Detection of Targetable Alterations in Non-small Cell Lung Cancer using Next-generation Sequencing.

Oncogene-addicted metastatic non-small-cell lung cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up

When should we order a next generation sequencing test in a patient with cancer?

Introduction: Next generation sequencing (NGS) is a highly sensitive method for detecting somatic mutations. Mutations in NRAS and KRAS may affect up to 50% of patients with colorectal cancer (CRC). Recent data suggest the clinical benefit of panitumumab is restricted to patients who have no tumor mutations in RAS, defined as codons 12, 13, 59, 61, 117 and 146 of KRAS and NRAS genes. An Extended RAS panel* targets these codons as a single multiplex assay. A dual strand approach distinguishes true mutations from artifacts commonly found in DNA from Formalin Fixed Paraffin Embedded (FFPE) tissue. Analytical performance of this panel was characterized. Methods: FFPE tissue was examined for tumor content and area. DNA was extracted from FFPE tissue, cell lines and plasmids. The DNA quality and quantity were simultaneously determined through a quantitative PCR measurement of amplifiability. TruSeq Custom Amplicon technology targeting each strand of DNA was used to construct NGS libraries. Sequencing was performed on the MiSeqDx, and the Illumina Somatic Variant Caller was used to call somatic variants. Results: The dual-strand amplicon approach results in high accuracy by eliminating false positives observed on one strand of DNA typical of an FFPE artifact, by requiring every variant to be detected on both strands. We evaluated performance on 500 FFPE CRC samples processed with the Extended RAS panel. Concordance against Qiagen TheraScreen and Sanger Sequencing results was evaluated. The somatic variant caller yields a specificity of 0.98 and sensitivity of 0.72 for variants in the 5%-10% range, and a specificity of 0.96 and sensitivity of 0.98, overall, where we take the alternate methods as truth. Cumulative FFPE tissue area influences performance of the RAS panel. FFPE of a range of tissue area were tested using 1, 3, 5, and 8 sections per sample. 80mm2 minimum (240mm2 recommended) cumulative tissue area using 5uM slices was optimal for performance and accuracy. DNA quantity and quality also influence test accuracy and sensitivity. A study with 40 FFPE specimens assayed over multiple operators and sequencers produced average coverage of 32,000 reads per target per sample, and over 93% alignable reads per sample. Reproducibility of mutations detected over multiple operators and instruments was 100%. The Extended RAS Panel demonstrates robust performance across many sources of variation, including FFPE DNA extraction method, DNA storage time, recommended DNA input range, qPCR/PCR thermocyclers with little to no impact on sample pass rate and agreement with Sanger sequencing. We also demonstrate detection of 56 distinct mutations in a single run. Conclusion: This multiplex assay achieves high accuracy for detection of somatic mutations from DNA extracted from FFPE colorectal tissue samples. *In development For Research Use Only. Not for use in diagnostic procedures. Citation Format: Tamsen Dunn, Anita Iyer, Margaret Porter, Robert Haigis, Shannon Smith, Desiree Lee, Nitin Udar. A rapid, multiplexed, and highly accurate next-generation sequencing RAS Panel for FFPE colorectal samples reporting on the absence or presence of low frequency somatic variants. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 3638.

MET Exon 14 Skipping Mutations: Essential Considerations for Current Management of Non–Small-Cell Lung Cancer

