1
Notes:
Notes:
Comprehensive Genomic Profiling to Optimize Precision‐Focused Cancer Management
A Year 2016 and Beyond Science‐to‐Best Practice Cancer Update and Scientific Exchange Meeting for the Asia‐Based Oncology,
Genomic, and Hematology Specialist
The Evidence‐Based Journey from Targets to Therapy
Welcome and Introduction
Professor Nir Peled, MD, PhD ‐ Program ChairHead, Thoracic Cancer Unit and Center for Precision Medicine
Davidoff Cancer CenterTel Aviv UniversityTel Aviv, Israel
2
Notes:
Notes:
Presenter Disclosure
Advisor for and honorarium from AZ, BI, BMS, Lilly , MSD, Novartis, Pfizer, Roche
Audience Response System
Getting to Know You and Your Practice Patterns
What is your practice specialty within the field of oncology?
1) General medical oncology
2) Organ specific medical oncology
3) Both medical and radiation oncology
4) Para‐medical (bioinformatics/lab/scientist)
5) Onco‐pathology
Please Enter Your Response On Your Keypad
3
Notes:
Notes:
How many years have you been in oncology practice?
1) Up to 3 years
2) 3‐5 years
3) 5‐10 years
4) 10‐20 years
5) 20 years or more
Please Enter Your Response On Your Keypad
Audience Response System
Getting to Know You and Your Practice Patterns
What percentage of your cancer patients do you treat with targeted therapy (i.e. precision‐focused cancer medicine)?
1) Less than 10%
2) 10% ‐ 20%
3) 20% ‐ 30%
4) 30% ‐ 50%
5) 50% ‐ 75%
6) 75% ‐ 100%
Please Enter Your Response On Your Keypad
Audience Response System
Getting to Know You and Your Practice Patterns
4
Notes:
Notes:
In your overall clinical experience, what percentage of your cancer patients do you feel have an actionable, druggable mutation?
1) Less than 10%
2) 10% ‐ 20%
3) 20% ‐ 30%
4) 30% ‐ 50%
5) 50% ‐ 75%
6) 75% ‐ 100%
Please Enter Your Response On Your Keypad
Audience Response System
Getting to Know You and Your Practice Patterns
Your targeted cancer therapy is most often based on:
1) Immunohistochemistry
2) PCR‐based mutation detection
3) FISH
4) I am not sure
Please Enter Your Response On Your Keypad
Audience Response System
Getting to Know You and Your Practice Patterns
5
Notes:
Notes:
In your view, the main advantage of Hybrid Capture‐Based, Next Generation Sequencing (HC‐NGS) over “standard” NGS is:
1) HC‐NGS requires less tissue
2) HC‐NGS is mainly good for common mutations
3) HC‐NGS is much faster
4) None of the above
5) I don’t know
Please Enter Your Response On Your Keypad
Audience Response System
Getting to Know You and Your Practice Patterns
8:10 AM –8:30 AM ‐Current and Evolving Technologies for Comprehensive Genomic Profiling in CancerDr. Brendan Pang , MBBS, FRCPath(UK)
8:30 AM – 8:50 AM ‐ Next‐Generation Sequencing Approaches for Understanding the Genetic Basis of Cancers for Personalized Medicine: A Review of Available Technologies and Methods
Professor Maria Li Lung, PhD
9:10 AM – 9:30 AM ‐ Treatment of Patients with Advanced NSCLC Based on Genomic Analysis of the Tumor Professor James CH Yang, MD, PhD
9:30 AM – 9:50 AM ‐ The Clinical Implications of Using Hybrid Capturing NGS in Lung CancerProfessor Nir Peled, MD, PhD, FCCP – Program Chair
10:10 AM – 10:30 AM ‐ The Role of Genomic Profiling to Optimize Clinical Outcomes in Patients with Breast Cancer
Professor Rebecca Dent, MD, FRCP
11:00 AM – 11:20 AM ‐ Comprehensive Genomic Profiling of Gynaecological Cancers to Optimize TherapyDr David SP Tan, BSc(Hons), MBBS(Hons)(London), MRCP(UK)(Medical Oncology), PhD(London)
11:30 AM – 11:50 AM ‐ Comprehensive Genomic Profiling for Gastrointestinal CancersDr. Wong Seng Weng, MBBS, MRCP(UK), FAMS (Medical Oncology)
Scientific Program –Morning Agenda
6
Notes:
Notes:
1:00 PM – 2:40 PM Interactive Mini‐Roundtables with Faculty Moderators: Participants and Faculty Analyze and Discuss Clinical Cases in Cancer‐Focused Precision Medicine
1:00 PM – 1:25 PM Case Management Session: Applying NGS and Actionable Genomic Alterations to Patients with Lung Cancer: Professor Peled
1:25 PM – 1:50 PM Case Management Session: Applying NGS and Actionable Genomic Alterations to Patients with Breast Cancer: Professor Dent
1:50 PM – 2:15 PM Case Management Session: Applying NGS and Actionable Genomic Alterations to Patients with Tumor of Unknown Etiology: Professors Lung and Pang
2:15 PM – 2:40 PM Case Management Session: Applying NGS and Actionable Genomic Alterations to Patients with Lung Cancer: Professor Yang
Scientific Program Agenda –Afternoon Roundtable Sessions
Germline Variation Somatic Mutation Personalized Care
Precision Medicine: Science Serving Patients
7
Notes:
Notes:
Cancer Genomics
Genetic subtypes of glioblastoma, gastric and other cancers identified
via TCGA research
2005
Hallmarks of Cancer published
2000 2010‐2014
• Advent of the “precision medicine” era
• But cancer’s biology is far more complex than we had imagined
Image Source: TCGA
The Cancer Genome Atlas (TCGA) launches
Cancer Genomics
Genomic Landscape of 5,000 Human Cancers
Source: MacConaill, L, et. al., J Mol Diagn 2014, 16: 660-672
8
Notes:
Notes:
Common Cancers Now Collections of Rare Cancers
Catherine B. Meador et al. Clin Cancer Res 2014;20:2264-2275
EGFR
KRAS
Unknown
What Impact Does Molecular Information Have? Evolving Clinical Practice Through Disease Biology
Unknown
MET Splice SiteMET Amplification
KRAS
NRAS
ROS1 Fusions
RET Fusions
EGFR
ALK Fusions
HER2BRAF
PIK3CA AKT1MAP2K1
Molecular profiling has changed the classification of lung cancer
Today, there are many known genomic alterations that can drive
the development of cancer...
2004 2015
Modified and updated from Pao and Hutchinson (2012) Chipping away at the lung cancer genome. Nature Medicine 18(3):349‐51.
A decade ago, genomic alterations were important in only around
1/3 of NSCLC cases
NSCLC: Non‐small cell lung cancer
9
Notes:
Notes:
What Impact Does Molecular Information Have? Evolving Clinical Practice Through Disease Biology
Molecular profiling has changed the classification of lung cancer
Trastuzumab, Afatinib
Vemurafenib, Dabrafenib
Crizotinib, Ceritinib
Erlotinib, Afatinib
Cabozantinib
Trametinib
2015
Modified and updated from Pao and Hutchinson (2012) Chipping away at the lung cancer genome. Nature Medicine 18(3):349-51.
Crizotinib
Crizotinib
…and many new targeted therapy options to treat patients harboring these genomic alterations
Unknown
MET Splice SiteMET Amplification
KRAS
NRAS
ROS1 Fusions
RET Fusions
EGFR
ALK Fusions
HER2BRAF
PIK3CA AKT1
MAP2K1
Cell 2011 144, 646‐674DOI: (10.1016/j.cell.2011.02.013)
Hallmarks of Cancer: Therapeutic Implications
10
Notes:
Notes:
Comprehensive Genomic ProfilingUnderstanding the Diagnostic Difference
Number of Targeted Therapeutics is RisingKnowing Which Tests to Order is the Challenge
FBXW7ROS1 KRAS
RET
VEGF/VEGFR
AURKA
CDK4
CCND1ERBB3
DDR2
DNMT3A
GNAQ
BRCA1
BRAF
CDK6
AKT1
TSC1/2
METNOTCH1
TSC2
PIK3CANF1
FLT3
CDKN2A
PTEN
HER2
KDR
GATA3RAF1
IGF1R
ALK TNF
STK11
IGF/IGFR
FGFR1
MAP2K1Year
2005 2012 2015 2020
~15 approved drugs hitting ~10 targets
TodayComing Soon
~700 compounds targeting~150 targets in development
2025IDH1/2
Extrapolated from BioCentury Online Intelligence Database
Target Markers
11
Notes:
Notes:
Current Testing ModelsMultiple Diagnostic Tests Exhaust Tissue and Resources
Single assays, «tissue is the issue» Comprehensive tumour analysis...
...and continuous monitoring
RNASequencing
DNASequencing
Blood DNA Sequencing
Imaging
8‐10 slides
Lung tissue biopsy
Enough tissue for only 2‐3 individual tests
Produces a single snapshot
8‐10 slides
Example: Lung cancer
Routine Single‐Marker Molecular TestThe Most Common Type of Molecular Testing
CATEGORY ONE
CATEGORY TWO
CATEGORY THREE
Routine single marker molecular tests such as IHC, PCR and FISH that have been used for decades and will continue to play an important role in cancer diagnosis
missedmissed
foundmissed
missed
FISH: fluorescence in situ hybridization; IHC: Immunohistochemistry; PCR: Polymerase chain reaction
12
Notes:
Notes:
Multi‐Gene “Hot Spot” TestBroader Testing Focused On a Narrow Subset Of Genes
CATEGORY ONE
CATEGORY TWO
CATEGORY THREE
The hot spot NGS panels identify pre‐specified mutations occurring in very limited areas of genes of interest and fail to detect all classes of genomic alterations
missed foundmissed
missedfound
Hybrid CaptureNGS‐Based, Comprehensive Genomic ProfilingThe Most Comprehensive Genomic Test Available
CATEGORY ONE
CATEGORY TWO
CATEGORY THREE
Hybrid capture NGS‐based comprehensivegenomic profiling approach of testing all of the known clinically relevant cancer genes for all classes of alterations
foundfound
foundfound
found
Hybrid Capture NGS‐Based, Comprehensive Genomic Profiling
13
Notes:
Notes:
The Rise of Targeted Therapy
Precision Medicine
• Cancers classified by molecular abnormalities and site of origin
• Exceptional success when treatment is matched to a driver mutation
20011998
First targeted drug: rituximab
Trastuzumabintroduced for HER2+ breast
cancer Imatinibintroduced
1997‐2016
100+ FDA‐approved targeted cancer drug
indications
1997
Image Sources: Slamon D, et al. Engl J Med 2001; 344:783‐792; NCI; FDA
14
Notes:
Notes:
Timeline of Selected Major Discoveries in Lung Cancer Treatment
Source: Katerina Politi, and Roy S. Herbst Clin Cancer Res 2015;21:2213-2220
Precision Medicine
Photographs were taken:
A. Before initiation of vemurafenib
B. After 15 weeks of therapywith vemurafenib
C. At relapse, after 23 weeks of therapy.
Source: Wagle, N et al. Dissecting Therapeutic Resistance to RAF Inhibition in Melanoma by Tumor Genomic Profiling. JCO August 1, 2011 vol. 29 no. 22 3085‐3096
But precision medicine has brought new complexity – and
challenges
15
Notes:
Notes:
The Range, Specificity, and Sensitivity of Genomic‐Focused Technologies Available for Identifying Molecular Drivers for Targeted Cancer Therapy
Hot Spot Panels, FISH, and Hybrid Capture‐Based Comprehensive Genomic Profiling—What is the Gold Standard? Why?
Dr. Brendan Pang , MBBS, FRCPath (UK)
Consultant, Department of Pathology Laboratory (Section) Director, Diagnostic Molecular Oncology Centre (DMOC) National University Hospital (NUH) Clinical Assistant Professor Clinician Tract, Department of Pathology Yong Loo Lin School of Medicine National
University Singapore (NUS) Singapore
I have received honorariums and/or speaker fees for consulting, speaking, and advisory boards from:
MSD, Roche, AstraZeneca
Presenter Disclosure
16
Notes:
Notes:
Milestones of DNA Sequencing Technology
Watson and Crick
Source: www.esciencecentral.org
AB370A: sequence 96 samples simultaneously, 500 kb per day
1950s Discovery of DNA replication mechanism
1970s Invention of Sanger DNA sequencing
1987 Invention of automated DNA
sequencer
Next-Generation
Sequencing
Comprehensive Genomic Profiling
1869 Discovery of DNA
1920-50s Discovery of DNA structure
Overview of NGS Comprehensive Genomic Profiling
Hanahan D and Weinberg RA, Cell 144, 2011
1. Library preparation
2. Cluster generation
3. Sequencing
4. Reference mapping
5. Variant calling
6. Bioinformatics analysis
DNA sample preparation
EGFRINHIBITORS
Cyclin-dependentKinase inhibitors
TelomeraseInhibitors
Immune activatingAnti-CTLA4 mAb
Selective anti-inflammatory drugs
Inhibitors of HGF/c-Met
Inhibitors of VEGF signaling
PARPinhibitors
ProapoptoticBH3 mimetics
Aerobic glycolysisinhibitors
ActivatingInvasion&
metastasisi
Inducingangiogenesis
GenimeInstability&mutation
Resistingcell
death
Deregulatingcelluar
energetice
SustainingProliferative
signaling
Evadinggrowth
suppressors
AvoidingImmune
destruction
EnablingReplicativeimmortality
Tumor-Promoting
inflammation
NGS Comprehensive
Genomic Profiling
17
Notes:
Notes:
Reference Mapping to Convert Raw Data to Meaningful DNA Sequence
Reference genome sequence
Sequencing results in a library of short DNA sequences
4. Reference mapping
5. Variant calling
Identify Abnormalities in the DNA Sequence (Variant Calling)
• Due to tumor heterogeneity, not all cancer cells contain every variant. Bioinformatics to identify true variants and separate from sequencing artifact.
• Variant annotation: identify variants that are clinically relevant.
Reference genome sequence
6. Bioinformatics analysis5. Variant calling
18
Notes:
Notes:
Number of Targeted Therapeutics is RisingKnowing Which Tests to Order is the Challenge
FBXW7ROS1 KRAS
RET
VEGF/VEGFR
AURKA
CDK4
CCND1ERBB3
DDR2
DNMT3A
GNAQ
BRCA1
BRAF
CDK6
AKT1
TSC1/2
METNOTCH1
TSC2
PIK3CANF1
FLT3
CDKN2A
PTEN
HER2
KDR
GATA3RAF1
IGF1R
ALK TNF
STK11
IGF/IGFR
FGFR1
MAP2K1Year
2005 2012 2015 2020
~15 approved drugs hitting ~10 targets
TodayComing Soon
~700 compounds targeting~150 targets in development
2025IDH1/2
Extrapolated from BioCentury Online Intelligence Database
Target Markers
Current Testing ModelsMultiple Diagnostic Tests Exhaust Tissue and Resources
Single assays, «tissue is the issue» Comprehensive tumour analysis...
...and continuous monitoring
RNASequencing
DNASequencing
Blood DNA Sequencing
Imaging
8‐10 slides
Lung tissue biopsy
Enough tissue for only 2‐3 individual tests
Produces a single snapshot
8‐10 slides
Example: Lung cancer
19
Notes:
Notes:
NCCN Guidelines for NSCLC Version 4.2016, p16
The NCCN NSCLC Guidelines Panel strongly endorses broader molecular profiling with the goal of identifying rare driver mutations for which effective drugs may already be available, or to appropriately counsel patients regarding the availability of clinical trials. Broad
molecular profiling is a key component of the improvement of care of patients with NSCLC.
Histologic Subtype Testing
Adapted from NCCN NSCLC v. 4.2016, summary slide NSCL‐16 "Systematic Therapy for Metastatic Disease“p29. Used with permission.
