Transcript Why Melanoma?

Question: Should genome sequencing of
multiple oncogenes surplant BRAF V600
mutation testing by an FDA approved test?
Answer: Yes
Jeffrey A Sosman MD
Ingram Chair for Cancer Research- Professor of Medicine,
Director, Melanoma Program
Why Melanoma?
• 2012- Target therapy
– Therapy for BRAFV600E melanoma
– Therapy for CKIT mutated melanoma (exon 11 mutations)
• Other BRAF mutations- V600K,M,R,D,E’ (20% of V600
mutations), L597 mutations
• Expansion in NRAS melanoma 15-20% of all
melanoma- targeted therapy in development
– MEK inhibitor+AKT inhibitor, MEK inhibitor+CDK4 inhibitor
• Expansion into alternate genes- NF1, MEK1, MEK2,
HRAS, CRAF (all components of MAP kinase pathway)
• RAC1, PPP6C, GRIN2A, targets?? Or modulating
• Other mutations which are activating same genes
Melanoma is Comprised of Clinically
Relevant Molecular Subsets
Arising from Skin
Without Chronic
Sun Damage
50% BRAF
20% NRAS
1-2%
KIT
Arising from Skin
With Chronic
Sun Damage
10% BRAF
10% NRAS
5% KIT
Arising from
Mucosal
Surfaces
5% BRAF
15% NRAS
20% KIT
Arising from
15% BRAF
Acral
15% NRAS
Surfaces
Curtin et al. NEJM 2005; Curtin et al. JCO 2006
15%
KIT
Goals of the VICC PCMI
• To establish ‘reflex’ testing of ‘common’ clinically
relevant genetic alterations in lung cancers and
melanomas.
• To develop a clinically-applicable high-throughput
molecular genotyping facility for ‘rarer’ genetic
variants.
• To develop bioinformatic algorithms to report
genetic results in the electronic medical record in
ways that are clinically useful for practicing
oncologists.
– Collaboration among Depts of Medicine, Pathology,
BioInformatics, and VICC
– Sounds simple, but…requires high level of
collaboration/coordination
43 Somatic Point Mutations in 6 Genes Relevant to
Targeted Therapy in Melanoma
BRAF
NRAS
Position
Position AA mutant
p.G12C
p.G12S
p.G12R
G12
p.G12V
p.G12A
p.G12D
p.G13A
p.G13V
G13
p.G13R
p.G13D
p.Q61E
p.Q61H
p.Q61H
p.Q61L
p.Q61L
Q61
p.Q61K
p.Q61P
p.Q61R
p.Q61R
Nucleotide mutant
c.34G>T
c.34G>A
c.34G>C
c.35G>T
c.35G>C
c.35G>A
c.38G>C
c.38G>T
c.37G>T
c.38G>A
c.181C>G
c. 183A>T
c.183A>C
c.182A>T
c.182_183AA>TG
c.181C>A
c.182A>C
c.182A>G
c.182_183AA>GG
GNAQ
Q209
p.Q209P
p.Q209L
p.Q209R
c.626A>C
c.626A>T
c.626A>G
V600
AA mutant
p.V600R
p.V600K
p.V600E
p.V600E
p.V600M
p.V600G
p.V600D
Nucleotide mutant
c.1798_1799GT>AG
c.1798_1799GT>AA
c.1799T>A
c.1799_1800TG>AA
c.1798G>A
c.1799T>G
c.1799_1800TG>AT
CTNNB1
S37
S45
p.S37F
p.S37Y
p.S45P
p.S45F
p.S45Y
c.110C>T
c.110C>A
c.133T>C
c.134C>T
c.134C>A
p.W557R
p.W557R
p.V559A
p.V559D
p.L576P
p.K642E
p.D816H
c.1669T>C
c.1669T>A
c.1676T>C
c.1676T>A
c.1727T>C
c.1924A>G
c.2446G>C
p.Q209P
p.Q209L
c.626A>C
c.626A>T
KIT
W557
V559
L576
K642
D816
GNA11
Q209
Fig 1B
NRAS_G13 (R)
38 G>A/T/C
BRAF _V600
1799T>A/G
BRAF_V600 (R)
1800 G>A/T
NRAS_Q61
182A>T/C/G
B-CAT_S45 (R)
133 T>C
BRAF _V600 (R)
1799 T> G/A
KIT_V559
1676 T>C/A
NRAS_G12 (R)
34G>A/T/C
B-CAT_S37
110C>A/G/T
KIT_W557
1669T>A/C
KIT_K642
1924A>G
NRAS_G12
35 G>T/C/A
NRAS_Q61
181C>A/G GNA11_Q209 (R)
626A/T/C
NRAS_G13 (R)
BRAF_V600
37G>T/C
1798G>A B-CATS45 (R)
134R C>A/T
KIT_L576
1727T>C
KIT_D816
2446G>C
NRAS_Q61 (R)
183 A>G/T/C
GNAQ_Q209
626A>T/C/G
First 150 Patients: 20% of BRAF V600 Mutations Would Have
Been Missed by Allele-Specific PCR
Lovly, Dahlman, Fohn, Su et al ‘12
First 150 Patients:
40% of Pts with Mutant Metastatic Disease  Genotype-Driven Treatment
Gene
BRAF
CTNNB1
GNAQ/GNA11
KIT
NRAS
No mutation detected
Total cases
# of metastatic cases
# patients placed on a
genotype-driven clinical trial (%)
32
12 (38%)
1*
1 (100%)
6
3 (50%)
1
1 (100%)
15
4 (27%)
28
N/A
82
21/54 (39%)
* This CTNNB1 mutation (CTNNB1 S45P) occurred concurrently with an NRAS Q61L mutation.
