Transcript Progress on Biomarkers of Cancer Diagnosis and Prognosis

Progress on Biomarkers of
Cancer Diagnosis and
Prognosis
William CS CHO
Queen Elizabeth Hospital, Hong Kong
May 22, 2010
Dual-specificity phosphatase 6 (DUSP6),
monocyte-to-macrophage differentiation associated protein (MMD),
signal transducer and activator of transcription 1 (STAT1),
v-erb-b2 avian erythroblastic leukemia viral oncogene homolog 3 (ERBB3),
lymphocyte-specific protein tyrosine kinase (LCK).
Conclusions
Our five-gene signature is closely associated with relapse-free and overall survival
among patients with NSCLC.
Kaplan–Meier Estimates of Survival of Patients
with NSCLC According to the Five-Gene Signatures as
Measured by RT-PCR.
Overall survival and relapse-free survival are shown for the 101 patients with
NSCLC (Panel A and Panel B, respectively) and for the 59 patients with stage I or
II disease (Panel C and Panel D, respectively). Overall survival is also shown for
the independent cohort of 60 patients (Panel E), for the 42 patients in this cohort
who had stage I or II disease (Panel F), and for the 86 patients described in an
independent set of published NSCLC microarray data10 (Panel G).
70 Gene Prognosis Profile
Supervised analysis
Tumor samples
70 significant prognosis genes
van´t Veer et al., Nature 415, p. 530-536, 2002
threshold set with 10% false negatives
91 % sensitivity, 73% specificity
70 prognosis genes are involved in all
aspects of tumor cell biology
proliferation
angiogenesis
intravasation, survival, extravasation
adhesion to extracellular matrix
local invasion
adhesion to extracellular matrix
proliferation
angiogenesis
Genes of unknown function (25)
Independent
validation:
Buyse et al. (2006)
JNCI. 98, 1183-1192.
307 patients
High reproducibility of microarray experiments
(99%)
Reproducibility; repeat of the experiment
Glas et al, BMC Genomics 2007.
No Recurrences in the Good Prognosis Group
MammaPrint:
Good Prognosis
(N=23)
Poor Prognosis
(N=144)
Marieke Straver et al.,
Br Cancer Res and Treat.
2009
Clinical Development of Oncotype Dx
• Development of a high-throughput, real time, RT-PCR method to quantify
gene expression from fixed tumor tissue samples
Published literature
• Selection of 250 candidate genes
Genomic databases
DNA array-based experiments
• Testing the relationship between the 250 candidate genes and risk of
recurrence in a series of 447 pts from three clinical studies
16 cancer-related genes + 5 reference genes → Oncotype DX (recurrence score)
Paik et al. NEJM. 2004.
How Do We Assess Risk
in Breast Cancer Patients?
Classic Pathological
Criteria
Lymph Node
Status
Tumor
Size
New tools in the
Genomic Era…
Oncotype DX®
Age
Tumor
Grade
ER/PR
HER2
Adjuvant!
Computer-based model
Oncotype DX 21-gene recurrence score
16 cancer genes and 5 reference genes make up the Oncotype DX
gene panel. The expression of these genes is used to calculate the
recurrence score:
PROLIFERATION
Ki-67
STK15
Survivin
Cyclin B1
MYBL2
REFERENCE
Beta-actin
GAPDH
RPLPO
GUS
TFRC
ESTROGEN
ER
PR
Bcl2
SCUBE2
RS =
BAG1
GSTM1
INVASION
Stromelysin 3
Cathepsin L2
CD68
HER2
GRB7
HER2
+ 0.47 x HER2 Group Score
- 0.34 x ER Group Score
+ 1.04 x Proliferation Group Score
+ 0.10 x Invasion Group Score
+ 0.05 x CD68
- 0.08 x GSTM1
Paik et al. N Engl J Med.
- 0.07 x BAG1
2004;351:2817-26.
Rate of Distant Recurrence at 10 years
Recurrence Score
Low
RS < 18
Rec. Rate = 6.8%
C.I. = 4.0% - 9.6%
40
35
Intermediate
RS 18 - 31
Rec. Rate = 14.3%
C.I. = 8.3% - 20.3%
High
RS  31
Rec. Rate = 30.5%
C.I. = 23.6% - 37.4%
30
25
20
15
Recurrence Rate
95% C.I.
