Oncotype DX

1
Oncotype DXBreast Cancer Assay

2
Agenda

• Introduction
• Development of Oncotype DX
• Clinical Studies
• Validation studies
• Hormonal therapy benefit study (NSABP B-14)
• Chemotherapy benefit study (NSABP B-20)
• Node study (SWOG 8814)
• Decision Impact Studies
• TAILORx
• Genomic Health Clinical Laboratory Experience
• Clinical Summary

3
Breast Cancer Treatment Planning History

• Treatment planning for N, ER disease is based
  on
• Traditional prognostic factors with limited
  predictive power (tumor size, patient age) or
  poor reproducibility (tumor grade)
• IHC markers (eg, Ki-67) lacking standardization
  and validation
• Limited insight into relative benefits of
  chemotherapy for different individuals

Bundred. Cancer Treat Rev. 200127137-142.
4
Breast Cancer Treatment Planning Not Optimized

• Chemotherapy treatment for N, ER disease
• Many women are offered chemotherapy, knowing that
  few benefit
• Prior to 2007, guidelines assumed all patients
  benefit equally
• Some patients are under-treated, many others are
  over-treated

5
What Would Your Treatment Strategy Be For This
Patient?

• Age 61
• ER 95
• PR 95
• Tumor Type IDC
• Tumor Size 0.6 cm
• Tumor Grade 2
• HER-2 neu Neg (FISH)
• Additional 6 mm on re-excision

www.adjuvantonline.com. Standard version 8.0.
Accessed 8/07
6
Recurrence Score 36Average Rate of Distant
Recurrence at 10 Yrs 25
7
Oncotype DX Unmet Clinical Need for Better
Markers
High risk/Large chemo benefit
Optimize chemotherapy local therapy hormonal
therapy
Biopsy or Resection
Robust markers
Optimize local therapy and hormonal therapy
Low risk/Little chemo benefit
8
Development and Validation of a 21-Gene Assay
for N, ER, Tam Patients
YEAR
Develop real-time RT-PCR method for paraffin block
2001
Select candidate genes (250 genes)
2002
Model building studies (N 447, including 233
from NSABP B-20)
2002
Commit to a single 21-gene assay
2003
Validation studies in NSABP B-14 and Kaiser
Permanente
2003
Paik et al. N Engl J Med. 20043512817-2826.
9
Oncotype DX Technology Final Gene Set Selection
Objective Gene expression and relapse-free
interval correlations across three independent
studies testing 250 genes in 447 patients
Study Site N Node Status ER Status Treatment
NSABP B-20, Pittsburgh, PA 233 N ER Tamoxifen (100)
Rush University, Chicago, IL 78 gt10 positive nodes ER/ Tamoxifen (54)Chemotherapy (80)
Providence St. Josephs Hospital, Burbank, CA 136 N/ ER/ Tamoxifen (41)Chemotherapy (39)
21 genes and Recurrence Score (RS) algorithm
Paik et al. SABCS 2003. Abstract 16. Cobleigh
et al. Clin Cancer Res. 200511(24 Pt
1)8623-8631. Esteban et al. Proc ASCO 2003.
Abstract 3416.
10
Oncotype DX 21-Gene Recurrence Score (RS) Assay
16 Cancer and 5 Reference Genes From 3 Studies
RS 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 - 0.07 x BAG1
PROLIFERATION Ki-67 STK15 Survivin Cyclin B1 MYBL2
ESTROGEN ER PR Bcl2 SCUBE2
BAG1
GSTM1
INVASION Stromelysin 3 Cathepsin L2
CD68
Category RS (0 -100)
Low risk RS lt18
Int risk RS 18 - 30
High risk RS 31
REFERENCE Beta-actin GAPDH RPLPO GUS TFRC
HER2 GRB7 HER2
Paik et al. N Engl J Med. 20043512817-2826.
11
Oncotype DX Clinical Validation RS as
Continuous Predictor
My RS is 30. What is the chance of recurrence
within 10 years?
95 CI
12

• Oncotype DX Clinical Validation
• The NSABP B-14 Study

Paik et al. N Engl J Med. 20043512817-2826.
13
Oncotype DX Clinical Validation Genomic Health
NSABP B-14

• Objective Prospectively validate RS as predictor
  of distant recurrence in N, ER patients
• Design
• Multicenter study with prespecified 21-gene
  assay, algorithm, endpoints, analysis plan

