Incurred Sample Reanalysis (ISR) is the repeated measurement of a portion of samples actually obtained from study participants or animals in a separate analytical run on a different day. It verifies whether the originally reported concentrations of an analyte are reproducible when a real study matrix is present instead of artificially spiked control samples.
Why real study samples are measured again
Calibrators and quality control samples are created during method validation by spiking a blank matrix. Study samples can differ significantly from these: protein binding, back-conversion of known or unknown metabolites, inhomogeneity, concomitant medication, or specific biological components can influence the concentration measurement. ICH M10 therefore uses ISR to verify the reliability of reported sample concentrations under real study conditions.
An ISR repeats the analysis of a subset of the same study in separate runs and on different days using the same bioanalytical method. The selection of samples is based on pre-established objective criteria in the protocol, study plan, or an SOP. Animals or subjects should be selected as randomly as possible from the treated population, and the concentration profile should be sufficiently covered.
Execution and evaluation of results
ICH M10 recommends samples in the range of maximum concentration and in the elimination phase. The reanalysis must take place within the stability window of the analyte and must not be performed on the same day as the original analysis. Pooling should be avoided because it can mask conspicuous individual values. ISR samples and quality controls are processed and evaluated as in the original run.
For chromatographic procedures, at least two-thirds of the repeats must be within a percentage difference of plus or minus 20%; for ligand-binding assays, plus or minus 30% applies. If the criterion is not met, ICH M10 requires an investigation and, if necessary, remedial action. ISR values do not replace the original study values. Large systematic deviations can be a reason for further examination even if the overall criterion is met.
Distinction from method validation
Incurred Sample Reanalysis is not a repetition of the full method validation. The latter uses calibrators and spiked quality controls to show whether the procedure is suitable for its intended purpose. ISR then checks whether the method provides reliably reproducible results for samples actually incurred, with their individual matrix properties.
The difference thus concerns both the sample material and the objective. In ISR, the samples originate from humans or animals and may contain metabolites or concomitant factors that a blank matrix does not reflect. Existing entries for validation and ‘Validierung’ deal with general or computer system validation; ISR is a narrowly defined requirement of bioanalytical study sample analysis.
The scope of the ISR depends on the number and character of the study samples. For up to 1,000 samples, ICH M10 requires at least a ten percent repetition. For larger data sets, ten percent of the first 1,000 samples and five percent of the samples exceeding that number are assessed. These specifications are minimum values; knowledge of the analyte and matrix may necessitate further testing.
Formal fulfillment of the two-thirds rule does not necessarily end the scientific evaluation. If differences cluster in a single analytical run, in one subject, or in a specific concentration range, they may indicate a specific problem. Investigating such patterns prevents an accepted overall result from masking local weaknesses in data quality.
The selection must be representative across all relevant subjects or animals. Limiting the selection to easily measurable or particularly clean samples would fail the objective of monitoring repeatability under the actual conditions of the study.
All selection and calculation steps must remain reconstructible in the Bioanalytical Report. This includes the originally measured values, repeat values, the calculated difference, and the results of any root cause analyses.
Relevance for clinical trials
ISR is relevant for the validity of PK, TK, and, where applicable, comparative bioavailability data. The study plan must determine in a timely manner when samples are available, who selects them, how the blinded re-measurement is organized, and how deviations are investigated. For the first study in humans, ICH M10 mentions ISR as well as for pivotal early patient studies and certain studies in impaired organ function.
Full-service CROs like Mediconomics support ISR planning through sample logistics, definition of selection rules, documentation of separately performed analytical runs, and data verification against acceptance criteria. Bioanalytics, Data Management, and Quality Assurance can thus transfer anomalies into the Bioanalytical Report and the associated investigation documents.
Frequently Asked Questions (FAQ)
Are new samples taken from the same participants for ISR?
No new samples are taken for ISR. ISR uses a portion of study samples already obtained and measures them again in a separate run.
Why are samples selected around Cmax?
Together with samples from the elimination phase, this selection covers relevant areas of the concentration profile.
Does an ISR result replace the original concentration value?
The original value remains. ICH M10 clarifies that the repeat values do not take the place of the original study data.
Regulatory References
- ICH M10, Bioanalytical Method Validation and Study Sample Analysis – defines ISR, sample selection, acceptance criteria, and investigations for deviations.
- ICH M10 Step 4 Presentation – explains the application of ISR for non-clinical and clinical study samples.
- ICH S3A, Toxicokinetics: The Assessment of Systemic Exposure in Toxicity Studies – contextualizes TK data in non-clinical toxicity studies.