10
5
0
0
5
10
15
20
25
30
Recurrence Score
35
40
45
50
Paik S. et al. N Engl J Med 2004;351:2817-26
Oncotype DXTM
Low RS associated with minimal chemotherapy benefit;
High RS associated with large chemotherapy benefit.
The Oncotype DX Recurrence Score provides precise,
quantitative information for individual patients on prognosis
across and statistically independent of information on patient
age, tumor size, and tumor grade.
Nobel Prize in Physiology or Medicine 2006
Andrew Z. Fire
Craig C. Mello
Cho WC. MicroRNAs in cancer - from research to therapy.
Biochim Biophys Acta - Rev Cancer 2010;1805(2):209-217.
C. elegans
Non-coding RNA: the NA formerly known as “junk”
RNA Transcripts
Protein-coding mRNA
Non-coding RNA Transcripts
Regulatory RNA
miRNA
siRNA
piRNA
Anti-sense RNA
•
•
•
•
Transcription/chromatin structure regulators
Translational regulators
Protein function modulators
RNA/Protein localization regulators
snoRNAs
Housekeeping RNAs
•tRNA
•rRNA
•snRNA
•tmRNA
•Rnase P RNA
•vRNAs
•gRNAs
•MRP RNA
•SRP RNAs
•Telomerase RNA
NC-RNAs compose majority of transcription in complex genomes
Unique MicroRNA Profile in Lung Cancer
Diagnosis and Prognosis
• miRNAs are small non-coding RNAs which
play key roles in regulating the translation
and degradation of mRNAs
• Genetic and epigenetic alteration may
affect miRNA expression, thereby
leading to aberrant target gene(s)
expression in cancers
• Yanaihara et al, Cancer Cell, 2006:
- miRNA profiles of 104 pairs of primary
lung cancers and corresponding noncancerous lung tissues were analyzed by
miRNA microarrays
- 43 miRNAs showed statistical differences
Unique MicroRNA Profile in Lung Cancer
Diagnosis and Prognosis
• A univariate Cox proportional hazard
regression model with a global permutation
test indicated that expression of the miRNAs
hsa-mir-155 and hsa-let-7a-2 was related to
adenocarcinoma patient outcome
• Lung adenocarcinoma patients with
either high hsa-mir-155 or reduced
hsa-let-7a-2 expression had poor survival
Yanaihara N, et al. Unique microRNA molecular profiles in lung cancer diagnosis and prognosis. Cancer Cell 2006, 9:189-198.
The role of microRNAs in cancer diagnosis
• With the application of in situ RT-PCR, it was shown that the
aberrantly expressed miR-221, miR-301 and miR-376a were
localized to pancreatic cancer cells but not to stroma or normal
acini or ducts.
• Aberrant miRNA expression offered new clues to pancreatic
tumorigenesis and might provide diagnostic biomarkers for
pancreatic cancer.
Lee EJ, et al. Expression profiling identifies microRNA signature in pancreatic cancer.
Int J Cancer 2007, 120:1046-1054.
Cho WC. MicroRNAs: potential biomarkers for cancer diagnosis, prognosis and targets for
therapy. Int J Biochem Cell Biol 2010.
Cho WC. MicroRNAs in cancer - from research to therapy. Biochim Biophys Acta - Rev Cancer
2010;1805(2):209-217.
The role of microRNAs in cancer prognosis
• Expression of let-7 miRNA was frequently reduced
in human lung cancers, and that reduced let-7
miRNA expression was significantly associated with
shorter postoperative survival.
• Overexpression of let-7 miRNA in A549 lung
adenocarcinoma cell line inhibited lung cancer cell
growth in vitro.
Takamizawa J, et al. Reduced expression of the let-7 microRNAs in
human lung cancers in association with shortened postoperative survival.
Cancer Res 2004, 64:3753-3756.
The role of microRNAs in cancer prognosis
• The expression pattern of miRNAs in pancreatic cancer
were compared with those of normal pancreas and chronic
pancreatitis using miRNA microarrays.
• Differentially expressed miRNAs were identified which
could differentiate pancreatic cancer from normal pancreas,
chronic pancreatitis, or both.
• High expression of miR-196a-2 was found to predict poor
survival of more than 24 months.
