David W. Salzman Yale School of Medicine Department of Therapeutic Radiology Fallopian Mucosa NSCLC Papillary Serous Ovarian Tumor Dicer mAb (13D6) miRNA-Target Site SNPs as Predictors of Cancer Risk and Treatment Response
Nov 28, 2014
David W. SalzmanYale School of Medicine
Department of Therapeutic Radiology
David W. SalzmanYale School of Medicine
Department of Therapeutic Radiology
Fallopian Mucosa NSCLC Papillary Serous Ovarian TumorD
ice
r m
Ab
(1
3D
6)
miRNA-Target Site SNPs as Predictors of Cancer Risk and Treatment Response
miRNA-Target Site SNPs as Predictors of Cancer Risk and Treatment Response
Need for new companion diagnostic markers in cancer therapyNeed for new companion diagnostic markers in cancer therapy
• Tumors contain a heterogeneous array of inherited and acquired mutations
• Best cure rates are achieved when specific drugs are used to target tumors with particular mutations (targeted therapeutics)
• Current strategies to identify targeted therapeutics rely heavily on the identification of tumor acquired mutations
• May only be represented in a small population of cells – elude identification• Short-term response is good, long-term response is poor due to drug resistance
Therefore: we need a new paradigm to identify companion diagnostics that do NOT rely on identifying tumor acquired mutations
but where do you find such mutations in the genome?
microRNA-Target Sites:uncharted territory to identify disease biomarkers
microRNA-Target Sites:uncharted territory to identify disease biomarkers
• The 3’UTR of mRNAs contain cis-regulatory elements that regulate the nature and timing of gene expression in conjunction with a requisite trans-acting factor
• MicroRNAs are a class of non-coding, trans-acting RNAs that negatively regulate gene expression by binding to complementary elements in the 3’UTR of a target mRNA
5'5' 3'3'AAAAAAAAAAAA
ORFORF5’UTR5’UTR 3’UTR3’UTR
ProteinProtein
miRNA-mRNA complementarity is required for target selectionmiRNA-mRNA complementarity is required for target selection
AGOAGO3'3' 5'5'
• Seed pairing = complementarity between nucleotides 2-7• ± complementarity of nucleotide 8• ± Adenosine residue opposite nucleotide 1• ± 3’ end complementarity
5'5' 3'3'AAAAAAAAAAAA
ORFORF5’UTR5’UTR 3’UTR3’UTR
ProteinProtein
3’ NNNNNNNNNNNNNNNNNNNNNN 5’ |||||||||| ||||||| 5’...NNNNNNNNNNNNNNNNNNNNNNNNNNNN...3’
3’ NNNNNNNNNNNNNNNNNNNNNN 5’ |||||||||| ||||||| 5’...NNNNNNNNNNNNNNNNNNNNNNNNNNNN...3’
miRNA-mRNA complementarity is required for target selectionmiRNA-mRNA complementarity is required for target selection
AGOAGO3'3' 5'5'
5'5' 3'3'AAAAAAAAAAAA
ORFORF5’UTR5’UTR 3’UTR3’UTR
ProteinProtein
Translation InhibitionTranslation Inhibition
XX
• Seed pairing = complementarity between nucleotides 2-7• ± complementarity of nucleotide 8• ± Adenosine residue opposite nucleotide 1• ± 3’ end complementarity
miRNA-mRNA complementarity is required for target selectionmiRNA-mRNA complementarity is required for target selection
AGOAGO3'3' 5'5'
• Centered pairing = complementarity between nucleotides (approx.) 6-16• Must include Watson-Crick base-pairing between nucleotides 9-12
5'5' 3'3'AAAAAAAAAAAA
ORFORF5’UTR5’UTR 3’UTR3’UTR
ProteinProtein
3’ NNNNNNNNNNNNNNNNNNNNNN 5’ |||||||||||||| 5’...NNNNNNNNNNNNNNNNNNNNNNNNNNNN...3’
3’ NNNNNNNNNNNNNNNNNNNNNN 5’ |||||||||||||| 5’...NNNNNNNNNNNNNNNNNNNNNNNNNNNN...3’
miRNA-mRNA complementarity is required for target selectionmiRNA-mRNA complementarity is required for target selection
5'5'3'3'
AAAAAAAAAAAA
ORFORF5’UTR5’UTR
3’UTR3’UTR
ProteinProtein
XX mRNA cleavagemRNA cleavage
