SUMMARY• The QuantideX® NGS RNA Lung Cancer Kit* is a research tool for analyzing recurrent
gene fusions, MET exon 14 skipping and gene expression in lung cancer specimens. • The kit demonstrated concordant results across two laboratories using 30 total
nucleic acid (TNA) isolates derived from clinically-relevant specimens such as FFPE tumor biopsies.
• Agreement was also achieved for 8 residual clinical FFPE samples comparing the QuantideX NGS RNA Lung Cancer Kit* and FISH.
• Results comparing the QuantideX NGS RNA Lung Cancer Kit* and the nCounter® Vantage™ Lung Gene Fusion Panel were in agreement for both the original cohort and residual FFPE samples.
INDEPENDENT EVALUATION OF A FULLY INTEGRATED NEXT-GENERATION SEQUENCING TECHNOLOGY FOR THE ACCURATE DETECTION OF ONCOGENIC RNA FUSIONS AND ABERRANT SPLICING EVENTS IN LUNG CANCERRichard Blidner1, Brian C Haynes1, Andrew Hadd1, Shobha Gokul1, Maria L Aguirre2, Gary J Latham1, and Léon C van Kempen2
1Asuragen, Inc., Austin, TX, 2The Molecular Pathology Centre, Jewish General Hospital, Montreal, QC, Canada
INTRODUCTIONRNA fusions and splice variants, such as MET exon 14 skipping, are recognized as important therapeutic targets in non-small cell lung cancer (NSCLC) and a growing number of other solid tumors. Despite the emerging importance of these targets to cancer research, NGS assays that analyze RNA markers currently lag behind DNA sequencing efforts in workflow efficiency and in the rigor of analytical performance evaluations. We present a new targeted RNA-Seq Lung Cancer assay that can report 107 known oncogenic gene fusions, MET exon 14 skipping, 26 expression markers, and 5 imbalance markers for novel fusion detection using a streamlined sample-to-answer workflow.
MATERIALS AND METHODSThe QuantideX NGS RNA Lung Cancer Kit* was evaluated at Asuragen (Austin, TX) and Jewish General Hospital (Montreal, QC, Canada). A set of 30 samples derived from banked tumors and cell lines plus two controls were analyzed at Asuragen using the MiSeq® System (Illumina) and used over three study phases. An additional set of FFPE-derived residual clinical research samples that were previously annotated by FISH were evaluated in the final study phase. A subset of samples were also evaluated using the nCounter Vantage Lung Gene Fusion Panel (NanoString Technologies; The Journal of Molecular Diagnostics (2014) 16(2):229-43).
CONCLUSIONS• An independent laboratory was trained to proficiency in two days using the QuantideX
NGS RNA Lung Cancer Kit*.• The kit workflow, from sample to sequencer, can be completed in about 9 hours.• High inter-operator, and intra- and inter-run reproducibility, was achieved at two
different sites with FFPE samples using inputs down to <10 ng. • NGS fusion results were in 100% agreement with ALK and ROS1 translocation calls
from orthogonal assays performed with 23 FFPE samples.Figure 1. QuantideX NGS RNA Lung Cancer Kit* content. The panel covers 107 recurrent gene fusions including ALK, RET and ROS1, MET ex14 skipping, 23 mRNA markers of prognostic and theranostic value and 3 internal control mRNA markers.
Figure 2. Overview of QuantideX NGS RNA Lung Cancer Kit* from wet lab to dry bench analytics. The workflow is designed to minimize the number of steps, operator hands-on-time, and overall turn-around-time. Time-Motion Analysis shows both operator hands-on-time and instrument time. The bundled QuantideX® Reporter software package runs locally on a standard computer.
Table 1. Jewish General Hospital NGS results compared to Asuragen sample annotations. Out of 64 libraries performed at Jewish General Hospital, 63 yielded the expected calls both for mutation and imbalance status (after resolving contamination events in reproducibility set 2). One sample failed analysis, which was identified at the QC step (not shown).
Table 2. Summary of QuantideX NGS RNA Lung Cancer Kit* FFPE sample agreement with orthogonal methods. Samples identified as fusion positive by either FISH (8 residual clinical samples) or NanoString (8 residual clinical samples + 15 Asuragen subset samples) demonstrated 100% agreement for fusion-positive calls. Imbalances for two samples (*) were not called but were near the positive threshold. One sample (†) was near threshold for an ALK imbalance for both NanoString and NGS technologies.
