Vol. 17 Issue 7
The Molecular Edge
Validating Lab Developed Tests
Editor's note: This is part one of a two-part article. Part two, to be published in the September issue, will examine assay performance and training issues.
Laboratory developed tests (LDTs) represent assays that have not been FDA-approved; rather, the performance characteristics and efficacy are demonstrated by medical professionals who offer the test as a clinical service.
LDTs are a vital component of patient care. Advantages include less dependence on specific sources of reagents, instrumentation and proprietary software (to minimize the risk of back orders and product recalls), and the opportunity to tweak the assay and offer the improved version in a timely fashion following re-validation. Patients benefit from LDTs when they would otherwise have no access to the information provided by the test, or when the LDT is faster, cheaper, less invasive or more informative than alternative tests. If the LDT is used for the same clinical indications as an existing FDA-approved test and on the same specimen type, it behooves the laboratory to show that their assay is at least as good as the FDA-approved alternative.
Validating the LDT
Typical steps involved in the validation of an LDT are shown in the Table. Reviewing the medical literature is the first step in any validation study. It is reasonable to consult with clinician clients to confirm their needs and define the minimum acceptable performance characteristics that will assure the test's acceptance as a legitimate aid to patient management.
Non-laboratorians tend to underestimate the amount of work it takes to validate a laboratory test. Analytic validity comprises determination of such factors as sensitivity and specificity for the target analyte, precision and "analytic measurement range" if the assay is quantitative, reproducibility and effect of interfering substances. One must establish a system of controls to verify assay performance on a day-to-day basis. For example, external controls show that a molecular test system provides intended results across the range of expected outcomes; internal controls show successful nucleic acid extraction and hybridization in each patient specimen by parallel testing of an endogenous or spiked sequence. The effects of partial versus complete interference must be considered for quantitative assays (e.g., viral load or tumor burden). One must develop guidance on analytic interpretation of raw data for controls and for each patient sample to generate each reportable result.
Clinical validity includes such factors as indications for testing, sensitivity and specificity for the medical condition, positive and negative predictive values in the target population(s), normal range, report format and clinical interpretation. Establishing clinical validity is usually beyond the scope of a single testing laboratory. It is reasonable to cite medical literature supporting the laboratory's stated indications for testing (e.g., which clinical settings are appropriate) and how the test assists in patient management.
Dr. Gulley is director of Molecular Pathology at the University of North Carolina at Chapel Hill. She serves as chair of the Strategic Planning Committee in the Association for Molecular Pathology.
Steps to Consider When Validating a LDT
- Assess clinical need; communicate with clients
- Anticipate test volume and institutional risks/benefits
- Literature review to choose analytic method and establish minimum acceptable performance characteristics
- Design pilot study of mock samples and actual patient specimens representing the various sample types and range of outcomes to assess performance characteristics: sensitivity, specificity, reproducibility, linearity, accuracy, precision, reportable range
- Consider IRB approval for testing of human subjects (if indicated)
- Obtain reagents, supplies, equipment, calibrators, controls and patient specimens
- Perform pilot study and conduct additional experiments to optimize performance characteristics and refine the step-by-step analytic procedure
- Gather and analyze data to make recommendations on clinical indications for ordering the test, acceptable specimen types and rejection criteria, other preanalytic factors (collection, processing, storage), batching and turn-around-time, guidance for analytic and clinical interpretation, reporting, finances, royalties and billing
- Compose and sign a Validation Report documenting work done to assess performance characteristics and other relevant parameters, and vetting the assay as ready for clinical implementation
- Train personnel and show competency
- Obtain adequate inventory of supplies
- Place signed Procedure in the Procedure Manual
- Notify clinicians of test availability; educate regarding indications for testing, acceptable specimen types, normal range, precision (if indicated) and clinical utility
- Revalidation when the procedure or its clinical indications are modified.