6.2 Assay Validation & Verification Standards (CLSI H42-A2, H62)

6.2 Assay Validation & Verification Standards (CLSI H42-A2, H62)

Clinical flow cytometry assays must undergo rigorous validation or verification before they can be used for patient testing. The Clinical and Laboratory Standards Institute (CLSI) provides foundational guidelines for these processes. CLSI H42-A2 specifically addresses the enumeration of immunologically defined cell populations (like CD4+ T-cells), while CLSI H62 provides comprehensive guidance on the validation of multi-parameter flow cytometry assays, focusing on the analytical characterization of the assay.

Validation vs. Verification

It is crucial to distinguish between validation and verification in a regulatory context (e.g., CLIA, CAP, FDA).

• Validation: The process of establishing the performance specifications of a new test, a Laboratory Developed Test (LDT), or a modified FDA-cleared test. It requires an extensive evaluation of accuracy, precision, analytical sensitivity, analytical specificity, reportable range, and reference intervals.
• Verification: The process of confirming that the laboratory can achieve the performance specifications established by the manufacturer of an unmodified, FDA-cleared/approved In Vitro Diagnostic (IVD) assay. Verification is generally less exhaustive than validation but still requires proving accuracy, precision, and the reportable range in the local laboratory environment.

Key Performance Characteristics (CLSI H62)

When validating a flow cytometry assay, the laboratory must evaluate and document several key performance characteristics as outlined in CLSI H62.

1. Accuracy (Method Comparison)

Accuracy refers to the closeness of agreement between the assay's result and a true or accepted reference value. In flow cytometry, accuracy is usually established by method comparison.

• Procedure: The laboratory tests a minimum number of patient samples (e.g., 20-40, spanning the reportable range) using the new assay and a reference method. The reference method could be an existing validated flow cytometry assay in the lab, a predicate FDA-cleared assay, or an alternative methodology (e.g., immunohistochemistry, though matrix differences make this challenging).
• Statistical Analysis: Results are analyzed using linear regression (e.g., Passing-Bablok or Deming regression), and Pearson's or Spearman's correlation coefficient (R) is calculated. A high R value (e.g., >0.95) and a slope close to 1.0 indicate acceptable accuracy. Bias must also be assessed using a Bland-Altman plot to ensure there is no systematic over- or under-estimation across the measurement range.

2. Precision (Repeatability and Reproducibility)

Precision measures the closeness of agreement between independent test results obtained under stipulated conditions.

• Repeatability (Intra-assay precision): Testing the same sample multiple times in the same run, by the same operator, on the same instrument.
• Reproducibility (Inter-assay precision): Testing the same sample across different runs, different days, different operators, and potentially different instruments.
• Acceptance Criteria: For flow cytometry, the Coefficient of Variation (CV) is the standard metric. Acceptable CVs depend on the population frequency. For a major population (e.g., CD3+ T-cells at 70%), a CV < 5% is typical. For a rare population (e.g., minimal residual disease at 0.05%), the CV will be inherently higher due to Poisson statistics, and acceptance criteria must account for this statistical variation.

3. Analytical Sensitivity (LOB, LOD, LOQ)

Analytical sensitivity defines the lowest amount of analyte (target cells) the assay can accurately and reliably detect. This is critical for assays like Minimal Residual Disease (MRD) or Paroxysmal Nocturnal Hemoglobinuria (PNH) testing.

• Limit of Blank (LOB): The highest measurement result likely to be observed for a blank sample (a sample devoid of the target cells). It is calculated by running normal, negative samples multiple times.
• Limit of Detection (LOD): The lowest concentration of target cells that can be consistently detected (typically 95% of the time) above the LOB. It is calculated by spiking low numbers of target cells into a negative matrix and analyzing the recovery.
• Limit of Quantitation (LOQ): The lowest concentration of target cells that can be quantified with acceptable precision (e.g., CV < 20%) and accuracy. The LOQ is always greater than or equal to the LOD and represents the lower bound of the reportable range. In flow cytometry, a minimum of 20 to 50 events in the final gate is often required to achieve an acceptable LOQ.

4. Analytical Specificity

Analytical specificity is the ability of the assay to exclusively identify the target cell population in the presence of other cells, interfering substances, or similar markers.

• Interference Testing: Assessing the impact of hemolysis, lipemia, icterus, or common drugs on the assay results. For example, severe hemolysis can increase background autofluorescence.
• Cross-reactivity: Ensuring that the antibodies used do not cross-react with non-target cell populations.
• Gating Strategy: Specificity is heavily dependent on the gating strategy. The validation must demonstrate that the Boolean gating logic effectively excludes non-target cells (e.g., excluding platelets and debris from a lymphocyte gate) and dead cells (using a viability dye).

5. Reference Intervals (Normal Values)

Every clinical assay must have established reference intervals (normal ranges) for the specific patient populations it serves.

• Establishment: Testing a minimum of 120 healthy individuals (stratified by age and sex if necessary) to establish the central 95% interval.
• Verification: If the lab adopts manufacturer-provided or published reference intervals, they must verify them by testing a smaller number of local healthy donors (e.g., 20 individuals). If no more than 2 out of 20 results fall outside the proposed range, the interval is considered verified.

Specimen Stability and Matrix Equivalency

• Specimen Stability: The laboratory must determine the maximum allowable time between sample collection and processing. This is done by testing a sample immediately (baseline) and then at various time points (e.g., 24h, 48h, 72h) stored at room temperature and 4°C. The assay is stable as long as the results remain within a predefined acceptable variation (e.g., ±10% relative to baseline) and the cell viability remains acceptable.
• Matrix Equivalency: If an assay is intended for multiple specimen types (e.g., peripheral blood, bone marrow, CSF), the lab must validate performance in each matrix. While full accuracy testing might be done on peripheral blood, smaller correlation studies must be performed to prove that bone marrow aspirates yield comparable analytical performance.

Carryover

Flow cytometers are fluidic systems, and sample-to-sample carryover is a significant risk, especially when a sample with a very high cell count or a strong positive MRD signal is followed by a negative sample. Carryover must be validated by running alternating sequences of high-concentration samples and true negative samples (e.g., High-High-High-Low-Low-Low). The acceptable carryover limit is typically <0.1% or <1.0%, depending on the assay's clinical sensitivity requirements.