Introduction: The oncoReveal™ Dx Lung and Colon Cancer Assay (oRDx-LCCA) is an amplicon-based targeted next generation sequencing (NGS), in vitro diagnostic test that detects DNA variants across 22 genes, including single nucleotide variants (SNVs) and insertions and deletions (INDELs) in KRAS and EGFR, from DNA isolated from formalin-fixed paraffin-embedded (FFPE) non-small cell lung cancer (NSCLC) and colorectal cancer (CRC) tumor tissue specimens. It is approved by the FDA as a companion diagnostic (CDx) for identifying approved therapeutic treatments for KRAS G12/G13 wild-type variants in CRC, and EGFR L858R variants and exon 19 in-frame deletions for NSCLC. oRDx-LCCA can detect non-CDx variants such as EGFR T790M, G719X and exon 20 insertions in NSCLC and KRAS Q61X and A146T and BRAF V600E in CRC. Variants are detected using Pillar Biosciences’ proprietary PiVAT® (Pillar Variant Analysis Toolkit) software. Methods: DNA input range was evaluated at 6 concentrations between 5 - 160ng in duplicate using DNA extracted from 15 clinical FFPE samples containing SNVs and INDELs from representative targeted mutations. Accuracy was determined by comparing detected variants with an externally validated comparator (CompO) across 208 samples (84 CRC, 124 NSCLC). Limit of detection (LoD) was determined by serial dilution of KRAS, EGFR, or BRAF positive FFPE samples in normal DNA to detect variant allele frequency (VAF) between 1 - 10%. Reproducibility tests were performed across 3 sites, 9 runs, 6 operators, 3 reagent lots and 3 thermocyclers, resulting in a total of 360 libraries. The non-inferiority (NI) statistical test was used to compare the oRDx-LCCA with an FDA approved CDx assay for EGFR exon 19 deletions and L858R mutations (CompQ) and for KRAS G12/G13 mutations (CompC). Synthetic FASTQ files were created at VAFs above and below the variant detection threshold for CDx mutations and used to validate quality control parameters, mean coverage, and on-target rate. Results: Effective DNA input range was determined to be between 10 - 80ng. LoD’s for KRAS G12/G13, EGFR L858 and exon 19 deletion, and BRAF V600E were established at < 4%. Positive percent agreement (PPA) and negative percent agreement (NPA) between oRDx-LCCA and CompO were > 99% each. Across all variables tested for reproducibility, the average percent positive agreement (APA) and average percent negative agreement (ANA) were each > 95%. NI margin was calculated to be < 10% between oRDx-LCCA and CompQ across all NSCLC samples and between oRDx-LCCA and CompC across all CRC samples. A total of 61 synthetic FASTQ datasets were created to validate the functionality and accuracy of PiVAT including VAF thresholds for 126 SNV and INDEL CDx variants. Conclusion: oRDx-LCCA is a highly accurate FDA approved IVD assay and PiVAT is a powerful FDA approved software tool for the detection of clinically relevant KRAS variants in CRC and EGFR variants in NSCLC and determination of approved therapy. Citation Format: Gloria Chan Johnson, Jordan Aldersley, Nicholas Lodato, Jamie Lim, Ian Flockhart, Mark Urbin, Yue Ke, Sean Polvino, Martin Zillmann, Gang Song, Zhaohui Wang. Validation of FDA approved oncoReveal Dx lung and colon cancer assay (oRDx-LCCA) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 5795.

11099 Background: Next-generation sequencing (NGS) allows for simultaneous detection of numerous actionable somatic variants in cancer. We have implemented a clinical NGS panel to detect genetic alterations in 25 genes with established roles in cancer and report here the frequency of clinically actionable genetic variants in a variety of cancer types. Methods: NGS testing was performed in a CAP-certified, CLIA-licensed environment on DNA extracted from FFPE tissue in 209 cases spanning 41 histologic tumor types. DNA was enriched by hybrid capture and sequenced to >1,000x average coverage on Illumina sequencers with 2x101bp or 2x150bp reads. Variants were called using clinically validated parameters using the Genome Analysis Toolkit, Pindel, and the custom-written Clinical Genomicist Workstation. Results: Non-small cell lung cancer (45%), pancreatic cancer (10%), and colorectal cancer (8%) were the most common tumors sent for NGS analysis. An average of 3 (range 1- 16) non-synonymous, non-SNP sequence variants per case (SNVs and indels) were detected in the 130kb exonic target. Variants were most commonly seen in TP53, KRAS, and EGFR. 27% of cases (56/209) had one or more variants with therapeutic implications for the tumor type tested (e.g., EGFR mutation in NSCLC). 15% of cases (32/209) showed actionable variants not generally associated with the malignancy tested (e.g., detection of an activating KITvariant in thymic carcinoma). 10% of cases (21/209) had variants that were prognostically significant but not directly targetable. Some cases (9%) had variants that were prognostic/diagnostic and targetable. In 117 cases (56% of total), no therapeutically or prognostically significant variants were identified. Overall, in 92 cases (44%), NGS testing yielded information with therapeutic (majority), prognostic, or diagnostic ramifications. Conclusions: We found that 44% of unselected cancer cases have clinically relevant sequence variants in a set of 25 commonly mutated cancer genes. Our data suggest that clinical NGS testing may serve as an integral tool in realizing the potential of precision medicine in oncology.

