ASCO 2012 Breast Cancer Updates Hope S Rugo

ASCO 2012: Breast Cancer Updates Hope S. Rugo, MD Professor of Medicine Director, Breast Oncology and Clinical Trials Education UCSF Helen Diller Family Comprehensive Cancer Center

Outline • What’s new in the treatment of HER 2 positive metastatic breast cancer? – Emilia - TDM 1 compared to lapatinib/capecitabine • Blackwell et al, #LBA 1 – MA. 31 - Is lapatinib as good or better than trastuzumab? • Gelmon et al, #LBA 671 – NSABP B-41 – lapatinib as neoadjuvant therapy • Robidoux et al, #LBA 506 • Do bisphonates improve breast cancer outcome? – MA. 27 – Impact of osteoporosis and osteoporosis therapy • Sheperd et al, #501 – AZURE – impact in postmenopausal women • Marshal, Coleman et al, #502

Outline (2) • DCIS – RTOG 9804 - does everyone with DCIS need radiation? • Mc. Cormick et al, #1004 • Evaluating chemotherapy combinations, doses – NSABP B-38 – adding gemcitabine, sequential versus combination chemotherapy in early stage disease? • Swain et al, #LBA 1000 – CALGB 40502 – comparing microtubule toxins in the metastatic setting • Rugo et al, #CRA 1002 – What is the role of maintenance chemotherapy for patients with metastatic disease? • Im et al, #1003 – Can we target the androgen receptor in TNBC? • Gucalp et al, #1006

HER 2 Positive MBC • The problem – Despite high response rates, almost all patients eventually develop progressive disease • How can we overcome resistance that we don’t understand? • Data before pertuzumab and T-DM 1: – Start with trastuzumab + chemotherapy/hormone rx – Continue HER 2 directed therapy through progression • Capecitabine/lapatinib > capecitabine (Geyer et al) • Capecitabine/trastuzumab > capecitabine (von Minckwitz et al) • Lapatinib/trastuzumab > lapatinib (Blackwell et al)

T-DM 1: Dual Mechanisms of Action Combined Targeted Therapy Trastuzumab Biologic Activity • Blocks downstream HER 2 signaling to inhibit proliferation of tumor cells • Flags HER 2 -positive tumor cells for destruction via antibody-dependent cellmediated cytotoxicity • Inhibits HER 2 shedding a. DM 1 is 25– 500 fold more potent than taxane in cytotoxic assays. Targeted Intracellular Delivery of DM 1 a • T-DM 1 binds to the HER 2 receptor and is internalized • DM 1 is released inside the cell, resulting in mitotic arrest and apoptosis • Systemic toxicity is limited due to low HER 2 expression in normal tissues

T-DM 1 selectively delivers DM 1 to HER 2 -positive tumor cells • Targeted intracellular delivery of a potent antimicrotubule agent, DM 1 T-DM 1 binds • Spares normal tissue from toxicity of free DM 1 to the HER 2 protein on cancer cells HER 2 • Trastuzumab-like activity by binding to HER 2 Receptor-T-DM 1 complex is internalized into HER 2 -positive cancer cell Potent antimicrotubule agent is released once inside the HER 2 -positive tumor cell

Phase II Randomized International Open Label Studyb Trastuzumab HER 2 -positive, recurrent locally advanced breast cancer or MBC (N=137) First line setting 8 mg/kg loading dose; 6 mg/kg q 3 w IV 1: 1 PDa + Docetaxel 75 or 100 mg/m 2 q 3 w Crossover to T-DM 1 (optional) (n=70) T-DM 1 3. 6 mg/kg q 3 w IV PDa (n=67) • Stratification factors: World region, prior adjuvant trastuzumab therapy, disease-free interval • Primary end points: PFS by investigator assessment, and safety • Data analyses were based on clinical data cut of Nov 15, 2010 prior to T-DM 1 crossover • Key secondary end points: OS, ORR, DOR, CBR, and QOL • Demographics imbalanced: 29 vs 34 % stage IV at diagnosis, 27 vs 18% exposed to prior trastuzumab, 40 vs 33% exposed to prior taxanes a. Patients were treated until PD or unacceptable toxicity. was a hypothesis generating study; the final PFS analysis was to take place after 72 events had occurred. b. This Hurvitz et al, ESMO 2011

