Oncotype DX testing in node-positive disease

Should gene array assays be routinely used in node positive disease? Yes Christy A. Russell, MD University of Southern California Oncotype DX testing in node-positive disease 1

Validity of the Oncotype DX assay has been demonstrated in multiple studies and guidelines along a continuum of nodal status 3 Information supporting Oncotype DX along a continuum of nodal status Node positive 1-3 Nodes 4 Nodes SWOG 8814 1 SWOG 8814 1 TransATAC 2 TransATAC 2 E2197 3 1. Albain KS, et al. SABCS. 2007; Abstract 10. 2. Dowsett M, et al. SABCS. 2008; Abstract 53. 3. Goldstein LJ, et al. ASCO. 2007; Abstract 526. Oncotype DX clinical validation in node-positive patients (ECOG trial 2197) ECOG TRIAL 2197 AC Doxorubicin 60 mg/m 2 Cyclophosphamide 600 mg/m 2 Every 3 weeks 4 cycles Tamoxifen 5 years If HR-positive (amended to allow Als) Plus RT if indicated Operable breast cancer 0-3 positive nodes T.1cm if node negative N = 2885 eligible patients AT Doxorubicin 60 mg/m 2 Docetaxel 60 mg/m 2 Every 3 weeks 4 cycles Tamoxifen 5 years If HR-positive (amended to allow Als) Plus RT if indicated 776 samples with genomic data, including Recurrence Score results Paraffin blocks with cancer cells occupying < 5% of the section area excluded Manual micro-dissection RNA extraction No difference between arms Median follow-up 76 months 96.8% reported follow-up until death or for at least 5 years Goldstein LJ, et al. ASCO 2007. Abstract 526. 4 2

Kaplan-Meier curve for disease-free survival Goldstein, L. J. et al. J Clin Oncol; 26:4092-4099 2008 Copyright American Society of Clinical Oncology Patients with 1-3 positive nodes and low Recurrence Score result do well without chemotherapy* 5-Year event rates by nodal status & Recurrence Score result RS Nodes RFI (%) DFS (%) OS (%) < 18 18-30 31 *Including micrometastases (pn1mi) Negative 96 93 95 Positive 95 91 97 Negative 86 87 97 Positive 87 77 86 Negative 87 80 92 Positive 75 61 72 Low Recurrence Score results (< 18) in patients with 1-3 positive axillary nodes may eventually be used to select individuals for a short course of chemotherapy plus hormonal therapy Elevated Recurrence Score results ( 18) may eventually be used to select individuals for participation in clinical trials evaluating novel treatment strategies or for more aggressive chemotherapy regimens RS, Recurrence Score result Goldstein LJ, et al. ASCO 2007. Abstract 526. 6 3

Oncotype DX clinical validation in node-positive patients (SWOG 8814 sub-analysis) Tamoxifen 5 yrs n = 361 SWOG 8814 Postmenopausal, node-positive, ER-positive breast cancer N = 1477 CAF 6 + tamoxifen n = 550 CAF 6 tamoxifen n = 566 SUB ANALYSIS Patients with samples (n = 666) RT-PCR obtained (n = 601) Tamoxifen alone (n = 148) CAF + T (n = 243) CAF T (n = 219) Sample for primary analysis 148 + 219 = 367 (40% of parent trial) Albain KS, et al. Lancet Oncol. 2009; [Epub ahead of print]. Superior disease-free survival and overall survival over 10 years 7 Recurrence Score result is prognostic for node-positive patients (tamoxifen arm) 1.00 DFS by risk group (tamoxifen-alone arm) 1.00 OS by risk group (tamoxifen-alone arm) 0.75 0.75 0.50 0.50 Stratified log-rank P = 0.017 at 10 years Stratified log-rank P = 0.003 at 10 years 0.25 RS < 18 (n = 55) 0.25 RS < 18 (n = 55) RS 18-30 (n = 46) RS 18-30 (n = 46) RS 31 (n = 47) RS 31 (n = 47) 0.00 0.00 0 2 4 6 8 10 0 2 4 6 8 10 Years since registration Years since registration 10-Year DFS: 60%, 49%, 43% 10-Year OS: 77%, 68%, 51% RS, Recurrence Score result Albain KS, et al. Lancet Oncol. 2009; [Epub ahead of print]. 8 4