Bloomston M, et al. MicroRNA expression patterns to differentiate pancreatic
adenocarcinoma from normal pancreas and chronic pancreatitis. JAMA 2007,
297:1901-1908.
microRNAs
Tumorigenesis
miR-9
Neuroblastoma
miR-10b
Breast cancer
miR-15, miR-15a
Leukemia, pituitary adenoma
miR-16, miR-16-1
Leukemia, pituitary adenoma
miR-17-5p, miR-17-92
Lung cancer, lymphoma
miR-20a
Lymphoma, lung cancer
miR-21
Breast cancer, cholangiocarcinoma, head & neck
cancer, leukemia
miR-29, miR-29b
Leukemia, cholangiocarcinoma
miR-31
Colorectal cancer
miR-34a
Pancreatic cancer
miR-96
Colorectal cancer
miR-98
Head & neck cancer
miR-103
Pancreatic cancer
miR-107
Leukemia, pancreatic cancer
miR-125a, miR-125b
Neuroblastoma, breast cancer
miR-128
Glioblastoma
miR-133b
Colorectal cancer
miR-135b
Colorectal cancer
miR-143
Colon cancer
miR-145
Breast cancer, colorectal cancer
miR-146
Thyroid carcinoma
Diagnosis
Prognosis
Pancreatic
cancer
Neuroblastoma
microRNAs
Tumorigenesis
miR-155, has-miR-155
Breast cancer, leukemia, pancreatic cancer
miR-181, imR-181a, imR-181b, imR-181c
Leukemia, glioblastoma, thyroid carcinoma
miR-183
Colorectal cancer
miR-184
Neuroblastoma
miR-193
Gastric cancer
Diagnosis
Lung cancer
miR-196a-2
Pancreatic cancer
miR-221
Glioblastoma, thyroid carcinoma
miR-222
Thyroid carcinoma
miR-223
Leukemia
Pancreatic cancer
miR-301
Pancreatic cancer
miR-376
Pancreatic cancer
let-7, let-7a, let-7a-1, has-let-7a-2, let-7a-3
Prognosis
Lung cancer, colon cancer
Lung cancer
Cho WC. MicroRNAs: potential biomarkers for cancer diagnosis, prognosis and targets for therapy.
Int J Biochem Cell Biol 2010.
Cho WC. OncomiRs: the discovery and progress of microRNAs in cancers. Mol Cancer. 2007;6:60.
Beyond the genome
Same genome
Different proteome
Characterizing proteins and DNA at the molecular
level is the key to understanding their function
Functional genomics
t-RNA
mRNA
t-RNA
Ribosome
t-RNA
(....)
Protein
t-RNA
DNA
Genomics
Post Translational
Modifications
X
(....)
X
Proteomics
Active Protein
CHO
PO4
Proteomics: leading biological science
in the 21st century
• Proteomics represents the effort to
establish the identities, quantities,
structures, biochemical and cellular
functions of all proteins in an
organism, organ, or organelle
• and how these properties vary in
space, time, or physiological state.
Cho WC. Proteomics – Leading biological science in the 21st century. Science J,
2004; 56(5):14-17.
Cho WC, Cheng CH. Oncoproteomics: current trends and future perspectives.
Expert Rev Proteomics 2007;4(3):401-410.
Traditional vs High-throughput approach
The emergence of proteomics
and its application
DNA
static genome
Transcriptional control
RNA
message variable: transcriptome
Translational control
Protein
product variable: proteome
Post-translational modification
Genome Era
Post-genome Era
Intrinsic factors:
physiological &
pathological
status, …
Sample
preparation
& processing
Cho WC, Cheng CH.
Oncoproteomics:
current trends and
future perspectives.
Expert Rev Proteomics
2007;4(3):401-410.