‐EGFRmutation testing (category 1)
‐ALK testing (category 1)
‐EGFR and ALK should be conducted as part of broad
molecular profiling
‐Consider EGFRmutation and ALK testing especially in never
smokers or small biopsy specimens, or mixed histology
‐EGFR and ALK should be conducted as part of broad
molecular profiling
Metastatic Disease:‐Establish histological
subtype with adequate tissue for molecular
testing, consider rebiopsyif appropriate)
‐Integrate palliative care (see guidelines)
‐Adenocarcinoma‐Large cell‐NSCLC NOS
‐Squamous Cell CA
20
Notes:
Notes:
Sensitive EGFR Mutation Testing – Single Gene Tests
MethodAnalytical selectivity
Coverage of
mutationsKey Reference for Method Equipment TAT
Sanger DNA Sequencing coupled withmacro/micro‐dissection
5‐10%(~ 20% without macro‐
dissection). Macro‐dissection is highly recommended for samples with less than 50% tumour
>99% (this method can detect novel changes)
•Eberhard DA et al. J Clin Oncol; 26 (6): 983‐993 2008•Ellison G, et al. J Exp Clin Cancer Res; 6(29)132 2010•Molina‐Vila, M, et al Journal of Thoracic Oncology: 3 (11) 1224‐1235, 2008•Macrodissection: EGFR Mutation expert training: EGFR‐mutation.com (video)
DNA sequencer 2‐3 days
Fragment length analysis
~ 5% >99% of indelsonly
•Molina‐Vila, M, et al Journal of Thoracic Oncology: 3(11) 1224‐1235, 2008
Capillary electrophoresis/ DNA sequencer
1 day
PNA LNA Clamp ~ 1% 90‐95% (detects 29 target mutations)
•Nagai Y et al. Cancer Res; 65, 7276‐7282 2005•Miyazawa, H, et al. Cancer Science, 99(3), 595‐600 2008•Rosell, R, et al N Engl J Med;361:958‐67 2009
Real‐time PCR machine
~3 hours
Cycleave ~ 1% 90‐95% •Yatabe Y, et al. J Mol Diagn; 8: 335‐341 2006•http://www.takarabioeurope.com/cpt.html
Real‐time PCR machine
~1 day
Invader ~ 1% 90‐95% •Hall JG et al. Proc Natl Acad Sci U S A; 97: 8272‐8277 2000•Naoki K, et al. Int J Clin Oncol 2011 (preview)
PCR machine, Fluorescent reader
~1 day
Pyrosequencing 5‐10% 90‐95% •Takano, T et al. Journal of Clinical Oncology. 23(28):6829‐37, 2005 •Dufort, S, et al. Journal of Experimental & Clinical Cancer Research, 30:57 2011
Pyrosequencer 1 day
Issues with allele specific RT‐PCR
• Cobas seemed to
suggest an exon
19 Deletion in
sample
• Patient did not
respond as hoped
to first line TKI
• Sanger ‐Due to a
cluster of 6 SNPs
at the deletion
start point which
may have led to a
false positive
signal PCR
21
Notes:
Notes:
Uncertain and/or false negative direct sequences benefit from verification by a more sensitive platform
Exon 20 insertion/deletion mutations are typically not sensitive to erlotinib, gefitinib, and afatinib
Chromosomal re‐arrangements of ALK are present in3% to 7% of NSCLC
• The resulting ALK
fusions, such as
EML4‐ALK, function
as potent
oncogenic drivers
and lead to a state
of oncogene
addiction
22
Notes:
Notes:
ALK Fluorescence In Situ Hybridization
2p23ALK Dual Color, Break Apart FISH assay, Abbott Molecular
normal3’ 5’
Rearrangement positive ‐ split Rearrangement positive ‐ single 3’ ALK
Inversion with EML4‐ALK fusion
23
Notes:
Notes:
ALK IHC for screening:Alectinibphase I/II study in JapanAF‐001JP
In contrast to the
earlier studies which
used ALK fluorescence
in situ hybridization
(FISH) only, patients
were identified as
ALK‐positive using ALK
immunohistochemistr
y (IHC), followed by
ALK FISH for
confirmation.
Paik et al, JTO 2011
Novocastra clone 5A4
Yi et al, JTO 2011
DAKO clone ALK1
Mino‐Kenudson et al., CCR 2010
Cell Signaling clone D5F3
An International Interpretation Study Using the ALK IHC Antibody D5F3 and a Sensitive Detection Kit Demonstrates High Concordance between ALK IHC and ALK FISH and between EvaluatorsWynes, Murry W. PhD*; Sholl, Lynette M. MD†; Dietel, Manfred MD‡; Schuuring, Ed PhD§; Tsao, Ming S. MD, FRCPC‖; Yatabe, Yasushi ME, PhD¶; Tubbs, Raymond R. DO#; Hirsch, Fred R. MD, PhD***††
Overall for the 100 evaluable cases the ALK IHC assay was highly sensitive (90%), specific (95%), and accurate relative (93%) to the ALK FISH results. Similar results were observed using a majority score.Journal of Thoracic Oncology:May 2014 - Volume 9 - Issue 5 - p 631–638
24
Notes:
Notes:
48 yomale patient• 3.2 x 4.4 cm
spiculated mass in the R lower lobe of the lung
• Liver metastases
• Large necrotic lymph nodes in the right retrocrural, upper abdominal (pericoeliac, retrocaval and portocaval, paracaval), retroperitoneal and also mesenteric lymph nodes are seen, surrounding the portal confluence, celiac axis and left renal vein, and indenting the IVC.
G1202R mutation has been reported to confer resistance to alectinib
Alectinib
reportedly
efficacious in 4
crizotinib‐
resistant ALK
mutations
L1196M, F1174L,
R1275Q and
C1156Y
25
Notes:
Notes:
Brigatinib has reduced susceptibility to ALK mutations compared with other first- and/or second-generation ALK TKIs in vitro.
Sen Zhang et al. Clin Cancer Res 2016;22:5527-5538
©2016 by American Association for Cancer Research
The Future – For a Single Patient…
We will test for many members of a molecular pathway
We will test for biomarkers associated with different drugs, to be given at different time points
This will require a different molecular diagnostic approach
TUMOR
TUMORRESISTANCE 1
TUMORRESISTANCE 2
(…) (…)
Drug 1
Drug 2
Drug 3
26
Notes:
Notes:
Routine Single‐Marker Molecular TestThe Most Common Type of Molecular Testing
CATEGORY ONE
CATEGORY TWO
CATEGORY THREE
Routine single marker molecular tests such as IHC, PCR and FISH that have been used for decades and will continue to play an important role in cancer diagnosis
missedmissed
foundmissed
missed
FISH: fluorescence in situ hybridization; IHC: Immunohistochemistry; PCR: Polymerase chain reaction
Multi‐Gene “Hot Spot” TestBroader Testing Focused On a Narrow Subset Of Genes
CATEGORY ONE
CATEGORY TWO
CATEGORY THREE
The hot spot NGS panels identify pre‐specified mutations occurring in very limited areas of genes of interest and fail to detect all classes of genomic alterations
missed foundmissed
missedfound
27
Notes:
Notes:
Hybrid CaptureNGS‐Based, Comprehensive Genomic ProfilingThe Most Comprehensive Genomic Test Available
CATEGORY ONE
CATEGORY TWO
CATEGORY THREE
Hybrid capture NGS‐based comprehensivegenomic profiling approach of testing all of the known clinically relevant cancer genes for all classes of alterations
foundfound
foundfound
found
Hybrid Capture NGS‐Based, Comprehensive Genomic Profiling
Hybrid CaptureNGS‐Based, Comprehensive Genomic Profiling DeliveringActionable Insights From the First Page
Patient and ordering physician information
Targeted therapies and clinical trials that may be relevant based on genomic alterations
identified
Summary of results and genomic alterations
identified
28
Notes:
Notes:
29
Notes:
Notes:
High performance was achieved for both high‐level amplifications (copy number ≥ 8) and homozygous deletions when tumor purity was as low as 30%: sensitivity was 99% (91/92) with PPV > 99% (127/127).
Performance was reduced for lower CNAs (6–7 copies) and at lower sample purities (20–30%), with overall sensitivity >80%.
Our results demonstrate that an optimized NGS‐based test can accurately detect most clinically actionable CNAs in a broad spectrum of patient specimens.
These results also highlight the
scope for further improvements in this methodology, including the robust detection of heterozygous loss.
Copy Number Alterations (CNAs)‐ Foundation
30
Notes:
Notes:
Durable Response to Crizotinib in a Patient withMET‐Amplified Carcinoma of Unknown Primary
59‐year‐old female
60‐pack‐year smoking history
New‐onset seizures
MRI identified 2 distinct masses; right frontal lobe of brain and left mid‐abdominalmass
Real‐time PCR demonstrated KRASmutation, FISH analysis negative for ALK rearrangement
Palma, N.A., et al (2014) Durable Response to Crizotinib in a MET-Amplified, KRAS-Mutated Carcinoma of Unknown Primary. Case Rep Oncol 7(2):503–8.
Patient Profile
Durable Response to Crizotinib in a Patient withMET‐Amplified Carcinoma of Unknown Primary
Adenocarcinoma of unknown primary with brain metastases
Disease progression after 4 cycles of carboplatin and docetaxel
FISH: Fluorescence in situ hybridization; MRI: Magnetic resonance imaging; PCR: Polymerase chain reaction
Palma, N.A., et al (2014) Durable Response to Crizotinib in a MET-Amplified, KRAS-Mutated Carcinoma of Unknown Primary. Case Rep Oncol 7(2):503–8.
Diagnosis
Treatment Status
31
Notes:
Notes:
Normalization of TumorMetabolic Activity Following MET Targeting with Crizotinib
Identification of genomic alterations in MET, CCND1, MYC, KRAS, TP53 and CARD11
Treatment with crizotinib was initiated
18F-FDG PET/CT: 18F-fluorodeoxyglucose positron emission tomography
Genomic position
Genomic profiling by NGS of resected brain specimenMET amplification
Lo
g2(
test
/ref
)
2.0
1.5
1.0
1 2
0.5
0.0
-0.5
-1.0
-1.5
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 x
Co
py n
um
ber
0
2
4
6
8
10
12
14
16
Palma, N.A., et al (2014) Durable Response to Crizotinib in a MET-Amplified, KRAS-Mutated Carcinoma of Unknown Primary. Case Rep Oncol 7(2):503–8.
Genomic Profiling and Subsequent Treatment
Normalization of TumorMetabolic Activity Following MET Targeting with Crizotinib
Complete normalization of tumor metabolic activity for > 19 months of crizotinib treatment
18F-FDG PET/CT: 18F-fluorodeoxyglucose positron emission tomography
18F‐FDG PET/CT fusion of the transaxial left mid‐abdominal mesenteric mass 1 month prior (*) and 3 months (**) after starting crizotinib
***
NGS comprehensive genomic profiling identified a MET amplification. The patient exhibited a sustained response to subsequent treatment
with a multi‐kinase inhibitor that targets MET, ROS1 and ALK
Palma, N.A., et al (2014) Durable Response to Crizotinib in a MET-Amplified, KRAS-Mutated Carcinoma of Unknown Primary. Case Rep Oncol 7(2):503–8.
32
Notes:
Notes:
ActivatingInvasion&
metastasisiInducing
angiogenesis
GenimeInstability&mutation
Resistingcell
death
Deregulatingcelluar
energetice
SustainingProliferative
signaling
Evadinggrowth
suppressors
EnablingReplicativeimmortality
Tumor‐Promoting
inflammation
AvoidingImmune
destruction
Immune Checkpoint Therapy
Hallmarks of Cancer
Signaling Pathways and Intervention of Immune Checkpoints in Tumor Microenvironment
Source: thasso.com, accessed in August 2016Pardoll DM, Nature Reviews Cancer 12, 252-264 (April 2012)
Molecule Company Target
Ipilimumab BMS CTLA4
Tremelimumab AstraZeneca CTLA4
Nivolumab BMS PD‐1
Pembrolizumab Merck PD‐1
Atezolizumab Roche PD‐L1
Durvalumab AstraZeneca PD‐L1
Avelumab Merck/Pifzer PD‐L1
33
Notes:
Notes:
MSI Status as Prognostic Indicator That Correlates to Higher TMB
• MSI is a condition of genetic hypermutability that is caused by a deficiency in DNA mismatch repair (MMR) in the tumor
• MSI is associated with better prognosis in multiple cancer types1‐3
Cancer Genome Atlas Network, Nature. 2012 Jul 18;487(7407):330-7.
1. Arzimanoglou II et al., Cancer. 15;82(10):1808-20 (1998)2. Zhu L et al., Mol. Clin. Oncol. 3, 699-705 (2015)3. Guastadisegni C et al., Eur. J. Cancer 46(15), 2788-2798 (2010)
Tumor Mutational Burden (TMB) and MSI are Biomarkers for Response to PD‐1/PD‐L1 Checkpoint Blockade
Le et al., NEJM, 2015
Snyder et al., NEJM, 2014
High TMB in Melanoma MSI‐High in Colorectal Cancer
34
Notes:
Notes:
TMB Measured for 315 CGP Test Genes Versus Whole Exome – TCGA Data
R²=0.9763
0
100
200
300
400
500
0 5,000 10,000 15,000
Mutationcount–
targetedtest
Mutationcount– wholeexome
1
10
100
1 10 100 1,000 10,000
Mutationcount–
targetedtest
Mutationcount– wholeexome
35 TCGA studies7,000 specimens
Cancer Res 2016;76(14 Suppl):Abstract nr 2629
Association of TMB With AtezolizumabEfficacy in 2L+ PD‐L1–Selected Patients From BIRCH and FIR
Kowanetz, M. et al. in Annual ESMO congress (Copenhagen, Denmark 2016). Poster #77P
35
Notes:
Notes:
Next‐Generation Sequencing Approaches to Understanding the Genetic Basis of Cancers
for Personalized Medicine
Maria Li Lung, PhD
Department of Clinical Oncology, University of Hong Kong
Center for Nasopharyngeal Carcinoma Research
OncoSeek Limited, Hong Kong Science Park
Singapore Scientific Exchange Meeting on Cancer Genetics
December 15, 2016
Presenter Disclosure
Non‐paid consultant for OncoSeek working on CTCs
36
Notes:
Notes:
EGFR
KRAS
Unknown
What Impact Does Molecular Information Have? Evolving Clinical Practice Through Disease Biology
Unknown
MET Splice SiteMET Amplification
KRAS
NRAS
ROS1 Fusions
RET Fusions
EGFR
ALK Fusions
HER2BRAF
PIK3CA AKT1MAP2K1
Molecular profiling has changed the classification of lung cancer
Today, there are many known genomic alterations that can drive
the development of cancer...
2004 2015
Modified and updated from Pao and Hutchinson (2012) Chipping away at the lung cancer genome. Nature Medicine 18(3):349‐51.
A decade ago, genomic alterations were important in only around
1/3 of NSCLC cases
NSCLC: Non‐small cell lung cancer
What Impact Does Molecular Information Have? Evolving Clinical Practice Through Disease Biology
Molecular profiling has changed the classification of lung cancer
Trastuzumab, Afatinib
Vemurafenib, Dabrafenib
Crizotinib, Ceritinib
Erlotinib, Afatinib
Cabozantinib
Trametinib
2015
Modified and updated from Pao and Hutchinson (2012) Chipping away at the lung cancer genome. Nature Medicine 18(3):349-51.
Crizotinib
Crizotinib
…and many new targeted therapy options to treat patients harboring these genomic alterations
Unknown
MET Splice SiteMET Amplification
KRAS
NRAS
ROS1 Fusions
RET Fusions
EGFR
ALK Fusions
HER2BRAF
PIK3CA AKT1
MAP2K1
37
Notes:
Notes:
Applications of NGS
Critical Reviews in Oncology/Hematology 96 (2015) 463–474
NGS Description Advantages Disadvantages
Genomesequencing
Determines the sequence of most of the DNA from the individual’s genome
• Provides information on non‐coding regions and structural variants as well as coding regions
• Expensive and time consuming
• Data can be more difficult to interpret
• Challenges of what to do about incidental findings
Exome sequencing
Ex: germline CPGs somatic tumor mutations
Determines the sequence of the protein –coding DNA regions(exons)
• Faster and cheaper than genome sequencing • The majority of known pathological abnormalities occur in the exome
• Functional consequences of variants are more easily understood
• Misses variations in non‐coding regions and some structural variants
• Challenges of what to do about incidental findings
Targeted panels
Ex: NGS of CTCs
Determines the sequence ofspecific genes or parts of genes
• Usually cheaper than exome or genome sequencing, but this depends on the size of the gene panel
• Focused on particular regions of interest and so data interpretation is easier
• No concern regarding incidental findings as only the regions of interest are sequenced
• Can optimize the panel to capture problematic regions that are difficult to sequence using exomeor genome approaches
• Does not provide information on regionsoutside of the gene panel
Nasopharyngeal Carcinoma (NPC)
Cancer associated with genetic, dietary/environmental factors, and EBV infection
Common cancer in SE Asia, nicknamed the “Guangdong cancer”
Salted fish
Located in center of head with lymphatic spread
38
Notes:
Notes:
NPC differs from many other cancers, since patients are relatively young at diagnosis and may survive many years after treatment. QoL
issues important!Sources from: Chien & Chen, 2003 and Hong Kong Cancer Registry, 2009
Relative Frequency of Five Most Common Cancers in Males by Age Group in 2012
% ofAges 20‐44 #/yr all cancersNasopharynx 130 17 Colorectum 84 11Liver 69 9Lung 58 7.6Testis 50 6.5
% of Ages 45‐64 #/yr all cancersLung 894 17.9Colorectum 880 17.6Liver 654 13.1Nasopharynx 370 7.4Prostate 323 6.5
Tumor Development
Year 0 Year 5
Inherited Risk
DysplasiaCa in situ
Invasion Angiogenesis
Detection Threshold
Onset of metastasis
Diagnosis period
Treatment period
Invasion prevention/ intervention window
Time
Tu
mor
Siz
e
Normal Hyperplasia Carcinoma in situ
Invasion angiogenesis Metastasis
Copyright / illustrated by Panda
39
Notes:
Notes:
NPC Genetic Predisposition?
