Cancer genomics and personalised medicine: a pathologist’s view
Stephen B FoxDepartment of Pathology
Peter MacCallum Cancer CentreMelbourne, Australia
!
What does pathology do?
• Pre-treatment characterisation of tumour
• diagnosis (core biopsy, cytology)
• staging (nodal involvement)
• treatment (neoadjuvant)
• Post-surgical assessment of primary
• is surgery complete?
• which adjuvant therapy?
• identification of special types e.g. BRCA
Pathology
Tumour size (T) Lymph nodes (N)
Mets (M)chemotherapy
Chemotherapy
Grade c-erb-B2
Herceptin™
ER
Tamoxifen
Margins-more surgery required?
Pathology- wide local excision+LN
Tumour size (T) Lymph nodes (N)
Mets (M)chemotherapy
Chemotherapy
Grade c-erb-B2
Herceptin™
ER
Tamoxifen
Margins-more surgery required?
Pathology- wide local excision+LN
Surgeon RadiologistPatient Oncologist
radiation medical
Pathology world view
SurgeonPathology
Conventional anatomical pathology
19th century 21st century
Cancer is a genetic disease
MyTumourGenome
• x
•140 genes recurrently mutated in cancer• 2–8 driver mutations per cancer• 12 signalling pathways
Diagnostic practice
Diagnostic practice
Future of tumour classification
Cell August 2014
The controversy over the microscope: Paris 1840s
Perspectives on Science 2004
• Would the microscope reveal any more about tumours than the naked-eye?
• The problem was......the microscope and tumour classification
• Should tumours be classified according to microscopy or the usual clinical observations?
• microscopy was threatening the foundations of medicine
Challenge of lung carcinoma classification
vs
Morphology Molecular pathology
Kerr J Clin Pathol 2013
Challenge of lung carcinoma classification
vs
Morphology Molecular pathology
Genomics colliding with oncology
Kerr J Clin Pathol 2013
Oncology milestones in personalised medicine
TrastuzumabHER2+ breast cancer
1998
Imatinib (kit & PDGFR) CML & GIST
2001
Gefitinib/erlotinib (2003/4)
EGFR+ NSCLC
2003
Cetuximab
KRAS wild type CRC
2004 2005
Irinotecan, CRC UGT1A1 assay
2006
Rituximab CD20+ NHL
2010
Vemurafenib
BRAF, melanoma
Crizotinib
ALK, ROS1, MET+ NSCLC
TrastuzumabHER2+ gastric cancer
2009
RET Cabozantinib
Drug Target Tumour
Antibodies
trastuzumab/pertuzumab HER2 breast, stomachcetuximab/panitumamab EGFR CRCSmall molecule inhibitorsimatinib KIT GIST, melanomagefitinib/erlotinib/afatinib EGFR lung calapatinib HER2 breast, stomach,
lungsorafenib/sunitinib TKs HCC, RCCVemurefenib BRAF MelanomaCritzotinib EML4-ALK/ROS1/RET lung ca
Targeted therapeutics
Old lung pathology
Small cell carcinoma
Non-small cell carcinoma
Kate Moodie, Peter Mac
Old lung pathology
Small cell carcinoma
Non-small cell carcinoma
Kate Moodie, Peter Mac
New lung pathology
Squamous cell carcinoma
Adenocarcinoma
NSCLC NOS
p63, probably SCC
TTF1, probably AdC
Moleculartesting
No moleculartesting..yet
NSCLC subtype
IHC
EGFR ALK ROS RET
NTRK1
Mutations in lung adenocarcinoma
TCGA Nature 2014
Frequency of mutations in lung adenocarcinoma
TCGA Nature 2014
Clinicopathological profile of ALK-Positive NSCLC
• ALK gene rearrangements occur in NSCLC more frequently in:
• Adenocarcinoma
• Never or light smokers
• Younger patients
• ALK rearrangements rarely co-exist with ROS1 or RET rearrangements, EGFR and KRAS mutations.
• ALK-EML4 first discovered in NSCLC in 2007 by Soda et al.
• Small inversion leading to the fusion with EML4.
• Multiple ALK-EML4 variants. • Rearrangements present in ~3-5%
NSCLC.