Paik et al. N Engl J Med. 20043512817-2826.
14
Oncotype DX Clinical ValidationB-14 Results
Distant Recurrence
Distant Recurrence Over Time All 668 Patients
100
90
80
70
60
Proportion Without Distant Recurrence at 10
years 85
50
Proportion without Distant Recurrence
40
30
20
10
0
0
2
4
6
8
10
12
14
16
Years
Paik et al. N Engl J Med. 20043512817-2826.
15
Oncotype DX Clinical Validation B-14 Results
Distant Recurrence
Distant Recurrence for the three distinct cohorts
identified
P lt0.001
Proportion without Distant Recurrence
RS lt18 n 338
RS 18-30 n 149
RS ?31 n 181
Paik et al. N Engl J Med. 20043512817-2826.
16
Oncotype DX Clinical ValidationB-14 Results
Distant Recurrence
Risk Group of 10-yr
Rate of Patients
Recurrence 95 CI Low (RS lt18) 51
6.8 4.0, 9.6 Intermediate (RS 18-30)
22 14.3 8.3, 20.3 High (RS 31)
27 30.5 23.6, 37.4
Test for the 10-year Distant Recurrence
comparison between the low-and high-risk groups
P lt0.001
Paik et al. N Engl J Med. 20043512817-2826.
17
Oncotype DX Clinical Validation B-14 Results
Multivariate Analysis Confirms Power of RS
Multivariate Cox Models Age, Size RS
P value
95 CI
Hazard Ratio
Variable
(0.48, 1.05)
0.71
Age 50
0.084
(0.86, 1.85)
1.26
Size gt2.0 cm
0.231
(2.23, 4.61)
3.21
Recurrence Score
lt0.001
Age at surgery used as a binary factor 0 lt50
yr, 1 50 yr. Clinical tumor size (CTS) used as
a binary factor 0 2 cm, 1 gt2 cm. Recurrence
Score used as a continuous variable, with HR
relative to an increment of 50 RS units.
Paik et al. N Engl J Med. 20043512817-2826.
18

• Oncotype DX Clinical Validation
• The Kaiser Permanente Study

Habel et al. Breast Cancer Res. 2006May
318(3)R25.
19
The Kaiser Permanente Study Methods
Matched Case-Control Kaiser Permanente patients
lt75 yr in 14 Northern California hospitals
diagnosed with node-negative BC 1985-94, no
chemotherapy (N 4964) Cases Deaths from BC (n
220) Controls Randomly selected, matched on
age, race, diagnosis year, KP facility, tamoxifen
(n 570) Cancer registry, medical records,
archived diagnostic slides, and tumor blocks

• Study Design
• Study Population
• Data Sources

Habel et al. Breast Cancer Res. May 2006.
20
The Kaiser Permanente Study Risk of BC Death
at 10 Years ER, Tam Patients

• 10-yr 10-yr
• Absolute Absolute
• Risk Classifier Risk1 Risk Kaiser NSABP
  B14
• Recurrence Score
• Low (lt18) 2.8 3.1
• Intermediate (18-30) 10.7 12.2
• High (gt31) 15.5 27.0

1Based on methods by Langholz and Borgan,
Biometrics 199753767-774.
Habel et al. Breast Cancer Res. May 2006.
21
The Kaiser Permanente Study Conclusions

• The RS has now been shown to be strongly
  associated with risk of breast cancer-specific
  mortality among LN, ER, tam-treated patients
  participating in a clinical trial and among
  similar patients from the community setting.
• Combining Recurrence Score, tumor grade, and
  tumor size provides better risk classification
  than any one of these factors alone.

Habel et al. Breast Cancer Res. May 2006.
22

• Oncotype DX Prediction of Tam Benefit
• NSABP B-14 Placebo and Tamoxifen Arms

Paik et al. ASCO 2004. Abstract 510.
23
Tamoxifen Benefit and Oncotype DX
NSABP B-14 Tam Benefit Study in N, ER Patients

• Design
• Objective Determine whether the 21-gene RS assay
  provides predictive information for patients who
  were treated with tamoxifen (likelihood of
  recurrence)