Age
Peak age: 45‐55 years
Early‐age onset cases: diagnosed with NPC at ≤ 20 years
Family clustering
5‐10% of NPC patients have a history of one or more affected family members
RR 12.8 for individuals with family history of NPC, compared to general population in endemic regions
Migrant studies
Offspring of migrants from endemic regions often show increased risk of NPC in low‐risk regions
Source: World Cancer Report 2008Zeng, Y. X., & Jia, W. H. (2002). Familial nasopharyngeal carcinoma. Sem Cancer BiolNg WT et al. (2005). Screening for family members of patients with nasopharyngeal carcinoma. IJC
Genetic Susceptibility
Family history
positive NPC
Early age onset NPC
FH+67 individuals from 56 families; 9 families with more than one affected individual sequenced, Southern Chinese
EAO39 NPC patients who have been diagnosed with NPC at or younger than 20, Southern Chinese4 cases have FH+ NPC
NPC Genetic Susceptibility
Sporadic NPC
Non‐cancer controls
Sporadic59 NPC patients who do not have family history
of NPC, Southern Chinese
Control cohorts895 Southern Chinese mainly from HK
Exome sequencing was performed using Illumina HiSeq (TruSeq capture)
Study Cohorts in Discovery Stage
Hypothesis: Important genes associated with NPC genetic susceptibility may be uncovered in EAO cases and FH+ cases because NPC occurs earlier in such patients compared to sporadic cases
40
Notes:
Notes:
Top CPG: Macrophage‐Stimulating 1 Receptor (MST1R)
• MST1R = RON. Maps to chr 3p21.3. Encodes a cell surface receptor with tyrosine kinase activity (Ronsin et al. Oncogene, 1993)
• Expressed in the tissue‐resident macrophages and functions to maintain inflammation homeostasis (Correll et al. J Leukoc Biol, 2004)
• Ligand: Macrophage‐Stimulating Receptor (MSP), encoded by MST1, also located in chromosome 3p21.3 (Wang et al. Science, 1994)
• SEMA domain harbors ligand‐binding pocket for MSP
• Two transcripts of MST1R initiated by 2 promoters, 2 CpG islands suggested to be molecular switch for expression of two transcripts, short isoform associated with aggressive tumor
• The MSP‐MST1R signaling has been implicated in several tumorigenic processes including cell proliferation, survival, migration, invasion, and angiogenesis (Yao et al. Nat Rev Cancer, 2013)
• Expressed in the ciliated epithelial cells in the nasal cavity. Activation of MST1R increases ciliary beat frequency, which moves the fluid over the surface of the epithelial cells and prevents chronic infection in the nasal cavity (Sakamoto et al. J Clin Invest ,1997)
MST1RDeleterious Variants in NPC
In total, we identified 11 rare deleterious variants (10 missense and 1 frameshift deletion) in NPC cases, accounting for 17.8% EAO cases, 7.1% FH+ cases and 5.1% sporadic cases
Dai et al, PNAS 113: 3317‐22, 2016
41
Notes:
Notes:
Somatic Landscape StudiesDiscovery WES specimens: 51 primary, 8 recurrent, 3 LN tumors
Validation targeted gene sequencing specimens: 73 primary tumors
Lymph node
metastaticSource: Cancer Research UK
T1 T2 T3 T4
Identify driver genes/pathways
Whole exome sequencing
Targeted resequencing
DNA methylome profiling
Transcriptome profiling
Prognosis and diagnosis
Patient stratification
New therapy
Landscape of genetic and epigenetic alterations
Explore new therapeutic targets
Chip‐seq analysis
Molecular basis of tumorigenesis
Mutation Landscape
Zheng et al, PNAS 2016
42
Notes:
Notes:
TP53Mutations in NPC
TP53 is the most frequently mutated gene in our cohort (9/124, 7.3%), as well as in the Singapore WES study (10/117, 8.5%, only consider somatic SNPs and INDELs)
Almost all somatic mutations fall into the DNA binding domain of TP53
Immunohistochemistry staining of p53 in three tumor samples with TP53missense mutations
missensestopgainsplicingframeshift INDELnonframeshift INDEL
Zheng et al, PNAS 2016
Mutations of NF‐κBRegulators
Verification of NFKBIA somatic INDELs
missensestopgainsplicingframeshift INDELnonframeshift INDEL
Zheng et al, PNAS 2016
43
Notes:
Notes:
Primary Tumor Cancer SpreadingCirculation
Circulating tumor cells “seeds” for cancer spread
What are Circulating Tumor cells and Why are CTCs So Important?
Alix‐Panabieres, C. and Pantel, K. Cancer Discov., 2016
Clinical Application of CTCs and ctDNAas Liquid Biopsies
44
Notes:
Notes:
Challenge for Isolation and Identification of CTCs in Blood
Just like finding you amongst 7.4 billion people!
Software analysis
Immunostaining‐CTC (e.g. CK)‐WBC (e.g. CD45)‐DNA stain
CTC Test Details
Image capture
Enrichment1
2
34
5 Reporting
45
Notes:
Notes:
CTC levels in several human cancers
NPC 42 (15QE5026)
Merged
20 µm
CK+ CD45‐DAPI
Representative CTC images
New diagnosisStaging and
baseline blood test
2‐3 weeks
Cycle I Chemo
Blood test
Cycle II Chemo
Blood test
Cycle III Chemo
Blood test
3 weeks 3weeks 3weeks
Cycle V Chemo
Blood test
Cycle IV Chemo
Blood test
3 weeks
Cycle VI Chemo
Blood test
3weeks
CTC1 at diagnosis: Baseline
CTC2 after week 6: early predictive biomarkers for
chemoRT response
CTC3 after week 9: Correlate with chemo interim
response
CTC4 at end of CT: Biomarker for Chemo final responseBoth to be done at Week 18 ‐ 20
Four Serial Time‐Point Schedule for CTC Analysis of Metastatic Stage IVC NPC
1PET
7 10
2PET
13
3PET
160
Palliative 1st line chemotherapyMedian survival ~ 11 monthsCommon chemotherapy regimens include:cisplatin/5FU, gemcitabine/cisplatin (or carboplatin) [capecitabine]
First line of treatment
46
Notes:
Notes:
CR: No significant soft tissue or uptake in nasopharynx, and remained same on all scans.
PET1 PET2 PET3 PET1 PET2 PET1 PET2
Similar size and metabolic activity with no new lesions
Reduction in size and metabolic activity of
indexed lesions
47
Notes:
Notes:
Why is Comprehensive Genomic Profiling So Important?
0
5
10
15
20
25
30
35%
of
pat
ien
tsPatients eligible for trastuzumab: using current standard testing
= rearrangements
= base substitutions
= insertions/deletions
= amplifications
ERBB2mutations
Data source: Foundation Medicine database
0
5
10
15
20
25
30
35
% o
f p
atie
nts
Patients eligible for trastuzumab: using a test for all solid tumor types
= rearrangements
= base substitutions
= insertions/deletions
= amplifications
ERBB2mutations
Data source: Foundation Medicine database
Why is Comprehensive Genomic Profiling So Important?
48
Notes:
Notes:
Comprehensive Genomic Profiling Can Identify Mutations that Would Have Been Missed by Standard Tests
0
5
10
15
20
25
30
35%
of
pat
ien
tsPatients eligible for trastuzumab: using a comprehensive profiling test
= rearrangements
= base substitutions
= insertions/deletions
= amplifications
ERBB2mutations
Data source: Foundation Medicine database
pre
NGS Study: Target Capture of 1373 Genes
Adapted from http://www.nimblegen.com/products/seqcap/ez/choice/index.html
49
Notes:
Notes:
Spike‐in DNA from Cancer Cell Line to Normal Individual for Evaluation of NGS Study by Target Capture
Only the protein‐altered somatic mutations were examined. We identified 102 protein‐altered mutations in KYSE270.
Sample summary
RC2202: blood sample from Red Cross
KYSE270: cell line KYSE270
10: RC2202 with 10% KYSE270
5: RC2202 with 5% KYSE270
2.5: RC2202 with 2.5% KYSE270
1.25: RC2202 with 1.25% KYSE270
Example: all three TP53 mutations can be identified in 10% and 5% spiked‐in
samples, but not in 2.5% and 1.25% samples.
Bioinformatics analysis done by Wei Dai
High specificity of protein‐altered somatic mutations with low false discovery rate.
0
500
1000
1500
2000
2500
KYSE270 10 5 2.5 1.25
TotalM
utations Compare to
RC2202
Compare to RC2202 KYSE270 KYSE270
(10%)
KYSE270
(5%)
KYSE270
(2.5%)
KYSE270
(1.25%)
Sensitivity reference 0.6078 0.2843 0.0686 0.0490
False discovery rate (FDR) reference 0.0606 0.0938 0.4615 0.5455
Positive prediction value
(PPV)reference 0.9394 0.9063 0.5385 0.4545
Preliminary data:
In ESCC13T, we identified 142 protein‐alteredmutations including 129 missense, 5 frameshiftinsertions, 4 splicing, 3 stopgain and 1 non‐frameshift deletion.
Identification of Somatic Mutations in ESCC CTC Sample
50
Notes:
Notes:
Acknowledgements
• Basic AoE lab team: J Ko, W Dai, H Zheng, A Cheung, B Wong
• Center for NPC Research Tissue Bank• AoE Clinical team: R Ngan, WT Ng, D Kwong, V Lee, KO Lam, CC Yau, S Tung
• AoE Imaging team: PL Khong, V Vardhanabhuti
• PRC collaborators: MF Ji (Zhongshan), J Pan (Fujian), X Peng (Shantou), ZF Zhang (Guangxi)
• OncoSeek: S Lam, V Wong
• AoE grant support from the Research Grants Council
• HK Health and Medical Bureau: HMRF grant
• HK Cancer Fund• OncoSeek, Ltd
CHALLENGES and QUESTIONS
Interactive Dialogue Session with Faculty‐ Facilitated Analysis and Discussion Focused on NGS Technologies
for Genomics‐Driven Cancer Medicine
51
Notes:
Notes:
Based on the growing number and combination of genomic alterations that are potentially actionable across a broad spectrum of tumor subtypes, hybrid capture NGS comprehensive genomic profiling should be used routinely, in preference of other strategies, as the primary methodology for precision‐focused tumor assessment and targeted treatment:
1) Strongly agree
2) Agree
3) Moderately agree
4) Agree somewhat
5) Disagree
Please Enter Your Response On Your Keypad
Audience Response System
Challenges and Questions
Based on my experience, laboratories that provide hybrid capture NGS comprehensive genomic profiling for cancer produce results of similar quality, i.e. they provide the same spectrum and level of coverage for genomic alterations, they have the same false‐negativity and false positivity rates, and they use technologies that are equally validated with respect to their value in aligning molecular targets with targeted therapy and resulting clinical outcomes:
1) Agree
2) Tend to agree
3) Disagree
4) Tend to disagree
Please Enter Your Response On Your Keypad
Audience Response System
Challenges and Questions
52
Notes:
Notes:
James Chih-Hsin Yang M.D., Ph.D.
楊志新教授Professor and Director, Graduate Institute of Oncology, NTU
台灣大學醫學院腫瘤醫學研究所所長Director, Department of Oncology, National Taiwan University Hospital
台大醫院腫瘤醫學部主任
Treatment of Advanced NSCLC Based onGenomic Analysis of the Tumor
James Chih‐Hsin Yang received honorarium for speech or participated in compensated advisory board of Boehringer Ingelheim, Eli Lilly, Bayer, Roche/Genentech/Chugai, Astellas, MSD, Merck Serono, Pfizer, Novartis, Clovis Oncology, Celgene, Merrimack, Yuhan Pharmaceuticals, BMS, Ono Pharmaceutical, Daiichi Sankyo and AstraZeneca
Presenter Disclosure
53
Notes:
Notes:
Advanced NSCLC Treatment Evolutions
Chemotherapy 1995 (mOS 6-12 months)
Platinum
Vinorelbine
Paclitaxel
Gemcitabine
Docetaxel
Irinotecan
S1
Pemetrexed
Targeted therapy 2002 (mOS 2-4 years)
Gefitinib #
Erlotinib #
Bevacizumab
Crizotinib #
Afatinib #
Ceritinib #
Alectinib #
Osimertinib #
Necitumumab
Ramucirumab
Nintendanib
Immunotherapy 2015
Nivolumab
Pembrolizumab
Histology Molecular driver profile Microenvironment
Red: only in non SCC# : driver mutation presencemOS: median overall survival
IDEAL: Objective Tumour Response Rates
Dose evaluation study for gefitinib in chemotherapy treated NSCLCNSCLC: Chemotherapy failure, no biomarker selection
Fukuoka M et al. JCO 2003
54
Notes:
Notes:
18
19
21
20
C-helix
P-loop
A-loop
Deletions 46%
L858R (39%)
Duplications/insertions (9%)N-lobe
C-lobe
Transmembraneregion
Extracellular domain
Regulatorydomain
ATP binding cleft
TKdomain
Distribution of Mutations in the TK domain of EGFR: Meta‐Analysis of Five Studies (n=1256)
Gefitinib Carboplatin / paclitaxel
EGFR M+ odds ratio (95% CI) = 2.75(1.65, 4.60), p=0.0001
EGFR M- odds ratio (95% CI) = 0.04(0.01, 0.27), p=0.0013
Overallresponserate (%)
(n=132) (n=129) (n=91) (n=85)
Odds ratio >1 implies greater chance of response on gefitinib
71.2%
47.3%
1.1%
23.5%
Iressa Pan‐Asia Survival Study (IPASS): Analysis based on EGFR mutations
55
Notes:
Notes:
PFS in EGFRMutation‐Positive and Wild‐Type Patients (IPASS)
EGFR mutation-positive EGFR wild-type
Probab
ility of PFS
1.0
0.8
0.6
0.4
0.2
0
Probab
ility of PFS
1.0
0.8
0.6
0.4
0.2
00 4 8 12 16 20 24 0 4 8 12 16 20 24
Months Months
132 108 71 31 11 3 0129 103 37 7 2 1 0
At risk:GefitinibC/P
91 21 4 2 1 0 085 58 14 1 0 0 0
Gefitinib (n=132)Carboplatin/paclitaxel (n=129)
HR (95% CI) = 0.48 (0.36, 0.64)p<0.0001
Gefitinib (n=91)Carboplatin/paclitaxel (n=85)
HR (95% CI) = 2.85 (2.05, 3.98)p<0.0001
Mok et. al. NEJM 2009
Treatment by subgroup interaction test p<0.0001
First‐line EGFR‐TKIs in EGFRMutation‐Positive: Summary
EGFR‐TKIMedian PFS
EGFR‐TKI vs Chemotherapy
IPASS[a] Gefitinib 9.6 vs 6.3 months (HR=0.48)
NEJ002[b] Gefitinib 10.8 vs 5.4 months (HR=0.30)
WJTOG3405[c] Gefitinib 9.2 vs 6.3 months (HR=0.49)
First‐SIGNAL[d] Gefitinib 8.4 vs 6.7 months (HR=0.61)
EURTAC[e] Erlotinib 9.7 vs 5.2 months (HR=0.37)
OPTIMAL[f] Erlotinib 13.1 vs 4.6 months (HR=0.16)
ENSURE[g] Erlotinib 11.0 vs 5.5 months (HR=0.33)
LUX‐Lung 3[h] Afatinib 11.1 vs 6.9 months (HR=0.58)
LUX‐Lung 6[i] Afatinib 11.0 vs 5.6 months (HR=0.28)
a. Mok T, et al. N Engl J Med. 2009;361:947‐957. b. Maemondo M, et al. N Engl J Med. 2010;362(25):2380‐2388.c. Mitsudomi T, et al. Lancet Oncol. 2010;11(2):121–128. d. Han JY, et al. J Clin Oncol. 2012;30(10):1122‐1128.e. Rosell R, et al. Lancet Oncol. 2012;13:239‐246. f. Zhou C, et al. Lancet Oncol. 2011;12:735‐742.g. Wu YL, et al. ELCC 2014. Oral Presentation 91O. h. Sequist LV, et al. J Clin Oncol. 2013;31(27):3327‐3334.i. Wu YL, et al. Lancet Oncol. 2014;15(2):213‐222.
56
Notes:
Notes:
Evidence to treat EGFRm+ NSCLC
with 1st line EGFR TKI (Gefitinib, Erlotinib, Afatinib )
compared to standard pt‐doublet
1. Higher PFS 2. Higher RR
3. Better HRQoL
Resistance Mechanisms in EGFR Mutant NSCLC
EGFR T790M
MET
Amplification
HGF
Production
57
Notes:
Notes:
AURA Study: Osimertinib (AZD9291) EGFR Mutant Specific Inhibitor
EscalationNot preselected by T790M status
ExpansionEnrollment by local testing
followed by central laboratory confirmation* of T790M status or by
central laboratory testing alone
*cobas® EGFR Mutation Test (Roche Molecular Systems)
Phase I, open‐label, multicenter study of AZD9291 administered once daily in Asian and Western patients with advanced NSCLC who have documented radiological progression while on prior
therapy with an EGFR‐TKI (AURA; NCT01802632)
Objectives
Primary: safety, tolerability and efficacy (objective response rate)
Secondary include: define maximum tolerated dose, pharmacokinetics
Extension
Jänne PA et al. ELCC 2015; abstract LBA3
Tumour Response to Osimertinib Treatment
AURA Ph I data cut-off 4 January 2016; population: evaluable for response set; assessment: investigator assessed; AURA pooled Ph II data cut-off 1 November 2015; population: evaluable for response set; assessment: BICR.*Represents imputed values: if it is known that the patient has died, has new lesions or progression of non-target lesions, has withdrawn due to disease progression, and has no evaluable target lesion (before or at progression) assessments, best change will be imputed as 20%; †Complete response, partial response, stable disease ≥6 weeks.
