The validation process for laboratory-developed tests (LDTs) can vary dramatically depending on the type of assay you wish to validate and can appear complicated and overwhelming to beginners. In this post, we provide an introduction to validation with a focus on next-generation sequencing (NGS) assays.
LDTs are diagnostic and prognostic assays developed by hospital,
academic, and clinical laboratories for use in clinical testing.
Although LDTs are exempt from clinical trials, they do need to be
validated prior to use. The validation process involves a series of
experiments that demonstrate clinical performance and is overseen by the
College of American Pathologists (CAP). Once validated, the assay is
performed under the guidance of the medical director.
Unlike LDTs, in-vitro diagnostics (IVDs) have undergone clinical
trials and require verification, not validation prior to use. For this
purpose, standards and controls are included within IVD kits.
The NGS workflow can be broken down into two main components: library
preparation and data analysis. Library preparation typically involves
PCR-based amplification of target nucleic acid and the addition of
sequencing indexes and barcodes. Following quality control (QC),
prepared libraries are loaded onto sequencers and run.
Downstream data processing begins with trimming, demultiplexing, and
filtering the sequencing reads. The reads are then aligned to a
reference genome, and variants are called. The variants are further
annotated and analyzed in conjunction with existing datasets for gene
discovery, variant interpretation, and clinical reporting. The
algorithms used for each step of data processing and interpretation have
significant impact on the assay’s utility.
There are two critical phases in NGS validation: the analytical validation and the clinical validation.
The purpose of the analytical validation is to optimize the NGS assay
for use in a particular lab. During this process, preliminary
performance specifications are developed using synthetic controls,
well-characterized cell lines, commercial reference materials and
potentially referenced samples. The analytical validation enables
researchers to determine a baseline for assay performance as well as
provide the opportunity to configure the bioinformatics pipeline.
If an NGS LDT performs well under analytical validation, it moves
into clinical validation. Clinical samples are inherently less
predictable and less consistent than established controls. It is,
therefore, necessary to establish clinical performance with regards to
factors such as concordance with established tests, reproducibility,
sensitivity, and specificity.
The clinical validation protocol must include the sequencing target,
which may consist of whole genomes, whole exomes, or specific gene
panels. Multianalyte NGS assays must include verification of prediction
algorithms defined during analytical validation.
After Validation—Validating Modified Components
Once a test or platform has undergone validation, the expectation is
that the LDT will be used precisely as validated. Any changes to
validated protocols, whether physical changes to the library preparation
process or changes to the bioinformatics pipeline, must be re-validated
via a supplemental validation process. Potential issues or
inconsistencies that may occur as a result of the modification should be
anticipated, noted, and tested during this process. The supplemental
validation should include re-testing of known samples and standards,
with an emphasis on those most likely to exhibit issues as a result of
Getting Started with NGS Assay Validation
Validation can be an intimidating process, especially if using a new or unfamiliar technique. As such, Fabric Genomics takes the guesswork out of data analysis by automating the entire process from alignment to clinical reporting with artificial intelligence (AI). Furthermore, Fabric can connect your laboratory with industry-leading professionals to plan and support every stage of your NGS validation.