Placebo-Eligible
Randomized
Tam-Eligible
Paik et al. ASCO 2004. Abstract 510.
24
B-14 Overall Benefit of Tamoxifen
All Patients (N 645)
1.0
0.9
0.8
0.7
0.6
0.5
Proportion without Distant Recurrence
0.4
0.3
0.2
Placebo
0.1
Tamoxifen
0.0
12
0
2
4
6
8
14
16
10
Years
Paik et al. ASCO 2004. Abstract 510.
25
B-14 Benefit of TamoxifenBy Recurrence Score
Risk Category
Proportion without Distant Recurrence
Proportion without Distant Recurrence
Low Risk (RSlt18)
Int Risk (RS 18-30)
N 171 142
N 85 69
Interaction P 0.06
Proportion without Distant Recurrence
High Risk (RS31)1
1 The results should not be used to conclude that
tamoxifen should not be given to the high-risk
group
N 99 79
Paik et al. ASCO 2004. Abstract 510.
26
Analysis of Placebo and Tam-Treated Patients in
NSABP B-14

• Results
• Subset of Oncotype DX genes is prognostic
• 5 proliferation genes
• Cyclin B1, Ki-67, MYBL2, Survivin, STK15
• PR
• Quantitative measurement of the ER gene
  expression by the Oncotype DX assay predicts the
  benefit of tamoxifen
• Quantitative ER and Recurrence Score are only
  modestly correlated

Paik et al. ASCO 2004. Abstract 510.
27
Analysis of Placebo and Tam-Treated Patients in
NSABP B-14

• Conclusions
• RS combines prognostic and predictive factors
  into one assay report
• RS performance is derived from measurement of
  expression of each of the 21 genes on a
  continuous scale with high precision and
  reproducibility

Paik et al. ASCO 2004. Abstract 510.
28

• Oncotype DX Prediction of Chemo Benefit
• NSABP B-20 Study

Paik et al. J Clin Oncol. 2006243726-3734
29
Chemotherapy Benefit and Oncotype DX
NSABP B-20 Chemo Benefit Study in N, ER Pts

• Design
• Objective Determine the magnitude of the chemo
  benefit as a function of the 21-gene RS assay

Tam MF
Randomized
Tam CMF
Tam
Paik et al. J Clin Oncol. 2006243726-3734.
30
B-20 Results
Tam vs Tam Chemo All 651 Patients
1.0
0.9
0.8
4.4 absolute benefit from tam chemo at 10
years
0.7
0.6
Proportion without Distant Recurrence
0.5
0.4
0.3
0.2
N Events 424 33 227
31
All Patients
Tam Chemo
0.1
P 0.02
Tam
0.0
0
2
4
6
8
10
12
Years
Paik et al. J Clin Oncol. 2006.
31
B-20 Results Tam vs Tam Chemo
p 0.61
Low RS
p 0.39
Int RS
Proportion without Distant Recurrence
28 absolute benefit from tam chemo
High RS
p lt 0.001
Paik et al. J Clin Oncol. 2006.
32
B-20 Results Absolute Increase in Proportion
Distant Recurrence-Free at 10 Years
n 353
Low RS lt18
n 134
Int RS 18-30
n 164
High RS 31
0 10 20 30
40
Increase in Proportion Distant Recurrence-Free
at 10 Yrs (mean SE)
Paik et al. J Clin Oncol. 2006.
33
Summary of Treatment Benefit Related to RS and
Breast Cancer Death in NSABP B-14 and B-20
34
Largest Tamoxifen Benefit Observed in Low- and
Intermediate-Risk Recurrence Score Groups
35
Largest Chemotherapy Benefit Observed in
High-Risk Recurrence Score Group
36
Standardized Quantitative Oncotype DX Assay

• Recurrence Score in N-, ER patients

• Lower RSs
• Lower likelihood of recurrence
• Greater magnitude of TAM benefit
• Minimal, if any, chemotherapy benefit

• Higher RSs
• Greater likelihood of recurrence
• Lower magnitude of TAM benefit
• Clear chemotherapy benefit

1) Paik et al NEJM 2004, 2) Habel et al Breast
Cancer Research 2006 3) Paik et al JCO 2006, 4)
Gianni et al JCO 2005
37
Correlation of RS with traditional prognostic
factors including age, tumor size, and tumor
grade