Tumor Response AURA Ph I (80 mg) n=61
AURA pooled Ph II (80 mg) n=397
Confirmed ORR 71% (95% CI: 57, 82) 66% (95% CI: 61, 71)
Disease control rate† 93% (95% CI: 84, 98) 91% (95% CI: 88, 94)
Best objective response Complete responsePartial responseStable disease ≥6 weeksProgressive disease
142142
62569925
–100
–80
–40
–60
–20
0
20
40
60
80
Complete responsePartial responseStable diseaseProgressive diseaseNot evaluable
Best
per
cent
age
chan
ge fr
om b
asel
ine
in
targ
et le
sion
siz
e (%
)
*********
100
–100
–80
–60
–40
0
20
40
80
–20
100
Best
per
cent
age
chan
ge fr
om b
asel
ine
in
targ
et le
sion
siz
e (%
)
Complete responsePartial responseStable diseaseProgressive diseaseNot evaluable
AURA Ph I AURA pooled Ph II
Proprietary and confidential AstraZeneca document – for internal use only.Yang JCH, et al. ELCC 2016; Abstract LBA2_PR.
58
Notes:
Notes:
Duration of Response With Osimertinib
AURA Ph I
AURA Ph I data cut-off 4 January 2016; population: evaluable for response set; assessment: investigator assessed; AURA pooled Ph II data cut-off 1 November 2015; population: evaluable for response set; assessment: BICR.Tick marks on the Kaplan-Meier plot denote censored observations *Duration of response is the time from first documentation of response until date of progression or death or last evaluable RECIST assessment for patients who do not progress.†Calculated using the Kaplan-Meier technique
Duration of Response AURA Ph I (80 mg) n=43
AURA pooled Ph II (80 mg) n=262
Median DoR*, months (95% CI) 9.6 (7.7, 15.6) 12.5 (11.1, NC)
Maximum DoR, months 26.3 ongoing 15.3 ongoing
Remaining in response†, % (95% CI)12 months18 months24 months
44 (29, 58)32 (19, 46)20 (8, 35)
53 (46, 59)NCNC
0Number of patients at risk:
Osimertinib 80 mg
Prob
abilit
y of
resp
onse
Month
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 3 6 9 12 15 18
43 40 36 22 18 16
21 24 27
11 5 3
Prob
abilit
y of
resp
onse
Number of patients at risk:Osimertinib 80 mg
Month262 240 185 142 69 4 0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 3 6 9 12 15 18 21 24 27
1.0 1.0
0.0
AURA pooled Ph II
Proprietary and confidential AstraZeneca document – for internal use only.
Yang JCH, et al. ELCC 2016; Abstract LBA2_PR.Yang JCH, et al. ELCC 2016; Abstract LBA2_PR.
EML4
EML4–ALK variant 1
HELP1 496 981
WDBasic
1 496 1059
1 1058 1620
TM
KinaseALK
Initially reported in 2007 as a result of an inversion in chromosome 2p, which results in the fusion of the N-terminal portion of the echinoderm microtubule-associated protein-like 4 (EML4) with the kinase domain of ALK
Soda et al., Nature 2007; 448:561-567
Discovery of the EML4‐ALK Fusion in NSCLC
59
Notes:
Notes:
Profile 1014 : Treatment Naïve ALK+ NSCLCCrizotinib Superior to Pemetrexed‐based Chemotherapy in Prolonging PFSa
Mok TS, et al. ASCO 2011
Tyrosine Kinase Inhibitors Activity Against VariousALK or ROS1 Fusion Kinases
60
Notes:
Notes:
Study Design (NCT01945021)
Phase II Study of Crizotinib in East Asian Patients With ROS1‐Positive Advanced Non‐Small Cell Lung Cancer
Goto K. et al. ASCO 2016 abstract 9022
IRR-assessed Best Percent Change <br />from Baseline in Target Lesion Size*
Phase II Study of Crizotinib in East Asian Patients With ROS1‐positive Advanced Non‐Small Cell Lung Cancer
Goto K. et al. ASCO 2016 abstract 9022
61
Notes:
Notes:
Efficacy and Safety of Crizotinib in Patients with Advanced MET‐Amplified NSCLC
Best percent change from baseline in target tumor lesionsa by patient
Camidge DR et al. ASCO 2014 abstract 8001
NSCLCcMET Skipping Exon 14 Mutation
Nele Van Der Steen et al. JTO 2016;11(9):1423-1432
62
Notes:
Notes:
NSCLCcMET Skipping Exon 14 Mutation
Nele Van Der Steen et al. JTO 2016;11(9):1423-1432
Actionable Targets in Lung Adenocarcinomas
Unknown75%
1999
2005-2016
EGFR
2004
Unknown60%
Modified from Kris M et al. IASLC 2012 Targeted Therapies Conference
SelumetinibTrametinib
GefitinibErlotinibAfatinib, DacomitinibOsimertinib, OlmutinibEGF816, ASP8273
Crizotinib, Ceritinib, Alectinib, BrigatinibLorlatinib
NRAS
CrizotinibCeritinib, Lorlatinib
Selumetinib?
Afatinib?Dacomitinib?
INC280MSC2156119jLY2801653Crizotinib
Dabrafenib, VemurafenibRegorafenib, Selumetinib, Trametinib
MK2206?BKM120?
HSP90 client oncoproteinAUY922PD1/PD-L1 expression Pembrolizumab, Nivolumab, Ateolizumab, Durvalumab, Avelumab etc. AntiangiogenesisBevacizumab, NintedanibRamucimumab
63
Notes:
Notes:
Patients with Driver MutationsCurrent Challenges
Detection of driver mutations: NGS
Detection and quantitation of T790M: plasma test
Selection of ALK inhibitors: ALK fusion mutation types
CNS metastasis
Poor performance patients
Selecting patients for combination therapy (bevacizumab, chemotherapy other targeted therapy)
Nivolumab2nd Line
Squamous cell carcinoma CheckMate17Nivolumab vs. Docetaxel 2nd line
Adenocarcinoma CheckMate57Nivolumab vs. Docetaxel 2nd line
64
Notes:
Notes:
Pembrolizumab1st, 2nd and Further Lines
Keynote 001 Keynote 010: pembro vs doc as 2nd line
Key End Points
Primary: PFS (RECIST v1.1 per blinded, independent central review)
Secondary: OS, ORR, safety
Exploratory: DOR
KEYNOTE‐024 Study Design (NCT02142738)
Key Eligibility Criteria
• Untreated stage IV NSCLC
• PD-L1 TPS ≥50%
• ECOG PS 0-1
• No activating EGFR mutation or ALK translocation
• No untreated brain metastases
• No active autoimmune disease requiring systemic therapy
Pembrolizumab 200 mg IV Q3W
(2 years)
R (1:1)N = 305
PDa Pembrolizumab 200 mg Q3W for 2 years
Platinum-Doublet Chemotherapy
(4-6 cycles)
aTo be eligible for crossover, progressive disease (PD) had to be confirmed by blinded, independent central radiology review and all safety criteria had to be met.
Reck M. et al. NEJM 2016:375(19);1823‐1833
65
Notes:
Notes:
Progression‐Free Survival
Assessed per RECIST v1.1 by blinded, independent central review.Data cut-off: May 9, 2016.
Events, n
Median, mo
HR (95% CI)
P
Pembro 73 10.3 0.50 (0.37-0.68)
<0.001Chemo 116 6.0
62%50%
0 3 6 9 12 15 180
10
2030
4050
6070
80
90
100
Time, months
PF
S,%
No. at risk
154 104 89 44 22 3 1151 99 70 18 9 1 0
48%15%
Reck M. et al. NEJM 2016:375(19);1823‐1833
Overall Survival
Data cut-off: May 9, 2016.
80%72%
0 3 6 9 12 15 18 210
10
20
30
40
50
6070
8090
100
Time, months
OS
,%
No. at risk
154 136 121 82 39 11 0151 123 106 64 34 7 0
21
70%54%
Events, n
Median, mo
HR (95% CI)
P
Pembro 44 NR 0.60 (0.41-0.89)
0.005Chemo 64 NR
DMC recommended stopping the trial because of superior efficacy observed with pembrolizumab
Reck M. et al. NEJM 2016:375(19);1823‐1833
66
Notes:
Notes:
Distribution of TMB Across All LC Cases
Comparison of Time on Anti‐PD‐1/PD‐L1 Therapy vs. TMB
Spigel DR et al. ASCO 2016
67
Notes:
Notes:
Comparison of Time on Anti‐PD‐1/PD‐L1 Therapy vs. TMB
Spigel DR et al. ASCO 2016
Slide 6
Spigel DR et al. ASCO 2016
68
Notes:
Notes:
2016 Oct UpdatedTreatment Paradigm
Squamous cell carcinoma(non‐adenocarcinoma)
Adenocarcinoma(Non‐SCC)
Small cell Lung cancer
platinum + gem/taxane/vnr
+necitumumab
nivolumab, atezolizumabor pembrolizumab(PDL1+)
docetaxelafatiniberlotinib? +ramucimumab
or Nintedanib(fast PD)
Molecular diagnosisDriver mut‐Or no TT* Driver mut+
platinum + pemor paclitaxel/car/bev
pem maintenance
nivolumab, atezolizumabor pembrolizumab(PDL1+)
EGFR mut
* K‐ras, N‐ras, HER2, EGFRexon20 ins, etc.
other not used agents or clinical trials
afatinib, erlotinib,gefitinib +bevacizumab
platinum + pem+/maintor paclitaxel/car/bev
ALK fusion
Other drivers#
#cMET exon14 skip, RET fusion, ROS1 fusion, TRK fusion etc.
crizotinib, alectinib
T790M+
osimertinib
T790M‐
Docetaxel +ramucimumabor Nintedanib(fast PD)
ceritinibalectinib
Corresponding targeted therapy, chemotherapy, immunotherapy
T790M‐
Pembrolizumab
PD L1>50%?
The Clinical Implications of Using Hybrid Capturing NGS in Lung Cancer
The Journey from Genomic Alterations to Optimizing Targeted Therapy at the Front Line of Oncology Practice
Professor Nir Peled, MD, PhD ‐ Program ChairHead, Thoracic Cancer Unit and Center for Precision Medicine
Davidoff Cancer CenterTel Aviv UniversityTel Aviv, Israel
69
Notes:
Notes:
NSCLC ‐ Survival Curves
Schiller et al. NEJM; 2002
Kris, M; 2014
Only ~ 20% of NSCLC Have A Driver Mutation
Most Patients Still Have Poor Outcome
Broadcasting News
1st-line combination
2nd line
Maintenance treatment
1st-line or unspecified setting single agent
1970 1980 1990 2000
Erlotinib2004
Docetaxel1999
Gefitinib†
2003
2010
Pemetrexed‡
2004
Erlotinib2010
Pemetrexed‡
2009
Crizotinib§2011 (US)/2012 (EU)
Erlotinib**2013Median OS, months
12+
~8–10~6
~2–4
13+
Carboplatin*1989
Gemcitabine1996
Vinorelbine1994
Docetaxel2002
Bevacizumab‡
2006
Pemetrexed‡
2008
Paclitaxel1998
Nab‐Paclitaxel2012
Cisplatin*1978
FDA Approval Dates for NSCLC
*Not approved in NSCLC, but commonly used; †Restricted to patients participating in a clinical trial or continuing to benefit from treatment already initiated; ‡Non‐squamous NSCLC only; §ALK‐positive NSCLC only; **EGFR exon 19 deletions or exon 21 (L858R) substitution mutations only; #Afatinib is approved for the treatment of patients with activating EGFRmutations but only PFS data have been published (May 2014).
U.S. Food and Drug Administration. Available at www.fda.gov. Accessed September 2014; European Medicines Agency. Available at http://www.ema.europa.eu. Accessed September 2014; NCCN Guidelines. Non‐small cell lung cancer. v3.2014.
Afatinib**,#
2013
Nivolumab 2015
Zykadia2014
Pembrolizumab 2015
AZD9291 2015
Alectinib2015
Atezolizumab 2016
70
Notes:
Notes:
Modified from Steven O'Day
Stratification for EGFR, ALK and histology
EGFRMut+EGFRWT
non‐squamous
EGFR TKI1+2 L
Platinum doublet + bevacizumab
ALK+
ALK TKI1+2 L
Crizotinib for
ROS1
Therapy Algorithm NSCLC 2016
EGFRWT squamous
Platinum‐based doublet
PD1 BlockadePD1 Blockade
71
Notes:
Notes:
KEYNOTE‐024: Pembrolizumab vs Platinum‐Based Chemotherapy as First‐Line Therapy for Advanced NSCLC With a PD‐L1 (IHC) ≥50%
Martin Reck, Delvys Rodríguez‐Abreu, Andrew G. Robinson, Rina Hui, Tibor Csőszi, Andrea Fülöp, Maya Gottfried,Nir Peled, Ali Tafreshi, Sinead Cuffe,Mary O’Brien, Suman Rao, Katsuyuki Hotta, Melanie A. Leiby, Gregory M. Lubiniecki, Yue Shentu, Reshma Rangwala, and Julie R. Brahmer, on behalf KEYNOTE‐024
PDL1≥50% in 30% of NSCLC patients
20% (EGFR/ALK) + 30% (PDL1++) = 50%!!
NEJM 2016
2017th Treatment Algorithm in Advanced NSCLC
Non‐squamous cellNon‐squamous cell
Locally Advanced or Metastatic NSCLCLocally Advanced or Metastatic NSCLC
Any histologyAny histology
EGFR PositiveEGFR
PositiveALK/ROS positiveALK/ROS positive
EGFR inhibitor
ALK/ROS inhibitor
Platinum Doublet(Excluding Pemetrexed in Squamous)
Platinum Doublet(Excluding Pemetrexed in Squamous)
Response
PD‐L1 ≥ 50% TPSPD‐L1 ≥ 50% TPSPD‐L1 < 50% TPSPD‐L1 < 50% TPS
TESTING: PD‐L1 EGFR, ALK, histologyTESTING: PD‐L1 EGFR, ALK, histology
PD‐L1 (+)Driver MutationDriver Mutation
Progression Progression
*Maintenance Pemetrexed (NSQ)
PD‐L1 ≥1% TPSPD‐L1 UnknwonPD‐L1 ≥1% TPSPD‐L1 Unknwon
Progression
Pembrolizumab
PD‐L1PD‐L1
PDL1 Therapy
72
Notes:
Notes:
Beyond EGFR, ALK & ROS1 (NCCN)
Journal of Thoracic Oncology 2016 11, 613‐638. DOI: (10.1016/j.jtho.2016.03.012)
73
Notes:
Notes:
What Do We Do Today?
H&E • ADC
EGFR • Negative
ALK • Negative
PDL1
ROS1; CMET; RET
Other Drivers?
What should we add?