Lovly, Dahlman, Fohn, Su et al ‘12
Vanderbilt-Ingram Cancer Center
Melanoma SNaPshot genotyping in CLIA Lab
(652 samples, from Jul 2010 to June 2012)
Distribution of all mutations detected
Distribution of BRAF V600 mutations
Melanoma
7/1/2010-11/1/2012
• 759 Specimens
– 65% specimens with mutation detected
– 16 specimens with 2 mutations
– 3 specimens with 3 mutations
• 715 patients
– 64% patients with mutation detected
11
Vemurafenib
(PLX4032)
Overall survival (February 01, 2012 cut-off)
censored at crossover
100
Vemurafenib (n=337)
Median f/u 12.5 months
Overall survival (%)
90
80
Hazard ratio 0.70
70
(95% CI: 0.57–0.87)
p<0.001 (post-hoc)
60
50
Dacarbazine (n=338)
Median f/u 9.5 months
40
30
20
BRIM2
10
9.7
0
0
13.6
6
12
15.9
18
24
Time (months)
No. at risk
Dacarbazine
Vemurafenib
338
337
244
326
173
280
111
231
79
178
50
109
24
44
4
7
0
1
c-KIT Mutations in Melanoma
• 4q12
– Selectively amplified in acral/mucosal
– Candidate genes → c-Kit amplifications
→ point mutations
• C-Kit by Subtype
– Acral
– Mucosal
– Cutaneous
• +CSD
11% Mt 25% Amp
21% Mt 29% Amp
1-18% Mt 6% Amp
• C-Kit: Melanoma vs GIST
–
–
–
–
Point mutations
↑ Exon 13 & 17 mutations
Amplified wild-type c-KIT
Lack of 2ndary mutations
Woodman, BCP, 2010
Phase II Studies of Imatinib 400 mg BID
in Advanced Melanoma
Imatinib
kit
c-abl
PDGFR-α
PDGFR-β
Three “large” studies have been embarked upon
include both KIT mutated and amplified
Hodi- DFCC central with imatinib, sunitinib, or nilotinib for imatinib fail
Carvajal- MSKCC central with imatinib
Guo-. Peking Univ, Beijing, China- imatinib
. Treatment Response Over Time by Melanoma Subtype and Genetic Alteration of KIT
Carvajal, R. D. et al. JAMA 2011;305:2327-2334
Copyright restrictions may apply.