• Results from NSABP B20 Results

38
NSABP B20 Results Many Younger Patients Have Low
Recurrence Scores
Paik et al. J Clin Oncol. 2006.
39
NSABP B20 Results Many Small Tumors Have
Intermediate to High RS
Paik et al. J Clin Oncol. 2006.
40
NSABP B20 Results Significant Proportion of
High-Grade Tumors Have Low RS
Grading by pathologist at local clinical trial
site
Grading by pathologist at central lab
Paik et al. J Clin Oncol. 2006.
41
Prospective multi-center study of the impact of
the 21-gene Recurrence Score (RS) assay on
medical oncologist (MO) and patient (pt) adjuvant
breast cancer (BC) treatment selection

• Shelly S. Lo1, John Norton1, Patricia B. Mumby1,
  Jeffrey Smerage2, Joseph Kash3, Helen K. Chew4,
  Daniel Hayes2, Andrew Epstein5, Kathy S. Albain1
• 1Loyola University, Maywood IL, 2University of
  Michigan, Ann Arbor MI, 3Edward Hospital,
  Naperville IL, 4UC Davis, Sacramento CA, 5Mount
  Sinai Medical Center, New York NY

ASCO 2007, Abstract 577
42
Background

• There is little data regarding the impact of the
  RS on medical oncologist (MO) and patient (pt)
  decision making. A multi-center study was
  designed to prospectively examine whether the RS
  can affect MO and pt adjuvant treatment
  selection.

ASCO 2007, Abstract 577
43
Methods

• 17 MOs at 1 community and 3 academic practices
  participated. Each participating MO
  consecutively offered enrollment to eligible
  women with N-, ER BC.
• Each participating MO and consenting patient
  completed pre- and post-RS assay questionnaires.
• MOs stated their adjuvant treatment
  recommendation and confidence in it pre and post
  RS assay.
• Pts indicated treatment choice pre and post RS
  assay. In addition, patients completed measures
  for quality of life, anxiety, and decisional
  conflict pre and post assay.
• RS assay results were returned to MO and shared
  with pt for routine clinical care.

ASCO 2007, Abstract 577
44
Change in MO Treatment Recommendation by RS
MO Treatment Recomm. Pre-RS Pre-RS Post- RS Post- RS
Number () Mean RS Number () Mean RS
CHT 42 (47.2) 21 23 (25.8) 29
HT alone 46 (51.7) 18 60 (67.4) 16
Equipoise 1 (1.1) 19 6 (6.7) 19
ASCO 2007, Abstract 577
45
MO Treatment Recommendations Changed 31.5 of the
Time
MO Pre to Post-RS Assay Treatment Recommendation Number of Cases()
CHT to HT 20 (22.5)
HT to CHT 3 (3.4)
CHT or HT to Equipoise 5 (5.6)
Treatment plan did not change 61 (68.5)
Total 89 (100)

• Treatment recommendation changed for 28 (31.5)
  cases after results of RS Assay known.
• The most common change was from a recommendation
  of CHT to HT in 22.5 of cases

ASCO 2007, Abstract 577
46
Other Findings

• Confidence was increased in 76 of the MO
  recommendations.
• RS Assay impacted patient adjuvant treatment
  decisions
• 95 of patients were glad they had the test
  performed
• 12 (13.5) patient treatment decisions did not
  match their MO treatment recommendation. This
  may be due to
• Patients choosing different treatment option than
  that recommended
• Inadequate communication between MOs and pts
• Patient misunderstanding of survey question

ASCO 2007, Abstract 577
47
Conclusions

• RS Assay changed physician adjuvant treatment
  recommendation 31.5 of the time
• Results from the RS assay were associated with
  less adjuvant chemotherapy administration
• The most common treatment recommendation change
  for MO was changing recommendation from CHT to HT
  in 22.5 of cases
• Of the 6 pts whose physicians thought their RS
  represented equipoise, 1 pt chose chemotherapy,
  3 chose HT, 1 chose observation, and 1
  understood the concept of equipoise.
• Results of the RS Assay increased physician
  confidence in treatment recommendation

ASCO 2007, Abstract 577
48

• TAILORx(PACCT-1 Trial)Sponsored by NCI
  Administered by ECOG

Participating cooperative groups include ECOG,
SWOG, NCCTG, CALGB, NCIC, ACOSOG, and NSABP
49
Trial Assigning IndividuaLized Options for
Treatment (Rx) (TAILORx)

