High-definition PCR (HDPCR) detection of DNA and RNA variants in non-small cell lung cancer (NSCLC) samples.

Abstract

e20603 Background: Digital PCR (dPCR) is an emerging technology platform for detecting genomic variants in cancer genomes due to its high sensitivity and fast time to results compared to sequencing. However, translational oncology applications often require the measurement of more biomarkers than there are color channels available on dPCR platforms. One approach to address this limitation with dPCR is to split a sample across many wells and profile a subset of variants in each well. For input-limited samples, however, this results in fewer molecules being profiled in each dPCR well, resulting in a reduction in sensitivity and fewer patient samples processed per instrument run. ChromaCode has developed a research use only (RUO) digital High Definition PCR (HDPCR) NSCLC assay, for multiplexed detection of 14 DNA variants and 15 RNA fusion variants relevant in non-small cell lung cancer samples. The assay is constructed using both amplitude modulation and multi-channel resilient signal encoding methods. Amplitude modulation enables different variants to generate a distinguishable signal at different intensity levels in a single color channel, allowing for multiple targets to be detected within that single-color channel. In addition, resilient encoding generates a signal in more than one color channel to create a form of error detection in the assay design. Methods: Assay benchmarking was performed using over 500 contrived human biological FFPE samples, consisting of synthetic DNA or RNA variants spiked into a background matrix of FFPE-extracted DNA or RNA; over 500 contrived human biological plasma samples, consisting of synthetic DNA or RNA variants spiked into a background matrix of plasma-extracted cell free DNA or RNA; and residual human biological FFPE and plasma NSCLC samples that were previously characterized using a targeted sequencing workflow. The samples were tested using the HDPCR NSCLC assay on the QuantStudio Absolute Q Digital PCR system, and data analysis was performed with custom analysis algorithms. Results: For the more than 500 contrived FFPE and plasma samples, the HDPCR NSCLC assay had high overall agreement with expectation across a range of mutant allele fractions for both DNA and RNA analytes (≥99% PPA and ≥99% NPA). For a set of N = 25 residual human biological FFPE samples, the assay was also highly concordant (100% PPA and 99% NPA) with a targeted panel sequencing comparator. The hands-on workflow time from isolation start to analysis complete was < 24 hours. Conclusions: The HDPCR NSCLC assay is a robust RUO tool for the sensitive and rapid detection of commonly targeted variants relevant to NSCLC samples. This technology could complement sequencing assays when there is a need for a rapid turnaround time or there are limited amounts of isolated nucleic acid.