20
20
20
20
20
20
13
20
6a
6b
15
14
V1V2
V3a
V3b
“V4”
V420 TFG-ALK
20 KIF5B-ALK
16%
22%
29%
33%
E13:A20 E6a/b:A20Unknown Others
ALK
EML4
Inversion
ALK-EML4 in NSCLC
Rapid response to Crizotinib in ALK-Positive NSCLC
Before Crizotinib After 2 cycles CrizotinibImage courtesy of Benjamin Solomon
48 y.o. female never smoker with stage IV ALK+ve NSCLC
Image courtesy of Benjamin Solomon
Before Crizotinib After 2 cycles Crizotinib
Tumour responses in patients with ALK-Positive NSCLC (n=106*)
% D
ecre
ase
or In
crea
se fr
om B
asel
ine
Progressive disease
Stable disease
Partial response
Complete response
100
80
60
40
20
0
–20
–40
–60
–80
–100
A8081001 clinical study
Objective Response Rate: 61%
*excludes patients with early death and indeterminate response
Does any of this matter?
Kris et al JAMA 2014
Shaw Lancet 2011
Shaw NEJM 2013
• Prognosis 73 months
• TreatmentA+B+C Omit D
Next generation sequencing in cancer pathology: (r)evolution
Solid tumour testing
How to bring molecular pathology to patients?
The Australian model: a vacuum• constraints
• capital costs• accreditation
• laboratory• scientific and clinical
staff• quality assurance
• bioinformatics• clinical reporting• TATs• input material (FFPE)
How to bring molecular pathology to patients: the international model
Cancer 2015• whole of health system “Framingham” type
cancer cohort on a scale to impact Victorian health outcomes-10,000 patients
• prospective collection of patients with newly diagnosed cancer
• Total cancer journey from diagnosis, agnostic to cancer type, age, gender, stage, locus of care• epidemiology• QoL questionnaires• clinical information, treatment and follow up• conventional and molecular pathology• clinical trials
• health economic modelling of cancer care and personalised medicine • trials, treatments, risk management, policy• linkages to Cancer Registry, State and Federal
databases
www.cancer2015.org
Cancer 2015
• Primary Aims• map variation in cancer care and outcomes • assess quality of life• measure health costs and societal value of cancer treatment• map the true prevalence of actionable mutations agnostic to
cancer type• Secondary Aims
• Facilitate the integration of molecular pathology into routine cancer care
• Provide a large-scale longitudinal cancer cohort for research
www.cancer2015.org
Cancer 2015• Recruiting sites
• PMCC
• Barwon Health
• Geelong
• Warnambool
• RMH
• Cabrini
www.cancer2015.org
RegistrationLIS NGS Testing Report
GeelongWar’bool
Cabrini
RMH
PMCC
2015 Dbase
Path data
System and infrastructure
FFPE
www.cancer2015.org
Research test Diagnostics
Molecular diagnostics accreditation
Illumina TruSeqTM custom panel
MiSeq
250ng DNA
14days
• FGFR3 • FLT3 • GNA11 • GNAQ • GNAS • HNF1A • HRAS • IDH1 • JAK2 • JAK3 • KDR • KIT • KRAS • MET • MLH1 • MPL1
• ABL1 • AKT1 • ALK • APC • ATM • BRAF • CDH1 • CDKN2A • CSF1R • CTNNB1 • EGFR • ERBB2 • ERBB4 • FBXW7 • FGFR1 • FGFR2
• NOTCH1 • NPM1 • NRAS • PDGFRA • PIK3CA • PTEN • PTPN11 • RB1 • RET • SMAD4 • SMARCB1 • SMO • SRC • STK11 • TP53 • VHL
Molecular pathology
1094
936 (86%)
854 (78%)
Total tumour samples received
DNA >10ng/ul
