an integrated system for targeted next-generation

AN INTEGRATED SYSTEM FOR TARGETED NEXT-GENERATION SEQUENCING THAT ENABLES SIMULTANEOUS ANALYSIS OF DNA
MUTATIONS, RNA FUSIONS AND GENE EXPRESSION IN RESIDUAL CLINICAL FFPE, FNA, AND LIQUID BIOPSIES
Brian C Haynes1, Richard Blidner1, Robert Zeigler1, Jason R Plyler1, Sachin Sah1, Liangjing Chen1, Huiping Zhu1, Andrew G Hadd1, Junya Fujimoto2, Vassiliki A Papadimitrakopoulou3, Ignacio I Wistuba2, and Gary J Latham1
1
Asuragen, Inc., Austin, TX; 2Department of Translational Molecular Pathology, 3Department of Thoracic/Head and Neck Medical Oncology, Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX
PROTOTYPE INTEGRATED DNA/RNA NGS THYROID PANEL
SUMMARY
• QuantideX® NGS technology can interrogate both DNA and RNA targets from total nucleic acid
(TNA) using a streamlined and unified workflow and customizable content.
• An evaluation of two prototype QuantideX NGS panels, a 146-marker panel for thyroid cancer
and a 187-marker panel for non-small cell lung cancer, demonstrated detection of DNA
mutations, gene fusions and quantitative measures of mRNA targets.
• This NGS approach can unveil mult-omic data from a single TNA specimen in pursuit of improved
diagnostic yields and more informed assessments of mutation-negative samples.
DNA Targets
AKT1
APC
The capability of NGS to interrogate a broad range of DNA mutations in a single assay has precipitated
a paradigm shift in precision medicine from single-target assays to highly multiplexed NGS panels.
Current NGS panels have increased the breadth of content but do not support the analysis of RNA and
DNA markers in a unified assay. We present QuantideX NGS, a comprehensive approach for targeted
clinical NGS that enables simultaneous quantification of DNA and RNA through a streamlined workflow
compatible with low-input total nucleic acid (TNA) derived from the most challenging clinical specimens
including FFPE, FNA, and liquid biopsies.
DNA Targets
ALK
mRNA Expression Targets
GNAS
BRAF
KRAS
IDH1
CTNNB1
MAP2K1
EGFR
SMAD4
DDR2
MET
HRAS
TP53
EGFR
NRAS
KRAS
TSHR
ERBB2
PIK3CA
MET
CDKN2A
ERBB4
PTEN
NRAS
EIF1AX
FBXW7
SMAD4
RET
CTNNB1
FGFR1
STK11
PIK3CA
PTEN
FGFR2
TP53
40 mRNA targets–
50 fusion targets–
25 Diagnostic biomarkers selected
from WT RNA-Seq discovery work
performed on 68 FFPE thyroid
lesions, and 15 theranostic
markers and ECs.
MET
20 DNA targets–
13 identified from whole transcriptome
RNA-Seq discovery on 68 FFPE
thyroid lesions (red). Other fusions
(blue) selected from thyroid cancer
publications and COSMIC.
Fusion Targets
FGFR3
ALK
56 hotspot regions selected from
NCCN guidelines, COSMIC and
thyroid cancer literature.
2A
Fusion Targets
BRAF
20 DNA targets–
INTRODUCTION
mRNA Expression Targets
PROTOTYPE INTEGRATED DNA/RNA NGS NON-SMALL CELL LUNG CANCER PANEL
RRM1
ERCC1
TLE3
ESR1
TOP1
FGFR1
TUBB3
PTEN
PDGFRA 3’/5’
ALK 3’/5’
RET 3’/5’
ROS1 3’/5’
NTRK1 3’/5’
24 mRNA targets–
55 hotspot regions selected from
NCCN guidelines, COSMIC
prevalence and NSCLC literature.
108 fusion targets–
9 prognostic and theranostic
markers predictive of response
to platinum and taxane therapies,
10 imbalance targets and 5
endogenous controls.
Fusions for ALK (yellow), RET (black),
ROS (red) and other (blue) were selected
based on COSMIC prevalence. All
fusions represented more than once in
COSMIC are covered.
2B
0.5
FFPE, FF, plasma and serum specimens used in this study were obtained through collaborations (NSCLC
FFPEs provided by MD Anderson Cancer Center) or acquired through tissue banks. Thyroid FNAs
were provided through collaborations or were residual clinical specimens from Asuragen’s CLIA lab.
Whole transcriptome (WT) RNA-Seq was performed on 68 thyroid FFPEs to discover novel mRNA and
fusion diagnostic markers. Targeted NGS Thyroid and NSCLC panels were developed using Asuragen’s
QuantideX NGS technology. Targeted NGS QC was performed with a novel qPCR assay that quantifies
functional DNA and RNA from TNA. PCR-based target enrichment was conducted using QuantideX
targeted NGS reagents and sequenced on a MiSeq® (Illumina). Library sequences were analyzed using
QuantideX NGS Reporter, a bioinformatic analysis suite that directly incorporates pre-analytical QC
information to improve the accuracy of variant calling, fusion detection and RNA quantification.