Example:Complete Response to Crizotinib in a 63 y/o Woman with a MET Amplification and MET Exon 14 Splice Site Mutation, DetectedVia FMI
74
Notes:
Notes:
Current Main Challenges
Diagnostic challenges:
Sample size & Turn around time
PDL1 IHC implementation
EGFR/ALK/ROS1 performance
CMET/RET/BRAF or Multiplex approach
Therapeutic challenges:
Patient selection
Drug availability & Reimbursement
The POWER to PREDICT
IPASS (EGFR; NEJM 2009) PROFILE 1014 (ALK; NEJM 2014)
KEYNOTE 024 (PDL1; NEJM 2016)
75
Notes:
Notes:
Case Presentation (6/2011)
43‐year‐old never smoker caucasian male
Shortness of breath
Pleural effusion and tamponade
Pleurocentesis with clinical relief
Cytology: Cancerous cells
Most probably – Adenocarcinoma
No further cells are available for analysis
CT/PET – RLL + pleural disease only
Case Presentation Continued
Histology: TTF1(+); p63(‐) Adenocarcinoma
EGFR (cobas®) – Negative
ALK FISH (abbott) – Negative
Cisplatin/Pemetrexed/Bevacizumab X 4 PD
76
Notes:
Notes:
A_Pre A_4 weeks A_4 months
B_Pre B_4 weeks B_4 months
Journal of Thoracic Oncology. 7(9):e14-e16, 2012
Case Continued (February/2013)
Crizotinib 250 mg BID
Mild Ataxia
CT/PET – Normal
Brain MRI
77
Notes:
Notes:
JTO 2013 (8): e112
Crizotinib250 mg X2
Post WBRT
Crizotinib
500 mg X1
78
Notes:
Notes:
Brain Response to Ceritinib
•Diagnosed: June 2011 Aug 2016
•Ceritinib since 4.2014
Peled et al; J Thoracic Oncology 2012
Peled et al; The Oncologist 2015
Peled et al; Journal of Clinical Neuroscience 2016
79
Notes:
Notes:
Nivolumab in Non‐SqNSCLC 2nd Line Checkmate 057
mOS (mos)Nivo 17.7Doc 9.0
mOS (mos)Nivo 19.4Doc 8.1
mOS (mos)Nivo 19.9Doc 8.0
mOS (mos)Nivo 10.5Doc 10.1
mOS (mos)Nivo 9.8Doc 10.1
mOS (mos)Nivo 9.9Doc 10.3
≥1% PD-L1 expression level
Time (mos)
100
90
80
70
60
50
40
30
10
0
20
3024211815129630 27
OS
(%)
NivoDoc
HR (95% CI) = 0.58 (0.43, 0.79)
≥5% PD-L1 expression level
100
90
80
70
60
50
40
30
10
0
20
3024211815129630 27Time (mos)
HR (95% CI) = 0.43 (0.30, 0.62)
≥10% PD-L1 expression level
100
90
80
70
60
50
40
30
10
0
20
3024211815129630 27Time (mos)
HR (95% CI) = 0.40 (0.27, 0.58)
<1% PD-L1 expression level100
90
80
70
60
50
40
30
10
0
20
3024211815129630 27Time (mos)
OS
(%)
NivoDoc
<10% PD-L1 expression level100
90
80
70
60
50
40
30
10
0
20
3024211815129630 27Time (mos)
<5% PD-L1 expression level100
90
80
70
60
50
40
30
10
0
20
3024211815129630 27Time (mos)
HR (95% CI) = 0.87 (0.63, 1.19) HR (95% CI) = 0.96 (0.73, 1.27) HR (95% CI) = 0.96 (0.74, 1.25)
157
"Copyright permission has been obtained for this figure. From The New England Journal of Medicine, Borghaei H, et al., Nivolumab versus Docetaxel in Advanced Nonsquamous Non–Small-Cell Lung Cancer [Epub ahead of print; DOI: 10.1056/NEJMoa1507643]. Copyright © 2015 Massachusetts Medical Society. Reprinted with permission from Massachusetts Medical Society."
POPLAR ‐ OS According to PD‐L1 Expression Level
80
Notes:
Notes:
Subgroups Analysis of Progression Free Survival
Wake Up Call!
An URGENT need for UPFRONT PDL1 staining
EGFR/ALK positives do not benefit from I‐O
Do ROS1 / RET / MET benefit from I‐O ?
Do we catch all patients with a driver mutation?
81
Notes:
Notes:
Molecularly targeted therapy based on tumour molecular profiling versus conventional therapy for advanced cancer (SHIVA): a multicentre, open‐label,
proof‐of‐concept, randomised, controlled phase 2 trial
Le Tourneau, C., Delord, JP., Gonçalves, A., et al. (2015) Lancet Oncol 16(13):1324‐34. http://www.ncbi.nlm.nih.gov/pubmed/26342236
40% of Patients Refractory Solid Tumors with ≥ 1 Genomic Alteration Could be Molecularly Targeted
Objective: Compare the PFS of patients with recurrent/ metastatic solid tumors refractory to standard care treated with MP‐directed or physician‐directed therapyMethods: • Open‐label, randomised, controlled phase 2 study (n = 195): experimental group (n = 99), control group (n = 96)
• MP by NGS (AmpliSeq cancer panel on an Ion Torrent/PGM system, Life Technologies, Carlsbad, CA, USA; appendix 2, Cytoscan HD (Affymetrix, Santa Clara, CA, USA); hormone receptors by IHC
• Patients with genomic alterations in hormone receptor, PI3K/AKT/mTOR or RAF/MEK pathways were included
Results: • 293 (40%) patients had at least 1 genomic alteration matching one of 10 available targeted regimens
Le Tourneau, C., et al. (2015). Molecularly targeted therapy based on tumour molecular profiling versus conventional therapy for advanced cancer (SHIVA): a multicentre, open‐label, proof‐of‐concept, randomised, controlled phase 2 trial. Lancet Oncol 16(13):1324‐34.
Age (years) 61 (54‐69) 63 (54‐69)
Sex
Female 60 (61%) 69 (72%)
Male 39 (39%) 27 (28%)
Previous lines of treatment 3 (2‐5) 3 (2‐5)
Molecular pathway targeted
Hormone receptor pathway
40 (40%) 42 (44%)
PI3K/AKT/mTOR pathway 46 (46%) 43 (45%)
RAF/MEK pathway 13 (13%) 11 (11%)
MP: Molecular profiling; NGS: Next‐generation sequencing; PFS: Progression‐free survival
Baseline patient characteristics
MP‐directed therapy (n= 99)
Physician‐directed therapy (n= 96)
82
Notes:
Notes:
No PFS Improvement Following Off‐label Matched Therapyin Patients with Metastatic Solid Tumors
Results: • Median follow‐up at time of primary analysis of PFS in both groups was 11.3 months
• Similar median progression‐free survival (p = 0.41) in:• Experimental group: 2.3 months (95% CI 1.7‐3.8)
• Control group: 2.0 months (95% CI 1.8‐2.1)• In both groups grade 3‐4 adverse events were observed (p = 0.30)
Conclusions: Use of targeted therapy outside indications of agents does not improve progression‐free survival compared with treatment at physician’s choice in heavily pre‐treated cancer patients
Le Tourneau, C., et al. (2015). Molecularly targeted therapy based on tumour molecular profiling versus conventional therapy for advanced cancer (SHIVA): a multicentre, open‐label, proof‐of‐concept, randomised, controlled phase 2 trial. Lancet Oncol 16(13):1324‐34.
MP: Molecular profiling; NGS: Next‐generation sequencing; PFS: Progression‐free survival
Progression‐free survival
HR 0.88 (95% CI 0.65‐1.19); p = 0.41
Molecularly targeted agentTreatment of physician’s choice
PFS (%)
Time (months)No. at risk
Molecularly targeted agent
Treatment at physician’s choice
0 62 4 8 10 120
20
80
40
60
100
99 62 20 10 5 2 0
95*
50 19 12 8 1 0*One patient was not followed‐up so is not shown here
Using multiplexed assays of oncogenic drivers in lung cancers to select targeted drugs
83
Notes:
Notes:
A Genomic Alteration was Identified in 64% of Patients with Lung Adenocarcinoma
Objectives: • Determine the frequency of genomic alterations (oncogenic drivers) in patients with lung adenocarcinomas
• Assess survival following a treatment strategy to target the identified drivers
Patients and methods: • Tumors from 1007 patients were tested for at least 1 gene, and from 733 patients for 10 genes (full genotyping)• MP by matrix‐assisted laser desorption/ionization time‐of‐flight mass spectrometry (Sequenom, Arizona Research Laboratories); multiplexed single‐nucleotide extension sequencing (SNaPshot, Applied Biosystem), or Sanger sequencing with peptide nucleic acid probes, as well as sizing electrophoresis• Selected, molecularly‐matched targeted therapies were initiated where genomic alterations were identified
Results: • Genomic alterations were identified in 466 of 733 (64%) patients with full genotyping performed• 44% of individuals with genomic alterations detected received a matched therapy
MP: Molecular profilingKris M.G., et al. (2014). Using Multiplexed Assays of Oncogenic Drivers in Lung Cancers to Select Targeted Drugs. JAMA 311(19):1998‐2006.
Patients with Lung Adenocarcinomas Receiving Matched Therapy Lived Longer
Results: • Median survival of the 938 patients with adequate data was 2.7 years (95% CI, 2.4–2.9)
• A higher median survival was observed in patients with a genomic alteration receiving a targeted therapy (3.5 years (IQR 1.96‐7.70), n = 260; p = 0.006)compared with patients with a genomic alteration but not receiving a targeted therapy (2.4 years (IQR 0.88‐6.20), n = 318)
Conclusions:• Patients with lung adenocarcinoma receiving molecular targeted therapy lived longer
• Molecular profiling may be clinically beneficial
Median survival (95% CI)
Kris M.G., et al. (2014). Using Multiplexed Assays of Oncogenic Drivers in Lung Cancers to Select Targeted Drugs. JAMA 311(19):1998‐2006.
1 2 3 4 5
Years
Survival probab
ility
0.2
0.4
0.6
0.8
1.0
00
No targeted therapy
Targeted therapy
No driver
Log‐rank p < 0.001
318 205 110 64 43 20
260 225 143 72 36 23
360 250 122 59 36 23
No. at riskPatients with oncogenic driver
No targeted therapy
Targeted therapy
No driver
IQR: Interquartile range
84
Notes:
Notes:
Identifying ALKRearrangements That are Not Detected by FISH with Targeted Next Generation Sequencing of Lung Carcinoma
Data analysis
MethodsGenomic profiling of 1,070 lung carcinomas was performed on DNA extracted from formalin fixed, paraffin embedded specimens, from either primary tumors or metastatic sites
47 patients with lung adenocarcinomas were identified whose tumors harbored no evidence of a genomic alteration via extensive, focused non‐NGS testing.
Ali, S., et al. (2014) Identifying ALK rearrangements that are not detected by FISH with targeted next generation sequencing of lung carcinoma. J Clin Oncol 32:5s, 2014 (suppl; abstr 8049).
ResultsHybrid Capture NGS‐Based, Comprehensive Genomic Profiling identified 47 ALKrearrangements (4.4% of cases) in 1070 advanced lung cancer cases
Of the 28 ALK rearranged specimens also tested by ALK FISH, 9 (32%) were negative, and 19 were positive
Identifying ALKRearrangements That are Not Detected by FISH with Targeted Next Generation Sequencing of Lung Carcinoma
Subsequent treatments
Conclusion‐ Targeted NGS may be more sensitive for the detection of ALK rearrangements than FISH. ‐ In light of the responsiveness of ALK NGS+/FISH tumors to crizotinib, the use of FISH as the gold standard for ALK detection in LC warrants prospective study
Treatment status‐ 22 patients were treated with crizotinib and had response data available‐ 19 responded by investigator assessmentOf the 9 cases negative by FISH:‐ 5 patients responded to crizotinib‐ 2 patients did not‐ The response data for the remaining 2 patients is unavailable
Hybrid Capture NGS‐Based, Comprehensive Genomic Profiling revealed ~32% of ALK fusions that are being MISSED by FISH
Ali, S., et al. (2014) Identifying ALK rearrangements that are not detected by FISH with targeted next generation sequencing of lung carcinoma. J ClinOncol 32:5s, 2014 (suppl; abstr 8049).
85
Notes:
Notes:
The Clinical Utility of NGS in Lung Cancer1N Peled,2A Dvir,2Gutman-Li Soussan,1FlexD,1E Dudnik,†1M Ilouze,†1AB Rozenblum
Evaluate the impact of hybrid NGS on AIM:treatment strategy in NSCLC
In PressJournal of Thoracic Oncology;
Methods ‐ Two
Commercial
Profiling Tests
1. Hybrid capture‐based NGS of solid tissue (Foundation‐
One, n=82/101)2. NGS of circulating cell‐free
DNA (Guardant360, n=19/101)
Study Population
The studied population is a focused group of patients ‐ the majority was negative on standard molecular testing
Age at Diagnosis (Years) 63 (Median), range: 20-84
Gender
M 46.5% (n=47) F 53.5% (n=54)
Cigarette Smoking Status Never 44.55% (n=45) Ever, Average Pack Years 52.48% (n=53), 38PY Not-Available 2.97% (n=3)
Histopathology
Adenocarcinoma 85.1% (n=86)
Other 14.9% (n=15)
Type of NGS Test Performed Solid Tissue (FoundationOne) 81.2% (n=82) Blood Circulating Cell-Free DNA (Guardant360) 18.8% (n=19)
Standard Testing Results EGFR % Negative or Inconclusive 85.2% (n=86) ALK
% Negative or Inconclusive 71.3% (n=72)
NGS Performance Timing Pre 1st Treatment Line 50.5% (n=51)
Post 1st Treatment Line 49.5% (n=50)
86
Notes:
Notes:
Seventeen patients were diagnosed with an EGFR/ALK alternation after previous negative/false approved standard molecular testing; Fifteen were treated with a targeted agent
Treatment Change 33% n=5/15 43% n=37/86
Driver No. of pts %
EGFR 11 29.7%
ALK 5 13.5%
RET 5 13.5%
2ERBB 5 13.5%
*MET 5 13.5%
ROS1 2 5.4%
BRAF 1 2.7%
KRAS 1 2.7%
NTRK1** 1 2.7%
NF1 1 2.7%
Previously TestedUpfront NGS
Driver No. of pts %
EGFR 3 60%
ALK 2 40%
Treatment Change
Response to Targeted Therapy
ORR: 64%
87
Notes:
Notes:
Duration of Treatment
Response to PD1 Therapy by Mutation Status
88
Notes:
Notes:
High Mutational Burden Predicts Response to Pembrolizumab in NSCLC
High Mutation Burden ≥200
Low Mutation Burden >200
Rizvi et al. 2015
Slide 3
89
Notes:
Notes:
Future Treatment Algorithm in Advanced NSCLC
Non‐squamous cellNon‐squamous cell
Locally Advanced or Metastatic NSCLCLocally Advanced or Metastatic NSCLC
Any histologyAny histology
EGFR PositiveEGFR
PositiveALK
positiveALK
positive
EGFR inhibitorEGFR
inhibitorALK
inhibitorALK
inhibitor
Platinum Doublet(Excluding Pemetrexed in Squamous)
Platinum Doublet(Excluding Pemetrexed in Squamous)
ResponseResponse
PD‐L1 ≥ 50% TPSPD‐L1 ≥ 50% TPSPD‐L1 < 50% TPSPD‐L1 < 50% TPS
TESTING: PD‐L1 EGFR, ALK, histologyTESTING: PD‐L1 EGFR, ALK, histology
PD‐L1 (+)Driver MutationDriver Mutation
Progression Progression Progression Progression
*Maintenance Pemetrexed (NSQ)*Maintenance
Pemetrexed (NSQ)
PD‐L1 ≥1% TPSPD‐L1 UnknwonPD‐L1 ≥1% TPSPD‐L1 Unknwon
Progression Progression
Pembrolizumab Pembrolizumab
PD‐L1PD‐L1
PDL1 TherapyPDL1 Therapy
Take Home Message
THANK YOU
/
90
Notes:
Notes:
CHALLENGES and QUESTIONS
Interactive Dialogue Session with Faculty‐Facilitated Analysis and Discussion Focused on NGS Technologies to Optimize Assessment and Targeted Therapy for
Patients with Lung Cancer
The multiplicity of potential driver mutations and associated genomic alterations that are being identified in NSCLS — including those whose targeting has been validated to improve clinical outcomes — are of such diversity and established clinical importance that hybrid capture‐based NGS comprehensive genomic profiling:
1) Should be utilized as the initial pathogenomic strategy in all patients with an established diagnosis of NSCLC
2) Should be utilized as the initial pathogenomic strategy in most patients with an established diagnosis of NSCLC
3) Should be utilized as the initial pathogenomic strategy in some patients with an established diagnosis of NSCLC
4) Should be utilized as a pathogenomic strategy in most patients with an established diagnosis of NSCLC, but only after hot spot panels have been employed
Please Enter Your Response On Your Keypad
Audience Response System
Challenges and Questions
91
Notes:
Notes:
The growing body of evidence suggesting responsiveness in NSCLC to PD‐1 checkpoint inhibitors based on mutational load supports the use of hybrid capture‐based NGS comprehensive genomic profiling in almost all patients in NSCLC:
1) Strongly agree
2) Agree
3) Moderately agree
4) Agree somewhat
5) Disagree
Please Enter Your Response On Your Keypad
Audience Response System
Challenges and Questions
The Role of Genomic Profiling to Optimize Clinical Outcomes in Patients
with Breast Cancer
Rebecca Dent, MD, FRCP (Canada)Senior Consultant, National Cancer Center Singapore
Associate Prof Duke‐NUS
92
Notes:
Notes:
I have received travel grants, honorariums and/or speaker fees from:
AstraZeneca, Celgene, Eisai, Genentech, Merck, Novartis, Pfizer, Roche
Presenter Disclosure
New Technologies
93
Notes:
Notes:
The Microenvironment
TCGA:The Cancer Genome Atlas
94
Notes:
Notes:
Heterogeneity
Tumor Evolution
Yates Nature Genetics 2012
95
Notes:
Notes:
Molecular Portraits of Breast Cancer
• Low HER2 cluster expression
• Low ER (and related genes) expression
Usually “triple negative”
• High basal cluster
– basal CK (5, 6, 14, 17)
– EGFR, c-kit
– others…
• Very proliferative
• Often p53 mutant
• Evidence of genomic instability
HER2 cluster
Basal gene cluster
Luminal (hormone receptor-related) cluster
Proliferation cluster
Basal- like Group
Perou C , Sorlie T et al. Nature 2000
METABRIC Validation
Ali Genome Biol 2014
Carlos Caldas from Cambridge group has given us 10 subtypes of breast cancer using over 10,000 samples
Evolution of increasing complexity ‐ but what is driving individual cancer
96
Notes:
Notes:
17%
35%48%
<1%
Hormone Receptor-Positive
HER2-E (n=26)
Lum A (n=54)
Lum B (n=75)
Basal (n=1)
Claudin-Low (n=0)
Normal-like (n=0)
51%
24%
4%
12%
3%6%
Hormone Receptor-Negative
HER2-E (n=56)
Lum A (n=26)
Lum B (n=5)
Basal (n=13)
Claudin-Low (n=3)
Normal-like (n=6)
31%
30%
30%
5%
2%
HER2+ tumors
HER2-E (n=82)
Lum A (n=80)
Lum B (n=80)
Basal (n=14)
Claudin-Low (n=3)
Normal-like (n=6)
Intrinsic Subtypes by Hormone Receptor
Carey, JCO 2015
NOAH: Response and Outcome Chemo/Trastuzumab vs Chemo alone
Perez et al, JCO 2014; Prat et al, CCR 2014
N9831/B31: Impact of H added to chemo
Subtype pCR 3y EFS
(all are cHER2+) ‐ H + H ‐ H + H
HER2‐Enriched (n=63) 28% 53% 61% 79%
Not HER2‐E (n=51) 18% 34% 48% 50%
NOAH: Impact of H by subtype
Small, hypothesis‐generating.Suggests that addition of HER2‐targeting especially important in HER2‐E subtype
97
Notes:
Notes:
pCR by Intrinsic Subtype
Other subtypes:3 Claudin‐low (0 pCR)14 basal‐like (36% pCR)Excluded “normal” (n=6)
70%
80%
71%
52%
34%37%
38%
9%
36%40% 41%
22%
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
Overall THL TH TL
HER2E (n=82)
LumA (n=80)
LumB (n=80)
P=0.04
Carey, JCO 2015
This may be why effect larger in HR-than HR+
Note ~ 70% pCRrate with 12 weeks TH (APT regimen) in HER2-E
N9831: immune signature associated with RFS only in H-treated cohorts
Perez EA et al. JCO 2015; Loi S et al, Ann Oncol 2013
Immune Signatures / TIL and Prognosis
FinHER: Tumorinfiltrating lymphocytes associated with greater
benefit of H and prognosis
98
Notes:
Notes:
How do we identify patients who will respond within a molecular subgroup?