PFS by Investigator Median PFS, mos N=137 Trastuzumab + docetaxel (n=70) 9. 2 T-DM 1 (n=67) 14. 2 Proportion progression-free 1. 0 0. 8 Hazard ratio 95% CI 0. 594 0. 364– 0. 968 Log-rank P value 0. 0353 0. 6 0. 4 Crossover allowed to TDM 1 Significantly less toxicity with TDM 1, better patient reported outcomes 0. 2 0. 0 0 2 4 6 8 10 12 Time (months) 14 16 18 20 63 51 53 46 43 42 4 15 2 6 2 3 0 0 Number of patients at risk T+D 70 T-DM 1 67 66 60 Hazard ratio and log-rank P value were from stratified analysis. 27 35 12 22 Hurvitz SA, et al. Abstract 5. 001. ESMO 2011.

Clinical Rationale for EMILIA • T-DM 1 – Two single-arm phase 2 trials in patients who received ≥ 1 HER 2 -directed therapies for MBC – ORR: 25. 9% (N=112)1 and 34. 5% (N=110)2 • Capecitabine + lapatinib – Randomized phase 3 trial in patients who received prior trastuzumab – Median TTP longer with capecitabine + lapatinib (n=163) vs. capecitabine (n=161), no difference in OS – 8. 4 vs. 4. 4 months (HR=0. 49; P<0. 001)4 1 Burris HA, et al. J Clin Oncol 2011; 2 Krop I, et al. J Clin Oncol 2012; 3 Hurvitz S, et al. ESMO 2011; 4 Geyer CE, et al. N Engl J Med 2006.

EMILIA Study Design HER 2+ (central) LABC or MBC (N=980) • Prior taxane and trastuzumab • Progression on metastatic tx or within 6 mos of adjuvant tx T-DM 1 3. 6 mg/kg q 3 w IV PD 1: 1 Capecitabine 1000 mg/m 2 orally bid, days 1– 14, q 3 w + Lapatinib PD 1250 mg/day orally qd • Stratification factors: World region, number of prior chemo regimens for MBC or unresectable LABC, presence of visceral disease • Primary end points: PFS by independent review, OS, and safety • Key secondary end points: PFS by investigator, ORR, duration of response, time to symptom progression Blackwell et al, ASCO 2011

EMILIA - Demographics • 496 vs 495 randomized • Median age 53, 27% from U. S. • Prior therapy – All treated with prior taxane – 16% received prior trastuzumab for early stage disease – 57% received at least one year of prior trastuzumab, 88% prior treatment for MBC – 79% measurable disease, 53 -57% HR+

Progression-Free Survival by Independent Review Median (mos) No. events Cap + Lap 6. 4 304 T-DM 1 9. 6 265 Stratified HR=0. 650 (95% CI, 0. 55, 0. 77) P<0. 0001 Subroup analysis: TDM 1 superior regardless of line of therapy (1 -3) and HR status No. at risk by independent review: Cap + Lap 496 404 310 176 129 73 53 35 25 14 9 8 5 1 0 0 T-DM 1 183 130 101 72 54 44 30 18 9 3 1 0 495 419 341 236 Unstratified HR=0. 66 (P<0. 0001).

Overall Survival: Interim Analysis Unstratified HR=0. 63 (P=0. 0005). NR=not reached.

Objective Response Rate (ORR) and Duration of Response (DOR) in Patients with Measurable Disease ORR DOR

Overview of Adverse Events Cap + Lap (n=488) T-DM 1 (n=490) All-grade AE, n (%) 477 (97. 7) 470 (95. 9) Grade ≥ 3 AE, n (%) 278 (57. 0) 200 (40. 8) AEs leading to treatment discontinuation (for any study drug), n (%) 52 (10. 7) 29 (5. 9) AEs leading to death on treatment, n (%)a 5 (1. 0) 1 (0. 2) LVEF <50% and ≥ 15 -point decrease from baseline, %b 7 (1. 6) 8 (1. 7) a. Cap + Lap: CAD, multiorgan failure, coma, hydrocephalus, ARDS; a. T-DM 1: metabolic encephalopathy. b. Evaluable pts: 445 (Cap + Lap); 481 (T-DM 1). • Cap and Lap: More grade 3 diarrhea (21 vs 1. 6%), hand foot syndrome (16 vs 0%) • TDM 1: More transaminiitis (4 vs 1%), grade 3/4 thrombocytopenia (13 vs 0%)