Next-generation sequencing (NGS) is a cost-effective technology capable of screening several genes simultaneously; however, its application in a clinical context requires an established workflow to acquire reliable sequencing results. Here, we report an optimized NGS workflow analyzing 22 lung cancer-related genes to sequence critical samples such as DNA from formalin-fixed paraffin-embedded (FFPE) blocks and circulating free DNA (cfDNA). Snap frozen and matched FFPE gDNA from 12 non-small cell lung cancer (NSCLC) patients, whose gDNA fragmentation status was previously evaluated using a multiplex PCR-based quality control, were successfully sequenced with Ion Torrent PGM™. The robust bioinformatic pipeline allowed us to correctly call both Single Nucleotide Variants (SNVs) and indels with a detection limit of 5%, achieving 100% specificity and 96% sensitivity. This workflow was also validated in 13 FFPE NSCLC biopsies. Furthermore, a specific protocol for low input gDNA capable of producing good sequencing data with high coverage, high uniformity, and a low error rate was also optimized. In conclusion, we demonstrate the feasibility of obtaining gDNA from FFPE samples suitable for NGS by performing appropriate quality controls. The optimized workflow, capable of screening low input gDNA, highlights NGS as a potential tool in the detection, disease monitoring, and treatment of NSCLC.

Implementation of next generation sequencing technology for somatic mutation detection in routine laboratory practice

e21030 Background: The management of NSCLC is rapidly evolving with the discovery of new driver mutations and the introduction of targeted therapies (TT). The list of targets with FDA-approved TT is estimated to cover over half of the cases of non-squamous NSCLC. The prevalence of driver mutations varies significantly depending on several demographic variables such as smoking status, race, and sex. This abstract presents data obtained from a unique, racially diverse cohort of patients in the Washington, DC metropolitan area. Methods: Retrospective review of circulating tumor DNA NGS data from 2 commercial testing companies (Guardant and Tempus) and demographic data from our health records on 66 patients with NSCLC treated at the George Washington University Medical Faculty Associates between October 2017 and November 2021 was performed. Demographic information including age, sex, race, and smoking history, as well as histology, stage at diagnosis, and driver mutations were collected. Only patients with non-squamous NSCLC and identifiable mutations were included. Results: Mean age was 67. Most patients in this cohort were female (56.1%, male 43.9%), African-American (62.1%, Caucasian 25.8%, African 9.1%, Asian 3.0%), had smoking history (former smoker 48.5%, current smoker 25.8%, never smoker 25.8%), had adenocarcinoma (78.8%, other NSCLC (not squamous cell carcinoma or adenocarcinoma) 21.2%) with advanced stage (stage 4 62.1%, stage 3 28.8%, stage 1 3.0%, unknown 6.1%). Mutations are listed as follows: KRAS 28.8% (G12C accounted for 58% of all KRAS mutations), EGFR 19.7% (exons 19 and 21 mutations such as del19 and L858R accounted for 69%), MET 9.1% (MET amplification made up 83%), Her2 3.0% (2/66), ALK 3.0% (2/66), RET 1.5% (1/66), non-targetable mutations 34.8% (23/66). A table at the end of this abstract shows the frequency of three most prevalent mutations in this cohort distinguished by race, sex, and smoking status. Conclusions: KRAS was the most prevalent driver mutation in this cohort, followed by EGFR and MET. In concordance with the existing literature, EGFR was more prevalent in non-smokers and females, while KRAS was more common among smokers. Interestingly, a larger percentage of Africans had EGFR mutations compared to African-Americans. Given the small sample size, however, a larger scale study is needed to investigate this further. These findings show that demographic factors influence the prevalence of driver mutations which can have therapeutic implications in patients with NSCLC.[Table: see text]

Introduction: Advance non-small cell lung cancer (NSCLC) is a disease with a high clinical and economic burden for healthcare systems. It exists few studies on the management costs of NSCLC patients. The main objective of this study was to identify healthcare resource consumption and costs related to management of advanced NSCLC in the context of the Catalan Health Service (CatSalut). Methodology: A systematic literature review (SLR) was performed to identify resource use for NSCLC management. Additionally was completed with observational data and expert’s opinion. Pharmacoeconomics evaluation was a costs analysis using a decision modelling for diagnosis and clinical - therapeutic pathway. for a time horizon of one year. Results: Total costs for advanced NSCLC patients (N= 4.026) would account for 141.054.270€/year: 9% for diagnostic and 91% for clinical - therapeutic management. The main cost driver could be associated to anticancer drugs (44-61% of the total medical costs) across the different line of treatment: 46.403.510€ in first line, 19.980.440€ in second line and 6.517.901€ in third line and after. The overall healthcare cost for advanced NSCLC could result in 1,65% of the total public health budget in Catalonia. Conclusion: Advanced NSCLC management could increase in healthcare spending. The identification of an appropriate therapeutic pathway, resource use and costs could support healthcare policies and interventions to guarantee an adequate and efficient disease management ensuring care accessibility for all patients. Keywords: Advanced Non-small cell lung cancer, NSCLC, diagnosis and disease management, cost of disease, resource use, CatSalut.