High Recurrence Score result predictive of chemotherapy benefit in node-positive patients DFS BY TREATMENT & RS GROUP RS < 18 RS 18-30 RS 31 1.00 0.75 0.50 0.25 Stratified log-rank P = 0.97 at 10 years Stratified log-rank P = 0.48 at 10 years Stratified log-rank P = 0.033 at 10 years 0.00 CAF T (n = 91, 26 events) Tam (n = 55, 15 events) CAF T (n = 46, 22 events) Tam (n = 57, 20 events) CAF T (n = 47, 26 events) Tam (n = 71, 28 events) 0 2 4 6 8 10 Years since registration RS, Recurrence Score result Albain KS, et al. Lancet Oncol. 2009; [Epub ahead of print]. 0 2 4 6 8 10 Years since registration No benefit to CAF over time if low or intermediate RS 0 2 4 6 8 10 Years since registration Strong benefit if high RS 9 SWOG 8814: Breast cancer-specific survival of node-positive patients by treatment and Recurrence Score group 10 100 BREAST CANCER-SPECIFIC SURVIVAL BY TREATMENT RS < 18 RS 18-30 RS 31 100 100 75 75 75 50 Stratified log-rank 25 P = 0.56 at 10 years 25 CAF T (n = 91, 10 events) e Tamoxifen (n = 55, 4 events) 0 0 2 4 6 8 10 Years since registration 10-yr BCSS T: 92% vs CAF T: 87% No benefit to CAF over time for low Recurrence Score 50 Stratified log-rank P = 0.89 at 10 years 25 CAF T (n = 46, 10 events) Tamoxifen (n = 57, 11 events) 0 0 2 4 6 8 10 Years since registration 10-yr BCSS T: 70% vs CAF T: 81% Interaction P = 0.021 50 Stratified log-rank P = 0.033 at 10 years CAF T (n = 47, 18 events) Tamoxifen (n = 71, 20 events) 0 0 2 4 6 8 10 Years since registration 10-yr BCSS T: 54% vs CAF T: 73% Strong benefit to CAF over time for high Recurrence Score Albain KS, et al. Lancet Oncol. 2009; [Epub ahead of print]. RS, Recurrence Score result 5

SWOG 8814: Chemotherapy benefit greatest with higher Recurrence Score values, regardless of number of positive nodes 11 5-YEAR PROBABILITY OF DEATH OR DISEASE RECURRENCE robability of an event 5-Year pr 1.0 0.8 0.6 0.4 0.2 Tamoxifen, 4 nodes (n = 54) CAF T, 4 nodes (n = 86) Tamoxifen, 1-3 nodes (n = 94) CAF T, 1-3 nodes (n = 133) Chemotherapy benefit ( 4 nodes) Chemotherapy benefit (1-3 nodes) 0.00 0 20 40 60 80 100 Recurrence Score Albain K, et al. SABCS. 2007; Abstract 10. Risk of Distant Recurrence Using Oncotype DX Assay in Postmenopausal Primary Breast Cancer Patients Treated with Anastrozole or Tamoxifen: a TransATAC Study Dowsett M et al on behalf of the ATAC Trialists Group San Antonio Breast Cancer Symposium. 2008; Abstract 53. 6

