Analytical validation of a next generation sequencing liquid biopsy assay for high sensitivity broad molecular profiling

Abstract

Circulating tumor DNA (ctDNA) analysis is being incorporated into cancer care; notably in profiling patients to guide treatment decisions. Responses to targeted therapies have been observed in patients with actionable mutations detected in plasma DNA at variant allele fractions (VAFs) below 0.5%. Highly sensitive methods are therefore required for optimal clinical use. To enable objective assessment of assay performance, detailed analytical validation is required. We developed the InVisionFirst™ assay, an assay based on enhanced tagged amplicon sequencing (eTAm-Seq™) technology to profile 36 genes commonly mutated in non-small cell lung cancer (NSCLC) and other cancer types for actionable genomic alterations in cell-free DNA. The assay has been developed to detect point mutations, indels, amplifications and gene fusions that commonly occur in NSCLC. For analytical validation, two 10mL blood tubes were collected from NSCLC patients and healthy volunteer donors. In addition, contrived samples were used to represent a wide spectrum of genetic aberrations and VAFs. Samples were analyzed by multiple operators, at different times and using different reagent Lots. Results were compared with digital PCR (dPCR). The InVisionFirst assay demonstrated an excellent limit of detection, with 99.48% sensitivity for SNVs present at VAF range 0.25%-0.33%, 92.46% sensitivity for indels at 0.25% VAF and a high rate of detection at lower frequencies while retaining high specificity (99.9997% per base). The assay also detected ALK and ROS1 gene fusions, and DNA amplifications in ERBB2, FGFR1, MET and EGFR with high sensitivity and specificity. Comparison between the InVisionFirst assay and dPCR in a series of cancer patients showed high concordance. This analytical validation demonstrated that the InVisionFirst assay is highly sensitive, specific and robust, and meets analytical requirements for clinical applications.

Liquid Biopsy for Advanced Non-Small Cell Lung Cancer (NSCLC): A Statement Paper from the IASLC

“…Patients should be drawn before any therapy, as even one or two weeks of effective treatment can render a previously plasma-positive patient undetectable. 61 Figure 1 depicts a flowchart which summarizes panel believes for the use of the liquid biopsy in tretament naive patients.…”

“…This scenario is most frequent in the case of very scant tissue, in patients for whom invasive procedures may be risky or contraindicated, or with bone biopsies, because although bone biopsies can be sufficient for a histologic diagnosis, they may not be compatible with molecular testing as some decalcification solutions can damage tumor DNA. 61 A positive finding of an actionable mutation in ctDNA, if using a validated assay, represents sufficient evidence to initiate targeted treatment. However, a negative result should be considered inconclusive and followed up with a secondary test.…”

Current and Future Molecular Testing in NSCLC, What Can We Expect from New Sequencing Technologies?

“…The sensitivity is related to the low amount of ctDNA among ccfDNA and to the global amount of ccfDNA that challenges the limit of detection of sequencing technologies. Bioinformatic methods are being developed to discriminate a true mutation at low frequency in ctDNA from background noise [107] and sequencing methods were adapted to improve sensitivity [108]. However, ctDNA cannot be detected in 20-30% of patients.…”

A novel high-throughput molecular counting method with single base-pair resolution enables accurate single-gene NIPT

“…In liquid biopsy, even though copy number variation is a hallmark of cancer, 13 current FDA-approved diagnostics are limited to resolving a 30% to 120% increase in the copy number found in cfDNA and therefore are only applicable to late-stage cancers. [14][15][16] These limitations are largely due to traditional library preparation methods for NGS that introduce significant levels of noise that obscure the relationship between DNA molecular counts of the input sample and the final sequencing read count. Without the ability to perform absolute quantification, NGS cannot reliably quantify fetal SNVs in prenatal testing, detect gene amplifications that are present in Stage 1-2 cancers, or sensitively detect rare cell-free tumor DNA sequences that may only be present at 1-10 molecules in the sample.…”