Introduction: In many instances in Precision Oncology Medicine fragmented DNA from formalin-fixed paraffin-embedded (FFPE) samples is the only option. The majority of such degradedsamples need to be sequenced using next generation sequencing (NGS) [1]. Standardizedguidelines highlight that a minimum of 500 × coverage depth is required for tissue analyses [2].However, coverage isn’t the only metric required for a successful NGS analysis. Highly uniformDNA libraries allow clinical labs to generate more hits from their screens, saving both time andmoney across the course of diagnosis therefore ultimately benefits patient care. The commerciallyavailable single stranded DNA library prep kit tested in this study renders all input moleculessingle-stranded before ligation therefore it recovers template molecules containing DNA nicks andlabile lesions thus maximizing library complexity. Method: 10 solid tumor FFPE derived DNA samples went through three different NGS workflows:100ng of DNA input in an amplicon-based assay 325 gene panel (Assay A), 60ng input for a 500gene mechanical sheared-based double strand DNA library preparation with hybridization captureassay (~500 gene panel, Assay B), 50ng and 10ng input for a enzymatic shear-based singlestranded DNA library preparation with hybridization capture assay (WES with 1000+ enhancedgene content, Assay C). Data from assays A and C was processed with in-house bioinformaticpipeline, while Assay B’s data was processed with a commercially available bioinformatic pipelinethat was bundled with the assay. NGS QC metrics were compared across the three datasets. Results: This novel single stranded DNA library preparation with hybridization capture workflowshowed superior QC metrics compared to the other two workflows, especially on coverageuniformity and percent target above 100X, with as low as 10ng input. Conclusion: Employing a novel commercially available DNA library preparation workflow in anNGS assay, NeoGenomics was able to rescue poor quality samples that had previously failed tomeet our internal uniformity metric on other NGS assays and generate robust and reportableresults with as low as 10ng DNA input.[1] Nagahashi et al., J Surg Res. 2017 Dec; 220: 125-132.[2] Cappello et al., J Pers Med. 2022 May; 12(5): 750. Citation Format: Jieying Chu, Rachel Schell, Faqiang Wu, Michal Krawczyk, Daniel Whang, Segun Jung, Hyunjun Nam, Kenneth B. Thomas, Fernando J. Lopez-Diaz, Vincent Funari, Julie A. Mayer, Shashi Kulkarni, Jiannan Guo, Steven P. Lau-Rivera. A single stranded DNA library preparation workflow used for hybridization capture based WES assay showed superior uniformity [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 949.