Ligand‐dependent Transcription
ER
ERE
ERER
Target‐Gene
MEK
MAPK
SOS
RAS
RAF
P
ER
Proliferation Cell SurvivalInvasion
P
PP
P
P
Adapted from Johnston S. CCR. 2005;11:889S‐899S.
Growth Factor
Estrogen/Tamoxifen
Ligand‐independent Transcription
EGFR/HER2
IGFR FGFR MET
Scr
Cyclin D
E2F
p16INK4A
Stat3
p21cip1
CDK4/6
P P
PP
P P
PP
PI3‐K
Akt
mTORC2
mTORC1
PI3K AKT mTOR Pathway
99
Notes:
Notes:
Median PFS (central)EXE + EVEROL: 11.0 moEXE + placebo: 4.1 mo
HR = 0.38 (95% CI: 0.31–0.48)Log-rank P value: < 0.0001
0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 90 96 102 108
0
20
40
60
80
100
Probab
ility of Event, %
Time, weeks
BOLERO-2
Baselga J et al, N Engl J Med 2011 366; 520-529Piccart M, et al. ASCO 2012; # 559.
84% prior endocrine response
EVE.PIK3CA.WT
EVE.PIK3CA.Alt
Study Arm
PIK3CA HR (95%CI)
EVE WT 0.36 (0.22 - 0.57)PBO WT
EVE Alt 0.44 (0.27 - 0.70)PBO Alt
0 100 200 300 400 500 600 700
0.0
0.2
0.4
0.6
0.8
1.0
Time, days
Probab
ility of PFS
Hortobagyi et al ASCO 2013 & 2014, Hortobagyi et al JCO. 2015 (in press)
Subgroup Definition
WT No alteration in any gene
Single Alteration in 1 gene only
Multiple Alterations in2 or more genes
Analysis of the following 4 genes: PIK3CA, PTEN, FGFR1/2, CCND1
Subgroup N HR* (95%CI)
EVE: WT27%
0.24 (0.11 - 0.54)PBO: WT
EVE: Single49%
0.26 (0.16 - 0.43)PBO: Single
EVE: multiple24%
0.78 (0.39 - 1.54) PBO: multiple
Everolimus benefit maintained in patients regardless of gene alterations in
PIK3CA
Patients with ≤1 genetic alteration in derive greater PFS benefit with EVE
PBO.PIK3CA.WT PBO.PIK3CA.Alt
100
Notes:
Notes:
4 main subtypes
Luminal AR (16‐23%)
Mesenchymal (18‐28%)
Basal immune suppressed (25‐30%)
Basal immune activated (29‐31%)
Burstein et al, CCR 2014
Triple Negative Breast Cancer MD Anderson and Baylor Collaboration
(DNA profiling and RNA Seq)
101
Notes:
Notes:
201
% o
f br
east
canc
ers
0
5
10
15
20
25
30
“Druggable” Genomic Alterations
Adapted from F. Andre
Rationale for Multi‐Gene Testing
Identify driver mutation(s) that promote survival or proliferation
Individualize treatment with targeted drugs that inhibit those key molecular pathways
Improve efficiency of screening for clinical trials with targeted drugs
102
Notes:
Notes:
Phase II GDC‐0032 + Fulvestrant
Dickler M, et al. ASCO 2016. Abstract P520.
ORR= 38.5% (PIK3CAmt) vs 10.5% (PIK3CAwt)
Identified in metastatic ER+ breast cancers after anti‐estrogen treatment
Rare in primary tumours, enriched in metastases Produce ligand‐independent activation of ER Polyclonal ESR1mutations may be detected in circulation;
produce resistance to AIs but not fulvestrant
Toy W et al Nature 2013Robinson DR et al Nature 2013Merenbakh‐Lamin K et al Cancer Res 2013
ESR1 Gene Mutations
103
Notes:
Notes:
Joseph JD et al. eLife 2016
Activity in ESR1mt Preclinical Models
Oral SERDs in Development
ARN-810/GDC-810RAD1901AZD9496SRN-927LSZ102
ERBB2mutation ~2% of non‐HER2 amplified breast cancers
May be enriched in lobular and metastatic breast cancers
Cell lines resistant to reversible HER2 TKI, sensitive to irreversible HER2 TKI
Bose R et al Cancer Discovery 2013. 3(2):1-14. Ross J et al CCR 2013. 19(10); 2668–76
ERBB2Mutation as an Oncogenic Driver in HER2 Non‐Amplified ER+ Breast Cancers
104
Notes:
Notes:
Neratinib in ERBB2‐Mutant, Non‐Amplified Breast Cancer
Ma CX et al ASCO 2016 Abstract 516
Washington University Phase II SUMMIT Basket Trial
Hyman DM et al SABCS 2015 Abstract PD5-05
AZD5363 (AKT inhibitor) Monotherapy in AKT1 E17K Mutant, ER+ Breast Cancer
Hyman DM, Smyth L et al AACR-EORTC-NCI Meeting 2015
105
Notes:
Notes:
15 3874
107
168
234
314370
449
531
619
722
795
0
100
200
300
400
500
600
700
800
900
Cumulative Accrual Projected Cumulative Accrual
High Level of Enthusiasm
NCI‐MATCH
Princess Margaret IMPACT/COMPACT
SHIVA Trial
Le Tourneau C et al Lancet Oncology 2015: 16:1324-34
106
Notes:
Notes:
423 patients signed
CGH arrays:288 patients
Biopsy of metastasis:403 patients
Targetablealterations in195 patients
27%Treatmentdriven by
genomics in 52 patients
+ 4 patients ERBB2 ampCGH array
13%
SAFIR‐01 Trial
Andre F et al Lancet Oncology 2014; 15(3):267-74
Mutation in Potentially Actionable Gene
Underwent Genomic TestingN = 2000
Genotype‐matched trial after genomic testing?
No (1211)
Genotype‐SelectedTrial N = 54
Genotype‐RelevantTrial N = 29
Yes (789)
No (706)Yes (83)
MD Anderson Institutional Profiling
54/2000 (3%) of pts who underwent genomic testing received genotype‐selected treatment
Meric-Bernstam F et al JCO 2015
107
Notes:
Notes:
NCI‐MATCH: Preliminary Phase
Activated 08/12/15; paused 11/11/15: 92 days
Patient cases registered for screening 795
Cases with samples submitted 739
Cases where labs were able to complete tumor testing
64587%
(N=739)
Cases with mutation matching 1 of 10 available treatment arms
569%
(N=645)
Patients matching specific eligibility criteria for, and assigned to, a treatment arm
335%
(N=645)
Patients who entered 7 of 10 available treatment arms
162.5%
(N=645)
Conley BA et al AACR 2016 Abstract CT101
32% 61%**
*Overall Response Rate p-value = 0.04**Any Reduction in Target Lesions p-value <0.0001
RECIST v1.1 ORR 20%*
Un‐matched Matched
RECIST v1.1 ORR 11%
Princess Margaret IMPACT/COMPACT Trials by Genotype‐Matching
Stockley TL et al Genome Medicine 2016; 8:109
108
Notes:
Notes:
Vall d’Hebron Breast Cancer Profiling
Oliveira M et al SABCS 2015 Abstract P2-08-13
N=280 patients with multi-gene panel results
N=127 (45%) no actionable
mutation(s)
N=153 (55%) ≥1 actionable
mutation(s)
N=62 (22%) genotype-directed
trials
Circulating Free DNA (cfDNA)
Opportunities
• Greater sensitivity of ddPCR & NGS cfDNA testing methods
• Alternative to metastatic biopsy for genomic characterization
• Detection of treatment‐emergent & polyclonal mutations
• Monitoring of treatment response
Challenges
• Assay standardization
• Optimal cut‐offs for clinical decision‐making
• Discordance with tissue‐based testing results
109
Notes:
Notes:
Development of MBC Precision Medicine
Arnedos, M. et al. (2015) Precision medicine for metastatic breast cancer—limitations and solutionsNat. Rev. Clin. Oncol. doi:10.1038/nrclinonc.2015.123
218
Take Home Messages
Driver mutations can be identified that are relevant for clinical trials with promising new investigational drugs
Likelihood of treatment‐matching following multi‐gene testing beyond these “druggable” driver mutations is low
Further evidence is needed before multi‐gene testing can be recommended in routine clinical practice
110
Notes:
Notes:
AR‐Driven Biology in TNBC Using a Gene Expression Profiling Assay
CBR16
PREDICT AR
AR IHC
YesNo
Not treated
Positive<1%≥ 1%
Data Cut‐off 01 July 2015
Parker, et al ASCO 2015
AR IHC PREDICT ARCBR16
Negative
• Hierarchical clustering according to biology
• Responders clustered within a recognized and distinct pattern that includes AR1‐5
– 521 genes significantly different in responders at 1% false discovery rate
• A diagnostic test was created and validated
Overall Survival by PREDICT AR Status
5662
5355
4946
4537
4227
4024
3213
156
116
32
Data cutoff 1Jul2015ITT = intent to treat; mOS = median survival; CI = confidence interval; .
Patients at riskPREDICT AR+PREDICT AR−
0
80
40
20
n = 118
PREDICT AR−mOS 32.3 weeks(95% CI: 20.7, 48.3)
PREDICT AR+mOS 75.6 weeks(95% CI: 51.6, 91.4)
0 8 16 24 33 41 49 61 64
Weeks
100
60
Ove
rall
Su
rviv
al (
%)
85
ITT Population
NCT01889238
PREDICT AR+ mOS 18.0 monthsPREDICT AR – mOS 7.5 months
111
Notes:
Notes:
ENDEAR: A Phase 3 Randomized, Placebo Controlled 3‐Armed Study
N = 780 patients with advanced, diagnostic‐positive TNBC who received either 0 or 1 prior line of systemic therapy for advanced disease (locally advanced or metastatic) will be randomized.
Acknowledgements
• Fabrice Andre
• Phil Bedard
• Andre Goncalves
• Paul Mainwaring
• Tira Tan
112
Notes:
Notes:
CHALLENGES and QUESTIONS
Interactive Dialogue Session with Faculty‐Facilitated Analysis and Discussion Focused on NGS Technologies to Optimize Assessment and Targeted Therapy for
Patients with Breast Cancer
Based on our evolving understanding of TNBC subtypes —including hypothesis related to polymerase inhibitors for BRCA‐mutated TNBC, antiandrogens for androgen receptor (AR)‐positive TNBC, fibroblast growth factor receptor (FGFR) inhibitors for TNBC harboring FGFR amplifications and so on — current treatment of TNBC based on molecular subsets and mutations, hybrid capture NGS comprehensive genomic profiling should be used routinely, in preference of other strategies, as a primary methodology for precision‐focused tumor assessment and targeted treatment in this patient population:
1) Strongly agree
2) Agree
3) Moderately agree
4) Agree somewhat
5) Disagree
Please Enter Your Response On Your Keypad
Audience Response System
Challenges and Questions
113
Notes:
Notes:
Given the preliminary studies recently presented at the SABC looking at immune checkpoint inhibition in TNBC, there us a rationale for looking at mutational loads using of hybrid capture‐based NGS comprehensive genomic profiling in this patient population:
1) Strongly agree
2) Agree
3) Moderately agree
4) Agree somewhat
5) Too early to form an opinions
Please Enter Your Response On Your Keypad
Audience Response System
Challenges and Questions
Comprehensive Genomic Profiling of Gynaecological Cancers to Optimize Therapy
A Science‐to‐Practice Update
Dr David SP TanBSc(Hons), MBBS(Hons)(London), PG Dip(Oncology), MRCP(UK)(Medical Oncology),
PhD(London), FRCP(Edin)Consultant Medical Oncologist :: National University Cancer Institute, Singapore :: National
University Hospital
114
Notes:
Notes:
Presenter Disclosure
Employment or Leadership Position: None
Consultant/Advisory Role: Astra Zeneca, Roche, D3 Singapore Stock Ownership: None
Honoraria: Astra Zeneca, Roche
Research Funding: Astra Zeneca, KaryopharmTherapeutics, National Medical Research Council (NMRC) Singapore, National University Hospital Cancer Institute Singapore (NCIS) Endowment fund, NCIS Centre Grant Expert Testimony: None Other Remuneration: Merck
Understanding the Complexity of Ovarian Cancers
115
Notes:
Notes:
Histological Subtypes of Epithelial Ovarian Cancer
Serous 75%High grade 80‐90%Low grade 10‐20%
Endometrioid 10%
Clear cell 10‐25%
Mucinous 5%
Different Subtypes = Different Outcomes
Ovarian Cancer
Chan et al Gynae Onc 2008
Stage I Stage III
Stage II Stage IV
116
Notes:
Notes:
Prat J Ann Oncol 2012
Ovarian Cancer
Different subtypes = Different Origins = Different Molecular Abnormalities
Mutations in Endometrial Cancers
Zhao et al. PNAS 2013Liang et al Genome Research 2012McConechy et al J Pathol 2012
Serous Endometrial Cancers (UPSC) CNAs
117
Notes:
Notes:
Mutations in Cervical Cancer
Ojesina et al Nature 2013
Ovarian Cancer: Evidence for Intratumoral Heterogeneity
Side population (SP) cells in ascites SP cellsNon ‐SP cells
Hu et al BJC 2010
SP cells
Non‐SP cells
118
Notes:
Notes:
Gynaecological Cancers AreHistologically, Molecularly, Intratumorally Heterogeneous:
Multiple Disease Entities
Therapeutic approaches need to start taking these issues into consideration
Targeting the Key Hallmarks of Cancer
Hanahan D & Weinberg RA (2011) Cell.
119
Notes:
Notes:
Targeting Homologous Recombination Repair Deficiency in Ovarian Cancer
Germline BRCA1/2Mutations: A Case of MultifactedEvolving Actionability in Ovarian Cancer
Associated with increased risk of breast and ovarian cancer
Fear of “genetic stigma” still a significant hindrance to uptake of testing in Asia
Concerns re insurance coverage in countries without nationalised health care system
BUT prognostic and therapeutic implications for patients
120
Notes:
Notes:
Chemotherapy Response in BMOC
Tan and Kaye 2015 ASCO Educational Book
BRCA+ vs Non-hereditary: Median OS 8.4 vs 2.9 years
HR = 0.3
BRCA mutants with ovarian cancer
Improved Responses and Increased Survival for BRCA1/2‐Mutant vs Non‐Hereditary EOC
Excellent responses to chemotherapy
Improved survival compared to non‐BRCA mutants
1 Tan et al JCO 20082 Bolton et al JAMA 20123 Sun et al PLoS One 2014
121
Notes:
Notes:
BRCA1/BRCA2 failure
Chromosomal instability
Cell death
Impaired HR repair
Alternative error prone repair
DNA replication fork arrest and collapse
HR‐based repair
Normal BRCA1/BRCA2
Chromosome stability
Cell survival
RAD51
Impaired HR Repair (BRCAness) and Platinum/PARP Inhibitor Sensitivity
HR, homologous recombination; DSB, double‐strand breakFarmer H et al. Nature 2005;434:917–921; Bryant HE et al. Nature 2005;434:913–917
H2AX
DSB
Alkylating agents (Platinum)
Genotoxic stress (Radiation)
PARP inhibitor
Phase I trial confirms excellent tolerance and expansion in 50 BRCA patients showed 46% response.
