Transcript Cancer Partners - Research at Oslo University Hospital

Programme for optimized treatment
of Norwegian cancer patients
Status, challenges and opportunities
Ola Myklebost
OUS - Norwegian Radium Hospital
Cancer Crosslinks Jan 12th 2012
All cancers are different
• Each cancer type is recognized as having many
subtypes, with different cellular origins or
reflecting different cellular phenotypes
• Within these subtypes there are vast
differences in mutation spectra and which
cellular pathways are deranged
• Even within each tumour and its metastases,
there is enormous heterogeneity in geno- and
phenotypes among cancer cell subpopulations
These give rise to resistant tumours
All patients are different
• Germ-line genetic variation impinges on
Cancer development or progression
immune responses
stroma interactions
Response to therapy
Adverse and late side effects
• Deep sequencing reveals a large number of
rare, “private” gene variants in each individual
Their impact is not yet understood
New targeted therapies
• Mechanistic understanding means
sensitive tumours can be identified and treated
resistant tumours can be allocated to better
treatment
• Expensive treatments may become cost
effective
• Cancers of other types having the targeted
properties may potentially have benefit
New diagnostic options
• Traditional pathology
 Testing a few markers for each cancer type
 Usually only certain mutations of each gene
 Each new assay needs effort
• High-throughput methods
 Can scan large numbers of relevant markers
 Can identify mutations genome-wide
 Sensitivity, speed and capacity improving dramatically
 Prices falling rapidly
 Easily scalable
Deep sequencing-based
tumour mutation detection
• Sample tumour and blood
• Sequence candidate genes in both
• Compare sequence and identify somatic
mutations in tumours
• Are “actionable” proteins mutated?
• Are the mutations likely or known to make the
phenotype targetable by an available drug?
Norwegian initiatives toward
personalized cancer care
• Oslo Cancer Cluster
 Initiative from Pfizer to make a Norwegian “Stratified
Medicine Initiative”
• National Collaboration Group for Health Research
 National proposal to use tumour gene profiles to determine
cancer treatment
• Norw. Radium Hospital pilot study
 Establishment of technology
 Determine the complete sequence of all kinases and a
number of cancer-related genes in 100 lung cancers
 Example results
Lung cancer study
• Project start late 2010
• 100 patients, DNA from tumour + blood
 Samples from Helland and Brustugun OUS
• Sequenced all kinase genes + 100 “cancer
genes”
 Meza-Zepeda et al. HSØ Core
 Eight first pairs processed
 The technology works perfectly
• Bioinformatic analysis
 Hovig HSØ et al. Core
 Multiple possibly actionable mutations detected
Kinases targeted in clinical trials
The majority of targets for
specific cancer therapy are kinases
Oleg Fedorov, Susanne Müller & Stefan Knapp
Genes included in “Kinome Set”
PIK3
Inositol
PROTEIN
REGULATORY polyphospha
KINASE GENES PI3K DOMAIN DIGLYCERIDE COMPONENTS te kinases
PIP4/PIP5
(517)
PROTEINS (12) KINASES (13) (6)
(9)
kinases (9)

AAK1
517
12
13
9
6
ADDITIONAL
MORE
BREAST
CANCER
CANCER
GENES (16) GENES (11)
CANCER
GENES (20)
9
16
20
10
PIK3C2A
AGK
PIK3R1
IP6K1
PIKFYVE
CDC6
COL1A1
CCND1
ATM
AATK
PIK3C2B
CERK
PIK3R2
IP6K2
PIP4K2A
CHD3
GAB1
CCND2
ABL1
AXL
ABL2
CDKs
PIK3C2G
DGKA
PIK3R3
IP6K3
PIP4K2B
HRAS
HAUS3
CCND3
PIK3C3
DGKB
PIK3R4
IPMK
PIP4K2C
KRAS
IRS2
ESR1
ACTR2
EGFR
ACVR1
FGFRs
ACVR1B
FLTs
ACVR1C
IGF1R
ACVR2A
PIK3CA
DGKD
PIK3R5
IPPK
PIP5K1A
NRAS
IRS4
ESR2
PIK3CB
DGKE
PIK3R6
ITPK1
PIP5K1B
PTEN
KIAA1468
FBXW7
PIK3CD
DGKG
ITPKA
PIP5K1C
CDH1
KLHL4
IDH1
PIK3CG
DGKH
ITPKB
PIP5KL1
TP53
NFKB1
IDH2
PI4KA
DGKI
ITPKC
PIPSL
CDKN2A
NFKBIA
MLH1
JAK1
ACVR2B
KIT
ACVRL1
PI4KB
DGKQ
CDKN2B
NFKBIE
TERT
PI4K2B
DGKZ
APC
PALB2
ADCK1
MAPKs
ADCK4
MET
PI4K2A
SPHK1
RB1
RHEB
SPHK2
CTNNB1
RNF220
BRCA1
SNX4
BRCA2
SP1
NF1
USP28
ADCK5
MTOR
ADRBK1
PDGFRs
ADRBK2
RAF1
AKT1
TGFBRs
AKT2
Total 3,2 Mbp
NF2
GATA3
AKT3
MYC
ALK
INPP4A
TOTAL GENES
612
Filtering of “private” genetic variation
Patient
Change from human
reference sequence
After exclusion of
known variants*
Real somatic
mutations
3
6684
5706
3
17
6157
5216
67
22
6209
5219
26
27
6100
5070
50
35
6137
5171
58
39
6065
5123
30
48
6318
5373
6
90
6344
5471
65
* dbSNP
(Including non-exonic flanking sequence)
Example findings
Tumour mutations in first 8 lung cancers:
Gene
PTEN
ERBB4
CDKN2A
DDR2
JAK2
Protein name
phosphatase and tensin
homolog
erythroblastic leukemia
viral oncogene homolog 4
cyclin-dependent kinase
inhibitor 2A (melanoma,
p16, inhibits CDK4)
discoidin domain receptor
tyrosine kinase 2
Janus kinase 2
epidermal growth factor
EGFR
receptor
c-abl oncogene 1, nonABL1
receptor tyrosine kinase
JAK1
Janus kinase 1
kinase insert domain
KDR (VEGFR) receptor
phosphoinositide-3PIK3CG
kinase, gamma peptide
Pati
ent
Mut Mut
type
AA AA Candidate therapy
chng
pos
17 stopgain GAG>TAG
E>X
sirolimus, imatinib, cisplatin, docetaxel, fluorouracil, leucov
256 levamisole, trastuzumab, pioglitazone, rosiglitazone
17 missense GAC>TAC
D>Y
376 trastuzumab, neratinib
27 missense GAA>GCA
E>A
35 stopgain GAA>TAA
E>X
35 missense TTG>TTC
L>F
48 missense CTG>CGG
L>R
Validated
858 Erlotinib, gefitinib,
Pertuzumab, Cetuximab
90 missense GAG>AAG
90 missense CCG>CTG
E>K
P>L
704 imatinib
430 CYT387,Tofacitinib,Ruxolitinib,LY3009104
90 missense GCA>TCA
A>S
90 missense GAC>GAA
D>E
94 busulfan, cisplatin, gemcitabine, troglitazone
786 imatinib
1082 Lestaurtinib,CYT387,Pacritinib,LY3009104,TG101348,Ruxoli
1324 pazopanib,sorafenib,vatalanib,axitinib,sunitinib
500 exenatide
Vanderbilt: MyCancerGenome.org
DDR2 mutation
National cooperation group for
health research (NSG)
• ”Cancer task force” (”Skrivegruppen”)