• Centered pairing = complementarity between nucleotides (approx.) 6-16• Must include Watson-Crick base-pairing between nucleotides 9-12
Hypothesis: variants in microRNA-target sites can deregulate gene expression and result in cancer
Hypothesis: variants in microRNA-target sites can deregulate gene expression and result in cancer
XXSNPSNP
inhibit oncogene targeting
inhibit oncogene targeting
enhance (or lead to aberrant) targeting of
a tumor suppressor
enhance (or lead to aberrant) targeting of
a tumor suppressor
XXSNPSNP
over expression ofoncogene
over expression ofoncogene
under expression of tumor suppressor
under expression of tumor suppressor
Identification of a germline let-7 target site variant (rs61764370)Identification of a germline let-7 target site variant (rs61764370)
5' 3'AAAAAAKRAS5’UTR 3’UTR
1 9 2 34 5 610
78
LCS6 Genotype Tumor and NAT Tumor
TT 35 24
TG/GG 8 7
n=74 (NCSLC patients)
G-allele present in (approx.) 20% of lung cancer patients (otherwise KRAS ORF WT)
Chin et al, Cancer Research (2008)
• let-7 targets the KRAS 3’UTR• 10 predicted let-7 complementary sites
rs61764370 associates with cancer riskrs61764370 associates with cancer risk
Cancer Subtype (subgroup) GenotypeFold-increased risk
(OR, 95% CI; p-value) Reference
Lung Non-small cell lung cancer (NSCLC) TG/GG 2.3 (1.1-4.6; p=0.02) Chin et al, 2008
Ovarian Hereditary breast ovarian (HBOC) TG/GG 2.46 (1.14-5.29; p=0.02) Ratner et al, 2010
Breast Triple negative (ER-/PR-/Her2-) TG/GG 2.307 (1.261-4.219; p=0.0067) Paranjape et al, 2011
multivariate analysis – adjusted for age and race
• Over 40,000+ individuals studies worldwide• Represented in (approx.) 6% of world populations• More frequently associated with cancer in women• Associated with later onset for most patients
rs61764370 associates with poor OS in HNSSCrs61764370 associates with poor OS in HNSSC
Christensen et al, Carcinogenesis (2009)
TT (wild type)TG/GG (variant)n=344
All sites (oral, pharyngeal, laryngeal)HR (CI 95%): 1.6 (1.0-2.5, p=0.20)
TT (wild type)TG/GG (variant)n=190
Oral cancerHR (CI 95%): 2.7 (1.4-5.3, p=0.06)
• rs61764370 is present in (approx.) 17.5% of this HNSSC cohort• Treatment not detailed
rs61764370 associates with poor OS in ovarian cancerrs61764370 associates with poor OS in ovarian cancer
Ratner et al, Oncogene (2011)
Variable HR 95% CI p-value
KRAS mutation 1.671 1.087 - 2.568 0.0192
Age 1.025 1.002-1.049 0.0307
Stage 1.380 1.185-1.607 <0.0001
Grade 1.341 0.912-1.972 0.1360
Histology 0.970 0.900-1.045 0.4168
Center (Yale vs non-Yale) 1.868 1.438-2.427 <0.0001
TT (wild type)TG/GG (variant)n=279
Ovarian cancerHR (CI 95%): 1.671 (1.087-2.568, p=0.0192)
rs61764370 confers platinum resistance in ovarian cancerrs61764370 confers platinum resistance in ovarian cancer
Adapted from Ratner et al, Oncogene (2011)
Genotype Univariate Multivariate
OR 95% CI p-value OR 95% CI p-value
Wild type (TT) (n=225) 1.00 1.00
Variant (TG/GG) (n=66) 2.45 1.08-5.53 0.0313 3.18 1.31-7.72 0.0106
Multivariate analysis: adjusted for age, stage, grade, histology, residual disease after cytoreductive surgery and treatment center
Cell LineGemcitabin
e Doxorubicin Topotecan
BG1 (TG variant) 30.4uM 307.5nM 161.8nM
CAOV3 (TT wild type) 2.2nM 75.9nM 30.8nM
p=<0.0001
p=<0.04 (TG variant)(TT wild type)
(TG variant/BRCA1 MT)
1st line therapy 2nd line therapy
rs61764370 is sufficient to up-regulate KRAS gene expressionrs61764370 is sufficient to up-regulate KRAS gene expression
p=0.007
p=0.036TT TG
KRAS
Actin
TT TG
LungNormal
Lung1o Tumor