• Premise
• Integration of a molecular profiling test
  (Oncotype DX) into the clinical decision-making
  process
• Potential Implications
• Reduce chemotherapy overtreatment in those likely
  to be appropriately treated with hormonal therapy
  alone
• Reduce inadequate treatment by identifying
  individuals who derive great benefit from
  chemotherapy
• Evaluate benefit of chemotherapy where
  uncertainty still exists about its utility

50
TAILORx Scientific Rationale for NCI and Breast
Intergroup Selecting Oncotype DX Assay for
First PACCT trial

• Validated prognostic test for tamoxifen-treated
  patients
• Predictive of distant recurrence
• May be used as categorical or continuous variable
• Paik et al. NEJM, 2004
• Also validated in population-based Kaiser study
• Habel et al. Breast Cancer Research, May 2006
• Lower RS predictive of tamoxifen benefit
• Paik et al. ASCO 2005, abstr 510
• Higher RS predictive of chemotherapy benefit
• Paik et al. JCO, August 2006
• Correlates more strongly with outcome than
  Adjuvant!
• Bryant et al. St. Gallen, 2005
• Predictive of local recurrence in tam-treated
  patients
• Mamounas, SABCS 2005, abstr 29

Sparano, Clinical Breast Cancer, 2006Sparano,
ASCO Educational Book 2007
51
TAILORx PrimaryStudy Group
52
Schema TAILORx
Node-Neg, ER-Pos Breast Cancer
Register Specimen banking
Oncotype DX Assay
RS 11-25 Randomize Hormone Rx vs Chemotherapy
Hormone Rx
RS lt10 Hormone Therapy Registry
RS gt25 Chemotherapy Hormone Rx
Primary study group
53
Study Design Primary Objectives

• To determine whether adjuvant hormonal therapy
  (ie, experimental arm) is not inferior to
  adjuvant chemohormonal (standard arm) for
  patients in the primary study group (Oncotype
  DX RS 11-25)
• To create a tissue and specimen bank for patients
  enrolled in this trial to learn more about breast
  cancer

54
TAILORx Key Points

• Participating groups
• ECOG, SWOG, NCCTG, CALGB, NCIC, ACOSOG, and NSABP
• Adjuvant therapy
• Choice of hormonal and/or chemotherapy regimen is
  at discretion of treating physician
• Other trials
• May enroll on other CTSU or other cooperative
  group studies if treatment assignment on other
  trial is consistent
• Payment for the Oncotype DX Assay
• Genomic Health will assist in securing
  reimbursement for patients who have health
  insurance
• By agreement with NCI to avoid bias in enrollment
  in the trial, patients who are uninsured or who
  have copayments or deductibles will not be
  responsible for the cost of the Oncotype DX

55
Oncotype DX Extensively StudiedStudy
Experience in gt3300 Patients
Study Type No. Pts
Providence Exploratory 136
Rush Exploratory 78
NSABP B-20 Exploratory 233
NSABP B-14 Prospective 668
MD Anderson Prospective 149
Kaiser Permanente Prospective Case-Control 790 Cases/Controls
NSABP B-14 Prospective Placebo vs Tam 645
Milan Exploratory 89
NSABP B-20 Prospective Tam vs TamChemo 651
ECOG 2197 Exploratory and Prospective 776
SWOG 8814 Prospective Tam vs TamChemo 367
Published studies
56
Conclusions

• Recurrence Score in N-, ER patients

• Lower RSs
• Lower likelihood of recurrence
• Greater magnitude of TAM benefit
• Minimal, if any, chemotherapy benefit

• Higher RSs
• Greater likelihood of recurrence
• Lower magnitude of TAM benefit
• Clear chemotherapy benefit

1) Paik et al NEJM 2004, 2) Habel et al Breast
Cancer Research 2006 3) Paik et al JCO 2006, 4)
Gianni et al JCO 2005
57
Genomic Health Today

• Physician Usage and Adoption
• 65,000 Oncotype DX test results delivered
• gt7,500 physicians have ordered the test
• Reimbursement for Oncotype DX
• Coverage for Medicare patients
• Coverage gt90 of privately insured lives
• As of August 5, 2008

58
Conclusions

• The Oncotype DX Recurrence Score assay predicts
  the likelihood of adjuvant chemotherapy benefit
• It also is a prognostic assay for the risk of
  distant recurrence at ten years assuming five
  years of adjuvant tamoxifen treatment
• Oncotype DX Recurrence Score assay shows
  consistent results across multiple independent
  studies