Massively Parallel Sequencing in NSCLC: Comparison to Traditional Hot Spot Analysis for Selection of Approved and Novel Targeted Therapies JS Ross, A Parker, M Jarosz, S Downing, R Yelensky, D Lipson, P Stephens, G Palmer, M Cronin, CE Sheehan Department of Pathology and Laboratory Medicine Albany Medical College Albany, NY Foundation Medicine, Inc. Cambridge, MA
26
Embed
Department of Pathology and Laboratory Medicine Albany Medical College Albany, NY
Massively Parallel Sequencing in NSCLC: Comparison to Traditional Hot Spot Analysis for Selection of Approved and Novel Targeted Therapies . JS Ross, A Parker, M Jarosz, S Downing, R Yelensky, D Lipson, P Stephens, G Palmer, M Cronin, CE Sheehan. Department of Pathology and Laboratory Medicine - PowerPoint PPT Presentation
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Massively Parallel Sequencing in NSCLC: Comparison to Traditional Hot Spot Analysis for Selection of Approved and Novel Targeted
Therapies
JS Ross, A Parker, M Jarosz, S Downing, R Yelensky, D Lipson, P Stephens, G Palmer,
M Cronin, CE Sheehan
Department of Pathology and Laboratory Medicine
Albany Medical CollegeAlbany, NY
Foundation Medicine, Inc.Cambridge, MA
Background (1)
• Next Generation DNA Sequencing (NGS) has recently been applied to FFPE cancer biopsies and major resections (Ross JS et al. J Clin Oncol 29: 2011)
• Current Hot-Spot Genotyping only detects:– Mutations restricted to specific exons and codons
• NGS detects:– Whole exome mutations in numerous cancer related genes– Insertions and deletions– Translocations and fusions– Copy number alterations (amplifications)
Background (2)• Recently, biomarker testing has emerged as a major driver of
the selection of therapy for non-small cell lung cancer (NSCLC)
• Currently, “hot-spot” DNA sequencing and FISH are used to select therapies for NSCLC:– EGFR genotyping for tyrosine kinase inhibitor (erlotinib)– EML4:ALK translocation testing for crizotinib
• The emergence of comprehensive genomic profiling by NGS has led investigators to question whether more thorough gene sequencing techniques could discover potential targets for the treatment of relapsed and metastatic NSCLC not currently searched for in current routine practice
Targeted Therapies for Cancer
Molecular profiling is driving many new targeted cancer therapeutics
Subset of analyzed targets listed; data from BioCentury Online Intelligence Database
• ~500 compounds hitting ~140 targets in development
• Growing number of newly identified potential targets
Design (1)
• DNA was extracted from 4 x 10 m FFPE sections from 49 primary NSCLC (28 female; 21 male; mean age 68 years; 24% Stage I; 13% Stage II; 5% Stage III; 16% Stage IV; 46% Stage unknown)
• The exons of 145 cancer-related genes were fully sequenced using the Illumina HiSeq 2000 (Illumina, San Diego, CA) and evaluated for point mutations, insertions/deletions (indels), specific genomic rearrangements and copy number alterations (CNA)
• A total of 606,676-bp content was sequenced and selected using solution phase hybridization, to an average coverage of 229×, with 84% of exons being sequenced at ≥100× coverage
• This assay captures and sequences 2,574 coding exons representing 145 cancer-relevant genes (genes that are associated with cancer-related pathways, targeted therapy or prognosis), plus 37 introns from 14 genes that are frequently rearranged in cancer
Design (2)
• To maximize mutation-detection sensitivity in heterogeneous NSCLC biopsies, the test was validated to detect base substitutions at a ≥10% mutant allele frequency with ≥99% sensitivity and to detect indels at a ≥20% mutant allele frequency with ≥95% sensitivity, with a false discovery rate of <1%
• Samples included 5% fluid cell blocks; 5% regional lymph nodes; 3% pericardial biopsy and 87% lung biopsies or resections
• There were 46 adenocarcinomas (34 acinar, 19 lepidic, 2 mucinous, 1 papillary), 1 large cell carcinoma, and 2 squamous cell carcinomas
• Results were compared with commercial laboratory allele-specific PCR genotyping on the same tissue blocks
Cancer Genome Profiling Workflow
<14-21 days
Increasing Coverage To 500x Allows For >99% Sensitivity To Detect Mutant Alleles >5%, With No False Positive Mutation Calls
Sensitivity vs Allele Frequency at 500X Coverage (1Mb panel)
0%
20%
40%
60%
80%
100%
Sens
itivi
ty (P
roba
bilit
y of
Fin
ding
Tru
e M
utati
ons
With
Zer
o Fa
lse
Posi
tives
)
Allele Frequency
0%
20%
40%
60%
80%
100%
Sens
itivi
ty (P
roba
bilit
y of
Fin
ding
Tru
e M
utati
ons
With
Zer
o Fa
lse
Positi
ves)
Allele Frequency
80X 0% Error
80X 0.5% Error
500X 0.5% Error
Deep coverage is required for clinical grade samples
5% 10%
Lower Coverage Misses Relevant Mutations
Mutant Allele frequency spectrum of known mutations found in a series of clinical samples
Fraction of mutations <5%
Fraction of mutations <10%
