Analytical validation of the Oncomine Pan-Cancer Cell-Free Assay in a CLIA- and CAP-regulated laboratory for detection of solid tumor-derived variants in blood plasma.

Abstract

e14614 Background: Assessing tumor-derived somatic variants from blood plasma provides minimally invasive tumor profiling, decreased cost relative to traditional tissue biopsy and rapid turn-around-time. We describe here the analytical validation of the Oncomine Pan-Cancer Cell-Free assay in a CAP-accredited, CLIA-certified laboratory. The assay is designed to detect somatic DNA single-nucleotide variants (SNV), insertions/deletions (INDEL), copy number variants (CNV), and gene fusions in cell-free total nucleic acid (cfTNA) across 52 genes. For research use only. Not for use in diagnostic procedures. Methods: We assessed the sensitivity, specificity, accuracy, and precision of the assay. Pre-characterized reference materials with alterations at known allelic frequencies (AF) were used to establish performance characteristics for each variant class followed by verification on clinical specimens. Whole blood (n = 73) from healthy (10), breast cancer (9), colorectal cancer (32), and lung cancer (22) donors were collected in K2EDTA tubes and plasma was separated within 8 hours of collection. cfTNA was isolated using the MagMAX Cell-Free Total Nucleic Acid Isolation Kit on the KingFisher Flex. Libraries were prepared following kit instructions. Templating and sequencing were performed using the Ion 550 Kit on the Ion Chef and S5 XL systems. Alignment to hg19 and variant calling were performed using Torrent Suite and Ion Reporter software. Results: We observed a sensitivity of 80% at 0.1% AF and > 99.9% at 0.5% AF for SNV/INDEL. Sensitivity was > 99.9% at 1.34 fold-change for CNV, and > 99.9% at 0.4% fusion fraction. Specificity and accuracy were > 99% for all variant classes. Precision was 98% for SNV/INDEL and > 99.9% for CNV/fusions. Analysis of donor plasma demonstrated clinical feasibility; on average, we detected 1 hotspot SNV/INDEL (range 0-4) across the 3 cancer types. For SNV/INDEL in plasma where matched tissue genotyping was available (n = 17), we observed > 99% concordance (AF range 0.09%-52%). Conclusions: We have demonstrated that this assay is a sensitive method for evaluating tumor-derived somatic variants in blood plasma. It may provide an alternative and minimally invasive method for mutation profiling.