* Research Use Only – Not For Use In Diagnostic Procedures Preliminary research data. The performance characteristics of this assay have not yet been established. Presented at AMP 2016 - S19
10
100
1,000
10,000
100,000
ERL0
1ER
L02
ERL0
3ER
L04
ERL0
5ER
L06
ERL0
7ER
L08
ERL0
9ER
L10
ERL1
1ER
L12
ERL1
3ER
L14
ERL1
5ER
L16
ERL1
7ER
L18
ERL1
9ER
L20
ERL2
1ER
L22
ERL2
3ER
L24
ERL2
5ER
L26
ERL2
7ER
L28
ERL2
9ER
L30
NTC
PosC
trl
Func
tion
al C
opie
s
Functional Copies Input per Library PreparationPhase II: Concordance
AsuragenJewish General
125 ng…….…..9 ng
lleCLines
FFPE Titration
400
200
10,000
100,000
1,000,000
1,000
100
ERL0
1
ERL0
2
ERL0
7
ERL0
8
ERL0
9
ERL1
1
ERL1
2
ERL1
3
ERL1
5
ERL1
6
ERL1
7
ERL1
8
ERL1
9
ERL2
0
Cell Lines FFPE Titration
10
Num
ber
Rea
ds
Pass
ing
Filt
er
Coverage of Positive TargetPhase II: Concordance
AsuragenJewish General
ALK
ALK
ALK
ALK
ALK
ALK
ALK A
LK
ALK
ALK
ALK
MET
e13
:e15
FGFR
3
ROS1
RESULTS
Figure 4. Quantification of both the RNA functional copy number and the on-target coverage for fusions and splice events was consistent between sites. A) Comparison of the functional copy input measured using the QuantideX RNA Assay (Lung). Results were consistent between sites (1.4-fold mean difference with a standard deviation of 0.5-fold). B) Comparison of the on-target coverage for each of the positive samples. Results were consistent between sites (1.1-fold mean difference with a standard deviation of 0.9-fold). Even coverage across titration series libraries demonstrated successful normalization and pooling.
4A 4B
Panel Content*
*Content sourced from: NCCN Guidelines, customer needs, COSMIC, Clinicaltrials.gov, publications & other databases.
mRNA Expression Targets
ALKROS1RET
NTRK1PDGFRA
3'/5' Imbalance 3' Fusion Genes # of Fusions ALK 53 ROS1 22 RET 12 FGFR3 7 NTRK3 3 NTRK1 4 NRG1 2 FGFR1 1 FGFR2 1 MBIP 1 PDGFRA 1
11 fusion genes–107 known and relevant fusion products
Exon Skipping EventMET e13:e14MET e14:e15MET e13:e15
ABCB1BRCA1
CD274 (PDL1)
CDKN2ACTLA4ERCC1
ESR1FGFR1FGFR2
IFNGRISG15MET
MSLNPDCD1
PDCD1LG2 (PDL2)
PTENRRM1TDP1
TERTTLE3TOP1
TUBB3TYMS
Endogenous Ctrls.
Figure 3. Summary of the site evaluation study design. The study was designed with three phases for NGS evaluation: Training (8 libraries), Proficiency & Concordance (32 libraries), and Reproducibility & Method Concordance (16 libraries in duplicate plus additional evaluator libraries). Specifically, Jewish General Hospital was trained within two days, and they achieved a successful proficiency run in the second phase (Table 1). A subset of samples were evaluated by NanoString’s Vantage Lung Gene Fusion Panel. Eight additional samples that were previously annotated by FISH were evaluated using both systems (Table 2). Analysis of mRNA expression was not part of this evaluation.
Core 8 Samples
Training
2 control samples (non-template control and a sample harboring a synthetic fusion target)
2 fusion-positive samples derived from 2 unique FFPE specimens
32 Concordance Samples(with Core 8)
16 Reproducibility Samples(1/2 of Concordance + 8 Residual Clinical Samples)
Pro�ciency/Concordance
13 Fusion Positive Samples (ALK, ROS1, FGFR3) derived from 5 unique FFPE specimens and 2 cell lines
1 cMET exon 14 aberrant spice event (cell line-derived)
6-point series of a fusion positive sample titrated down to ~9 ng mass input (Qubit)
Reproducibility/Method Comparison
7 Fusion Positive Samples (ALK, ROS1, FGFR3) derived from 5 unique FFPE specimens and 2 cell lines
1 cMET exon 14 aberrant spice event (cell line-derived)
8 FISH annotated samples derived from residual clinical FFPEs
Time-Motion Analysis
• 8 Libraries• 9 Hours from Sample to Sequencer
SAMPLEQC
TARGETENRICHMENT
LIBRARYPURIFICATION
LIBRARYQT
NORM/POOLING
TOTALRT
