Fattori prognostici e predittivi

1
Prognostic and predictive markers and the role
of genomics proteomics
Stefano Iacobelli Medical Oncology University
G. DAnnunzio, Chieti-Pescara
2

• Prognostic predictive
• biomarkers General concepts
• Novel technologies
• Genomics
• (Proteomics)
• 3. Application in breast cancer

3
Human cancer a series of genetic and epigenetic
alterations that can be classified into 6 main
classes and are responsable of the
characteristics of the neoplastic phenotype
4
These genetic and/or functional alterations may
play an important role as tumor biomarkers
Biomarkers are important tools for cancer
management

• Monitoring patients with established disease for
• - Recurrence/Prognosis assessment
• - Prediction of response to drugs

• Early detection of asymptomatic patients
• - Aiding in the diagnosis
• - Surveillance of individuals known to be at
  risk of cancer
• - Surrogate endpoint markers for primary
  prevention
• strategies (i.e. chemoprevention)

5
The ideal tumor biomarker
It must give clinically useful information for
The recognition of subgroups of pts who differ in
disease outcome

• The identification of pts
• at minimal risk of disease recurrence
• at elevated risk, who may benefit from systemic
• treatment
• more likely to respond to specific treatment

6
BIOMARKER
Predictive NOT prognostic (bax and chemotherapy)
Prognostic NOT predictive (lymphnode status)
Separates poor from favorable groups independent
of therapy
Outcome in the absence of therapy is the same
regardless the marker is or -
Adapted from Hayes et al., BCRT 52, 1998
7
Prognostic AND Predictive
Separates groups to some extent but much more in
the presence of specific treatment
Adapted from Hayes et al., BCRT 52, 1998
8
Three categories of prognostic factors
strong
moderate
weak
High risk
Disease recurrence ()
medium risk
Low risk
adapted from Hayes et al., BCRT 52, 1998
9
PROGNOSTIC factors in breast cancer
College of American Pathologists Consensus
Statement 1999, Arch Pathol Lab Med, 2000
I Prognostic relevance Clinical stage
(T,N,M) proven clinical usefulness
histologic grade mitotic index, histotype,

steroid hormone receptorsu (uPA
PAI-1)
II Many studies Proliferation
indices biological clinical, Peritumoral
invasion but need of HER-2 /neu,
p53 statistical evaluation
III Prognostic relevance ploidy,
neo-angiogenesis but not proven clinical
apoptosis (bcl-2), usefulness based on GF and
their Rec, available
information pS2, Cathepsin D
10
1800 patients with N- breast cancer undergoing
loco-regional treatment only Incidence of
distant metastases
Estrogen Receptors
Progesterone Receptors
Follow-up (years)
Silvestrini et al., JCO 1995 CCR 1997
11
Cell Proliferation (TLI)
Incidence of distant metastases (1800
N-)
Response to treatment (281 N- TLI gt3)
Follow-up (years)
Silvestrini et al., J Clin Oncol 1995 CCR 1997
Amadori et al., J Clin Oncol 18, 2000
12
Angiogenesis (Intratumoral VEGF)
Disease recurrence (226 N- pts, surgery only)
Response to treatment (212 N/ER pts treated
with Tamoxifen)
VEGF-
VEGF-
VEGF
VEGF
HR2.46 (1.29-4.65), P0.0059
Coradini et al., Br J Cancer 2001
Coradini et al., Br J Cancer 2003
13
Prognostic relevance of uPA and PAI-1 4676
patients - Incidence of distant metastases
Look et al. JNCI, 2002
14
The score A comprehensive view that helps

• To identify patients
• at low risk of disease recurrence who do not
  need
• adjuvant treatment
• at high risk of disease recurrence who may
  benefit
• from adjuvant systemic treatment

