Low-Level EGFR Variants in NSCLC Detectable by Liquid Biopsy Assays

Target capture next-generation sequencing (NGS), MassARRAY, and real-time quantitative polymerase chain reaction (qPCR) all effectively detected low frequency somatic epidermal growth factor receptor (EGFR) mutations in cell-free (cf) circulating tumor (ct) DNA from individuals with non–small cell lung cancer (NSCLC) with a concordance of 83.3%, according to a presentation by Lei Zhang, PhD, University of Alberta, Edmonton, Alberta, Canada, at the AMP 2020 Annual Meeting and Expo.

Target capture sequencing using the AVENIO ctDNA Expanded Kit (Roche) was the most sensitive of the 3 methods. The real-time qPCR-based Cobas EGFR Mutation Test (Roche) required the shortest testing turnaround time. The MassARRAY-based UltraSEEK Lung Panel (Agena Bioscience), which has been clinically validated for testing EGFR in ctDNA, was used as a reference method.

The leading cause of cancer-related deaths worldwide is lung cancer, with the majority of cases of lung cancer being NSCLC. Many targeted therapies have been developed for the treatment of NSCLC, including tyrosine kinase inhibitors (TKIs). EGFR is one of the driver mutations in NSCLC targeted by TKIs.

TKIs are effective in the treatment of patients with NSCLC who harbor sensitizing mutations in the EGFR gene. Molecular analysis of driver mutations in genes known to respond to targeted therapies is critical for the management of NSCLC. For example, in metastatic NSCLC with sensitizing EGFR mutations, treatment with TKIs is associated with improved progression-free survival compared with treatment with systemic chemotherapy.

However, tumors treated with TKIs over a period of time can develop mutations conferring TKI resistance. Monitoring for EGFR mutations is desirable because the presence of some (eg, T790M), confer resistance to first-generation TKIs. Tissue biopsy is generally considered the “gold standard” for EGFR testing. Biopsies involving invasive tissue sampling are time-consuming and expensive, and are associated with the risk of adverse events for patients. A cost-effective and non-invasive alternative to tissue biopsy is liquid biopsy.

When patients experience NSCLC tumor progression on first-line TKI therapy, blood plasma samples could be collected for ctDNA testing. If results indicate the presence of the EGFR T790M mutation, treatment with a third-generation TKI would be indicated. A negative ctDNA result, however, would necessitate a tissue biopsy for confirmation.

Although liquid biopsy methods have been developed as a low-risk alternative to tissue biopsy for testing tumor biomarkers, Zhang cautioned that a high level of sensitivity is needed for the accurate detection of extremely low levels of gene variants in ctDNA.

The main objective of this study was to compare the performance of the leading testing methods for EGFR variants in plasma ctDNA from patients with NSCLC, namely target capture sequencing NGS, Mass-ARRAY, and real-time qPCR.

A total of 23 plasma samples were tested in parallel by the target capture NGS and MassARRAY-based assays. Among these, 12 plasma samples were also tested by real-time qPCR. An EGFR multiplex ctDNA reference standard set (Horizon Discovery) was used for evaluating the limit of detection of the assays.

cfDNA can be effectively extracted from the plasma of patients with NSCLC using the AVENIO ctDNA Isolation Kit. Good quality cfDNA was defined as having a major peak at 160 to 200 base pairs (bp) as per the High Sensitivity DNA Assay Agilent 2100 bioanalyzer. A smaller secondary peak for ctDNA may be visible at 320 to 400 bp. cfDNA in plasma samples was highly variable and for some samples, only a limited amount of cfDNA could be extracted.

The DNA library was prepared using the AVENIO ctDNA Expanded Kit and analyzed using an Agilent Bioanalyzer. There is a major DNA peak at about 300 bp, with additional larger DNA peaks observed. Several distinguishable cfDNA peaks with a size shift indicated the successful ligation of adapters.

After ligation, the libraries were sequenced on the Illumina NextSeq platform. Typical metrics for biomarker sequencing of ctDNA show >20 ´ 106 read pairs per sample, which allows detection of the targeted gene variants. The typical error rate for the sequencing was <7 ´ 10-5. Zhang said this indicates the sequencing quality is high. Fragment length ranges from 150 to 300 bp. The theoretical sensitivity is >99%.

After sequencing and a bioinformatic analysis, somatic mutations in ctDNA were tested by NGS. The presence of EGFR mutations in plasma samples was highly variable. The median variant allele fraction of EGFR somatic mutations in ctDNA by NGS was 2% (range, 0.08%-45.16%). Median variant copy number concentration of EGFR gene variants was 66.70 copies/mL of plasma (range, 0.679-4950).

Two types of EGFR variants of interest—EGFR TKI–sensitizing (G719A, S768I, L861G, and exon 19 deletions) and EGFR TKI–resistant (T790M)—were detected in plasma ctDNA from patients with NSCLC by the NGS-based AVENIO panel. The median variant allele fraction of EGFR TKI–sensitive mutations was 4% (range, 0.11%-45.16%). The median variant allele fraction of the TKI-resistant EGFR T790M mutation was 2% (range, 0.08%-8.28%).

The median variant copy number for EGFR TKI–sensitive mutations in ctDNA by NGS was 135.00 copies/mL plasma (range, 1.79-4950). The median copy number for TKI-resistant EGFR T790M mutations was 36.10 copies/mL plasma (range, 0.68-561).

The NGS assay showed high concordance with the MassARRAY test (91.3%) for detecting the TKI-resistant EGFR T790M mutation. The concordance of NGS with MassARRAY for detecting TKI-sensitizing mutations was also high (95.7%-100%). The qPCR method showed 83.3% and 100% concordance with the MassARRAY test for detecting EGFR sensitizing mutations and the T790M resistance mutation, respectively. Concordance for the NGS assay in comparison with the qPCR method was 91.7% for detecting the TKI-resistant EGFR T790M mutation.

EGFR TKI–sensitizing mutations with a variant allele fraction of <0.2% were easily missed by the MassARRAY-based test. EGFR T790M mutations with a variant allele fraction of <5% were missed by both the MassARRAY and qPCR-based assays.

In addition to the variant allele fraction, the copy number of gene variants can also be an important factor that can impact the sensitivity of different detection assays on testing tumor biomarker genes in plasma ctDNA, the study authors noted.

Zhang concluded, “With the advantages of high throughput and an excellent limit of detection, the NGS-based assay shows high competitiveness to be clinically used for the targeted treatment of NSCLC.”

_Reference:

_{Zhang L, Coffin J, Formenti K, et al. Comparative Study of Three Assays: Target Capture Sequencing, MassARRAY and Real-Time qPCR for Testing Somatic Mutations in Plasma Cell-Free Circulation Tumour DNA of Non-small Cell Lung Cancer. Presented at: the virtual Association for Molecular Pathology (AMP) 2020 Annual Meeting and Expo; November 16-20, 2020. Abstract #TT10}