Total Nucleic
Acid Input
(TNA)
4-Point DNA
Standard Curve
FNA Benign
FFPE Malignant
FNA Malignant
Type
Biomarker
Specimen
-0.5
0.0
0.3
0.6
Figure 4. Multi-omic characterization of NSCLC FFPE specimens: 61 adenocarcinomas (AD) and 36 squamous (SQ). Inset table compares
study results for rearrangements and mutations with reference TCGA data. Specimen AD-54, positive for an EML4-ALK fusion, showed a strong
3’/5’ imbalance of ALK. Another specimen, AD-42, was negative for other mutations, but presented a high ALK imbalance ratio, suggestive of a
latent ALK fusion.
PC1 (22.5% of variance)
MOLECULAR READOUT
DNA mutations
Quant Reporter
FFPE Benign
Inhibition Reporter
Figure 2. Analysis of targeted RNA-Seq expression data. A) QuantideX targeted RNA-Seq is analytically concordant with WT RNA-Seq. WT
RNA-Seq expression (y-axis) relative to QuantideX targeted RNA-Seq (x-axis) for the 25 diagnostic mRNAs selected from the 68 FFPE WT RNASeq discovery cohort. B) PCA analysis of QuantideX targeted RNA-Seq expression distinguishes malignant specimens. Analysis of FFPEs (N=123)
and FNAs (N=65) reveals separation of benign and malignant specimens by PC1 (x-axis).
FFPE or FNA
DNA
100
% variant
QuantideX DNA Assay
Liquid Biopsy
Group
Biomarker
0.0
-0.3
QUANTIDEX NGS SYSTEM
SAMPLES
PC2 (15.5% of variance)
METHODS
88
68.1
55
50
0
NA
NA
1.4
CRC KRAS G12A
3*
NA
1.75
Sequencing and Bioinformatics
17.5
11
Melanoma BRAF V600E
Serum
• The interrogation of both DNA and RNA markers in thyroid cancer using QuantideX NGS
technology enabled 95% sensitivity for malignant thyroid FNA biopsies and improved the
interpretation of mutation-negative FNA specimens.
• A 187-marker DNA/RNA NGS panel for NSCLC reported mutations, RNA fusions, and mRNA
transcripts from 97 lung cancer specimens with results consistent with TCGA reference data.
• The described NGS technologies are versatile, offer high analytical sensitivity, and can be
applied to liquid biopsies to reveal low-abundance mutations in circulating tumor DNA.
RESULTS
Preliminary research data. The performance characteristics of this assay have not yet been established.
Presented at AMP 2015
Plasma
NA
CONCLUSIONS
Figure 1. The QuantideX NGS system is a streamlined workflow from QC to informatics that enables simultaneous quantification of DNA
point mutations, indels, structural variants, RNA expression and gene fusions from a total nucleic acid (TNA) isolated from low-input,
low-quality samples.
*Research Use Only – Not For Use In Diagnostic Procedures
Melanoma BRAF V600E
FF
4.45
*Interestingly, one CRC specimen had positive calls for PIK3CA E545K at ~1.8% in the plasma and serum and 3% variant (below threshold) in the
FFPE primary tumor. The FFPE tumor, however, was positive for PIK3CA H1047R, which was not detected in the biofluids, suggesting a latent
metastatic subclone positive for PIK3CA E545K.
Expression
markers
We present two unified DNA/RNA cancer panels developed using QuantideX NGS technology: 1) a
thyroid cancer panel, and 2) a NSCLC panel. The thyroid panel was evaluated with 123 FFPE thyroid
lesions and 65 FNAs. A diagnostic score based on RNA expression enabled accurate classification of
mutation negative samples. The lung panel was assessed with 97 NSCLC FFPE specimens. Evaluation
of matched FFPE, fresh-frozen (FF) and liquid biopsies using the same NGS technology but with pancancer content demonstrated detection of ctDNA mutations down to 1.4%.
NA
Figure 5. Four selected tumors with a detected DNA mutation in both the primary tissue and matched liquid biopsy. Of the 28 tumors
that were profiled using the QuantideX NGS Pan Cancer Panel*, all with a mutation detected at >66% abundance in the primary tissue also
demonstrated a matching positive in the liquid biopsy. “NA” in the plot indicates that the matched sample was not available.
Gene fusions
Targeted multiplex PCR and tagging
FNA
1.8
CRC PIK3CA E545K
FFPE
FFPE
53.6
Figure 3. Multi-omic characterization of 123 FFPEs and 65 FNAs. Mutation-negative FFPE and FNA specimens are accurately classified as
benign or malignant using the Malignancy Score. The Malignancy Score is a composite of the 25 diagnostic mRNAs obtained by training a
DLDA classifier on the 123 FFPE specimens under cross-validation. Classification AUC-ROC statistics for FFPE and FNA specimens were 0.94
and 0.91 respectively. Malignant FNAs were distinguished with 95% sensitivity and 61% specificity enabling accurate interpretation of mutationnegative FNAs.