rs61764370 positive tumors display unique transcription patternsrs61764370 positive tumors display unique transcription patterns
Expression SignatureTT vs TG/GGTNBC Tumor pK-S Test
NRAS up 0.02
BRCA mutant-like up 0.04
Luminal Progenitor up 0.04
MAPK Creighton up 0.06
PCA Estrogen down 0.04
Adapted from Paranjape et al, Lancet Oncology (2010)
The let-7 family of microRNAs is also consistently and significantly down-regulated in rs61764370 positive (NSCLC, TNBC, ovarian and HNSSC) tumors
Adapted from Ratner et al, Oncogene (2011)
TTTG/GGn=10, p=0.095
TTTG/GGn=10, p=0.095
TTTG/GGn=10, p=0.05
Targeted therapeutics in the EGFR signaling pathwayTargeted therapeutics in the EGFR signaling pathway
EGFR
Proliferation
RASRAS
RAFRAF
MEKMEK
MAPKMAPK
anti-EGFRGefitinibErlotinib
Cetuximab
T790ML858R
G12DQ61L
V600Eanti-BRAFSorafenib
anti-MEKSelumetinibTrametinibAZD6244
rs61764370 associates with poor OS in mCRC patients undergoing cetuximab-irinotecan salvage therapy
rs61764370 associates with poor OS in mCRC patients undergoing cetuximab-irinotecan salvage therapy
anti-EGFRCetuximab
TT (wild type) n=100TG/GG (variant) n=34p=0.001
Graziano et al, Pharmacogenomics J. (2010)
*Patient cohort was otherwise KRAS ORF WT and BRAF ORF WT
EGFR
Proliferation
RASRAS
RAFRAF
MEKMEK
MAPKMAPK
rs61764370
The germline rs61764370 3’UTR variant phenocopies a tumor acquired KRAS ORF mutation
The germline rs61764370 3’UTR variant phenocopies a tumor acquired KRAS ORF mutation
KRASKRAS5' 3'
Tumor AcquiredKRAS ORF Mutation
X
X
Canceranti-EGFR Rx resistance
KRASKRAS5' 3'
Germ-lineKRAS rs61764370 T>G
X
X
Canceranti-EGFR Rx resistance
KRASKRAS5' 3'
WT KRAS
Normal anti-EGFR Rx sensitive
let-7 RISC let-7 RISClet-7 RISC
Future directions
Clinicopathology
Hypotheses
Genotype
Summary and Current Work FlowSummary and Current Work Flow
rs61764370rs61764370
Increased cancer riskIncreased cancer risk Altered Responseto Therapy
Altered Responseto Therapy
rs61764370 positive cells will display
enhanced cancer-associated phenotypes
rs61764370 positive cells will display
enhanced cancer-associated phenotypes
We can selectively target rs61764370
positive cells
We can selectively target rs61764370
positive cells
rs61764370 is a predictive biomarker to direct cancer therapy
rs61764370 is a predictive biomarker to direct cancer therapy
Clinical trialsClinical trials
Isogenic cell lines
High throughput screening of FDA-
approved compounds
High throughput screening of FDA-
approved compounds
Cell biology: Transformation, growth, Mobility, Invation, EMT
Cell biology: Transformation, growth, Mobility, Invation, EMT
Generation of Isogenic KRAST/T and KRAST/G Cell Lines(workflow)
Generation of Isogenic KRAST/T and KRAST/G Cell Lines(workflow)
Obtain cell line from NCI (KRAST/T)
Transfect with•zinc-finger plasmids (x2)•donor plasmid
A B C3 2 1
{
(3) zinc-fingers/nuclease(highly specific DNA binding)
bidentate nucleases(dsDNA cleavage)
KRAS 3’UTR
dsDNA cleavage
DNA repair
Homologousrecombination
Donor(mutant KRAS 3’UTR)
A B C3 2 1
DNA binding
Generation of Isogenic KRAST/T and KRAST/G Cell Lines(workflow)
Generation of Isogenic KRAST/T and KRAST/G Cell Lines(workflow)
Obtain cell line from NCI (KRAST/T)
Transfect with•zinc-finger plasmids (x2)•donor plasmid
Single cell clone
Expand
1 2 3 4 5 6 7 8 9 10 11 12
A
B
C
D
E
F
G
H
1 2 3 4 5 6 7 8 9 10 11 12
A
B
C
D
E
F
G
H
X XXX X
Exclude wells w/ >1 cell
Expand
Extract gDNA
Generation of Isogenic KRAST/T and KRAST/G Cell Lines(workflow)
Generation of Isogenic KRAST/T and KRAST/G Cell Lines(workflow)
Obtain cell line from NCI (KRAST/T)
Transfect with•zinc-finger plasmids (x2)•donor plasmid