Fraction of mutations <20%
Fraction of mutations <25%
Fraction of mutations <50%
Fraction of mutations <100%
11% 32% 55% 67% 93% 100%
0-5%6-10%
11-15%
16-20%
21-25%
26-30%
31-35%
36-40%
41-45%
46-50%
51-55%
56-60%
61-65%
66-70%
71-75%
76-80%
81-85%
86-90%
91-95%
96-100%
0
5
10
15
20
25
30
35
40
Mutant Allele Frequency
Num
ber o
f Mut
ation
s
Genomic Alteration Categories
Highly Actionable“Page 1”
Actionable in Principle“Page 2”
Prognostic“Page 3”
Biologically Significant“Page 4”
Category A: Approved / standard alterations that predict sensitivity or resistance to approved / standard therapiesCategory B: Alterations that are inclusion or exclusion criteria for specific experimental therapies
Category C: Alterations with limited evidence that predict sensitivity or resistance to standard or experimental therapies
Category D: Alterations with prognostic or diagnostic utility
Category E: Alterations with clear biological significance in cancer (i.e. driver mutations) without clear clinical implications
Initial Cohort Results (1)• For EGFR status, the NGS result was concordant with commercial
laboratory genotyping in 23/23 (100%) cases
• In 22/23 (96%) NSCLC samples, NGS revealed 53 total genomic alterations, including :– 14 (64%) base substitutions– 2 (9%) INDELs– 6 (27%) CNA– 0 (0%) rearrangements
• Genomic alterations associated with sensitivity or resistance to targeted therapies for NSCLC were found in 16/22 (73%) of cases including: 10 KRAS 4 STK11 3 JAK2 2 PIK3CA11 BRAF 2 EGFR 1 NF1 1 TSC11 TSC2 1 CCNE1 1 PTCH 1 CDK41 CCND1 1 BRCA2 1 CDKN2A 1 ATM
•
Initial Cohort Results (2)• In comparison with the COSMIC database,
NGS results were similar for most genes except for– a lower rate of EGFR mutations (9% vs. 21%)– a higher rate of KRAS mutations (41% vs. 16%) – an unprecedented rate of JAK2 mutations (14% vs.
0%)• 7/22 (32%) of the NSCLC had 2 or more
potentially actionable alterations after NGS
NSCLC: Actionable Genomic AlterationsKR
AS
TP53
EGFR
STK1
1
LRP1
B
PIK3
CA
CTNN
B1 NF1
MDM
2
JAK2
DNM
T3A
CDKN
2A
ATM
TSC1
CCNE
1
BRAF
SMAR
CA4
SMAD
4
RUNX
1
RB1
PTPR
D
NOTC
H1
MYC
MSH
6
MAP
2K1
MLH
1
MCL
1
GNAS
FGFR
2
CDKN
2B
CDK4
BRCA
1
APC
0%
10%
20%
30%
40%
50%
60%
Tumor Type
Perc
enta
ge o
f Cas
es w
ith M
utati
on
Cetuximab/Panitum resist.
Tubulin
s.
Vemuraf
enib se
ns./Cetu
ximab
resis
t.
CDK inhibito
rs
MEK in
hibitors
(sens.
and re
sist.)
/Vem
urafen
ib resis
t.
Nutlins
MEK/E
RK inhibito
rs
CDK4/6 in
hibitors
mTOR in
hibitors
Dasati
nib
DNMT inhibito
rs
FGFR
inhib
PARP in
hibitors
Gefitinib, E
rlotinib, o
thers
PI3 kinas
e, mTO
R inhibito
rsJA
K2 inhibito
rs
mTOR in
hibitors
Notch in
hibitors
CDK4/6 in
hibitors
PARP in
hib
Genes with Actionable AlterationsGenes with Alterations, Actionability Unknown
Multiple ‘Potentially Actionable Alterations in a Single Sample
* Novel alterations discovered in tumor cell (somatic) sequence only as determined by comparison with the COSMIC database. Gene variants of undetermined significance which may represent germline variants are not included in this list.
NSCLC EGFR Activating Mutation
• Sample: SM58• Mutation: EGFR_c.2573T>G_p.L858R• Freq=32%, depth=53• 79 year old white female non-smoker• FNA of lung mass: NSCLC
• FNA sample cytocentrifuged and converted to an FFPE section• Very small numbers of viable tumor cells• Extensive tumor cell necrosis• Genotyping by allele-specific PCR showed identical activating EGFR mutation
Acquired Resistance to EGFR-TKI
By NGS, the resistance clone was seen in 6% of cells and the
sensitizing mutation in 25%
Sensitivity to gefitinib and erlotinib
Resistance to gefitinib and erlotinib
Nutlins
NSCLC: JAK2 Mutation Detected by NGS
• Sample: SM86• Mutation: JAK2_c.1849G>T_p.V617F• Freq=4%, depth=205• 77 year old white female• Lung adenocarcinoma diagnosed by pleural biopsy• Patient diagnosed with polycythemia vera
Low power of pleural biopsy positive for adenocarcinoma
High power view shows adenocarcinoma of the lung. Rare capillaries not blood filled. No nucleated RBC or blasts seen.
G T A T G T G T C T G T G G A
Val Cys Val Cys Gly
c.1849G>T p.V617F
Multiple CNAs in Adenosquamous Carcinoma• Sample: SM92• Mutation: CDK4 amp (6.6x), MDM2 amp (3.3x)• 77 year-old white male• Left lower lobe• Adenosquamous Carcinoma (composite tumor)• pT2 pN0 pMx
CDK4MDM2
Low power view of lobectomy
specimenHigh power view of tumor with adenocarcinoma glands to the left and squamous carcinoma
to the right
Conclusions• Deep sequencing (NGS) of clinical NSCLC samples is
completely concordant with traditional hot-spot genotyping
• NGS uncovers an unexpected number of genomic alterations that could influence therapy selection for NSCLC
• Broad-based, deep sequencing of cancer-related genes results in sensitive detection of all classes of genomic alterations in NSCLC and can reveal actionable genomic abnormalities that inform treatment decisions