Introduction: The FDA has approved 11 therapies for non-small cell lung cancer (NSCLC) patients diagnosed with gene fusions. Importantly, these life-saving targeted therapies remain underutilized due to a lack of cost-effective, comprehensive, and simple-to-use technologies for gene fusion detection. Our previous work using ASPYRE technology (allele-specific pyrophosphorolysis reaction; Silva et al, 2020) demonstrated detection of somatic variants at single-molecule level from DNA extracted from tumor tissue. However, DNA-based fusion detection requires tiling across large introns, which limits sensitivity particularly if breakpoints are located in repetitive regions of the genome. Here, we show that ASPYRE technology can also be utilized for highly sensitive detection of RNA fusions, scaling to over 30 fusion targets covering the main 3’ fusion partners (ALK, ROS1, RET, NTRK1, NTRK3) and from a single reverse transcription PCR (RT-PCR) amplification reaction. Methods: The ASPYRE assay consists of four sequential enzymatic stages: RT-PCR; enzymatic clean-up; combined exonuclease digestion, hybridization, pyrophosphorolysis and ligation; rolling circle isothermal amplification and detection. All targets are detected across two reaction wells using four channels on a standard real-time PCR instrument. A panel of synthetic RNA oligonucleotides comprising gene fusion boundaries was spiked into a background pool of lung-RNA, quantified using digital PCR (dPCR) and compared against a commercial reference standard to validate its suitability for assessing assay performance. RNA was extracted from formalin-fixed paraffin-embedded (FFPE) lung tissue from healthy donors to determine assay specificity, and from FFPE lung tissue from NSCLC patients with known fusions to exemplify end-to-end sample analysis. Results: We demonstrate equivalence of our in-house reference samples to commercial reference standards. Using serial dilutions of input RNA standards, we show that detection by ASPYRE is consistent with detection of single molecules as input for common ROS1, ALK, RET and NTRK3 fusions. In addition, ASPYRE was able to detect previously confirmed ROS1 and ALK fusions and MET exon skipping variants in FFPE samples. ASPYRE had high specificity, correctly confirming the absence of RNA fusions and exon skipping variants in healthy control FFPE samples. Conclusions: ASPYRE approached single-molecule sensitivity for over 30 RNA fusion targets in a single reaction. The assay required only 1 ng RNA input, with a turn-around time of <4 hours from extracted RNA to result, opening the door to routine detection of RNA fusions from limited specimens. The simplicity and ease of use of ASPYRE, combined with its rapid turnaround time and analytical performance characteristics, provides a new tool for rapid and comprehensive treatment selection in NSCLC. Citation Format: Eleanor Gray, Justyna Mordaka, Efthimia Christoforou, Kristine von Bargen, Nicola Potts, Rebecca Palmer, Barnaby Balmforth. Low-cost, simple and rapid assay for single-molecule detection of gene fusions from RNA with ASPYRE technology [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 4104.

T790M in epidermal growth factor receptor (EGFR) accounts for about 50% of the acquired resistance to EGFR tyrosine kinase inhibitor (TKI) in patients with non-small cell lung cancer (NSCLC) carrying sensitive EGFR mutations. Earlier studies suggested that T790M mutation was also detected in about 2% of TKI-naive NSCLCs. Recently, three groups reported that, by using highly sensitive assays, T790M mutation was detected in about 40% of TKI-naive NSCLC with sensitive EGFR mutations. When we carefully studied these reports, we realized that all of those data were generated from formalin-fixed paraffin embedded (FFPE) tumor tissue samples, which raised a concern that the high mutation positivity could be consequence of formalin fixation. To address this, we assessed the T790M mutation in 36 pairs of frozen and FFPE tumor tissues of TKI-naïve NSCLC with sensitive EGFR mutations using an enzyme-based mutant-enriched PCR assay (ME-PCR) with 0.1% sensitivity. In addition, frozen and FFPE adjacent normal tissues from the same patients were also assessed for a further comparison. While 41.7% (15/36) T790M positive rate was detected in the tumor FFPE samples which was consistent with previous reports, only 1 of the 36 (2.8%) frozen counterparts was T790M positive. As suspected, 48.5% (16/33) of T790M mutation rate was also identified in FFPE adjacent normal tissues but none of the 35 frozen adjacent normal tissues was T790M positive. Our results indicate that the high T790M positivity detected in TKI-naïve NSCLC using some highly sensitive methods may in some cases be FFPE-derived artifacts. Citation Information: Mol Cancer Ther 2013;12(11 Suppl):A36. Citation Format: Xin Ye, Zhong-zheng Zhu, Lei Zhong, Yachao Lu, Yun Sun, Xiaolu Yin, Zhenfan Yang, Guanshan ZHU, Qunsheng Ji. High T790M detection rate in TKI-naïve NSCLC with EGFR sensitive mutation: Truth or artifact?. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2013 Oct 19-23; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2013;12(11 Suppl):Abstract nr A36.

Lung cancer arises as a result of multiple genetic alterations and environmental influences such as cigarette smoking, radiation and air pollution. Molecular classification of these alterations may help in the development of individualised anticancer therapies. In this study, Ion Torrent technology was used to sequence genomic material extracted from formalin fixed paraffin embedded (FFPE) tumours. Multiple genetic variants were identified in each tumour sample. About 65% of the identified mutations occurred in the epidermal growth factor receptor ( EGFR ) and phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha ( PIK3CA ) genes. Results from this study demonstrated the feasibility of using FFPE material in next generation sequencing (NGS). In conclusion, several key mutations associated with human cancers were identified.

Exploiting the Full Potential of Novel Agents Targeting EGFR Exon 20 Insertions in Advanced NSCLC: Next-Generation Sequencing Outperforms Polymerase Chain Reaction–Based Testing