Summary and Ongoing Trials • T-DM 1 superior to cap/lap – PFS, interim OS, response, safety – Will clearly be a new standard in this setting – Approval expected towards the end of 2012 • Marianne (n=1092, untreated HER 2+ MBC) – Three arms • Trastuzumab + taxane • TDM 1 + pertuzumab • TDM 1 plus placebo • Th 3 RESA (n=795) – Prior trastuzumab/lapatinib/anthra/taxane/cape – 2: 1 randomization to TDM 1 v TPC

Early Stage Trials • Post-neoadjuvant cooperative group • Neoadjuvant company sponsored • Adjuvant small tumors: ATTEMPT Trial Stage I BC HER 2+ N=500 Randomize 3: 1 – Tolaney PI (DFCI) 3 1 Trastuzumab emtansine q 3 weeks x 17 N=375 Paclitaxel + Trastuzumab weekly x 12 Trastuzumab every 3 weeks x 13 N=125

Two Questions • Can lapatinib overcome trastuzumab resistance? – Doesn’t require externalized receptor – Can this agent overcome resistance mediated by alterations of the PI 3 K pathway? • Can dual targeting of the HER 2 Receptor overcome resistance more effectively? – Maintains benefits of trastuzumab while targeting the pathway through an alternate mechansim

Trastuzumab T 1 1 2 2 1 2 L L L Downstream signaling pathways Cell proliferation Cell survival Lapatinib

MA. 31/ EGF 108919: Metastatic Disease Randomize EXPERIMENTAL ARM 24 Weeks: Lapatinib plus Taxane Until PD: Lapatinib STANDARD ARM 24 Weeks: Until PD: Trastuzumab plus Trastuzumab Taxane Primary Outcome: PFS Sample Size: ~ 600 (536 centrally confirmed HER 2+ patients) Gelmon et al, LBA 671, ASCO 2012

Progression Free Survival Centrally-confirmed HER 2+ Analysis Median PFS TTAX/T= 13. 7 months Median PFS LTAX/L = 9. 0 months HR = 1. 48 (95% CI = 1. 15 – 1. 92), P = 0. 003 TTAX/T LTAX/L TTAX/ T LTAX/L

Serious Adverse Events LTAX/L (Total SAE reports = 136) EVENT TTAX/T (Total SAE reports = 78) Total Number* Number post amendment ** EVENT Total Number* Number post amendment ** Diarrhea 32 25 Diarrhea 5 3 Febrile Neutropenia 17 7 Febrile Neutropenia 7 6 ** Protocol Amendment after first 189 patients were randomized mandated * Included as one of the adverse event terms within a single SAE report primary GCSF prophylaxis for patients on docetaxel and lapatinib • Discontinuation rates significantly higher for lapatinib arm Gelmon et al, LBA 671, ASCO 2012

Neo-Altto (N=455) Baselga J et al. Lancet 2012 Untch M et al. Lancet Oncology 2012 Gepar. Quinto (N=615) Pertuzumab + trastuzumab: improved p. CR (Neo. Sphere, Tryphaena)

NSABP B-41 Schema (n=529) Tissue for Biomarkers Operable Breast Cancer R HER-2 neu Positive T > 2 cm AC→ WP + T AC→ WP + L 1250 AC→ WP + T + L 750 WP=Weekly Paclitaxel S U R Trastuzumab for a total of G 1 year E R Y Endpoints: p. CR, cardiac events, DFS, OS WP: d 1, 8, 15 q 28 d x 4 Robidoux et al, #LBA 506

Results • p. CR breast – 53% (T) vs 53% (L)vs 62% (TL) (P 0. 095 for TL vs T) • p. CR breast and nodes – 49% (T) vs 47% (L) vs 60% (TL) (p=0. 056 TL vs T) • p. CR based on HER 2 (IHC 3+, n=421, 81%) – 55% (T) vs 53% (L) vs 71% (TL) (p=0. 006 TL vs T) • Completion of protocol defined neoadjuvant Rx – 78% (T) vs 68% (L) vs 63% (TL) (p=0. 01) • Toxicity similar except diarrhea and overall – Grade 3: 2% (T) vs 20% (L) vs 27% (TL) (p<0. 001)

