Article Text

Download PDFPDF

978 Concordance of the FDA-approved companion diagnostic and a next-generation sequencing assay kit for assessing homologous recombination deficiency in ovarian cancer
Free
  1. W Weichert1,
  2. P Qiu2,
  3. J Lunceford2,
  4. A Wehn2,
  5. A Yarunin3 and
  6. M Marton2
  1. 1Institute of Pathology, Technical University of Munich, Munich, Germany
  2. 2Merck and Co., Inc., Kenilworth, USA
  3. 3AstraZeneca, Alderley Park, UK

Abstract

Introduction/Background*Olaparib+bevacizumab is approved as first-line maintenance treatment of advanced homologous recombination deficiency (HRD)-positive ovarian cancer (OC), defined by the presence of a deleterious or suspected deleterious BRCA mutation (BRCAm) and/or genomic instability (evaluated with a United States Food and Drug Administration-approved companion diagnostic). We evaluated the performance of an in-development next-generation sequencing assay, based on Illumina’s RUO TSO 500 content, that identifies variants in tumour tissue and HRD genomic scars (Illumina test). Herein, we report the performances of the in-development Illumina test versus the Myriad myChoice PLUS assay (Myriad test).

Methodology OC tissue samples were analysed with Illumina (n=227; 40ng DNA) and Myriad tests (n=254; 200ng DNA). Samples that failed QC during the first run using the Illumina test were retested with higher DNA input. Agreement rates for BRCAm, genomic instability score (GIS), and HRD status (includes BRCA and GIS [cutoff, 42]) were analysed. For the overall and the non-BRCAm cohorts, correlation between the continuous GIS of the Illumina and Myriad tests was determined. The analytical sensitivity and specificity of the Illumina-derived GIS to correctly classify genomic instability status as determined by the Myriad test (cutoff, 42) was evaluated using area under the receiver operating characteristic (AUROC).

Result(s)*Agreement rates are reported in the Table. The GIS correlation between the 2 tests was 0.980 (all samples) and 0.975 (non-BRCAm cohort). AUROC was 0.992 (all samples) and 0.988 (non-BRCAm cohort). Prevalence (Illumina and Myriad tests) was 51.0% and 49.2% for overall HRD and 27.6% and 25.5% for BRCAm. Success rates (Illumina and Myriad tests) were 86.8% (197/227) and 94.1% (239/254) (overall HRD), 88.1% (200/227) and 97.6% (248/254) (BRCAm), and 91.2% (207/227) and 94.1% (239/254) (GIS); after re-running the failed samples with the Illumina test, rates were 90.3%, 92.5%, and 93.4%, respectively.

Abstract 978 Table 1

Agreement rates for the illumina test versus the myriad test

Conclusion*Illumina test and Myriad test HRD, BRCAm, and GIS detection results were in >91% agreement. With both tests, GIS was highly correlated (0.98), and prevalence estimates of HRD and BRCAm rates were similar. Data suggest that a distributable solution such as the Illumina test may replicate the performance of the Myriad myChoice HRD assay.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.