15
NOTTINGHAM PROGNOSTIC INDEX (NPI)   Tumor Size
(cm) x 0.2 points Tumor Grade from 1 (better)
to 3 (worse) points Axillary Lymph Nodes
negative nodes 1 point positive nodes,
1 to 3 positive 2 points positive
nodes, gt4 3 points
by whatever system
Size grade lymph-node Total NPI points
Groups
80 OS _at_ 15 yrsif NPIlt3.4 sum
42 OS _at_ 15 yrsif NPI3.4-5.4 sum
13 OS _at_ 15 yrsif NPIgt5.4 sum
Conclusionsas to need foradjuvant CT
need is doubtful
MAY BENEFITwith CT
chemo needed
16
Six-year recurrence rate as a function of
bio-pathological score
17
The innovation
From a reductionistic approach (gene by gene)
to an olistic approach (global genomic
analysis)
Novel analytical tools Microarray
technology Proteomics . .
Molecular signature of cancer
18
Genomica
Farmacogenomica
Ricerca
Prevenzione
Clinica
Diagnostica
Proteomica
19
Genomics
Novel analytical tools

• Gene Sequencing
• Conventional Karyotyping
• FISH (Fluorescent in Situ Hybridization)
• CISH (Chromogenic in Situ Hybridization)
• CGH (Comparative Genomic Hybridization)
• SKY (Spectral Karyotyping)
• Real Time RT-PCR
• cDNA Microarrays

20

• 2D-PAGE
• MS (Mass Spectrometry)
• HPLC (High Performance Liquid Chromatography)
• CA (Capillary Array)
• MALDI (Matrix Associated Laser Desorption/Ionisati
  on)
• MALDI-TOF MS (Time of Flight)
• MALDI ION TRAP- TOF MS
• ESI (Electron Spray Ionisation) Tandem MS
• Quadrupole
• Functional proteomics
• TWO YEAST HYBRID SYSTEM
• PROTEIN MICROARRAY
• FRET (Fluorescence Resonance)
• SELDI (Surface-Enhanced Laser Desorption/Ionisatio
  n)
• TISSUE MICROARRAY

Proteomics
21
Multispot Arrays

• Sonde
• antigeni
• anticorpi
• cDNA
• oligonucleotidi
• prodotti di PCR
• plasmidi
• BACs (Bacterial Artificial Chromos.)
• YACs (Yeast Artificial Chromos.)

• Sonde
• (DNA, oligonucleotidi, proteine, anticorpi)

• Deposito
• blotting
• printing
• elettrodipendente
• Sintesi in situ
• meccanica
• fotolitografica
• elettrodi
• printing di precisione
• deposito sulla superficie tensione-dipendente

Deposito o sintesi

• Substrato
• vetro
• nitrocellulosa
• nylon
• vetro rivestito di poliacrilammide
• polipropilene
• silicone
• polistirene

Spots sulla superficie di un substrato solido
Gene chip, DNA chip, DNA array, Protein chip..
22
MICROARRAYS a cDNA o OLIGONUCLEOTIDI

• Sistemi utilizzati per confrontare i livelli di
  espressione genica in due campioni diversi.
• Estrazione RNA cellulare
• Trasformazione in cDNA
• Marcatura del cDNA
• Ibridazione (DNA/nucleotidi)
• Lettura laser
• Analisi dati

De Risi et al Science 278680 (1997) Heller et al
PNAS 942150 (1997)
23
Un microarray è costituito da una superficie
sulla quale sono depositate migliaia di sequenze
specifiche di nucleotidi, ciascuna delle quali
identifica un particolare gene. Le diverse
migliaia di cDNA sono poste in spot separati.
Ciascuno spot rappresenta un gene, in quanto
contiene numerose copie di un cDNA corrispondente
a tale gene.
1.28 cm
1.28 cm
Milioni di catene di DNA in ciascuno spot
500.000 spot
25 basi in ogni catena
GeneChip array
24
RNA del tumore
RNA normale
Ibridando tale superficie con cDNA ottenuti dalla
retro-trascrizione dellRNA estratto da due
campioni diversi è possibile determinare il
livello di espressione dei singoli geni per
confronto diretto tra labbondanza relativa di
RNA prodotto.
cDNA del tumore
cDNA normale
Ibridazione
Analisi statistica
Plot multidimensionale
25
Per effettuare tale confronto, i cDNA
corrispondenti ai due differenti campioni vengono
marcati con sostanze fluorescenti diverse e, ad
ibridazione avvenuta, il microarray viene esposto
ad una sorgente di luce laser. Gli spettri di
emissione vengono quindi raccolti da uno scanner
e le immagini monocromatiche indicanti i livelli
diversi di espressione genica vengono
pseudocolorate da un software di acquisizione
dimmagine.
De Risi J.L. et al Science 1997
278680-686. Heller R.A. et al PNAS 1997
942150-2155.
26
(No Transcript)
27
Utilizzo dei microarrays
Fattori Prognostici e predittivi Markers
Diagnostici Targets per farmaci Attività
farmaci Per lo studio di Tumori Patologie su
base genetica Malattie infettive
28
I cDNA MICROARRAYS nel 2004