ALTTO Paclitaxel weekly or TCH Surgery Lapatinib Closed due to futility +/- Anthracycline containing CT Trastuzumab Lapatinib Trastuzumab 1 year

Clinical Context • What does this data mean to our clinical practice? – TDM-1 is a new and effective treatment for HER 2+ metastatic disease progressing on trastuzumab – Toxicities are unique but generally well tolerated – Likely to be FDA approved by the end of the year • Pertuzumab approved in the first-line setting in combination with trastuzumab/docetaxel (PFS 18. 5 mo. ) • Lapatinib is associated with more toxicity and less efficacy than trastuzumab in the first-line metastatic setting • Lapatinib/capecitabine is still an option for later line therapy or in special settings (low EF (? ), brain metastases) Pertuzumab/ Trastuzumab/ Taxane Pertuzumab/ Trastuzumab TDM-1 Lapatinib/ Capecitabine Multiple drugs in clinical trials: inhibitors of m. TOR, HSP 90, TKs, vaccines ?

Osteoporosis • Osteoporosis is characterized by decreased bone mineral density. • The increased bone resorption associated with osteoporosis may provide fertile “soil” for cancer growth. • Will osteoporosis or therapy for osteoporosis affect outcome in patients with early stage breast cancer?

Observational Studies of Bisphonate Use and Breast Cancer Incidence Author Study Design Breast Patient Cases Control Subjects (% bisphosphonate) Cancer Association Rennert Casecontrol 1822 (10. 5%) 2207 (14. 8%) OR 0. 72 (0. 57 -0. 90) Newcomb Casecontrol 2336 (4. 4%) 2975 (6. 2%) OR 0. 67 (0. 51 -0. 89) Chlebowski Cohort 154, 768 women, 2816 (1. 8%) bisphonate users, 5, 092 breast cancer cases Rennert G, Pinchev M, Rennert HS JCO, 2010 June 12 Epub ahead of print Newcomb PA, Trentham-Dietz A, Hampton JM Bristish J of Cancer 2010; 102: 799 -802 Chlebowski RT, Chen Z, Cauley JA, et al JCO 2010 June 12 Epub ahead of print HR 0. 68 (0. 52 -0. 88)

• Osteoporosis: 17% (1294) • Osteoporosis therapy: 36% (2711) – 116 (0. 2%) took raloxifene prior to study – 39 started after randomization • Of those with osteoporosis – 85% (1101) took osteoporosis therapy • Of those taking osteoporosis therapy – 25. 6% (1610) did not report osteoporosis Shepherd LE et al, ASCO 2012, #501

EFS by Osteoporosis Therapy HR (Yes/No) = 0. 70 p<. 00001 Shepherd LE et al, ASCO 2012, #501

Conclusions • Patients with self-reported osteoporosis and osteoporosis therapy had a lower incidence of breast cancer relapse • Patients receiving osteoporosis therapy, regardless of report of osteoporosis, had a lower incidence of relapse • Limitations – Self reporting – Variations in osteoporosis therapy and duration – Event rate (9. 2%) and distant relapse rate low (4. 1%)

The seed and soil hypothesis 'While many researchers have been studying ‘the seed’, the properties of ‘the soil’ may reveal valuable insights into the ‘metastatic peculiarities’ in cancer cases. ' The Distribution of Secondary Growths in Cancer of the Breast. The Lancet. 1889 Stephen Paget 1855– 1926

AZURE: Study Design Accrual September 2003 - February 2006 Standard therapy 3, 360 Breast Cancer Patients R Stage II/III Standard therapy + Zoledronic acid 4 mg No difference in disease free or invasive disease free survival (IDFS) 6 doses 8 doses 5 doses Q 3 -4 weeks Q 3 months Q 6 months Months 6 30 60 Coleman et al. N Engl J Med 2011; 365: 1396 -1405 Zoledronic acid treatment duration 5 years