Extrinsic factors:
environment, pathogens, drug, …
Functional protein
expressed
Automation sample application
ESI-TOF MS
MALDI-TOF MS
Low-throughput
High-throughput
Peptide ions
(MS)
Peptide fragment ions
(MS-MS)
Protein chip,
e.g. SELDITOF MS
ESI: Electrospray ionization
MALDI: Matrix-assisted
laser desorption ionization
Bioinformatics
Experimental or
clinical results
Database interrogation
Protein identification
SELDI: Surface-enhanced
laser desorption ionization
TOF: Time of flight
Validation and application
Surface-enhanced laser
desorption/ionization (SELDI)
Chemical Surfaces – Protein Expression Profiling:
Hydrophobic
Cationic
Anionic
IMAC
Normal Phase
H50 – C9 chains
WCX2 -
SAX2 –
Chelates metals
NP20 –
H4 – C16 chains
Carboxylate
4O Ammonium
(Cu, Ni, Zn, Ga, Mn, …)
SiO2
Biological Surfaces – Protein Interaction Assays:
PS-10 or PS-20
Protein conjugation
Antibody - Antigen
Receptor - Ligand
DNA - Protein
Cho WC. Proteinchip. In: Encyclopedia of Cancer: 2nd Edition. 2009. Springer.
HTP automation
Biomek 2000 (Beckman)
Programmed protocols for highly
reproducible sample processing
Proteinchip System PCS4000
Aquarius (Tecan)
Sample fractionation, chip binding and
data acquisition in SELDI-TOF MS
Serum / lysate sample
Disease samples
Control samples
vs
Cho WC, et al. Clin Cancer Res
2004;10:43-52.
+ Urea / CHAPS / TrisHCl pH 9
Sample anion exchange
pre-fractionation
Strong anion exchange resin
Q HyperD resin
Organic
eluant
Fractionation
pH 9/
flow through
pH 3
eluant
Cho WC, et al. J Cell Biochem
2006;99(1):256-68.
pH 7
eluant
pH 5
eluant
pH 4
eluant
Chip binding
Cho WC. Chin J Biotech
2006;22(6):871-876.
Weak cation exchange (WCX2 / CM10)
100 mM NaAc, pH 4
Cu(II) (IMAC3 / IMAC30)
100 mM phosphate, 0.5 M NaCl, pH 7
Data acquisition
T O FM S
D etector
L a ser
Protein Biology System (PBS) IIc SELDI-TOF mass spectrometer
Cho WC, et al. Dis Markers
2006;22(3):153-66.
Cho WC, et al. J Ethnopharmacol
2006;108(2):272-9.
Cho WC, et al. Clin Chem
2007;53(2):241-250.
Biomarker discovery
• Markers can be easily
found by comparing
protein maps.
• SELDI is faster and more
reproducible than 2D
PAGE.
(Normal)
(Cancer)
• Has been being used to
discover protein
biomarkers of diseases
such as ovarian cancer,
breast cancer, prostate and
bladder cancers.
Cho WC. Contribution of oncoproteomics to cancer
biomarker discovery. Mol Cancer 2007;6:25.
Proteins as biomarkers
The protein composition may be associated with disease
processes in the organism and thus have potential utility as
diagnostic markers.
• Proteins are closer to the actual disease process,
in most cases, than parent genes
• Proteins are ultimate regulators of cellular function
• Most cancer markers are proteins
• The vast majority of drug targets are proteins
Cho WC. Cancer biomarkers (an overview). In Hayat MA (ed): Methods of cancer
diagnosis, therapy and prognosis. Volume 7. New York, NY: Springer, 5 Jan 2010.
Nasopharyngeal cancer (NPC)
Normal nasopharynx
• 7th most prevalent cancer in Hong Kong.
• Problems in clinical management of NPC:1. Diagnosis at late stage (at stage 3/4)
2. Frequent relapse (>50% for CR
patients)
Nasopharynx with tumor
Tumor on the right
eustachian cushion
Cho WC. Most common cancers in Asia-Pacific region: nasopharyngeal carcinoma.
In: Cancer report of Asian-Pacific region 2010. 284-289.
