Clinical Performance of Comprehensive Genomic Profiling Using a Wide Range of Input Quantities and Solid Tumor Tissue Types
Document Type
Presentation
Publication Date
12-1-2022
Keywords
oregon; chiles; genomics
Abstract
Introduction: Identification of solid tumor mutation profiles using comprehensive genomic profiling (CGP) has shown significant utility in predicting response to targeted and immuno-oncology therapies. In clinical practice, tumor samples are primarily preserved as formalinfixed, paraffin-embedded (FFPE) tissues and can vary highly in nucleic acid quantity and quality. The Illumina TruSight Oncology 500 Highthroughput assay (TSO 500 HT) is a CGP next-generation sequencing (NGS) assay designed to detect somatic DNA variants in 523 genes, RNA fusions in 55 genes, tumor mutational burden, and microsatellite instability. Although accumulating evidence supports robust TSO 500 HT performance using a variety of FFPE tissues, few studies have explored how variable initial sample quantity predicts reflex testing success in a high-volume, large healthcare system. Methods: We retrospectively assessed TSO 500 HT clinical sequencing results (n = 5,576) from 33 cancer types for nucleic acid quantity, sequencing quality, and the presence of clinically significant DNA or RNA alterations. Adult cancer patients (stages I-IV) in the Providence Healthcare system received CGP testing from January 2021-April 2022 (n = 5,332 patients). DNA and RNA was extracted from FFPE tissues using Qiagen or Promega FFPE Extraction Kit procedures. NGS libraries were prepared using the TSO 500 HT protocol and sequenced on the Illumina NovaSeq 6000 platform. Variant clinical significance was determined using interpretation assistance from the OncoKB precision oncology knowledgebase. Results: Overall, 95% of samples had a DNA and RNA result with TSO 500 HT assay metrics in acceptable quality ranges and 91% of samples had at least one clinically significant DNA/RNA biomarker. Initial DNA and RNA stock concentrations were significantly higher in successful assays (DNA mean = 34.1 ng/µl, RNA mean = 135 ng/µl) compared to failed assays (DNA mean = 11.3 ng/µl, RNA mean = 91.1 ng/µl). Notably, acceptable test results and clinically significant biomarkers were obtained in an appreciable number of cases (n = 5,200) that had DNA or RNA stock concentrations within the range of failed tests (failed DNA range 0.018-325 ng/µl, failed RNA range 0.69-1186 ng/µl). Finally, we show several case studies with scant tissue that revealed clinically actionable sequencing results. Conclusions: Our results found that initial DNA and RNA concentrations following FFPE extraction are a significant predictor of TSO 500 HT procedure test success. However, testing success occurred even at the lowest DNA and RNA stock concentrations. We therefore propose clinical value in using CGP testing for all non-zero nucleic acid stock concentrations in FFPE solid tumor samples.
Clinical Institute
Cancer
Specialty/Research Institute
Oncology
Specialty/Research Institute
Earle A. Chiles Research Institute