AZURE: Invasive DFS and OS by Menopausal Status IDFS: Pre, Peri, and Unknown Menopause IDFS: > 5 Yrs Postmenopausal 1. 0 Proportion Alive and invasive Disease Free 1. 0 0. 8 Adjusted HR: 1. 15 0. 6 (95% CI: 0. 97 -1. 36; P =. 11) 0. 4 288 vs 256 events Pts at Risk, n ZOL: 1162 1088 No ZOL: 1156 1092 0. 2 0 0 996 919 829 393 57 0 995 920 853 388 47 0 0. 8 0. 6 0. 4 0. 2 0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 Adjusted HR: 0. 75 (95% CI: 0. 59 -0. 96; P =. 02) 116 vs 147 events Pts at Risk, n ZOL: 519 490 No ZOL: 522 482 0 Time From Randomization (Mos) 418 393 177 25 0 431 396 368 156 21 0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 Time From Randomization (Mos) N=2318 N=1041 OS: Pre, Peri, and Unknown Menopause OS: > 5 Yrs Postmenopausal 1. 0 0. 8 0. 6 Adjusted HR: 0. 97 (95% CI: 0. 78 -1. 21; P =. 81) 0. 4 161 vs 165 events Pts at Risk, n 0. 2 0 ZOL: 1162 1131 No ZOL: 1156 1123 0 1078 1020 955 466 71 0 1076 1032 963 446 60 Proportion Alive 1. 0 Proportion Alive 447 0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 Time From Randomization (Mos) Coleman RE, et al. N Engl J Med. 2011; 365: 1396 -1405. 0. 6 0. 4 0. 2 0 Adjusted HR: 0. 74 (95% CI: 0. 55 -0. 98; P =. 04) 82 vs 111 events Pts at Risk, n ZOL: 519 502 No ZOL: 522 509 0 482 448 422 190 29 0 475 441 401 177 26 0 6 12 18 24 30 36 42 48 54 60 66 72 78 84 Time From Randomization (Mos)

Treatment Effects on First IDFS Recurrence Outside Bone by Menopausal Status Menopausal Group Odds Ratio Pre, Peri and unknown menopause HR: 1. 32 (95% CI: 1. 09 -1. 59) > 5 yrs postmenopause HR: 0. 70 (95% CI: 0. 54 -0. 92) TOTAL: 6% +/- 8 Z =. 79, P =. 43 0. 2 0. 4 0. 6 0. 8 1. 0 1. 2 1. 4 1. 6 1. 8 2. 0 21 (heterogeneity) = 14. 00; P = <. 001 Adjusted for imbalances in ER, lymph node status, and T stage. No significant differences in bone recurrence by menopausal status or age

SABCS 2011

Conclusions • In this population of patients, primarily treated with adjuvant chemotherapy – No effect of zoledronate on breast cancer outcome in the unselected overall population – Suggestion of improved outcome in pts with >5 years of menopause, due to recurrence outside bone • Unplanned subset analysis of interest but not sufficient to influence patient care – Variable results in 7 published trials – Bisphonates should be used to prevent or treat bone loss in patients with breast cancer – Bisphonates should not be used only for the purpose of preventing breast cancer recurrence • Ongoing studies are evaluating the impact of denosumab on breast cancer outcome in women with early stage disease – D-Care, ABCSG-18 – UCSF phase II trial in pts with DTCs

RTOG 9804: Does Good Risk DCIS Always Need Radiation? • Common disease, low long-term risk • Primary goal of treatment is to prevent invasive IBTR • Clearly variable risk exist based on biology, age, extent of disease • We have transitioned from mastectomy to lumpectomy and radiation with essentially equivalent outcomes – Does everyone need radiation? Mc. Cormick et al, #1004

RTOG 9804: Eligibility and Schema No palpable mass, low or intermediate grade, < 2. 5 cm, margins > 3 mm Age 1. < 50 2 ≥ 50 Final Path Margins 1. Negative (re-excision) 2. 3 -9 mm 3. 10 mm Mammographic/Pathologic Size of Primary 1. ≤ 1 cm 2. > 1 cm to ≤ 2. 5 cm Nuclei Grade 1. Low 2. Intermediate Tamoxifen Use 1. No 2. Yes N=585 Arm 1 Observation ± tamoxifen, 20 mg per day for 5 years Results at 5 years • Local failure in ipsilateral breast • 3. 2 vs 0. 4% • HR 0. 14, p=0. 0022 • 15 vs 2 events Arm 2 • DFS • Identical Radiation therapy to the Mc. Cormick et al, #1004 whole breast, ± tamoxifen, 20 mg per day for 5 years • HR 0. 84, p=0. 48