Study overview ATAC study population (N = 9366) Tamoxifen Anastrozole Tamoxifen + Anastrozole (combination arm not examined) Primary Analysis: To determine whether Oncotype DX assay significantly adds to a proportional hazards model for time to distant recurrence (age, tumor size, grade, treatment) in node-negative, HR+, patients with no adjuvant chemotherapy Secondary analyses: Determine whether the relationship between continuous Recurrence Score result and time to distant recurrence differs by nodal status or treatment arm Determine the relationship of predefined Recurrence Score groups with time to distant recurrence by nodal status and treatment arm Evaluate whether Recurrence Score result adds to the Adjuvant! Online estimate of risk Dowsett M, et al. SABCS 2008; abstract 53. 13 Number of evaluable patients and distant events by nodal status Node Node Node negative positive unknown Total All 890 363 55 1308 Adjuvant chemo 9 55 1 65 HR negative 4 0 0 4 Didn t start T or A 5 2 1 8 Evaluable patients Number of distant events 872 306 53 1231 (71%) (25%) (4%) (100%) 72 74 6 152 Distributions of the clinical variables in the 1231 evaluable (non-n Am) patients were similar to those in the 2929 ATAC (non-n Am) patients who were not included in this study Dowsett M, et al. SABCS 2008; abstract 53. 14 7

Time to distant recurrence by Recurrence Score group Pro oportion distant recurrence-free Node negative (n = 872) (both treatment arms) 10 1.0 0.9 0.8 Log-rank P < 0.001 0.7 0.6 0.5 0.4 0.3 N (%) Events 0.2 Low 513 (59%) 20 0.1 Int 229 (26%) 24 High 130 (15%) 28 0.0 0 1 2 3 4 5 6 7 8 9 Years RS group HR* (95% CI) High vs Low 5.2 (2.7-10.1) Int vs Low 2.5 (1.3-4.5) 96% 88% 75% Pro oportion distant recurrence-free Node positive (n = 306) (both treatment arms) 1.0 0.9 0.8 83% 0.7 Log-rank P < 0.001 72% 0.6 0.5 51% 0.4 0.3 0.2 N (%) Events Low 160 (52%) 25 01 0.1 Int 94 (31%) 25 0.0 High 52 (17%) 24 0 1 2 3 4 5 6 7 8 9 Years RS group HR* (95% CI) High vs Low 2.7 (1.5-5.1) Int vs Low 1.8 (1.0-3.2) *Hazard ratio for RS group adjusted for tumor size, grade, age and treatment Dowsett M, et al. SABCS 2008; abstract 53. RS, Recurrence Score result 15 Rate of distant recurrence increases with the number of positive nodes for all Recurrence Score results 9-Year risk of distant recurrence (% %) 90 100 40 50 60 70 80 0 10 20 30 Mean 95% CI 4 Positive nodes 1-3 Positive nodes Node negative 0 5 10 15 20 25 30 35 40 45 50 Recurrence Score value Dowsett M, et al. SABCS 2008; abstract 53. 16 8

S1007: A Phase III, Randomized Clinical Trial of Standard Adjuvant Endocrine Therapy +/- Chemotherapy in Patients with 1-3 Positive Nodes, Hormone Receptor-Positive and HER2-Negative Breast Cancer with Recurrence Score of 25 or Less Ana M. Gonzalez-Angulo, M.D. Specific Hypotheses The 21-gene RS will predict the benefit of chemotherapy in node positive (1-3 nodes), HR+ breast cancer patients with RS 25 treated t with state-of t of-the the-art t endocrine therapy. Chemotherapy benefit (if it exists) will increase as the RS increases. Chemotherapy is not beneficial for some patients in the range of RS 0-25 and that the point of equivalence between chemotherapy and no chemotherapy can be identified in this range. Above this point, patients begin to benefit from the addition of chemotherapy. We will identify a RS cutpoint t for which there is clinically significant benefit of chemotherapy for all RS values above this cutpoint. 9

Rationale for Approach and Trial Design SWOG 8814 Patients with RS 25 Patients with by RS 14-25 Rationale for Approach and Trial Design SWOG 8814 Cox model and allow for a linear interaction of RS and treatment 10