Proteinchip application: nasopharyngeal
carcinoma biomarkers discovery
• Serum samples from 149 NPC patients (undifferentiated
carcinoma of the nasopharyngeal type or poorly
differentiated squamous cell type)
• 35 normal individuals
10000
11000
12000
13000
14000
15000
GC10 A3 (P1F6)
GC10 A5 (P1F6)
GC17 A3 (P1F6)
GC29 A3 (P1F6)
GC6 A1 (P1F6)
GC1 A7 (P2F6)
GC25 A5 (P2F6)
GC37 A7 (P2F6)
GC3 A7 (P3F6)
GC24 A1 (P3F6)
GC23 BT1 (P3F6)
GC20 BT1 (P3F6)
GC11 A4 (P3F6)
GC11 A9 (P3F6)
GC9 BT1 (P4F6)
GC13 BT1 (P5F6)
GC8 A3 (P5F6)
GC15 BT1 (P5F6)
GC14 BT1 (P5F6)
GC21 A3 (P6F6)
GC18 A3 (P6F6)
GC12 A4 (P6F6)
GC26 A6 (P7F6)
GC22 BT1 (P7F6)
GC27 A3 (P8F6)
GC27 A10 (P8F6)
GC28 A7 (P8F6)
GC36 A8 (P9F6)
GC35 A3 (P9F6)
GC32 BT1 (P10F6)
GC31 A13 (P10F6)
GC34 A3 (P11V6)
PS165 A8 (P3F6)
PS165 A15 (P3F6)
PS178 A8 (P3F6)
PS178 A15 (P3F6)
PS192 A8 (P4F6)
PS192 A15 (P4F6)
PS205 A8 (P4F6)
PS205 A15 (P4F6)
PS213 A8 (P2F6)
PS213 A15(P2F6)
PS217 A8 (P5F6)
PS217 A15 (P5F6)
PS223 A8 (P6F6)
PS223 A15 (P6F6)
PS250 A8 (P4F6)
PS250 A15 (P4F6)
PS253 A8 (P2F6)
PS253 A15 (P2F6)
PS260 A9 (P4F6)
PS260 A16 (P4F6)
PS279 A8 (P5F6)
PS279 A15 (P5F6)
10000
11000
12000
13000
14000
15000
Mass data collection for protein
identification
1021
1386
854
1524
sample
tryptic digestion
2-D gel purification
identification
1600
600
mass spectrometry
(peptide mapping information)
database
Protein search
Identification of marker by MS/MS
+TOF Product (2176.9): 70 MCA scans from spot B 2177.wiff, Smoothed
34/37 ions matched with
Serum Amyloid A
931.4332
484
Max. 484.0 counts.
450
400
I n t e n s it y , c o u n t s
350
300
250
200
914.4099
418.2148
1317.6205
150
100
617.2782
315.1216
497.1998
461.2252
50
0
200
400
600
861.3582
661.2993
800
2178.9799
896.3904
1000
1247.5223
1200
m/z, amu
1400
1600
1800
2000
2200
Longitudinal follow up of biomarker, 11,695 Da in
3 relapsed NPC patients & 11 remission patients
Cho WC, et al. Clin Cancer Res 2004, 10(1):43-52
Serum biomarkers with changes before and after
chemotherapy in relapsed NPC patients
EP: Biomarker: 7,659 Da
% Diff = 206.8%
increased
4
p = 4.8E-6
decreased
1.727
+/- 0.781
GC: Biomarker: 7,765 Da
Mean +/- SD
0.835
+/- 0.753
3
% Diff = 194.4%
increased
6
p = 6.1E-6
decreased
Mean +/- SD
2.382
+/- 1.128
5
4
2
1.225
+/- 0.734
3
1
2
0
1
-1
0
Before
After
Chemotherapy (N=35)
EP, Etoposide and Cisplatinum;
Before
After
Chemotherapy (N=29)
GC, Gemcitabine and Cisplatinum.
Cho WC et al. ProteinChip array profiling for identification of disease- and
chemotherapy-associated biomarkers of nasopharyngeal carcinoma. Clin Chem. 2007;53(2):241-50.