Summary • In patients with good risk DCIS, the addition of radiation – Added little to local control – Added nothing to survival • Use of genomic tests, such as the GH DCIS score, may help refine treatment decisions • Longer follow-up of this trial will be interesting.

NSABP B-38 Schema (n=4894, med FU 5. 3 yrs) Stratification: # nodes, HR status + or neg, Surgery and RT TAC q 3 wk All arms pegfilgrastim or filgrastim AC q 2 wk P q 2 wk AC q 2 wk PG q 2 wk N+ Swain et al, LBA#1000 ER positive: hormonal therapy for 5 yrs after chemo 80% ER+ 65% 1 -3+ nodes P = paclitaxel 175 mg/m 2 G = gemcitabine 1000 mg/m 2

NSABP B-38 Disease-Free Survival # at risk 1610 1532 1618 1554 1613 1533 1424 1331 1452 1348 1453 1350 1217 1240 1244 * Stratified log-rank test adjusting for randomization factors 719 754 730

Survival, Toxicity and Conclusions • Overall survival no different between arms • More deaths on treatment in the TAC arm than AC/P or AC/PG (13/5/7, p=0. 2) • Addition of gemcitabine did not improve outcome • DD AC/P similar to TAC – More FN, anemia, diarrhea, less neuropathy with TAC • This trial started in 2004 and completed accrual in 2007 – About 10 years from initial planning to negative results – We can no longer do large trials testing therapy based solely on extent of disease

CALGB 40502 - NCCTG N 063 H - CTSU 40502 An Open Label Phase III Trial of Firstline Therapy for Locally Recurrent or Metastatic Breast Cancer Exp 1 1 nab-paclitaxel 150 mg/m 2 weekly + bevacizumab 10 mg/kg q 2 wks 2 N = 900 (planned) Control paclitaxel 90 mg/m 2 weekly + Randomize 1: 1: 1 bevacizumab 10 mg/kg q 2 wks 1 Strata: Adj taxanes ER/PR status Exp 2 ixabepilone 16 mg/m 2 weekly + bevacizumab 10 mg/kg q 2 wks 3 Restage q 2 cycles until disease progression • All chemotherapy was given on a 3 week on, one week off schedule • Patients could discontinue chemotherapy and continue bevacizumab alone after 6 cycles if stable or responding disease 1. Miller et al, N Engl J Med, 2007 2. Gradishar et al, J Clin Oncol, 2009 3. Dickson et al, Proc ASCO 2006.

CALGB 40502 Progression-Free Survival By Treatment Arm Comparison HR P-value 95% CI nab vs. pac 1. 19 0. 12 0. 96 -1. 49 ixa vs. pac 1. 53 < 0. 0001 1. 24 -1. 90 paclitaxel nab-paclitaxel ixabepilone Agent N Median PFS paclitaxel 283 10. 6 nab-Paclitaxel 271 9. 2 ixabepilone 245 7. 6

CALGB 40502 Overall Survival Comparison HR P-value 95% CI nab vs. pac 1. 02 0. 92 0. 75 -1. 38 ixa vs. pac 1. 28 0. 10 0. 95 -1. 72 Agent N Median OS paclitaxel 283 26 nab-paclitaxel 271 27 ixabepilone 245 21

Dose Reductions by Cycle 3 All Cause Cumulative Discontinuation by Cycle 45% 60 50 paclitaxel nab-paclitaxel ixabepilone Percent 40 15% 30 20 10 nab pac Cycle 3 ixa 0 1 2 3 Cycle number 4 5