• Tre ditte stanno lanciando dei chip per lintero
  genoma umano
• Applied Biosystems, 30.000 geni
• tecnologia chemiluminescenza
• NimbleGenSystems, 190.540 probes
• con una media di 5 pobes per gene
• tecnologia fotolitografica
• Agilent Technologies, 44000 probes
• per 36.000 geni e trascritti
• tecnologia injk-jet

Agilent's microarray, con 36.000 geni e
transcritti su un singolo vetrino 1 x 3". I
probes sono sintetizzati in situ usando la
tecnologia ink-jet
29
Limiti dei Microarrays

• Disponibilità di tessuto fresco
• Esame dellespressione genica limitato alla
  valutazione della presenza di mRNA
• Riproducibilità dei dati
• Eterogeneità delle cellule presenti nel campione

30
Laser capture microdissection (LCM)
Campione eterogeneo
Campione A
Campione B
Microdissezione
Microdissezione
Popolazione omogenea
Popolazione omogenea
Popolazione eterogenea
Popolazione eterogenea
31
Siero
ANALISI PROTEOMICA
Proteine
Peptidi
Possibilità di individuare markers molecolari di
tumori (o altre patologie)
Frazionamento
Digestione con enzimi proteolitici
Cromatografia o 2D-PAGE
Analisi con algoritmi specifici
Spettrometria di massa
Sidransky D. Emerging molecular markers of
cancer. Nature Cancer Rev 2002 2210-9.
32
PROTEIN MICROARRAYS (ProteinChip)

• Sono utilizzati per esaminare
• i livelli di espressione delle proteine
• le interazioni proteina-proteina
• le interazioni proteina-piccole molecole
  (farmaci, etc)
• le attività enzimatiche

Page, M. J. et al. Proteomic definition of
normal human luminal and myoepithelial breast
cells purified from reduction mammoplasties. PNAS
1999 961258912594.
33
PROTEIN MICROARRAYS

• Esistono due tipi principali di chip
• antibody arrays
• Ab Microarray 500 - BD Biosciences' Clontech
  division, Palo Alto, CA
• gt 500 anticorpi per quantificare proteine in
  lisati cellulari o altri campioni biologici
• TranSignal Human Cytokine Antibody Array 2.0
  (Redwood City, CA)
• gt 21 anticorpi per misurare citochine
• general protein arrays
• Yeast ProtoArray from Protometrix, Branford,
  CT, con circa 5.000 polipeptidi da Saccharomyces
  cervisiae
• per monitorare le interazioni proteina-proteina
  e proteina-piccole molecole (farmaci.)