Single cell clone
Expand
Screen for rs61764370 insertion into KRAS 3’UTR
ID KRAST/G
•sequence verify•expand•store
TaqMan genotype for rs61764370 (ID positive clones)
PCR amplify KRAS 3’UTR(1kb +/- donor sequence)
Topo clone PCR amplicon
TaqMan genotype bacterial colonies for rs61764370
Cell Line Cell type rs61764370 (T:G)
Cal27 (+ control) Lung 12:16
MCF10a-luc Parental Normal breast epithelial 12:0
MCF10a-luc Isogenic Normal breast epithelial 15:15
HCC1937 Parental TNBC (BRCA1-/-) 12:0
HCC1937 Isogenic TNBC (BRCA1-/-) 22:19
H1299 Parental Lung (P53-/-) 10:0
H1299 Isogenic Lung (P53-/-) 15:16
If KRAST/G then allele frequency = 50:50 (T:G)
14 cell lines failed to make isogenic pairs2isogenic cell line generation is ongoing process
anal
ysis
of
posi
tiona
l ins
ertio
n
Sequence Verification of MCF10a Isogenic Cell LinesSequence Verification of MCF10a Isogenic Cell Lines
ReverseSequence
Reads
Allele-1 Allele-2
MCF10a WT (KRAS 3’UTRT/T)
ForwardSequence
Reads
‘T-allele’Allele-1
‘G-allele’Allele-2
MCF10a MT (KRAS 3’UTRT/G)
Allele-1 A T C
Allele-2 T TC
Allele-1 A T C
Allele-2 G TC
* * * *
** *
*
rs612587 rs61764370 rs2966 rs612587 rs61764370 rs2966
MCF10a rs61764370 positive cells senesce in cultureand display a mesenchymal-like morphology
MCF10a rs61764370 positive cells senesce in cultureand display a mesenchymal-like morphology
Knock-in of rs61764370 into MCF10a cells causedan epithelial-to-mesenchymal transition (EMT)
Knock-in of rs61764370 into MCF10a cells causedan epithelial-to-mesenchymal transition (EMT)
Knock-in of rs61764370 into MCF10a caused a mild growth defectKnock-in of rs61764370 into MCF10a caused a mild growth defect
0
0.2
0.4
0.6
0.8
1
1.2
0 20000 40000 60000 80000 100000
Cell number
Absorbance
MCF10a WT MCF10a MT
Sequence verification of HCC1937 isogenic cell linesSequence verification of HCC1937 isogenic cell lines
‘T-allele’Allele-1
‘G-allele’Allele-2
HCC1937 MT (KRAS 3’UTRT/G)
ReverseSequence
Reads
Allele-1 Allele-2
HCC1937 WT (KRAS 3’UTRT/T)
ForwardSequence
Reads
Allele-1 T C
Allele-2 T TC
T-Allele T C
G-Allele G TC
CC
* ** *
*
***
rs612587 rs61764370 rs2966 rs612587 rs61764370 rs2966
HCC1937 rs61764370 positive cells display altered platting efficiency and cell growth
HCC1937 rs61764370 positive cells display altered platting efficiency and cell growth
2D
Nu
mb
er o
f co
lon
ies
HCC1937 WT
HCC1937 MT
HCC1937 WT
HCC1937 MT
100
3D
Nu
mb
er o
f co
lon
ies
HCC1937 WT
HCC1937 MT
HCC1937 WT HCC1937 MT
MEK inhibitors target HCC1937 rs61764370 positive cellsMEK inhibitors target HCC1937 rs61764370 positive cells
anti-MEKSelumetinibTrametinibAZD6244
HCC1937 MT (TG)
HCC1937 WT (TT)
EGFR
Proliferation
RASRAS
RAFRAF
MEKMEK
MAPKMAPK
rs61764370
BATTLE1: Biomarker-integrated Approaches of Targeted Therapy of Lung Cancer Elimination
BATTLE1: Biomarker-integrated Approaches of Targeted Therapy of Lung Cancer Elimination
0-wt (Event/N = 32/32)1-Variant (Event/N = 4/4)p=0.001
(Erlotinib)
0-wt (Event/N = 39/61)1-Variant (Event/N = 3/8)p=0.056
(Sorafenib)
anti-EGFRErlotinib
anti-BRAFSorafenib
EGFR
Proliferation
RASRAS
RAFRAF
MEKMEK
MAPKMAPK
rs61764370
AcknowledgementsAcknowledgements
The Weidhaas Lab
CollaboratorsFrank Slack, Roy Herbst – Yale UniversityNicola Miller, Michael Kerin – National University of IrelandKim Smits, Manon van England - Maastricht UniversityJeffrey Weitzel – City of HopeRob Pilarski – OHSUKen Offit – MSKCCChristine Chung – JHMISabine Tejpar – LeuvenXifeng Wu, Hai Tran - MDACC
FundingNIH/NCI RO1NIH K08Yale Cancer CenterMary K. Ashe FoundationShanon FoundationRTOG Seed GrantsCT State Funding