Kit Inhibition in Melanoma
 Kit Inhibitors can produce dramatic effects in patients
with melanomas containing a variety of C-kit mutations
 Kit mutations are seen in 2% of all melanomas
 Role of Kit inhibition in Kit amplified tumors has yet to
be established
 Multiple studies currently underway
 Imatinib, sunitinib, dasatinib, nilotinib
 International Phase II trial (nilotinib )- comp[leted
 Exciting, but not the answer for the majority of patients
with melanoma
Index Case: Using NGS to Find Novel Drivers
• 75 year old male presented with ulcerated right ear
melanoma  resected
• 4 mos later – local recurrence  re-resection and
radiation; BRAF V600E and KIT mutations not detected
• 12 mos later – widespread mets  palliative
thyroidectomy; no mutations detected by SNaPshot
• Whole genome sequencing performed on thyroid
metastasis (90% tumor) and matched normal blood
Dahlman, Xia, Hutchinson et al ‘12
Vanderbilt-Ingram Cancer Center
WGS Analysis of “Pan-Negative” Melanoma
GAIIx
Paired-end
SAMtools
Pindel
CREST
FREEC
Dahlman, Xia, Hutchinson et al ‘12
SNaPshot Limitation Example:
Melanoma Patient with BRAF L597 Mutation
• Melanoma SNaPshot Negative Patient
• Whole-genome sequencing  BRAF L597R
• Sensitive to MEK inhibition in vitro
Patient with BRAF L597S, treated with TAK-733
Dahlman, Xia, Hutchinson et al, Cancer Discov, 2012)
8% of “Pan-Negative” Samples Harbor non-V600E
BRAF Exon 15 Mutations
7/1/10-12/31/11
Cosmic: 0.1% of
BRAF mutations
Melanoma Panel: 538 Samples
Of 49: 2 L597,
1 D594, 1 K601
(8%)
Dahlman, Xia, Hutchinson et al ‘12
Vanderbilt-Ingram Cancer Center
Mutations in the BRAF gene
PRESENTED BY:
MEK 162: Best percentage change from baseline
and best overall response (NRAS mut)
45 mg NRAS
N=28*
Progressive Disease (PD)
Response rate: 21% (6 of 28
pts)
Disease control rate: 68%
Stable Disease (SD)
Partial Response (PR)
Unconfirmed PR
*Patients with missing best % change from baseline and unknown overall response are not included.
Ongoing pts
Ascierto, Berking, Agarwala et al. ASCO 2012
Actionable Mutations- MAPKinase Pathway
Summary
• Routine multiplex mutational profiling of
melanoma with a disease-specific panel
– Identifies patients with clinically relevant driver
mutations
– Enables genetically-informed cancer medicine in the
clinic
– Facilitates clinical trial enrollment
– Allows for rapid discovery of potentially targetable
novel drivers in ‘pan-negative’ cases
• BRAF L597 mutations and MEK inhibitors
Vanderbilt-Ingram Cancer Center
16 Cancers
ALL
ALCL
Basal Cell Carcinoma
Breast
Colorectal
Gastric
GIST
IMT
Lung
Medulloblastoma
Melanoma
Neuroblastoma
Ovarian
Rhabdomyosarcoma
Thymic
Thyroid
24 Genes
271 Disease-GeneVariant Relationships
Vanderbilt-Ingram Cancer Center
Vanderbilt-Ingram Cancer Center
More Comprehensive Profiling
with Illumina MiSeq
1000x read coverage
Automated Alignment & Analysis
Amplicon Target Enrichment
Illumina.com
“Vanderbilt Cancer Panel” for MiSeq
• Design: Illumina Design Studio
• Targets: All exons of 66 genes
Panel 1 = 34 genes
Panel 2 = 32 genes
Targets = 594 (exons)
Target bp = 195838 bp
# Amplicons = 1494 (max 1536)
Coverage = 95%
Low-Scoring Targets = 13
Targets = 457 (exons)
Target bp = 210570 bp
# Amplicons = 1448 (max 1536)
Coverage = 93%
Low-Scoring Targets = 13
AKT1
ALK
BRAF
CDK4
DDR2
EGFR
ERBB2
FGFR1
FGFR2
FGFR3
GNA11
GNAQ
AKT2
AKT3
ARAF
BCL2
BCL2L1
ERBB3
ERBB4
FGFR4
HRAS
JAK3
KDR
MCL1
MYC
MYCL1
MYCN
NOTCH1
NOTCH2
NOTCH3
NTRK3
PTCH1
PTCH2
RAF1
RB1
RET
SMAD4
STK11
TP53
JAK1
JAK2
NTRK1
NTRK2
IGF1R
IDH1
IDH2
KIT
KRAS
MAP2K1
MAP2K2
MET
MLH1
MLH3
MSH2
MTOR
NF1
NF2
NRAS
PDGFRA
PIK3CA
PTEN
RICTOR
RPTOR
SMO
TSC1
TSC2
Vanderbilt Cancer Panel Plans
• Validate with samples with known mutations:
– FFPE Patient Tissue
– Frozen Patient Tissue
– Cell Lines
• Expand to SNaPshot-negatives/unknowns
• Design a capture method/panel for fusion genes
• Now seeking interesting clinical samples! Please
contact me ([email protected])
• Implement into Clinical Molecular Diagnostics Lab???
(Cindy Vnencak-Jones)
The major issues critical to personalized
cancer care in melanoma
• Acquired resistance to BRAF inhibitors– mechanisms and overcoming resistance
• Targeting other mutations (NRAS) effectively with new or old
drugs
• Defining new genetic mutations, amplifications, or
translocations
• Need for both clinical and translational collaboration to speed
up the discoveries needed for clinical progress
• Transmitting genetic information to the oncologist in a clinically
relevant language