Yeast ProtoArray
34
PROTEIN MICROARRAYS
Siero
Proteine
Bio-chip
Individuazione nuovi biomarkers
SELDI-TOF MS
m/z
Pattern proteico
Riconoscimento del pattern
Conrads TM et al. Cancer diagnosis using
proteomic patterns. Expert Rev Mol Diagn 3411-20
(2003)
35
Nuovi Biomarkers individuati con il ProteinChip e
tecnologia SELDI-TOF-MS
36
Gene profiling of breast cancer
37
Sorlie et al., PNAS 98, 10869-10874, 2001

38
Hierarchical clustering of 78 primary breast
cancers and 4 normal breast tissue
Dendrogramma
Alberi di unione tra i vari casi che si
assomigliano (i.e intensità di colore
relativo ad un gene o a gruppi di geni)
5 differenti fenotipi
39
(No Transcript)
40
Cluster analysis
Sorlie et al., PNAS, 98, 2001
41
Van t Veer et al. Nature 415, 530, January 2002
42
20 BRCA1/2 BC
78 sporadic BC (Tlt5cm, N-)
34 pts w metastases lt5 y
44 pts NED gt5 y
25,000 genes of microarray
Unsupervised hierarchical clustering
5000 genes significantly regulated (in gt 3
tumors)
Supervised hierarchical classification
231 genes correlated w disease outcome
Rank-ordered based on p-value
70 genes Poor/Good prognostic
signature correctly predicted disease outcome in
65/78 sporadic tumors
43
Good prognosis signature
Tumori clinici studio pilota
Poor prognosis signature
Tumori clinici serie di validazione (N19)
Vant Veer et al., Nature 415, 2002
44
(No Transcript)
45
(No Transcript)
46
295 carcinomi mammari sporadici Tlt5 cm, N-/N lt
53 anni
Decorso clinico in base al profilo di
espressione genica (70-gene prognosis signature)
Van de Vijver et al., NEJM, 347, 25, 2002
47
295 sporadic breast cancers Clinical course
according to gene expression profile
Van de Vijver et al., N Engl J Med, 2002
48
151 N- patients
Van de Vijver et al., N Engl J Med, 2002
49
144 N patients
Van de Vijver et al., N Engl J Med, 2002
50
151 N- patients Clinical course according to
molecular signature (A) or clinico-patological
classification (B, C)
Van de Vijver et al., N Engl J Med, 2002
51
Therapeutic benefit
According to usual selection criteria (EBCTCG)
over 100 pts N- pre-menopausal pts receiving
adjuvant chemotherapy, 83.5 are alive even
without chemotherapy and 13.5 die despite
chemotherapy at 5 years FU.
Using gene expression profile, only 22.5 of pts
will be over-treated
52
Clin Cancer Res 10 2272-83, 2004
53
(No Transcript)
54
(No Transcript)
55
(No Transcript)
56
Clin Cancer Res 9 6326-34, 2003
57
PE, Ab to CK HER-2 FISH
BM, Ab to CK
BM, Ab to CK nuclear Counterstaining w
d-p-indole
Same as in C Ab to uPAI-R
58
(No Transcript)
59
AdnaGen CancerSelect Genzyme Virotech GmbH
Test system for the early detection of
disseminated cells in blood for a better
diagnostic and monitoring of colon and
breast cancer patients
60
Will the new molecular knowledge be applied to
bedside?
61
EORTC/TRANSBIG MINDACT TRIAL
The first large-scale independent trial to
prospectively validate the 70-gene expression
signature (MammaPrint) in breast cancer.
Adequate Processed Core Biopsy Prognostic Risk
Evaluation
Randomize
Clinico-pathological
Microarray
Low Risk
Low Risk
Average/High Risk
Chemotherapy Possible further randomization
Endocrine therapy Possible further randomization
MammaPrint has reached level 3 in Evidence Based
Medicine
62
Other ongoing trials incorporating translational
research in BC
Evaluating predictive factors for
response BIG p53 (EORTC 10994) pts with
LABC Tax vs Non-Tax CT
(Neoadjuvant) Evaluating prognostic factors
(uPA/PAI-1) EORTC-RBG High-Risk,
Node-negative (NNBC-3) FEC vs FEC
Docetaxel ADEBAR ? 4 lymph nodes
Adjuvant Epirubicin Docetaxel
(Wilex's uPA inhibitor WX-UK1)
63
Setting The Gene Expression Base-Line For Breast
Cancer ResearchDate May 05, 2004
Affymetrix Launches ENCODE Array to Uncover
Hidden Function of Human GenomeOctober 22, 2004
Toray Develops Ultra Sensitive, Quick DNA
ChipDate September 20, 2004
Agilent Partners with TGen to Develop CGH Arrays
for Cancer ResearchJune 8
SIRS-Lab releases new biochipDate September
22nd, 2004
Velcura to use custom Affymetrix
technologyAugust 03, 2004
Affymetrix and Immusol to Collaborate on Cancer
Drug DiscoveryJune 22, 2004
Predicting cancer patient survival with gene-
expression dataDate May 06, 2004
Agilent Acquire Silicon GeneticsAugust 29, 2004
BioTrove Announces OpenArray Transcription
Analysis SystemDate September 20, 2004
Illumina Announces 100,000 SNPs on a Single
BeadChipDate April 21, 2004
Toshiba to Develop DNA Chip with Osaka University
July 20, 2004
NCI awards grant for gene expression
researchDATE Thursday, April 15, 2004
Gene Logic Provides Data from GeneExpress System
to FDAMay 13, 2004
Expression Analysis Launches Affymetrix
Microarray-Based Genotyping ServicesDate
September 14, 2004
64
Can the novel technologies be used to predict the
therapeutic response?
65
Nature Clin Pract Oncol 1 44-50, 2004
66
Early BC
New Engl J. Med. 351 2817-2826 (2004)
67
Panel of the 21 genes and the Recurrence Score
Algorithm
Oncotype DX? from Genomic Health
68
(No Transcript)
69
(No Transcript)
70
Advanced BC
J. Clin. Oncol. 23 732-740 (2005)
71
(No Transcript)
72
(No Transcript)
73
(No Transcript)
74
GENES GENE PRODUCTS INVOLVED IN DRUG
RESISTANCE/SENSITIVITY (Cancer literature)
Anthracyclines Topoisomerase II?, MDR,
MRP, ErbB-2 5-FU, Capecitabine Cyclin D1,
Thymidilate synthase, Thymidine
phosphorylase, NFkB, p53, Bax/Bcl2 Gemcitabine
Ribonucleotide reductase, 5-nucleotidase,
b-tubilin III deoxycytidine-kinase Vinca
Alkaloids MAP4, Topoisomerase I Taxol b-tubul
in III, MDR, MAP4, survivin Platin
compounds ERCC1, MDR1, b-tubulin III, XPA, XPD,
cJun XPG, p53, cyclinD1, GSTpi, MLH1,
MSH6 Irinotecan, Topotecan Topoisomerase I,
p14ARF, carboxylesterase, MDR Small TRK
inhibitors Akt, MAPK, ecc
75
   