Other AEs – Grade 3+ Sensory Neuropathy nab (N = 258) pac (N = 262) ixa (N = 237) Grade 3+ Leukopenia 25% p=0. 012 Neutropenia Hypertension 16% Fatigue 25% Pain p=0. 022 Motor neuropathy pac ixa (N = 258) Arm nab (N = 262) (N = 237) 17% p = 0. 0004 47% p = 0. 0001 7% 16% p = 0. 010 10% p = 0. 0003 7% 18% 8% 9% 4% 2% 3% p = 0. 042 7% p = 0. 0002 11% 15% p = 0. 036 4% 6% p = 0. 021

Summary and interpretation • Neither weekly nab-paclitaxel or ixabepilone are superior to weekly paclitaxel • Weekly paclitaxel appears to offer better progression-free survival than ixabepilone • Hematologic toxicity was greater with nabpaclitaxel; sensory neuropathy was greater in both experimental arms compared to paclitaxel • Paclitaxel is a reasonable choice in a similar setting • 100 mg/m 2 is the most appropriate dose for weekly nab-paclitaxel

Study Design: Role of Maintenance Therapy Prospective, phase III, multi-center, randomized study Enroll period: 2007. 05 – 2010. 09 Off the study PD 324 MBC patients with no prior chemotherapy till PD 6 cycles of PG PG regimen Paclitaxel 175 mg/m 2 Day 1 Gemcitabine 1, 250 mg/m 2 Day 1 & 8 every 3 weeks CR/PR/SD N=231 R N=115 Observation till PD Stratification 1. Visceral diseases 2. Prior adjuvant taxane 3. Response(CR/PR vs. SD) 4. HR(+) vs. HR(-) N=116 Primary Endpoint; PFS from Randomization Secondary Endpoints; OS, Toxicities, Qo. L, and Response Duration Im et al, #1003 75% HR positive, ~20% received hormone therapy in the metastatic setting

• Subgroup analysis – Primary benefit in HR negative (n=59), and in premenopausal women • Toxicity – More toxicity in the maintenance arm > cycle 6 – QOL similar between the two arms • Interpretation complicated by lack of use of hormone therapy in the control arm – This should not change our general management of patients in this setting – For hormone receptor negative disease, better to continue at least a low dose of chemotherapy

Other Interesting Data • Targeting the androgen receptor in ER/PR negative disease (Gucalp et al, #1006, TBCRC) – 12% of screened patients were AR+ – Treated with bicalutamide 150 mg/day • 26 patients – Median age 66 – 15% visceral metastases – Median 1 prior chemotherapy regimen for MBC • 24% clinical benefit rate (5/24) – Disease in LN, breast and bone, one GI • A larger trial is planned

Summary • New HER 2 directed therapy provides increased efficacy without significant toxicity – Adjuvant and neoadjuvant trials are ongoing or planned • Zoledronate may improve outcome in subsets of women with early stage breast cancer and high bone turnover – This is not practice changing – The current recommendation is to use potent bisphonates to treat osteoporosis, or osteopenia in high risk women • Radiation appears to add little benefit to patients with low risk DCIS treated with BCT

Summary (2) • Early stage disease – Gemcitabine does not add to DD AC/P but increases toxicity – Approach to clinical trials needs to change • Its all about biology and understanding the drivers of disease • Metastatic disease – Paclitaxel is the preferred microtubule targeting agent in 1 st line MBC – Maintenance chemotherapy may provide benefit, but this is likely to be in HR negative or resistant disease – More studies targeting the androgen receptor are warranted

Phase 1 b Study: all BC PLX 3397 oral daily dosing Eribulin: 1. 4 mg/m 2 iv, day 1 and 8 Each cycle of treatment lasts 21 days First Cohort = 600 mg/day 3 -6 patients KOMEN Promise Grant: Can inhibition of macrophages Second Cohort = 800 mg/day reverse chemotherapy 3 -6 patients resistance? Phase II Primary Endpoint: PFS at 12 weeks Third Cohort = 1000 mg/day 3 -6 patients Phase II Study: Metastatic TNBC Lead in period of 5 -7 d with PLX 3397 at MTD oral daily dosing (day -7/5 to day 0) Biopsy for immune profiling Starting Day 1 Add Eribulin 1. 4 mg/m 2 iv day 1 and 8 Each cycle of treatment lasts 21 days PI: Hope S. Rugo, UCSF, Lisa Coussens, OHSU, Shelley Hwang, Duke.