P2.03b-093 Validation and Performance of a Standardized ctDNA NGS Assay across Two Laboratories: Topic: Biomarkers

Abstract

Molecular profiling of tumors using circulating tumor DNA (ctDNA) in the blood of cancer patients, as a liquid biopsy, is rapidly becoming established as a useful source of information to aid clinical decision-making when a solid tumor biopsy is not available, or is limited in amount or quality. DNA alterations are often found in a small fraction of the total cell free DNA in plasma, and their detection requires specially designed assays that are sensitive and reproducible. Individual hotspot mutations can be assayed using technologies such as droplet/digital PCR, but multiplexing such assays is limited by the small amount of clinical material. This can be addressed by assays based on next generation sequencing (NGS), to create sensitive panels for ctDNA analysis. For clinical application, it is essential that such NGS assays be standardized and reproducible, both intra-and inter-laboratory. Standardization for tissue-based NGS assays has only recently been implemented, after much discussion. We describe a strategy for validation and standardization of a high sensitivity NGS-based ctDNA assay between two laboratories, based in the US and UK. The enhanced TAm-Seq® assay (eTAm-Seq™) uses efficient library preparations and bespoke algorithms to identify cancer mutations within a panel of 34 genes, covering cancer hotspots as well as entire coding regions of selected genes. To ensure this complex process is standardised and controlled, a high level ISO and CLIA quality management system is implemented. To perform analytical validation of this assay, we used reference standards and plasma controls to demonstrate the sensitivity, specificity and quantitative accuracy of this ctDNA analysis platform. We compared performance of the assay between two laboratories, finding a high rate of concordance and reproducibility. Using DNA quantities typical of those found in up to 4ml of plasma from cancer patients, our assay provides high sensitivity for variants that are present at allele fraction 0.25% or higher in plasma, and retains substantial sensitivity at allele fractions as low as 0.1%. Standard dilution curves of well-characterized reference samples show that the accuracy of the eTAm-Seq® assay is predominantly limited by stochastic sampling. Analysis of plasma samples from control individuals demonstrates a low false positive rate. Additional data with associated clinical data will be presented at the meeting. Our data demonstrates eTAm-SeqTM assy's robustness and performance in two labs, supporting its use as a basis for clinical applications globally, allowing a high degree of standardization and comparability for molecular profiling of tumors using liquid biopsy.