Breast Cancer goes Molecular: Understanding Molecular Taxonomy and Application in Routine Practice - PowerPoint PPT Presentation

Loading...

PPT – Breast Cancer goes Molecular: Understanding Molecular Taxonomy and Application in Routine Practice PowerPoint presentation | free to download - id: 443f09-ODQ0O



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Breast Cancer goes Molecular: Understanding Molecular Taxonomy and Application in Routine Practice

Description:

Breast Cancer goes Molecular: Understanding Molecular Taxonomy and Application in Routine Practice Rohit Bhargava, M.D. Department of Pathology Magee-Womens Hospital – PowerPoint PPT presentation

Number of Views:202
Avg rating:3.0/5.0
Slides: 57
Provided by: RohitBha
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Breast Cancer goes Molecular: Understanding Molecular Taxonomy and Application in Routine Practice


1
Breast Cancer goes MolecularUnderstanding
Molecular Taxonomy and Application in Routine
Practice
  • Rohit Bhargava, M.D.
  • Department of Pathology
  • Magee-Womens Hospital
  • University of Pittsburgh Medical Center
  • Pittsburgh, PA

2
Disclosures
NOTHING TO DISCLOSE
3
Background
  • Morphologic classification
  • Ductal and lobular carcinomas
  • Special subtypes of breast carcinoma
  • Tumor grading
  • Drawbacks of morphologic classification
  • No difference in disease free and overall
    survival between ductal and lobular tumors
  • Different criteria used to define special
    sub-types, and most criteria are arbitrary
  • Poor inter-observer reproducibility

4
Background
  • Factors that prompted new classification
  • Availability of gene expression profiling assays
  • Desire to identify new prognostic and predictive
    biomarkers
  • Clinicians and researchers fed up with the
    nuances of surgical pathology!

5
Outline
  • Molecular classification
  • Brief overview of intrinsic gene set based
    classes of breast cancer
  • Other molecular (multi-gene prediction) assays
  • Recurrence score model (oncotype DX)
  • Seventy gene profile (Mammaprint)
  • Two gene ratio (Theros H/I indexSM)
  • Molecular grade index (Theros MGISM)
  • How to use the above information in routine
    practice?

6
Molecular Classification
  • Molecular portraits of human breast cancer
    (Nature 2000406747752)
  • Forty-two patients, 65 specimens (36 invasive
    carcinomas)
  • Twenty tumors sampled twice before and after 16
    weeks of chemotherapy
  • Two tumors paired with lymph node metastasis from
    the same patient
  • c-DNA microarrays used to study 8102 genes
  • Complementary DNA clones used as probes to
    hybridize a target under high stringency
    conditions

7
Molecular Classification(Nature 2000406747752)
  • Cy5 labeled cDNA prepared by labeling
    experimental sample mRNA
  • Reference sample against which experimental
    sample was compared
  • Cy3 labeled cDNA prepared from pooled mRNA of 11
    different cultured cell lines
  • Hybridization to cDNA microarray
  • Relative abundance of 2 transcripts visualized
    using pseudo-colored image
  • Ratio of red to green fluorescence intensity
    at each spot

8
Molecular Classification(Nature 2000406747752)
  • Ratios are log transformed
  • Data further analyzed by hierarchical clustering
    algorithm
  • Classifies samples based on overall similarity of
    gene expression
  • Focused on 1753 genes
  • Fourfold change from the median abundance in the
    sample set in at least 3 samples
  • Created dendrogram using hierarchical clustering
    analysis
  • Pattern and length of branches reflects the
    relatedness of samples

9
15/20 tumor samples before and after chemotherapy
clustered together Paired (primary tumor and LN)
samples also clustered together
Nature. 2000406747-752
10
Molecular Classification
  • Implications of relatedness of the paired sample
  • Each tumor is unique and has distinct gene
    expression signature
  • Type and number of non-epithelial cells remain
    constant and do not interfere with expression
    analysis

11
Molecular Classification
  • Selection of alternative gene subset
  • Consisted of 496 genes that showed significantly
    greater variation in expression between different
    tumors than between paired samples of the same
    tumor
  • Now known as intrinsic gene set
  • Ideally suited for classification
  • Consists of genes whose expression patterns were
    characteristic of an individual tumor as opposed
    to those which vary as a function of tissue
    sampling