“this is nothing like chemotherapy
Fong P et al. N Engl J Med, 2009; 361, 123‐134;Fong P et al. J Clin Oncol, 2010; 28, 2512‐2519
Pre-clinical Early Clinical Trials(Phase I, incl. IB)
Randomised Clinical Trials (Phase II and III)
Randomised trial (maintenance therapy) showed marked PFS benefit particularly in BMOC
Farmer et al, Nature 2005 Ledermann et al, NEJM 2012 366 1382‐92Ledermann et al Lancet Oncology 2014
PARP : poly(ADP) ribose polymerase
Exquisite preclinical
efficacy of PARPi
PARP Inhibitors for BRCA1/2 Mutant Patients
2005 2015
122
Notes:
Notes:
Olaparib Monotherapy for BRCA1/2Mutant OC – The Route to Registration in the USA
• From the ongoing pooled analysis of almost 300 patients, data on subgroup of 137 patients who received ≥ 3 lines of chemo presented to FDA for accelerated approval.– response rate 34%; response duration 7.9m.
Note: approval does not distinguish between platinum‐sensitive and platinum resistant BMOC
Study 42: Matulonis et al, SGO 2015
Plat sens 35/74 47% resp
Plat resi 31/115 27% resp
Other PARP Inhibitors in Clinical Trials
AGO14699 (Rucaparib) Clovis/Pfizer IV/oral
MK 4827 (Niraparib) Merck Oral
ABT 888 (Veliparib) Abbott Oral
INO‐1001 Inotek IV
GP1201 Eisai Oral
CEP 9722 Cephalon Oral
BMN 673 (700x more active than olaparibin vitro)
BioMarin Oral
Iniparib (BSI‐201) is NOT a PARP inhibitor
123
Notes:
Notes:
HR DNA REPAIR PATHWAY
HR Repair: Not Just BRCA1/2
Levine et al 2013 TCGA
HRD Defects in High Grade Serous Ovarian Cancers
~50% of HGSOC have HRD related genetic aberrations
124
Notes:
Notes:
Pennington et al. Clin Cancer Res 2014;20:764‐775
115 of 367 (31.3%) EOC pts had deleterious mutations in 13 homologous recombination genes:
Germline homologous recombination mutations included 49 (13.4%) in BRCA1, 17 (4.6%) in BRCA2, and 22 (6%) in other homologous recombination genes, including BARD1, BRIP1, CHEK1, CHEK2, FAM175A, NBN, PALB2, RAD51C, and RAD51D.
Mutations in HR Genes and Platinum Response in EOC
BRCA1/2 Deficiency is Not Restricted to Ovarian and Breast Cancer
• Percentage of tumours showing loss of BRCA1 or BRCA2 (copy number) in some different tumour types
http://cancer.sanger.ac.uk/cosmic/gene/analysis?ln=BRCA1#dist Accessed 03/07/14;http://cancer.sanger.ac.uk/cosmic/gene/analysis?ln=BRCA2#dist Accessed 03/07/14
72.9
31.8
16.7
14.1
13
7.2 5.3
Loss of BRCA1
Ovary
Breast
Endometrium
Lung
Large intestine
CNS
Kidney
56.7
53.9
38.5
28.8
24.6
12.410.6 3.3
Loss of BRCA2
Lung
Ovary
Breast
CNS
Pancreas
Kidney
Endometrium
Large intestine
Tumours with BRCA defects represent potential targets for PARP inhibitors. extend use of PARP inhibitors to other tumour types with BRCAness phenotype
Abbreviation: CNS, central nervous system
125
Notes:
Notes:
Apart From Somatic/ Germline Mutation Testing for Individual HR genes, How Else Can We Identify HR Deficient Tumours?
BRCAmut
BRCA-like
Chromosome No.
Biomarker Negative
Hypothesis 1: Ovarian cancer patients with high genomic LOH suggesting BRCA-like signature will respond to rucaparib.
Hypothesis 2: Ovarian cancer patients who are “Biomarker Negative” (ie, with low genomic LOH) will not respond to rucaparib.
250
NGS=next-generation sequencing; mut=mutation; wt=wild type.
HRD Causes Genome‐Wide Loss of Heterozygosity that can be Measured by Comprehensive Genomic Profiling Based on NGS
BRCAwt
McNeish et al ASCO 2015
ARIEL 2 Study
126
Notes:
Notes:
Primary Efficacy Analysis: PFS in BRCAmut and BRCA‐like versus Biomarker Negative Patients
HRD SubgroupMedian PFS, mo (90% CI)
BRCAmut 9.4 (7.3, Not Reached)
BRCA‐like 7.1 (3.7, 10.8)
BiomarkerNegative
3.7 (3.5, 5.5)
SubgroupComparison
Hazard Ratio (90% CI)
BRCAmut vsBiomarkerNegative
0.47 (0.35, 0.64)
BRCA‐like vsBiomarkerNegative
0.61 (0.41, 0.92)
Progression-free survival by HRD molecular subgroup
BRCAmut
BRCA-likeBiomarker Negative
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
1.00.90.80.70.60.50.40.30.20.1
0
Time (months)
PF
S
CI=confidence interval.
McNeish et al ASCO 2015
NOVA Trial: Niraparib Maintenance in High‐Grade Serious Ovarian Cancer
127
Notes:
Notes:
NOVA Trial: Niraparib Maintenance in High‐Grade Serious Ovarian Cancer
The Myriad HRD Test
128
Notes:
Notes:
NOVA Trial: PFS in non‐gBRCAmut Subgroups
Key Practical Issues When Using PARP Inhibitors for the Treatment of BRCAMutated Ovarian Cancer
Testing for BRCA mutation ‐ when ?
Tumour testing for somatic mutation?
Which line ‐ 2nd or 3rd line maintenance?
129
Notes:
Notes:
How are Patients with BRCA1/2Mutations Identified?
Approximately 35% patients with BRCA mutated ovarian cancer have NO family history of cancer
25% BRCA mutated ovarian cancer occurs in > 60 years old
Somatic mutations are found in 5‐6 % HGSOC
All patients with high grade ovarian cancers should be tested for a BRCAmutation
Histological Type is NOT a Good Predictor of BRCAMutation
Frequency of germline BRCA mutation
Total population
Serous Endometrioid Clear Cell Mucinous
Risch HA et al1 13.2% 18.0% 7.1% 0%
Soegaard M et al2 5.8% 5.4% 5.4% 9.1% 0%
Malander S et al3 8% 7.6% 13.0% 12.5% 0%
Alsop4 14.1% 16.6% 6.3% 8.4% ‐
1. Risch HA, et al. J Natl Cancer Inst 2006;98:1694‒1706; 2. Soegaard M et al. Clin Cancer Rev 2008;14:3761‒3767; 3. Malander S et al. Eur J Cancer 2004;40:422‒428. i4 Alsop et al J Clin Oncol 2012 30:2654-2663
130
Notes:
Notes:
Trial PopulationFrequency of
somatic mutations
Hennessy et al1 235 unselected ovarian cancers 11/235: 5%
TCGA network2489 high‐grade serous ovarian
cancers19/316: 6.1%
Pennington et al3 390 ovarian carcinomas 20/367: 5.5%
Ledermann et al4265 high‐grade, recurrent
ovarian carcinomas, platinum‐sensitive
18/265: 6.8%
BRCA Somatic Mutations in Ovarian Cancer
1. Hennessy BT et al. J Clin Oncol 2010;28:3570–3576; 2. The Cancer Genome Atlas Research Network (Suppl.). Nature. 2011;474(7353):609–6153. Pennington P et al. Clin Cancer Res 2014;20(3):764–775; 4. Ledermann J et al. Lancet Oncol 2014;15(8):852–861.
BRCA Somatic Mutations in Ovarian Cancer
Knowledge of BRCA1/2 or other HRD status in patients with ovarian cancer will have crucial implications for choice of therapy in the primary and recurrent disease setting.
All patients diagnosed with non‐mucinous ovarian cancer should be referred for germ‐line or somatic testing of ovarian cancer for BRCA1/2 mutations
?Clinical relevance of other HRD defects – e.g. RAD51C/ ATM
131
Notes:
Notes:
• Key growth factor‐mediated signal transduction pathway
• PIK3CAmutations reported in:
18% of breast ca
17%‐33% of cervical ca
39% of endometrial ca
40% of ovarian ca
• Associated with platinum and taxaneresistance
PI3‐kinase‐AKT‐mTOR Pathway
Tan et al BJC 2013Janku et al J Clin Oncol 2012 From Weigelt and Downward; Front. Oncol., 2012
CBio Portal Oncoprint of mTOR pathway genes in ovarian cancerRed = gene amplifications, Blue = deletions and Green dots = mutations.
Musa & Schneider Transl Cancer Res 2015;4(1):97-106
Frequency of Gene Alterations Noted in mTOR Pathway in Epithelial Ovarian Cancer (n=580) from TCGA
132
Notes:
Notes:
Efficacy of Targeted Sequencing with Matched PI3K/AKT/mTOR inhibitor Therapy
Waterfall plot of patients with breast and gynae cancer harbouring PIK3CA mutations treated with phosphatidylinositol 3‐kinase/AKT mammalian target
of rapamycin inhibitors
Janku et al J Clin Oncol. 2012 Mar 10;30(8):777‐82.
ORR = 30%
Evidence for efficacy in targeting this pathway in EOC remains limited
Good predictive biomarkers urgently needed PIK3CA and PTEN mutations alone do not appear sufficient ?Genotype specificity e.g. H1047R vs E545K, AKT1E17KmutationsContext may be crucial e.g. endometrioid/ clear cell vs high grade serous EOC
Inhibitors of this pathway may have more of a cytostatic effect leading to disease stabilisation vs tumor shrinkage
Duration of response usually short PI3K pathway inhibition alone may be insufficient, and combination strategies may be required.
Results from combination studies using PIK3CA/Akt with chemotherapy and other pathway inhibitors in EOC currently awaited
PIK3CA AKT mTOR Pathway Summary
133
Notes:
Notes:
What’s Next for Precision Medicine in Gynaecological Cancer?
Epi-AMesStem-AStem-B
Epi-B
Mesenchymal
IFN-inducibleMHC Class IIIgs
Stem-A
Epi-A/Stem
Epithelial
VCAM1
ZEB1
FN1
PDGFRA
TWIST1
OAS1/2
HLA-Ds
IGH/K/Ls
MYCNCDH3/2
NCAM1TOP2A/BUB1/CCNE1/CENPA
LGR5
WNT5A
DKK3
SFRP1
PROM1
XIST
KRT6A/17/14/19/7
EPCAMCD24
CDH1
MUC1
Epi-A
Epi-B
Mes
Stem-A
Stem-B
p < 0.0001
Five major subgroups were identified within EOC that harbor distinctive signatures of Epithelial, Mesenchymal, or Stem‐like, which confer to different
clinical survival outcomes.
EOC Gene Expression Molecular Subtypes
Tan, Miow & Huang et al., EMBO Mol Med, 2013
Meta-analysis (Tan, Miow, & Huang et al., EMBO MM, 2013) of 1,538 EOCs including all histological types.
5 subtypesshowing survival differences
134
Notes:
Notes:
Molecular Subtypes Likely to Reflect Aggregate Phenotype of Tumour Cells and Microenvironmental Factors
Balkwill et al Journal of Cell Science, 2012
Antiangiogenic Therapy with Bevacizumab Improves PFS in First Line Treatment of Ovarian Cancer
135
Notes:
Notes:
Is the response to bevacizumab affected by the molecular subtype of ovarian cancer?
Is Response to Bevacizumab Dependent on Molecular Subtype? – ICON 7 data
Gourley et al ASCO 2014
Outcome of ‘Immune’ and ‘Proangiogenic’ Groups of Ovarian Cancer in ICON 7
Bev had adverse effect on PFS in immune subgroup
Benefit in pro-angiogenic subgroup
Control arm ICON7Immune and proangiogenic groups
Gourley et al ASCO 2014
136
Notes:
Notes:
Anti‐Microtubule Agents for Stem‐A Subtype
Dr. Clare Scott WEHI, PDXs from Mayo Clinic
Tan, Miow & Huang et al., EMBO Mol Med, 2013
Stem-A/C5-like PDX Stem-A/C5-like Subtype Clinical Trial
Immunotherapy:Anti‐tumour Immune Activation and Checkpoint Inhibition
Gettinger et al 2016
137
Notes:
Notes:
Immunotherapy and Gynaecological Cancers
Immune microenvironment affects disease outcomes in Gynae cancers:
• Ovarian:
– Intraepithelial CD8 TILs and a high CD8+/Treg ratio associated with favorable prognosis (Sato et al, PNAS 2005)
• Cervix:
– Largely HPV driven and reversed CD4/CD8 ratios of tumor‐infiltrating lymphocytes correlated with disease progression (Sheu et al, Cancer 1999)
• Endometrium:
– Presence of tumor‐infiltrating lymphocytes is independent prognostic factor in type I and II endometrial cancer (de Jong et al, Gynaeol Onc 2009)
Herzog et al SGO 2015
CD8+ Killer T Cells (CD8 TIL) are Associated with Survival in HGSC
100
50
0
% overall survival
Years
Dense CD8+ TIL Sparse CD8+ TIL
BCCA/VGH cohorthigh-grade serousoptimally de-bulkedn = 200p = 0.0008
5 10 150 Clarke, B. et al. 2009Milne, K. et al. 2009
138
Notes:
Notes:
Partial Response to PDL1 Inhibitor Avelumab in Metastatic Clear Cell Ovarian Cancer
DisisetalASCO2015
Baseline: 69 mm RLL lesion Week 25: 41 mm (‐40.6%)
• 65 years old; 6 prior lines for metastatic disease• 4th assessment cycle, still on treatment
Ovarian cancers developed at mean 48 years of age
Histologically, endometrioid (35%) and clear cell (17%) tumors were overrepresented.
The underlying MMR gene mutations in these families affected MSH2 in 49%, MSH6 in 33% and MLH1 in 17%.
Immunohistochemical loss of the corresponding MMR protein was demonstrated in 92% of tumors
Gynecologic Oncology 121 (2011) 462–465
139
Notes:
Notes:
MOCCA Trial: A Multicentre Phase II randomised trial of MEDI4736 (DURVALUMAB) versus physician’s choice chemotherapy in recurrent
ovarian clear cell adenocarcinomas (MOCCA)
Relapsed Clear Cell Cancer Ovarian Cancer
(>70% clear cell)
Inclusion-Histologically confirmed- WT1 negative-Relapsed after at least 1 line of platinum-based chemotherapy-Measurable disease by RECIST 1.1 -ECOG 0 / 1
Exclusion-Concurrent use of experimental anti-cancer agent-Untreated brain mets
MEDI-4736 (Durvalumab)
1500mg 4 weekly for max.12mths
CAELYX or Investigator’s choice if
prior Rx included caelyx (but antiangiogenic
therapy not permitted)
Crossover on PD
2:1 randomisation
2
1
Primary Endpoint: MOCCA: median PFS improvement from 10 weeks to 20 weeks
Secondary Endpoint: RECIST/ GCIG response
Singapore (GOGS)Korea (KGOG)
Australia (ANZGOG)
N = 31
N = 15
Checkpoint Inhibitors for Endometrial Cancer
MSI‐High tumours:
• ~20–30% of endometrial cancers are characterized by high microsatellite instability (MSI‐H) due to defects in DNA mismatch repair (MMR) pathway.