Inclusion Criteria Histologically i ll confirmed diagnosis i of node positive (1-3 positive nodes) Invasive breast carcinoma with positive ER or PR and HER2-negative RS of equal or less than 25 (after rounding to the nearest integer value) Schema and Patient Flow Node-positive (1-3 nodes) HR-positive and HER2-negative breast cancer (N= 8,800) Patients consent to study-sponsored sponsored RS testing, discussion of potential trials, tumor tissue submission and linkage to cancer registry data (N= 600) RS already Available Physician and patients discuss randomization knowing the RS STEP 1 REGISTRATION Tumor tissue submission for RS STEP 2 RANDOMIZATION RS > 25 (N= 3,800) Discuss alternative ti trials for high risk patients RECURRENCE SCORE RS < 25 N= 5,600 Physician i and patients t discuss randomization knowing the RS Refuse Accept STEP 2 REGISTRATION/ RANDOMIZATION N= 4,000 Randomization stratified by 1. RS 0-13 vs. 14-25 2. Menopausal status 3. Axillary node dissection vs. Sentinel node biopsy N= 2,000 Chemotherapy; appropriate endocrine therapy N= 2,000 No Chemotherapy; appropriate endocrine therapy N= 1,600 Record chosen therapy and followed for vital status through cancer registry 11

Primary Objective To determine the effect of chemotherapy in patients with node positive breast cancer who do not have high RS by Oncotype DX. Patients with 1-3 positive nodes, and HR+ and HER2- breast cancer with RS 25 DFS for patients treated with chemotherapy compared to no chemotherapy and dependence on the magnitude of RS. Determine the optimal cutpoint for recommending chemotherapy or not. Secondary Objectives OS, DDFS, LDFI by receipt of chemotherapy or not and its interaction with RS. Toxicity Other molecular assays (PAM50 ROR) Impact of management with Oncotype DX on patient-reported anxiety Impact of Oncotype DX on the initial management cost Patient-reported utilities (QOL) for those randomized to chemotherapy vs. not Estimate the cost-effectiveness of management with Oncotype DX vs. usual care Role of other assays as predictors of DFS, DDFS and LDFI for patients randomized to chemotherapy vs not Impact of treatment with chemotherapy versus not on patient-reported fatigue and cognitive concerns Impact of management with Oncotype DX on patient-reported decision conflict, perceptions regarding Oncotype DX testing, and survivor concerns prior to screening, after disclosure of test results, and during the randomized trial 12

Chemotherapy Third Generation Regimens Second Generation Regimens Endocrine Therapy Premenopausal Patients Postmenopausal Patients 13

Statistical Considerations Primary question: to test whether chemo benefit (if it exists) depends on the RS; interaction of chemo and RS Based on simulation work, power to find a significant ifi interaction with an equivalence point is 81% if the interaction of chemo and RS (linear) is statistically significant (two-sided α) Estimated cutpoint is the upper bound of the 95% confidence interval on the point of equivalence Kaplan-Meier curves comparing arms will be generated separately for RS values below and above this cutpoint and tested t with stratified tifi log-rank kt tests. t Accrual rate: 6 years, at 56 patients per month Statistical Considerations Early termination for futility: After 2 years of accrual, a committee of five statisticians will meet and discuss viability of the trial based on accrual, acceptance of randomization, and crossover rate. The result of this discussion will be presented to CTEP and the DSMC for action. Crossover rate (randomized participants receiving the opposite treatment) -- upper limit is set at 15% in the first year and 12% after 2 years The trial will terminate if the crossover rate exceeds this unless NCI determines differently Interim analysis: after 24% of the events have been observed or approximately 6 years after initiation There would be subsequent annual interim analyses thereafter with 36%, 50%, 65%, and 82% in years 7-10 with the final analysis in year 11 14

Correlative Studies Tissue Collections Tumor Block Positive Node Negative Node Blood PAM50 ROR Other planned prognostic/predictive scores and pharmacogenomic studies 15

Conclusions If you believe that all women with ER (+), lymph node (+) breast cancer should receive chemotherapy, then then trial is UNETHICAL!!!! 31 16