12
Blue Luminal epithelial Green Normal
breast-like Pink ERBB2 Orange Basal-like
Transcript Levels Green Below median Red Above
median Black Equal to median Grey Technically
inadequate
Luminal epithelial/ER gene cluster
ERBB2 overexpression cluster
Basal epithelial cell associated cluster
A 2nd basal epithelial cell enriched gene cluster
Nature. 2000406747-752
13
Basal cell
Luminal cell
14
PNAS. 200198-10869-10874
  • Refinement of previous classification
  • Using larger number of cases (total 78
    carcinomas)
  • Clinical correlations (with respect to relapse
    free and overall survival)
  • Subdivision of ER tumors into at least 2 groups

15
ERBB2 amplicon cluster
Novel unknown cluster
Basal epithelial cluster
Normal breast like cluster
Luminal epithelial gene cluster
Sorlie T et al. PNAS. 200198-10869-10874
16
Overall Survival
Relapse free Survival
Sorlie T et al. PNAS. 200198-10869-10874
17
PNAS. 20031008418-8423
  • Analysis expanded to 115 tumors
  • Intrinsic gene list slightly expanded to 534
    genes
  • Similar groups with similar outcomes
  • Elimination of luminal C tumors
  • BRCA1 mutation carriers develop basal-like tumors

18
Other Multi-gene Assays for Breast
Cancer (2004-2009)
.some of which are commercially available
19
Multigene Prediction Assays
IHC based Assays ProExBr Mammostrat
RT-PCR based Assays Oncotype DX H/ISM index and
MGISM BioClassifier
FISH based Assay eXagenBC
Microarray based Assays MammaPrint Rotterdam
assay NuvoSelect Roche AmpliChip
20
Recurrence Score Model(oncotype DX)
  • RT-PCR based assay
  • Calculates the risk of distant recurrence in ER,
    node negative breast cancer
  • Test development and validation
  • Developed using stored 447 samples including 233
    samples from the tamoxifen treated arm of NSABP
    trial B-20
  • Test validation was performed on samples from 668
    patients of the tamoxifen-only treated arm in the
    NSABP trial B-14

21
Recurrence Score Model(oncotype DX)
  • Measures gene expression for 16 target genes
  • Estrogen group (ER, PGR, BCL2, SCUBE2)
  • HER2 group (GRB7, HER2)
  • Proliferation group (Ki67, STK15, Survivin,
    CCNB1, MYBL2)
  • Invasion group (MMP11, CTSL2)
  • Others (GSTM1, BAG1, CD68)
  • Five reference genes

22
oncotype DX
N Engl J Med 20043512817-2826
Unscaled RS 0.47 x GRB7 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
0.07 x BAG1
Highly dependent on ER expression, proliferation
and HER2 related genes
23
oncotype DX
  • Reported as distant disease recurrence score (RS)
  • Low-risk RS 0-17
  • Risk of recurrence 7
  • Intermediate risk RS 18-30
  • Risk of recurrence 14
  • High risk RS 31-100
  • Risk of recurrence 31
  • Successfully marketed by Genomic Health Inc.

24
oncotype DX
  • Can we predict oncotype DX RS using routine
    histology and IHC?
  • RS significantly correlates with
  • Tubule formation
  • Nuclear grade
  • Mitotic count
  • ER IHC score
  • PR IHC score
  • HER2 status

Flanagan MB et al. Histopathologic variables
predict oncotype DX Recurrence Score. Mod
Pathol. 200821363-368
25
Flanagan MB et al. Histopathologic variables
predict oncotype DX Recurrence Score. Mod
Pathol. 200821363-368
26
ER and PR IHC Score
  • ER and PR results can be made objective using an
    H-score like method
  • Percentage cellular staining for intensity levels
    0, 1, 2, 3
  • Example
  • 0 10 1 20 2 20 3 50


0
20
40
150


210
27
00, 10, 20, 3100 0x0 1x0 2x0
3x100 300
28
035, 140, 225, 30 0x35 1x40 2x25
3x0 90
29
Flanagan MB et al. Histopathologic variables
predict oncotype DX Recurrence Score. Mod
Pathol. 200821363-368
30
oncotype DX
Regression Equation RS 13.424 5.420 (nuclear
grade) 5.538 (mitotic count) 0.045 (ER
immunohistochemical score) 0.030 (PR
immunohistochemical score) 9.486
(HER-2/neu) Predicts the recurrence score with
an R2 of 0.66, indicating that the full model
accounts for 66 of the data variability
  • Data is based only on 42 pilot cases, however
    following could be inferred
  • Well differentiated carcinomas with strong ER
    staining generally have low RS
  • Poorly differentiated carcinomas with weak ER
    staining have high RS
  • HER2 status could be decisive in moderately
    differentiated tumors