Stein Kvaløy, OUS (Head)
Roy Bremnes, UNN and UiTromsø
Stein Kaasa, St Olav and NTNU
Ragnhild A. Lothe, OUS and UiO
Per Eystein Lønning, Haukeland and UiB
• Proposed national initiative ”Development of diagnostics and
personalized cancer treatment based on genetic analyses”
 Based on deep sequencing of tumours
• Endorsed by the National cooperation group for health research
• Personalized cancer therapy chosen by the National council for
health priorities for grant call through the Research council
Norwegian Cancer Genomics
Consortium
• Joint application from Oslo, Bergen and Trondheim
• Key investigators
 Ola Myklebost, PI, OUS
 Ragnhild Lothe, OUS
 Harald Holte, OUS
Translational
 Per Eystein Lønning, Haukeland
medicine group
 Anders Waage, St Olavs
 Giske Ursin, Norw. Cancer Registry
 Leonardo Meza-Zepeda, OUS, Sequencing Technology
 Eivind Hovig, OUS, Bioinformatics
Plans
• Applying for a budget for 4000 tumour/blood pairs
• Will design custom gene capture set
Kinome + 5-600 others
• Cancer types
Breast cancer
Lymphoma
Leukemia
Colorectal cancer
Malignant melanoma
Sarcoma
Multiple myeloma
Gynecological cancers
Prostate cancer
Main objectives I
• Provide a national network for
implementation of personalised cancer
medicine in Norway
• Provide and disseminate methodology for
deep sequencing of tumour material and
identification of somatic mutations
• Initiate a number of research projects to
determine the applicability of mutation
profiles from the individual tumour for
therapeutic decisions
Main objectives II
• Provide the bioinformatics tools necessary to
make mutation spectra clinically interpretable
and for national data logistics
• Establish a nation-wide cancer mutation
database in collaboration with the Cancer
Registry
• Provide a dialogue on the clinical implementation
of personal mutation data
• Analyse the health economic impacts of
improved and standardised access to molecular
targeted therapies nation-wide
NCGC Web Page
Project plan
Phase Activity
Issues
1
Retrospective studies
3000 sample pairs
Technology dissemination
Standardization of technologies and results
Sample and data logistics
Stepwise introduction of molecular pathology
Communication of relevant findings to clinicians
2
Prospective studies
1000 sample pairs
Application of standardized protocols
Scientific hypotheses, approved treatments
Close interaction with clinicians
Clinical monitors
Clinical trial units (OUS, Haukeland)
Data transfer to Cancer registry
3
Evaluation
Frequency of “actionable” mutations
Potential for targeted treatment
Possible impact on health economy
Possible health benefits
4
Clinical
implementation
What is learned, and what should be implemented
Transfer of technology to routine cancer pathology
NCGC data logistics
Myeloma
Bioinformatics
Sequence
Lymphoma
OUS
Leukemia
Colon
Melanoma
National
mutation
database
Haukeland
Prostate
Breast
St Olavs
Cancer registry
Sarcoma
Clinical data
Others?
Accumulated
national data
Added values from national collaboration
•
•
•
•
•
Population-based data
Standardized genome analysis
Equal patient access nation-wide
Standardized treatment choices
National follow-up of multiple N=1 trials
(compassionate use)
• National evaluation of mutation frequencies and
health-economic consequences
• Over time accumulation of outcome data of
personalized treatments and other therapies
stratified by mutation profiles
National Cancer Genomics Consortium
Haukeland
Hospital
Pharma?
St Olavs
Hospital
NCE
Oslo Cancer
Cluster
Norwegian
Cancer
Registry