Analytical Validation

The lowest concentration of an analyte that can be reliably distinguished from the absence of that analyte, often called the Limit of Detection (LoD). This parameter defines the assay's ability to detect minute quantities of a biomarker.

• Verification Protocol: Serial dilution of a high-titer sample to extinction
• Reporting: Concentration units (e.g., pg/mL, copies/mL)
• Clinical Relevance: Critical for early-stage disease detection and minimal residual disease monitoring
• Confounders: Matrix effects from different sample types (serum vs. plasma) can shift LoD

The ability of an assay to measure only the intended analyte without cross-reacting with structurally similar molecules or interfering substances. High specificity ensures the signal originates exclusively from the target.

• Cross-Reactivity Studies: Testing against homologous proteins, isoforms, or common interferents
• Interfering Substances: Hemoglobin (hemolysis), lipids (lipemia), bilirubin (icterus), and common pharmaceuticals
• Hook Effect: A phenomenon where excess analyte saturates both capture and detection reagents, producing falsely low results—must be ruled out at high concentrations

The degree of agreement among replicate measurements under specified conditions. Precision is decomposed into repeatability (same operator, instrument, day) and intermediate precision (varying operators, days, or reagent lots).

• Intra-Assay CV: Coefficient of variation within a single run; typically < 5% for clinical assays
• Inter-Assay CV: Variation across multiple runs over days; typically < 10%
• CLSI EP05-A3: The standard guideline for precision evaluation protocols
• Acceptance Criteria: Defined by total allowable error (TAE) budgets derived from biological variation

The range of analyte concentrations over which the assay produces results directly proportional to the true concentration. Linearity confirms the method's ability to provide accurate measurements without dilution or concentration adjustments.

• Lower Limit of Quantification (LLoQ): The lowest concentration measurable with acceptable precision and accuracy
• Upper Limit of Quantification (ULoQ): The highest concentration before signal saturation
• Polynomial Regression Analysis: Fitting data to linear, quadratic, and cubic models to assess deviation from linearity
• CLSI EP06-A: Guideline for evaluating linearity of quantitative measurement procedures

The closeness of agreement between a measured value and the true value of the measurand. Trueness is expressed as bias—the systematic difference between the average of replicate measurements and a reference standard.

• Reference Material Comparison: Testing against certified reference materials (CRMs) or a gold-standard method
• Recovery Studies: Spiking a known quantity of analyte into a sample matrix and measuring the percentage recovered
• Bland-Altman Analysis: A graphical method to assess agreement between two measurement techniques
• Acceptable Bias: Typically within ±10% of the reference value for ligand-binding assays

The capacity of the assay to remain unaffected by small but deliberate variations in method parameters, and the analyte's resilience under various storage conditions. Robustness testing challenges the method's operational boundaries.

• Sample Stability: Bench-top (room temperature), refrigerated (2-8°C), frozen (-20°C, -80°C), and freeze-thaw cycles
• Reagent Stability: On-board stability, lot-to-lot variability, and reconstituted shelf-life
• Youden Ruggedness Testing: A factorial experimental design to evaluate the influence of minor environmental or procedural perturbations
• Critical Parameters: Incubation time, temperature, pH, and pipetting precision

The fundamental measure of a test's ability to correctly differentiate between patients with and without a target condition. It is a composite metric derived from sensitivity and specificity, representing the proportion of all tests that yield a correct result.

• Formula: (True Positives + True Negatives) / Total Population
• Context: A prerequisite for any 510(k) submission comparing a new SaMD to a predicate device.
• Limitation: Highly dependent on disease prevalence in the study population.

The true positive rate, quantifying the test's ability to correctly identify patients who do have the target condition. A highly sensitive test minimizes false negatives.

• Clinical Impact: Critical for screening applications where missing a disease is catastrophic (e.g., cancer screening).
• Calculation: True Positives / (True Positives + False Negatives)
• Trade-off: Often inversely related to specificity; adjusting the decision threshold shifts the balance.

The true negative rate, quantifying the test's ability to correctly identify patients who do not have the target condition. A highly specific test minimizes false positives.

• Clinical Impact: Essential for confirmatory diagnostics where a false positive leads to unnecessary invasive procedures.
• Calculation: True Negatives / (True Negatives + False Positives)
• Regulatory Note: The FDA requires clear justification for the chosen specificity threshold in the intended use statement.

The Receiver Operating Characteristic curve is a graphical plot illustrating the diagnostic ability of a binary classifier across all possible discrimination thresholds. It plots the true positive rate (sensitivity) against the false positive rate (1 - specificity).

• AUC: The Area Under the Curve summarizes overall performance; an AUC of 1.0 represents perfect discrimination, while 0.5 represents random chance.
• Utility: Used to select the optimal operating point that balances clinical sensitivity and specificity for the intended use.

The combined, documented processes mandated by ISO 13485 and IEC 62304 to prove a SaMD works correctly. In analytical validation, this is strictly separated from clinical evaluation.

• Verification: Confirms design outputs meet design inputs. Does the algorithm correctly calculate the biomarker value?
• Validation: Confirms the final device meets user needs. Does the measured biomarker accurately reflect the true biological state?
• Artifacts: These activities populate the Design History File (DHF) for regulatory review.

The lowest concentration of an analyte that can be reliably distinguished from the absence of that analyte (a blank sample) with a stated confidence level, typically 95%.

• Relevance: Critical for molecular diagnostics and biomarker identification systems where trace quantities of a marker must be detected.
• Methodology: Determined by repeatedly testing low-concentration samples and calculating the signal-to-noise ratio.
• Distinction: Not to be confused with the Limit of Quantification (LoQ), which is the lowest concentration that can be measured with acceptable precision.