31
oncotype DX
  • Test is widely utilized by oncologists
  • Cost of assay 3400
  • Test is performed on FFPE tissue
  • Ongoing clinical trial
  • Trial Assigning Individual Options for Treatment
    (TAILORx)
  • NCI-funded cooperative groups, coordinated by
    ECOG
  • Accrual began in 2006 and the results wont be
    available until 2013
  • Goal is to determine chemotherapy benefits when
    the RS ranges from 11-25

32
oncotype DX
  • Example of a case where oncotype created more
    confusion than providing any additional
    information
  • 57 year old woman, simple mastectomy
  • Invasive carcinoma, Nottingham grade I (score 5)
  • Three foci 1.5 cm, 0.6 cm, and 0.3 cm
  • Micrometastasis (2 mm) in 1/11 lymph nodes
  • Tumor reported as ER, PR, HER2 negative
  • Tumor block sent for oncotype DX
  • High recurrence score, tumor reported as ER
    negative

33
lymphocytes
Large sclerotic area
lymphocytes
TUMOR
lymphocytes
34
ER
35
Seventy Gene profile (Mammaprint)
  • Gene expression microarray based assay
  • Test was developed at the Netherlands Cancer
    Institute
  • Used oligonucleotide arrays to identify genes
    associated with breast cancer prognosis
  • Commercial assay utilizes 70 genes to assign good
    versus poor prognosis signature
  • Poor prognosis signature genes
  • Cell cycle
  • Invasion
  • Metastasis
  • Angiogenesis

36
Mammaprint
  • Test was validated on 295 breast cancers from
    young patients with or without lymph node
    involvement
  • Overall 10 year survival rates were 94.5 for
    good and 54.6 for poor prognostic signature
  • Estimated hazard ratio for distant metastasis
    with poor versus good signature was 5.1
  • Ratio remained significant when groups were
    analyzed with respect to lymph node status

37
Mammaprint
  • Commercially offered by Agendia BV, Netherlands
  • US contact Agendia, Huntington Beach, CA
  • Fresh/frozen tissue required for analysis
  • FDA approved
  • Patients lt61years
  • Stage I or II
  • Tumor size 5 cm
  • Lymph node negative disease
  • ASCO guideline committee on breast cancer tumor
    marker
  • Suggests more evidence is required before its use
    is recommended

38
Mammaprint
  • Ongoing clinical trial
  • The Microarray In Node-Negative Disease May Avoid
    Chemotherapy (MINDACT)
  • Sponsored by European Organizations for Research
    and Treatment for Cancer
  • Trial will compare Mammaprint assay with
    adjuvant! Online
  • Accrual began in 2007 and accrued lt100 patients
    in one year

39
Two Gene Ratio
  • Measures expression of 2 genes
  • Genes first identified by expression analysis of
    hormone receptor positive tumors from 60 patients
    treated with adjuvant tamoxifen
  • Expression signature predictive of disease free
    survival was reduced to 2 gene ratio
  • HOXB13IL17BR
  • HOXB13 associated with shorter recurrence free
    survival
  • IL17BR associated with longer recurrence free
    survival

Ma XJ et al. Cancer Cell. 2004 Jun5(6)607-16
40
Two Gene Ratio
  • Further validation of HOXB13IL17BR
  • RT-PCR on 852 FFPE primary breast cancers from
    566 untreated and 286 tamoxifen-treated patients
  • HOXB13IL17BR index predicted clinical outcome
    independently of treatment, but more strongly in
    node-negative patients
  • In multivariate analysis, the index remained a
    significant predictor of recurrence free survival

Ma XJ et al J Clin Oncol. 2006 Oct
124(28)4611-9
41
N Engl J Med. 2006355560-569
  • Concordance among 4 expression profiling based
    assays
  • Intrinsic subtypes
  • 70 gene profile (Mammaprint)
  • Wound response
  • Recurrence Score (oncotype Dx)

