Performance validation of a next generation sequencing myeloid assay on an integrated nucleic acid purification and sequencing system

The large diversity of myeloid neoplasms requires in-depth characterization of genetic abnormalities, including single-nucleotide variants (SNVs), small insertions and deletions (INDELs), and fusions and translocations for diagnosis and management. NGS can identify mutations associated with myeloid disorders to diagnose specific subtypes or identify a personalized and patient-specific treatment. We validated a targeted NGS myeloid assay for use in our CAP/CLIA accredited clinical laboratory. The myeloid NGS panel can identify up to 45 key DNA targets and 34 fusion driver genes covering over 800 unique fusions. The assay utilizes an automated purification system and sequencer with an on-board bioinformatics pipeline and report generation capabilities. This automated workflow has a turnaround time of approximately 24 hours from loading of the test specimen on the purification system to results ready for review. All steps were hands-off except for sample loading and transfer of the nucleic acid purification plate to the sequencer. Commercially available reference materials consisting of contrived DNA and RNA from cell lines with SNVs, INDELS, and fusions common to myeloid disorders were used for analytic sensitivity and precision studies. Precision was evaluated in triplicate over 2 instruments, 3 days, 2 operators, and 3 runs. For analytic specificity, ten normal healthy donor DNA and RNA were extracted from blood collection tubes designed to stabilize nucleic acid. For accuracy, 8 peripheral blood lymphocytes specimens from patients with a variety of myeloid disorders were extracted and assayed on both the myeloid panel under validation and a reference NGS myeloid panel. The 8 myeloid specimens' primary diagnosis was chronic myeloid leukemia (n=3) and myelodysplastic syndrome (n=5). For accuracy, the NGS myeloid assay panel had 100% concordance to the reference NGS myeloid panel. All ten normal healthy donors had no variants detected indicating high specificity. Limit of detection was established as ≥5% Variant Allele Frequency (VAF) for DNA and ≥250 fusion mapped reads for RNA fusions. For precision, the %CV between all conditions was 4.8-28.6% for the 23 DNA variants and 17.9-36.0% for the 9 fusion variants. Our data supports the use of this targeted myeloid NGS assay and the integrated purification/sequencer system for the detection of genetic lesions in myeloid neoplasms. The streamlined workflow accelerated result generation, enabling faster treatment decisions. Automation reduces the need for qualified operators and facilitates efficient clinical management of patient samples for disease diagnosis and treatment guidance. Additionally, availability of timely results can benefit patients by enabling earlier treatment decisions. These qualities support the use of this system in clinical settings for more efficient management of patients' disease.