Basic statistics of ovarian cancer
• Prevalence 40/100,000 (1 in 2500)
• 23,000 new cases diagnosed annually
• 14,000 deaths annually
• Overall 5 year survival 20-30%
• 75% of cases are diagnosed in late stage (stage III/IV)
• 90% cure rate in stage I/IIa
• Therefore, detection in earlier stages critical in improving
overall survival
Study design for biomarker discovery
Stage I/II (20)
Site 1
(100)
Benign (50)
Multivariate
Model
Derivation
Discovery 1
Control (30)
Cross
Comparison
Stage I/II (35)
Site 2
(176)
Candidate
Markers
Stage III/IV (2)
Discovery 2
Multivariate
Models
Benign (90)
Control (49)
Stage I/II (35)
Site 3
(164)
Stage III/IV (103)
Benign (26)
Independent
Validation
Control 63
Protein ID
Ca (41)
Other Ca 1 (20)
Site 5
(142)
Other Ca 2 (20)
Independent Validation by
Immunoassay
Results:
• Descriptive statistics
• Two-group t-tests
• Performance
• ROC curve analysis
ROC curve, area=0.94327, std = 0.094973, alpha= 2.791, beta= 0.47154
1
0.9
Other Ca 3 (20)
0.8
0.7
0.6
Sensitivity
Site 4
(63)
0.5
0.4
0.3
Control (41)
0.2
0.1
0
0
0.1
0.2
0.3
0.4
0.5
0.6
1 - Specificity
0.7
0.8
0.9
1
Summary of performance
• Markers for Stage I/II ovarian cancer discovered using ProteinChip system
• 503 samples from 5 institutions
• Rigorous cross-validation and independent validation study design
• Fixed specificity (97%)
• 3 marker panel (Apolipoprotein A1, inter alpha trypsin inhibitor IV and
Transthyretin) : 74% sensitivity
• CA125: 65% sensitivity
• Fixed sensitivity (83%)
• 3 marker panel: 94% specificity
• CA125: 54% specificity
Pioneers in multimarker research
Peak A
Criteria
Peak B
Criteria
Cancer
Normal
Peak C
Criteria
Cancer
ID the biomarkers,
Link to biology of disease
Normal
Sensitivity
“True
Positives”
Specificity
“True
Negatives”
Single Marker
65%
35%
Biomarker Pattern
>90%
>90%
FDA Cleared the OVA1 Test
on Sep 11, 2009
● Translating biomarker discovery from lab to clinic
● Based on a prospective double-blind clinical trial
involved 516 patients from 27 institutions
• 269 patients were evaluated by pre-surgical information
alone
• 247 patients were evaluated by pre-surgical information
with OVA1 results
● OVA1 identified additional patients with potential
malignancies
● Help to guide surgical decisions
OVA1
● First FDA-cleared protein-based in vitro diagnostic
multivariate index assay
● First FDA-cleared prognostic test for ovarian cancer in
the pre- and post-surgical setting
● Test 5 proteins in blood sample
• β2-microglobulin,
transferrin,
transthyretin identified by SELDI
apolipoprotein
• CA125
● Indicate the likelihood of benign or malignant
A1,
Scientific American
Cho WC. Proteomic approaches to cancer target identification. Drug Discov Today: Ther Strategies 2007;4(4):245-250.
Targets of Cancer Therapy
1
2
5
4
Plasma Membrane
6
Growth
Factor
Signaling
3
7
Microtubule Dynamics
7
12
RNA Translation
7
8
Nuclear Membrane
1. Growth factors
2. Growth factor
receptors
3. Adaptor proteins
4. Docking
proteins/binding
proteins
5. Guanine nucleotide
exchange factors
9
11
10
Gene Transcription
Cell Growth
Motility
DNA Replication and Repair
Survival
Proliferation
Angiogenesis
6. Phosphatases and
phospholipases
7. Signaling kinases
8. Ribosomes
9. Transcription
factors
10. Histones
11. DNA
12. Microtubules
In colon cancer KRAS mutation determines response
to EGFR therapy
Mutant KRAS
+EGFR
-EGFR
Wild type KRAS
+EGFR
-EGFR
51
Amado et al. J Clin Oncol; 26:1626-1634 2008
In colon cancer KRAS mutation determines response
to EGFR therapy
KRAS mut: 32%
PIK3CA mut: 13%
BRAF mut: 10%
Mutant KRAS
+EGFR
-EGFR
Wild type KRAS
+EGFR
-EGFR
52
Amado et al. J Clin Oncol; 26:1626-1634 2008
Conventional cancer treatment:
Dx
Rx
Personalized cancer treatment:
→
Rx
Treatment:
Diagnosis
Stage, Grade,
IHC
Treatment
Chemotherapy
Pathway
targeted
therapy
A 159-gene signature of activated PI3K
pathway in colon cancer
Pathway & network analysis
Cho WC. Proteomics technologies and challenges.
Genomics Proteomics Bioinformatics 2007;5(2):77-85.
Cho WC (ed): An omics perspective on cancer research. New York, NY: Springer 2010
Thank You
E-mail: [email protected]