Samples passed QC metric
• ABL1
• AKT1
• ALK
• APC
• ATM
• BRAF
• CDH1
• CDKN2A
• CSF1R
• CTNNB1
• EGFR
• ERBB2
• ERBB4
• FBXW7
• FGFR1
• FGFR2
• NOTCH1
• NPM1
• NRAS
• PDGFRA
• PIK3CA
• PTEN
• PTPN11
• RB1
• RET
• SMAD4
• SMARCB1
• SMO
• SRC
• STK11
• TP53
• VHL
• FGFR3
• FLT3
• GNA11
• GNAQ
• GNAS
• HNF1A
• HRAS
• IDH1
• JAK2
• JAK3
• KDR
• KIT
• KRAS
• MET
• MLH1
• MPL1
Br J Cancer 2015
400
300
200
0"100"
200"
300"
400"
JAK2
"GN
AS"
FGFR1"
ALK"
CSF1R"
IDH1
"NOTC
H1"
MLH
1"CD
H1"
PTPN
11"
CTNNB1
"MPL"
ERBB
2"SRC"
NRA
S"AK
T1"
HRAS
"VH
L"HN
F1A"
PDGF
RA"
SMO"
STK1
1"AB
L1"
FGFR2"
JAK3
"SM
ARCB
1"RE
T"BR
AF"
RB1"
FLT3"
PTEN
"FBXW
7"GN
AQ"
EGFR"
SMAD
4"FG
FR3"
KIT"
KDR"
MET"
GNA1
1"ER
BB4"
KRAS
"AT
M"
PIK3
CA"
APC"
TP53"
Total"
100
0
Mutations per gene
www.cancer2015.org
Landscape of actionable mutations
Curated variants 2,013
Mutation frequency comparison
Classification of Curated Mutations
At least one clinically relevant mutationNo mutation detected
Sensitivity/resistance to approved/preclinical drugPrognostic/diagnostic significanceUnknown clinical significance
Actionability Class
New pathways for investigation
www.cancer2015.org
Muta%on Class by Tumour type
BreastCervical
MelanomaOesophagogastric
PancreasCNS
OvarianAnal
0% 25% 50% 75% 100%
C>T / G>A G>T / C>AC>G / G>C T>C / A>GT>A / A>T T>G / A>Cindel
Clinical trials
Lim
ited
gene
pan
el
Extended gene panel & ancillary
testing
2015
pat
ient
s
Trials
Real-time reporting
www.cancer2015.org
Cancer 2015 investigatorsChief InvestigatorsJoe Sambrook
Michael Wright
InvestigatorsUniversity of Melbourne: Paul Waring, Graham Taylor, Mark Jenkins, Melissa Southey
WEHI: Melanie Bahlo, Tony Papenfus
RCH: Paul Ekert, Francois Mechinaud, Richard Sullivan
RMH: Lara Lipton
RWH: Orla McNally, Michael Quinn
Cabrini: Gary Richardson
St Vincent’s Hospital: Ray Snyder
Barwon Health: David Ashley
Peter Mac: David Thomas, Paul James
Alfred Hospital: Andrew Wei
Austin: Jonathan Cebon, Tom John, Alex Dobrovic
Monash: Neil Watkins, John McNeil, Paula Lorgelly, Gail Risbridger
Bendigo: Rob Blum
Cancer 2015 Staff PeterMac
Kate CroughKristy Barnes-‐CullenJess McDonaldHeather Thorne (KConFab)KConFab staff
Mandy Ballinger /Kim Riddell/Jasmine Marr (ISKS)Ann Officer/Renee Webb (Lung Cohort)Anne Fennessy/Sonia Mailer (MMP)Andrew Fellowes/Anthony Bell (MolP))
Cabrini Health Laura ZamursKate HurfordKate RichardsBarbara Scher
Barwon Health Judi BroadLea-‐Anne HarrisonCarolyn WielensAnne WoollettSandra RobinsonMarcelle Hennig
Melbourne Health Stefanie HartleyLidia Vecia (Uro-‐Onc)Pat Bugeja (Uro-‐Onc)Christopher Bates
DBase Development Team Suvarna Joshi Mark Lucas
David Morrison/Ravi RavindranChris Reid (Monash CIDMU) Julie Johns (BioGrid)
Funded by The VCA
Future reporting of cancer pathology?
Primary metastasis first relapse second relapse
Measure of heterogeneity
Circulating DNAmethylationmiRNAsproteomics
Future reporting of cancer pathology?
Primary metastasis first relapse second relapse
Measure of heterogeneity
Circulating DNAmethylationmiRNAsproteomics
but beware.....
PazopanibFGFR3 rearrangement
AFP
Jan 1, 2013 March 7, 2013
Pazopanib
Questions
49
Top Related