Abstract 4262: Analytical validation of TruSight Oncology 500, a comprehensive genomic profiling LDT for use in routine patient care

Abstract

Abstract We report the analytical performance of Illumina's TruSight™ Oncology 500 (TSO500) NGS assay for key oncologic variants/ biomarkers. A combination of commercially available reference materials and de-identified clinical FFPE samples were used as the truth set to assess accuracy, analytical sensitivity, and precision. The TSO500 assay utilizes FFPE-extracted DNA and RNA for enrichment-based, targeted next-generation sequencing to detect a comprehensive list of biomarkers relevant to solid tumors and hematological malignancies: single nucleotide variants (SNVs), copy number variants (CNVs), indels, fusions, splice variants, tumor mutational burden (TMB), and microsatellite instability (MSI). Targeted sequencing reduces sequencing data on irrelevant regions in the genome thus greatly increasing coverage on relevant genetic variants and regions. The assay covers small variants in 523 genes, fusions/spice variants in 55 genes, and CNVs in 59 genes, thus providing comprehensive coverage of the variants that play a role in tumorigenesis. The assay also provides a measure of two additional biomarkers, MSI and TMB, which are important for predicting response to cancer immunotherapy agents. De-identified, FFPE patient samples were processed and analyzed in the Illumina Lab Services (ILS, CAP-accredited laboratory). Samples that passed QC metrics were evaluated for accuracy for variants that were known to be present in the truth set. Accuracy was evaluated between the truth set and ILS for TMB using linear regression analysis and for microsatellite stability (MSS/MSI) using a cutoff of 20%. Additionally, accuracy, analytical sensitivity, and precision for the biomarkers covered by TSO500 were assessed using reference materials with known content. The assay demonstrated ≥ 99% concordance between the truth set and ILS for small variants (present at an allele frequency of ≥ 5%), fusions/splice variants, and CNVs. For the immunotherapy biomarkers, we observed high concordance between the truth set and ILS. There was 100% concordance for MSS/MSI calls with most samples being MSS. The linear regression R-squared value for TMB was 0.99 with scores ranging from very low (0) to high (370); most samples had scores between 0-20. A high level of repeatability and reproducibility was observed across three reference materials each for small variants, CNVs, and fusions/splice variants with values of ≥ 98%. For SNVs, the sensitivity of detection was 100% at the Lower Limit of Detection of 5%. Collectively we demonstrated high accuracy and precision for the variant types found in the TSO500 panel. Validation of this assay will allow the implementation of this laboratory developed test (LDT) for use in routine patient care. Citation Format: Karena Kosco, Brad Sickler, Silvia Chan, Mawadda Alnaeeli, Sue Beruti. Analytical validation of TruSight Oncology 500, a comprehensive genomic profiling LDT for use in routine patient care [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 4262.