ASURAGEN 12min
17min
28min
10min
20min
20min
107min
60min
90min
195min
0min
90min
0min
435min
NGS Reporter
Analytics and Reporting
RT CopyNumber Readout
Gene-Speci�cPCR
Fusion PosCtrl
Barcode Tagging
(Dual index)
SequencingPrimers Includedfor Post-Library
Puri�cation
qPCR LibraryQuanti�cation
Bead LibraryPuri�cation
Push-Button Analyticsand Reporting Suite
Sequencing
Sample QC cDNA Quanti�cation
Single-Tube Gene-Speci�c PCR Library Prep
Sample Indexing and Sequencing Primers
Library Puri�cationand Quanti�cation
Reverse Transcription
ReverseTranscription
RNA Assay(Lung)
NGS Lung Cancer Panel
NGS Codes NGS Library PurePrep & Quant
Asuragen Annotations Concordance (32) Reproducibility Set 1 (16) Reproducibility Set 2 (16)
Sample ID Source Mutation Status Imbalance Status
Expected Fusion
Expected Imbalance Met e13:e15 Expected
FusionExpected Imbalance Met e13:e15 Expected
FusionExpected Imbalance Met e13:e15
NTC Control Neg Neg none none none none none none none none none
Pos Control Synthetic ALK Neg ✓ none none ✓ none none ✓ none none
ERL1 FFPE EML4:ALK ALK ✓ ✓ none ✓ ✓ none ✓† ✓ none
ERL2 FFPE EML4:ALK ALK ✓ ✓ none ✓ ✓ none Failed Failed Failed
ERL3 FFPE Neg Neg none none none none none none none none none
ERL4 FFPE Neg Neg none none none none none none none none none
ERL5 FFPE Neg Neg none none none none none none none none none
ERL6 FFPE Neg Neg none none none none none none none none none
ERL7 FFPE EML4:ALK ALK ✓ ✓ none ✓ ✓ none ✓† ✓ none
ERL8 FFPE EML4:ALK ALK ✓ ✓ none ✓ ✓ none ✓† ✓ none
ERL9 FFPE EML4:ALK ALK ✓ ✓ none ✓ ✓ none ✓† ✓ none
ERL10 FFPE Neg Neg none none none none none none none none none
ERL11 Cell MET e13:e15 Neg none none ✓ none none ✓ none none ✓
ERL12 Cell FGFR3:TACC3 Neg ✓ none none ✓ none none ✓ none none
ERL13 Cell SLC32A2:ROS1 ROS1 ✓ ✓ none ✓ ✓ none ✓ ✓ none
ERL14 FFPE Neg Neg none none none
Expected fusions, imbalance and splice variant calls were accurately called in 63 of 64 libraries
One library, ERL2, failed both QC and NGSfor Reproducibility Set 2
†Samples had low coverage false positive contaminants that were resolved based on plate position
ERL15 FFPE EML4:ALK ALK ✓ ✓ none
ERL16 FFPE EML4:ALK ALK ✓ ✓ none
ERL17 FFPE EML4:ALK ALK ✓ ✓ none
ERL18 FFPE EML4:ALK ALK ✓ ✓ none
ERL19 FFPE EML4:ALK ALK ✓ ✓ none
ERL20 FFPE EML4:ALK ALK ✓ ✓ none
ERL21 FFPE Neg Neg none none none
ERL22 FFPE Neg Neg none none none
ERL23 FFPE Neg Neg none none none
ERL24 FFPE Neg Neg none none none
ERL25 FFPE Neg Neg none none none
ERL26 FFPE Neg Neg none none none
ERL27 FFPE Neg Neg none none none
ERL28 FFPE Neg Neg none none none
ERL29 FFPE Neg Neg none none none
ERL30 FFPE Neg Neg none none none
Concordance with Orthogonal Test Methods
Sample ID FISH NanoString Vantage Calls Asuragen QuantideX Fusion Calls Fusion Concordance Asuragen QuantideXImbalance Calls Imbalance Concordance
ERL2 Neg EML4(13):ALK(20) EML4(13):ALK(20) ✓ ALK ✓
ERL7 Neg EML4(6):ALK(20) EML4(6):ALK(20) ✓ ALK ✓
ERL10 Neg Neg Neg - Neg -
ERL13 Neg SLC34A2(4):ROS1(32) SLC34A2(4):ROS1(32) ✓ ROS1 ✓
ERL14 Neg Neg Neg - Neg -
ERL15 Neg EML4(6):ALK(20) EML4(6):ALK(20) ✓ ALK ✓
ERL16 Neg EML4(6):ALK(20) EML4(6):ALK(20) ✓ ALK ✓
ERL17 Neg EML4(6):ALK(20) EML4(6):ALK(20) ✓ ALK ✓
ERL18 Neg EML4(6):ALK(20) EML4(6):ALK(20) ✓ ALK ✓
ERL19 Neg EML4(6):ALK(20) EML4(6):ALK(20) ✓ ALK ✓
ERL20 Neg unknown ALK-fusion EML4(6):ALK(20) ✓ ALK ✓
ERL21 Neg Neg Neg - Neg -
ERL22 Neg Neg Neg - Neg -
ERL23 Neg Neg Neg - Neg -
ERL24 Neg Neg Neg - Neg -
JGH1 Neg Neg Neg - Neg -
JGH2 Neg Neg† Neg - Neg† -
JGH3 ROS1 split SLC34A2(4):ROS1(32) SLC34A2(4):ROS1(32) ✓ Neg* X
JGH4 Neg Neg Neg - Neg -
JGH5 Neg Neg Neg - Neg -
JGH6 Neg Neg Neg - Neg -
JGH7 ROS1 split EZR(10)-ROS1(34) EZR(10)-ROS1(34) ✓ ROS1 ✓
JGH8 ALK split EML4(6):ALK(20) EML4(6):ALK(20) ✓ Neg* X