ERBB2/Basal-like/Luminal B poor 70 gene profile
activated wound response High recurrence score
Non-concordant group was the two gene index!
42
Five Gene Molecular Grade Index
  • Using a previously published database
  • Proc Natl Acad Sci U S A 200310059749
  • Differential gene expression in high versus low
    grade tumors
  • Subset of 39 genes in high-grade tumors that were
    more highly expressed in invasive carcinomas
    versus in-situ carcinomas
  • Narrowed this list to five genes based on
  • Functional annotation of the genes
  • Association with clinical outcome
  • Correlation with tumor grade in another
    independent cohort
  • Created a score called molecular grade index (MGI)

43
MGI-genes
affy_id Gene Peak of Expression Role in Cell
Cycle 203755_at BUB1B G2/M mitotic spindle
assembly checkpoint 204962_s_at CENPA
G2/M centromere assembly 204641_at NEK2
G2/M centrosome duplication 222077_s_at
RACGAP1 G2/M Initiation of
cytokinesis 209773_s_at RRM2 S DNA
replication
44
Five Gene Molecular Grade Index
  • Prognostic performance
  • Tested using two independent publicly available
    microarray data set to predict clinical outcome
  • Compared MGI with previously published 97-gene
    genomic grade index (GGI)
  • Four of five genes of MGI were part of 97 genes
    of GGI!
  • Developed an RT-PCR assay for these five genes of
    MGI

Sotiriou C et al. Gene Expression Profiling in
Breast Cancer Understanding the Molecular Basis
of Histologic Grade To Improve Prognosis. J Natl
Cancer Inst. 2006 98(4)262-272
45
MGI versus tumor grade and distant metastasis
free survival
Ma XJ et al. Clin Cancer Res. 2008142601-2608
46
Complementary Prognostic Value MGIHOXB13IL17BR
Risk Group Explanation 10 year distant metastasis free survival probability
Low risk low MGI and low or high HOXB13IL17BR 98
Intermediate high MGI and low HOXB13IL17BR 78
High risk high for both MGI and HOXB13IL17BR 60
Hazard ratio high versus low-risk group 26.4
95 CI, 5.8-121
47
MGI and HOXB13IL17BR
  • Commercially available tests
  • Theros H/ISM
  • Assessment of endocrine resistance and risk of
    recurrence in ER, node negative breast cancer
  • Theros MGISM
  • Objective measurement of tumor grade
  • Prediction of chemosensitivity in ER tumors
  • Theros Breast Cancer IndexSM
  • Combined risk analysis for better stratification
    of ER-positive, node-negative breast cancer
    patients

BioTheranostics (previously known as AviaraDx) is
now a part of BioMérieux
48
How to use this new information in routine
practice?
49
Clinical Prognosticators
  • Nottingham Prognostic Index (NPI)
  • 0.2 x tumor size (cm) grade (1-3) lymph node
    stage (1-3)
  • Good NPI 3.4
  • Moderate NPI gt3.4 but 5.4
  • Poor NPI gt5.4
  • St. Gallen criteria
  • Two risk categories for node negative breast
    cancers
  • Low risk ER and/or PR, grade 1, pT1, age 35
    years
  • High risk ER and PR-, grade 2-3, size gt2 cm, age
    lt35 years
  • NIH consensus statement

50
Clinical Prognosticators
  • Adjuvant! Online
  • SEER data
  • EBCTCG data
  • Proprietary formula
  • To estimate prognosis
  • To estimate benefit from systemic adjuvant
    therapy
  • Program is continuously updated since 2001
  • Takes other co-morbid conditions into account

Ravdin PM et al. J Clin Oncol. 200119980-991
51
Clinical Prognosticators-Adjuvant! online
52
Multigene Expression Assays versus Clinical
Prognosticators
  • Both provide somewhat similar information
  • Prospective trials are underway to compare one
    versus the other (MINDACT)
  • Integration would be more valuable than using
    them independently
  • Adjuvant! online is currently being updated to
    include oncotype DX into the model

53
Genet Med. 200911(1)66-73
  • Tests reviewed
  • Oncotype DX
  • Mammaprint
  • H/I indexSM

54
EGAPP Recommendations
Genet Med. 200911(1)66-73
55
The Future.
  • Multigene expression assays will be used more
    frequently in the future
  • Additional molecular assays for assessing the
    effectiveness of therapy will become available
  • Pathologists will be asked for additional
    biomarker testing by IHC

56
Thank you
About PowerShow.com