Replicative DNA polymerase epsilon (POLE) ultra‐mutated tumours:
• 5% of endometrial cancers ‐ predominantly endometrioid, grade III, and associated with peritumoral and tumor infiltrating lymphocytes
Both POLE and MSI‐H tumors exhibit higher numbers of CD8+ TIL characterized by PD1 overexpression
POLE tumours have highest number of predicted neoantigens per tumor sample, followed by MSI‐H tumors, and microsatellite stable tumors
Howitt et al JAMA oncol 2015Billingsley et al Cancer 2015
140
Notes:
Notes:
Endometrial Cancer Molecular Stratification
From TCGA Levine et al
MSI‐H
Clinical Responses to Pembrolizumab
Treatment in MMR Deficient vs Proficient
Cancers
Le DT et al. N Engl J Med 2015;372:2509-2520
RR of 40% and 71% in MSI‐H colorectal and non‐colorectal cohorts (including 2 patients with endometrial cancer) respectively
141
Notes:
Notes:
Immunotherapy for Endometrial Cancer
• A subset of highly immunogenic endometrial cancers characterized by ultra‐high somatic mutations rates resulting from defects in the proof‐reading function of POLE
From TCGA Levine et al
Mutations in the Replicative DNA Polymerase Epsilon (POLE) – the Ultramutated POLE Subtype
Landscape of TMB in >60,000 Clinical Cases
Immunotherapy approval Heavily investigated immunotherapy indication
Fabrizio DA et.al., ESMO 2016 Presentation #: 52O
142
Notes:
Notes:
Frequent Cases with High Mutation Burden in Nearly Every Tumor Type
20 mutations / megabase
Immunotherapy approval Heavily investigated immunotherapy indication
Fabrizio DA et.al., ESMO 2016 Presentation #: 52O
MSI‐High Specimens are a Subset of High TMB Specimens
Microsatellite stable
Microsatellite ambiguous
Microsatellite instable
All specimensn = 46,465
MSI and TMB High n = 550
TMB-Highn = 3,531
Fabrizio DA et.al., ESMO 2016 Presentation #: 52O
143
Notes:
Notes:
Key Questions for Immune Checkpoint Inhibitor Therapy in Gynaecological Cancers
1)What are the predictive biomarkers for checkpoint inhibitors in gynae cancers? PD‐1/PD‐L1 – not consistently predictive MMR/ MSI‐H – but POLE tumours are MSS Tumour mutational burden?
2) How can we overcome immune resistance to enhance the efficacy of checkpoint inhibitors in gynae cancers?
Case Study –Molecular Profiling of Metastatic Endometrioid Ovarian Cancer
50 yr old female with platinum resistant endometrioid ovarian cancer with lung, liver and sacral metastases
Progressed on 3 prior lines of chemotherapy
Referred to NCIS phase I unit ?targeted therapy
144
Notes:
Notes:
• Screening for actionable targets using NGS and copy number analysis in tumour tissue
• All cancer types
• Establish prevalence of mutations in local patients
• Evaluate clinical impact of molecular profiling (RR/PFS/OS)
Integrated Molecular Analysis of Cancer (IMAC) at NCIS
Patient with progressive disease
Consent, screening, tissue collection
Next generation sequencing using Ion Torrent PGM Sequencer withminimum requirement of 10‐
40ng DNA
FISH and CISH/ Gene expression profiling/ immunohistochemistry
Final Profiling Report
Results reviewed by IMAC tumour
board
Interpretation and
recommendation
Medical Oncologist
FFPE or fresh tissue
?Matched Therapy
• FFPE tissue ok
• Minimum of 10ng DNA required
145
Notes:
Notes:
AKT1 inhibitor
Patient with AKT1 E17K mutant endometrioid ovarian cancer with lung, liver and sacral metastases matched to an experimental AKT1 inhibitor
Mol Cancer Ther; 14(11) November 2015
Aug 2015 Oct 2015
Summary: Targeted Therapy for Gynaecological Cancer
New era of targeted therapy in gynaecological cancer Improved outcomes Increase the benefit:risk ratio Improved understanding of disease biology acceleration of new therapeutic developments
Challenges: Predictive biomarkers: Bevacizumab/ PARP inhibitors/ Immunotherapy Optimal timing of new agents in treatment Increase efficacy and durability of response ?combination studies Management of toxicity Cost
Translational studies important to understand reasons for success and failure and to gain new insights in tumour biology that may provide new therapeutic opportunities
146
Notes:
Notes:
THANK YOU
CHALLENGES and QUESTIONS
Interactive Dialogue Session with Faculty‐Facilitated Analysis and Discussion Focused on NGS Technologies to Optimize Assessment and
Targeted Therapy for Patients with Gynaecological Malignancies
147
Notes:
Notes:
Based on our evolving understanding of molecular markers and genomic alterations in gynecological cancers — potentially actionable markers extending beyond BRCA1/2, but also including HRD defects, PI3K signaling, and others — hybrid capture NGS comprehensive genomic profiling should be used routinely as a primary methodology for precision‐focused tumor assessment and targeted treatment in patients with advanced ovarian cancer and other gynecological malignancies:
1) Strongly agree
2) Agree
3) Moderately agree
4) Agree somewhat
5) Disagree
Please Enter Your Response On Your Keypad
Audience Response System
Challenges and Questions
ESMO Asia 2016
An Oncologist’s Bid to Personalise Patient Care with
Genomic Profiling
Dr Wong Seng WengMedical Director & Consultant Medical Oncologist
The Cancer CentreSingapore Medical Group
Paragon Medical & Mt Elizabeth Novena Specialist CentreVisiting Consultant
Mt Elizabeth Hospital & Mt Elizabeth Novena HospitalAdjunct Clinician Scientist, Institute of Bioengineering and Nanotechnology,
Agency for Science, Technology and Research (A*STAR)
148
Notes:
Notes:
Comprehensive Genomic Profiling for Gastrointestinal Cancers
The Foundational Role of Mutation‐Based Molecular Targets to Optimize Cancer Management
Presenter Disclosure
HonorariaAstra Zeneca
Bayer Schering Pharma
Bristol‐Myers Squibb
Eisai
Eli Lilly
Hospira
Merck Serono
Merck Sharp & Dohme (MSD)
Mundipharma
Novartis
Orient Europharma
Pfizer
Roche
Advisory BoardsAstra Zeneca
Bayer Schering Pharma
Bristol‐Myers Squibb
Eli Lilly
Hospira
Merck Serono
Merck Sharp & Dohme (MSD)
Novartis
Orient Europharma
Pfizer
Roche
149
Notes:
Notes:
What Oncologists Want…
150
Notes:
Notes:
What Oncologists Want…
Patient
What Oncologists Want…
Patient
Family
151
Notes:
Notes:
What Oncologists Want…
• Pick therapy destined to work
Patient
Family
What Oncologists Want…
• Pick therapy destined to work
• Drop therapies destined to fail
Patient
Family
152
Notes:
Notes:
What Oncologists Want…
• Pick therapy destined to work
• Proven therapies
• Drop therapies destined to fail
Patient
Family
What Oncologists Want…
• Pick therapy destined to work
• Proven therapies
• Investigational therapies
• Drop therapies destined to fail
Patient
Family
153
Notes:
Notes:
What Oncologists Want…
• Pick therapy destined to work
• Proven therapies
• Investigational therapies
• Drop therapies destined to fail
Patient
• PickupfamilialcancersyndromesFamily
So Many Tools…
Mikail et al, Cancer Lett., 2016; 374(2): 187-91.
154
Notes:
Notes:
So Many Genes…The Cancer Genome Atlas (TCGA)
So Many Targets…
155
Notes:
Notes:
The Cancer Genome Atlas (TCGA)
The Cancer Genome Atlas (TCGA)
156
Notes:
Notes:
Upfront Molecular Testing in Advanced GI Cancers
Lights, Camera, Action…
S. Mikhail, et al. Oncotarget. 2015;6:22206‐13.
89% (50/56) with at least 1 actionable mutation.
Prevalence
Cell‐cycle abnormalities: 58%
HER2 amplification: 30%
PIK3CAmutations: 14%
MCL1 amplification: 11%
PTEN loss: 9%
MET amplification: 5%
Picking the Winners…Proven Therapies…Is It Time to Drop IHC, FISH?
157
Notes:
Notes:
Picking the Winners…Proven Therapies…Is It Time to Drop IHC, FISH?
66% (12/18) HER2+ by IHC/FISH showed amplification by NGS.
92% (12/13) HER2 amplification by NGS were HER2+ by IHC/FISH.
S. Mikhail, et al. Oncotarget. 2015;6:22206-13.
Experience from MD Anderson
Total cohort: 2000
GI cancers: 19%
F. Meric-Bernstam, et al. J. Clin. Oncol. 2015;33:2753-62.
Therapy Pancreas Colorectal Gastroeso
W KRAS 79% 67% 16%
W/O KRAS 16% 31% 11%
Picking the Winners…Investigational Therapies…
Is There Always Action after the Lights and Camera?
158
Notes:
Notes:
Experience from MD Anderson
Total cohort: 2000
GI cancers: 19%
39% (789) with at least 1 actionable mutation
Number enrolled in genotype‐matched trials:__________
F. Meric‐Bernstam, et al. J. Clin. Oncol. 2015;33:2753‐62.
Picking the Winners…Investigational Therapies…
Is There Always Action after the Lights and Camera?
Experience from MD Anderson
Total cohort: 2000
GI cancers: 19%
39% (789) with at least 1 actionable mutation
Number enrolled in genotype‐matched trials:__________
Poll:1) 10‐20%
2) 20‐50%
3) 50‐80%
4) 80‐100%
F. Meric‐Bernstam, et al. J. Clin. Oncol. 2015;33:2753‐62.
Picking the Winners…Investigational Therapies…
Is There Always Action after the Lights and Camera?
159
Notes:
Notes:
Experience from MD Anderson
Total cohort: 2000
GI cancers: 19%
39% (789) with at least 1 actionable mutation
Number enrolled in genotype‐matched trials:
83 patients
11% of those with actionable mutations
4% of cohort
F. Meric-Bernstam, et al. J. Clin. Oncol. 2015;33:2753-62.
Picking the Winners…Investigational Therapies…
Is There Always Action after the Lights and Camera?
Experience from MD Anderson
Total cohort: 2000
Number enrolled in genotype‐matched trials: 83 patients
11% of those with actionable mutations
4% of cohort
Reasons cited: Patient preference
Poor PS (median delay 26 days)
Lack of trials/trial slots
Trial ineligibility
Insurance denialF. Meric‐Bernstam, et al. J. Clin. Oncol. 2015;33:2753‐62.
Picking the Winners…Investigational Therapies…
Is There Always Action after the Lights and Camera?
160
Notes:
Notes:
Experience from France (SHIVA trial)
8‐centres phase II trial
195 patients (actionable mutations by NGS with available targeted therapies) – 17% with GI cancers
99 in experimental arm (targeted therapy)
96 in control arm (investigators’ choice of conventional therapy)
Median PFS
Experimental arm:_____
Control arm:______
C. Le Tourneau, et al. Lancet Oncol. 2015;16:1324‐34
Picking the Winners…Investigational Therapies…
Is There Always an Oscar After the Action?
Experience from France (SHIVA trial)
8‐centres phase II trial
195 patients (actionable mutations by NGS with available targeted therapies) – 17% with GI cancers
• 99 in experimental arm (targeted therapy)
• 96 in control arm (investigators’ choice of conventional therapy)
Median PFS• Experimental arm: 2.3 months
• Control arm: 2.0 months (HR: 0.88, p=0.41)
• Grade 3/4 toxicity: No difference
C. Le Tourneau, et al. Lancet Oncol. 2015;16:1324‐34
Picking the Winners…Investigational Therapies…
Is There Always an Oscar After the Action?
161
Notes:
Notes:
One man’s experience from Singapore
68‐year‐old Chinese lady with advanced pancreatic cancer
Peritoneal metastases
PS: 2
Treatment:
• Nab‐paclitaxel/gemcitabine: PR
• mFOLFOX6: SD
• Irinotecan: PD
Picking the Winners…Investigational Therapies…
Is There Always Action after the Lights and Camera?
Picking the Winners…Investigational Therapies…
Is There Always Action after the Lights and Camera?
162
Notes:
Notes:
Is There Hope?
One man’s experience from Singapore
50‐year‐old Chinese gentleman with stage III colon cancer
K/N‐RAS wild type
MSI: low
Treatment:
Adjuvant XELOX
Peritoneal metastases 2 months post adjuvant
FOLFIRI/cetuximab: PD
Regorafenib: PD
mFOLFOX/bevacizumab: PD
Is There Hope?
163
Notes:
Notes:
VemurafenibMonotherapy: Not Effective in BRAFmCRC
Corcoran et al, Cancer Discov. 2012; Montero‐Conde et al, Cancer Discov. 2012; Prahallad et al, Nature 2012.
RAS
BRAF
CRAF
MEK
ERK
EGFR
PROLIFERATION & SURVIVAL
NODRUG
‐100
‐75
‐50
‐25
0
25
50
75
100
%Chan
ge From Baselin
e
(Sum of Lesion Size)
5% Response Rate
Kopetz et al, J Clin Oncol, 2010.
‐100
‐75
‐50
‐25
0
25
50
75
100
%Chan
ge From Baselin
e
(Sum of Lesion Size)
MEK116833 Phase 1/2 Study Design
Dabrafenib+Panitumumab
D 150 mg BID P 6 mg/kg Q2W N = 20
Dabrafenib+Panitumumab+ Trametinib
D 150 mg BID P 6 mg/kg Q2WT 2 mg QDN = 35 total; N = 24 RP2R
Panitumumab+ Trametinib
EGFR AcquiredResistance
P + T
N = 12/20
N = 2/20
P 6mg/kg Q2WT 2mg QD
N = 5 N = 1
March 16, 2015 Data Cut
164
Notes:
Notes:
Best Response With Confirmation Percent Change from Baseline at Maximum Reduction in Tumor Measurement
• D+P (N = 20)• CR+PR: 2 (10%)• Stable disease: 16 (80%)
• D+P+T (N = 35) • CR+PR: 9 (26%) • Stable disease: 21 (60%)
Color: confirmed responseHeight of bar: best unconfirmed response
‐80
‐60
‐40
‐20
0
20
40
60
80
100
‐100
‐80
‐60
‐40
‐20
0
20
40
60
80
100
‐100
*Maximum reduction from baseline is 0%+RP2R cohort
+
++
+ + + + + + + + + + + + + + + + + + + + +*
Maxim
um % Chan
ge from Baseline
*Maximum reduction from baseline is 0%
Progressive diseaseStable diseasePartial responseComplete response
Maxim
um % Chan
ge from Baseline
* *
Duration on Study
• D+P (N = 20)• > 6 months: 5 (25%)
• D+P+T (N = 35)• > 6 months: 9 (26%) • > 1 year: 4 (11%)
87654321 9 10 11 12 13 14 15 16 17 180
Progressive diseaseStable diseasePartial responseComplete response
First responseDisease ProgressedCrossover
Progressive diseaseStable diseasePartial responseComplete response
First responseDisease Progressed
Ongoing
Treatment Duration (Months)
Median Duration of Response: 5.4 mos (2.7, not available)
165
Notes:
Notes:
Is There Hope?
One man’s experience from Singapore
50‐year‐old Chinese gentleman with stage III colon cancer
• K/N‐RAS wild type
• MSI: low
Treatment:
• Adjuvant XELOX
• Peritoneal metastases 2 months post adjuvant
• FOLFIRI/cetuximab: PD
• Regorafenib: PD
• mFOLFOX/bevacizumab: PD
• Dabrafenib/Panitumumab/Trametinib: PR
• PFS: 6 months
Dropping the Losers…
Resistance to anti‐EGFR monoclonal antibodies in mCRC
KRAS, NRAS exons 2, 3, 4
BRAF
PIK3CA
PTEN loss
MET amplification
High Resolution Melting
KRAS (codons 12 & 13)• Direct sequencing: 45%
• HRM: 60%
• Resistant mutation (inclusive of BRAF, PIK3CA, PTEN) detected in 87% of non‐responders
J.G. Guedes, et al. BMC Cancer. 2013;13:169
166
Notes:
Notes:
Dropping the Losers…
Resistance to Trastuzumab in advanced gastroesophageal cancers
MET amplification
PTEN loss
A.K. Paulson, et al. Mol. Cancer Res. 2013;11:1112‐21
X. Zhang, et al. Oncology. 2015; 88: 76‐85
Colorectal cancer
Familial CRC syndromes: 5%
Familial Adenomatous Polyposis (FAP): APC gene mutation
Polyposis with MUTYH gene mutation
Hereditary Nonpolyposis CRC: MLH1, MSH2, MSH6, PMS2
Familial component: 30%
NGS vs Sanger sequencing
Miss‐rate: 4%
Homopolymeric DNA sequences
M. Simbolo, et al. Heredit. Cancer Clin. Pract. 2015;13:18
We Are Family…
167
Notes:
Notes:
We Are Family…
Gastric cancerHereditary diffuse gastric cancer: CDH1mutation
Other hereditary cancer syndromes: BRCA2
S. Mikhail, et al. Oncotarget. 2015;6:22206‐13.
CHALLENGES and QUESTIONS
Interactive Dialogue Session with Faculty‐Facilitated Analysis and Discussion Focused on NGS Technologies to Optimize Assessment and Targeted Therapy for
Patients with GI Malignancies
168
Notes:
Notes:
Based on our evolving understanding of molecular markers and genomic alterations in GI malignancies, hybrid capture NGS comprehensive genomic profiling should be used routinely as a primary methodology for precision‐focused tumor assessment and targeted treatment in patients with advanced GI cancers.
1) Strongly agree
2) Agree
3) Moderately agree
4) Agree somewhat
5) Disagree
Please Enter Your Response On Your Keypad
Audience Response System
Challenges and Questions
Top Related