Integrity Q&A addresses actual samples used in SST, test, preparation, or equilibration runs to disguise testing into compliance. The concern is result-driven testing and incomplete records, not the mere presence of a documented standard check before a sequence.
FDA recommends written SST procedures that identify the preparation to be injected and explain its selection, consistent with the approved application or applicable monograph. Its answer distinguishes standard solutions from the conditional use of a sample as a secondary standard:
Standard preparations are the default for SST: FDA Q&A 13, referring to USP <621>, describes including replicate injections of a standard preparation or other standard solutions to determine whether requirements for precision are satisfied.
A sample used as a secondary standard needs defined controls. FDA recommends that it be properly characterized, covered by established written procedures, and taken from a different batch from the samples being tested. This conditional allowance does not authorize previewing the current test batch and hiding its results under a system-suitability label.
The record-retention point appears in Question 2 of the same FDA guidance: excluding an invalidated result from conformance decisions needs a documented scientific justification, while the original result and investigation remain in the full batch record. Question 14 separately addresses reprocessed chromatography and retaining each result. Reporting status and retention status are different decisions; withholding suspect data from use does not make those data disposable.
What the Quality Unit Signs Before Release
Gate 3 sits with the quality unit before release. A certificate of analysis (CoA) and an analytical worksheet serve different purposes; neither should be assumed to replace the complete supporting record. WHO 2024 sections 6.33–6.43 and 6.88–6.90 distinguish internal analytical documentation from reports and certificates. This article does not prescribe a universal CoA layout. The relevant release question is whether the records behind the reported result can be reconstructed and reviewed.
Under 21 CFR 211.194 (Laboratory records) and 21 CFR 211.192 (Production record review), US finished-pharmaceutical manufacturers must maintain and review specific evidence layers before releasing or distributing any batch:
Verification under actual conditions of use (21 CFR 211.194(a)(2)): The laboratory record must state each method used and verify the suitability of all testing methods under actual conditions of use.
Complete record of all data secured (21 CFR 211.194(a)(4)): Records must include a complete record of all data secured in the course of each test—explicitly including all graphs, charts, and spectra from laboratory instrumentation, properly identified to the specific material and lot tested.
Comparison to standards and second-person review (21 CFR 211.194(a)(6)–(8)): Test results must be compared with established standards, and original records must undergo documented second-person review for accuracy, completeness, and compliance with established standards.
Quality unit record review and discrepancy investigation (21 CFR 211.192): All drug product production and control records must be reviewed and approved by the quality control unit to determine compliance with established written procedures before a batch is released or distributed. Any unexplained discrepancy or failure of a batch or its components to meet any specification must be thoroughly investigated whether or not the batch has already been distributed, and the investigation must extend to other batches of the same drug product and other products that may have been associated with the specific failure or discrepancy. (To contrast routine commercial batch release packages with process validation documentation, see how PPQ protocols differ from commercial batch release records.)
WHO 2024 sections 6.37–6.38 recommend recording SST results, where applicable, and analytical acceptance criteria on the worksheet. All test values, including blanks, should be entered, with graphical data attached or traceable to electronic records. For a reviewer, that means a summary should lead back to the source data rather than force reconstruction from isolated screenshots or a final result alone.
| Record layer | Worksheet or report linkage | Supporting record to review |
|---|---|---|
| Sample result and specification comparison | Record the result, its method reference, and conformance assessment. Report or CoA contents follow the applicable requirements. | Individual values, preparations, calculations, and comparison with standards under 21 CFR 211.194(a). |
| SST results and criteria | WHO section 6.37 identifies these as worksheet information where applicable. | The standard chromatograms, criteria, calculations, and bracketing checks supporting the stated assessment. |
| Blank and other interference checks | WHO section 6.38 includes blank values and traceable graphical data. | Retained blank or placebo data where generated, and the method-based assessment of interference. |
| Sequence and processing history | Use a traceable link to the electronic sequence and processing records. This is a practical implementation choice. | Original records and relevant audit trails, including exceptional or aborted injections and reprocessing, under FDA Data Integrity Q&A and WHO chromatography guidance. |
| Suspect or invalidated results and investigations | Cross-reference the controlled investigation so its status and scope are clear. The reference format is a practical choice. | Original data, scientific justification for any invalidation, impact assessment, and documented review; exclusion from conformance does not remove retention obligations. |
What to Verify in the Applicable Method Before Testing
A reviewer should be able to connect each SST decision to the controlled procedure used on the testing date. WHO section 11.10 refers to the respective pharmacopoeial monograph or validated in-house specification and procedure; 21 CFR 211.194(a)(2) requires method identification and verification of suitability under actual conditions of use. A familiar chapter number or a screenshot of a passing status does not establish those links.
For compendial methods, check the effective chapter and product-specific monograph rather than treating older summaries as current operating instructions. An SOP should identify the criteria, solution identities, and sequence controls that apply to the particular procedure. The operational task is to identify the controlling requirements, show that they were applied, and retain the evidence for review.
Before executing or auditing commercial release sequences, QC laboratory managers and regulatory affairs teams should run a four-point verification check:
Verify the applicable version and requirements. Identify the currently effective monograph, chapter, method, and approved changes used for the test. Reconcile the sequence setup with those documents before interpreting any pass flag.
Separate Gate 1 and Gate 2 criteria in the written test method: Ensure the approved analytical procedure explicitly separates sequence-level SST acceptance criteria (standard replicate %RSD, resolution, tailing/symmetry, S/N, bracketing drift, blank limits) from sample-level validity criteria (number of sample preparations/injections and maximum replicate variability limits per FDA's 2022 OOS guidance).
Define suspect-period assessment and control solution identity. Establish how failed standard checks are investigated, which data are withheld from use, and how earlier data are assessed under FDA's OOS guidance. Document nonconforming-work impact assessment using the applicable quality system. SST procedures should identify the solution and rationale; follow FDA Question 13 for any characterized secondary-standard use and prevent disguised current-batch preview testing.
Enforce complete-record QU review before batch disposition: Verify that second-person reviewers and the quality unit inspect the complete electronic sequence, integration audit trails, bracketing results, and any OOS or nonconforming-work investigations under 21 CFR 211.194 and 21 CFR 211.192. A passing SST supports the system-performance assessment; sample and batch decisions need their own supporting evidence.
Sources
U.S. Food and Drug Administration (CDER/CBER). Analytical Procedures and Methods Validation for Drugs and Biologics - Guidance for Industry (Final, July 2015).
U.S. Food and Drug Administration (CDER). Investigating Out-of-Specification (OOS) Test Results for Pharmaceutical Production - Guidance for Industry, Revision 1 (May 2022).
U.S. Food and Drug Administration (CDER/CBER/CVM). Data Integrity and Compliance With Drug CGMP - Questions and Answers Guidance for Industry (December 2018).
Office of the Federal Register / U.S. Food and Drug Administration. 21 CFR 211.194 - Laboratory records (eCFR).
Office of the Federal Register / U.S. Food and Drug Administration. 21 CFR 211.192 - Production record review (eCFR).
World Health Organization (WHO). WHO good practices for pharmaceutical quality control laboratories (WHO Technical Report Series, No. 1052, Annex 4, 2024).
World Health Organization (WHO). WHO good chromatography practices (WHO Technical Report Series, No. 1025, Annex 4, 2020).
U.S. Food and Drug Administration. FDA Warning Letter to Unipack LLC (Warning Letter 320-26-32, December 19, 2025).
U.S. Food and Drug Administration. FDA Warning Letter to Nortec Quimica SA (320-23-07; MARCS-CMS 639894; December 8, 2022).
