Abstract 340: Validation of TSO500 NGS panel for comprehensive genomic profiling from FFPE tumor samples

Abstract

Abstract Comprehensive genomic profiling (CGP) of tumor sample using large targeted NGS panels (e.g. MSK-IMPACT) has been shown to allow the detection of biomarkers from hundreds of cancer relevant genes involved in targeted and immunotherapy. Illumina TruSight Oncology 500 (TSO500), a commercial CGP assay, allows concurrent analysis of DNA and RNA to simultaneously detect multiple types of variants across 523 genes. The assay is intended to detect single nucleotide variants (SNV), insertions and deletions (indel), and multinucleotide variants (MNV), copy number gain (CNV), tumor mutational burden (TMB) and microsatellite instability (MSI) in DNA, and fusions and splicing variants (exon skipping) in RNA. Here we aimed to validate TSO500 assay for its analytical performance on FFPE tumor samples. The validation samples include commercial controls, cell lines, clinical FFPE from NSCLC, CRC, Uterus and GIST (tumor content ≥20%). Library preparation was performed from minimum of 40 ng DNA and 40ng RNA. The data analysis is performed using TruSight Oncology 500 Local App software (LRM v2.2). A custom script (v1.0) was used to add SNP annotation for small variants and combine data output into cumulative format. For SNV and indel detection QC, 61 of 64 DNA libraries (95%) passed the QC for library size (median insert size ≥70bp) and coverage (median exon coverage ≥150). The percentage of DNA libraries passing QC for CNV and MSI are 95% and 88%, respectively. All 64 RNA libraries passed QC. The analytical performance of TSO500 was summarized in the Table below. In summary, the results showed that all the acceptance criteria were met and the TSO500 assay can be utilized for genomic profiling of FFPE tumor samples. The study also identified a few limitations of the Illumina LRM software e.g. potential false positive SNV within a indel background, potential false negative of complex indel, and the challenge of long indel detection. Summary of analytical performance of TSO500 Sensitivity and specificity Assay LoD Accuracy by concordance Inter-run reproducibility SNV/Small indel Sensitivity for SNV and indel detection are 99.24% and 99.36%. The LoD for SNV is 3-5%, for DEL is 5-10% and for INS is 3-5%. PPA for SNV detections is 96.64% (115/119; 95% CI: 91.68%, 98.69%) and for small indel is 92.86% (13/14; 95% CI: 68.53%, 98.73%) between TSO500 and comparator assays. The positive call rate across all SNV and small indel variants is 98.7% (42842/43419). Specificity for SNV and small indel detection are 99.9998% and 100.00%, respectively. MSI Sensitivity and Specificity for MSI detection are both 100%. NA MSI status for all samples were detected with 100% of concordance by three assays; MSI concordance for MSI-H is 100% (95% CI: 60.97%, 100.0%) and for MSS is 100% (87.13%, 100.0%). PPA is 100% (95% CI: 51.01%, 100%) and NPA is 100% (95% CI: 34.24%, 100%). TMB NA NA TSO500 reported TMB is well correlated and comparable to exome determined TMB. The CV values for samples are acceptable. CNV NA NA All positive CNVs detected by the comparator assay were detected by the TSO500 assay with similar fold changes; The overall reproducible call rate is 91.5% (43/47). PPA between TSO500 and OCAV3 assays is 100% (95% CI: 51.01%, 100%). Fusion Sensitivity and specificity for fusion detection is 91.94% and 100%, respectively. NA All TSO500 assay detected fusions were confirmed by other assays;PPA is 100% (95% CI: 60.97%, 100%) and NPA is 100% (95% CI: 75.75%, 100%) The reproducibility of fusion call is 92.59% (25/27; 95% CI: 76.63%, 97.94%). Splicing Sensitivity and Specificity for splicing variant detection are 91.67% and 100.00%, respectively. NA The splicing variant detections were confirmed by OCAV3, OFA and Archer assays; The overall reproducibility of positive call is 100% with 95% CI (43.85%, 100%) and negative call reproducibility is 100% with 95% CI (87.13%, 100%). PPA is 100% (95%% CI: 20.65%, 100%) and NPA is 100% (95% CI: 83.18%, 100%) Citation Format: Jin Li, Zhenyu Yan, Weihua Liu, Lin Shi, Cynthia Spittle, Chad Galderisi. Validation of TSO500 NGS panel for comprehensive genomic profiling from FFPE tumor samples [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 340.