                 
       
Smart Chip (antibody array) A CINBOs Project
   
1.
Topo II
MDR1
ErbB-2
1. Anthracyclines
2.
Bcl-2/bax
Cyclin D1
TS
2. Fluoropyrimidines
 
3.
3. Taxanes
p14ARF
Topo I
Breast cancer
Colorectal cancer
1.
       
Topo I
p14ARF

• Irinotecan

 
TS
2.
2. 5-FU, capecitabine
Cyclin D1
Bcl-2/bax
       
3.
cJun
ERCC1
p53
3. Platin compounds
76
Smart Chip (antibody array) A CINBOs Project
FFIA (Fluorescent Immuno Assay)
2. Rh-Fusion-GFP proteins
1. Biopsia del paziente (Biomarkers)
Fluorescence intensity

Amount of Biomarker
Biotin-Antibody coated chip
77
Conclusion
The emerging fields of genomics (and proteomics)
offer the ability to precisely analyze the
molecular portrait of a particular patient
tumor
These approaches appear extremely useful for
defining individual patients prognosis and
assessing responiveness to anti-cancer therapy
A new era will come soon, wherein we will treat
each patient with a prescription based on the
molecular profile of its tumor resulting in more
rationale use of the therapy
78
(No Transcript)
79
Good prognostic signature
White ED pts Black NED pts
Poor prognostic signature