HOW FAR WE VE COME: TREATING HER2- POSITIVE BREAST CANCER WITH TARGETED THERAPIES

HOW FAR WE VE COME: TREATING HER2- POSITIVE BREAST CANCER WITH TARGETED THERAPIES Javier Cortes, Vall d Hebron Institute of Oncology (VHIO), Medica Scientia Innovation Research (MedSIR) Barcelona, Spain Ramon y Cajal University Hospital, Madrid, Spain

Targeted therapy trials for HER2-positive Breast Cancer EBC MBC Neo Surgery Adjuvant Relapse First line Progression Second + lines NOAH MDACC GeparQuattro GeparQuinto NeoALLTO Neosphere TRYPHAENA HannaH NeoALLTO HERA NSABP-B31 NCCTG N9831 BCIRG 006 APHINITY ALLTO HO648g M77001 US Oncology TanDEM BCIRG 007 CHAT HERNATA RHEA AVEREL CLEOPATRA MARIANNE GBG-26 EGF104900 Numerous Phase II Studies BO17929 PHEREXA TDM4374g EMILIA Th3resa EBC, early breast cancer; MBC, metastatic breast cancer; Neo, neoadjuvant

Trastuzumab with chemotherapy in HER2 positive MBC Study Design Overall survival Eligible patients (n=469) No prior anthracyclines l Metastatic breast cancer l HER2 overexpression 2/3+ l No prior CT for MBC l Measurable disease l KPS ³60% Prior anthracyclines Probability of survival 1.0 0.8 0.6 0.4 0.2 20.3 ms H + CT CT HR=0.80 p=0.046 25.1 ms Trastuzumab + AC AC (n=143) (n=138) Trastuzumab + paclitaxel Paclitaxel (n=92) (n=96) 0 0 5 15 25 35 45 Time (months) Slamon DJ et al. NEJM 2001

Efficacy: Capecitabine ± Lapatinib Lapatinib plus Capecitabine Hazard capecitabine alone ratio End point (N = 163) (N = 161) (95% CI) Median time to progression - mo 8.4 4.4 0.49 (0.34-0.71) P value <0.001* Median progression-free survival - mo 8.4 4.1 0.47 (0.33-0.67) <0.001 Overall response % (95% CI) 22 (16-29) 14 (9-21) 0.09 Clinical benefit - no (%) 44 (27) 29 (18) Death - no (%) 36 (22) 35 (22) *End points are based on evaluation by the independent review committee under blinded conditions The p value was calculated with the log-rank test The p value was calculated with Fisher s exact test Geyer CE, et al. N Engl J Med 2006

Continuation of trastuzumab prolongs median TTP in the GBG-26 study Probability 1.0 0.8 0.6 0.4 0.2 + + + + + + + + + + + + + + + + Trastuzumab + capecitabine (n=78) Capecitabine (n=78) HR=0.69 (two-sided p=0.034; one-sided p=0.015) 0.0 0 5.6* 8.2* 10 20 30 Time from first progression (months) + 40 + + 74 77 40 55 15 29 8 12 5 4 3 3 2 1 1 1 1 1 *Median TTP in months von Minckwitz, et al. JCO 2009

Lapatinib or trastuzumab-based therapy as first-line? MA 31 Trial Gelmon K, et al. JCO 2015

CEREBEL Trial Key eligibility: HER2+ MBC* Prior anthracyclines or taxanes Any line therapy No CNS metastases* Evaluable systemic dx Stratification: Prior trastuzumab -yes vs no Prior MBC tx -0 vs >1 *No CNS mets at baseline confirmed by independently reviewed MRI scan R A N D O M I S E Phase III Planned N=650 Lapatinib 1250 mg/day + Capecitabine 2000 mg/m 2 Trastuzumab 6 mg/kg q21 days + Capecitabine 2500 mg/m 2 Study was a Specific Obligation measure required by CHMP in 2008 Pivot X, et al. JCO 2015

CEREBEL Trial Secondary endpoints: Clinical relevance Is lapatinib as good as trastuzumab? ITT Population Trastuzumab-pretreated Alive without progression (%) HR: 1.13 (0.85-1.50) Time from randomisation (months) Trastuzumab-naive Lap + Cap (N=271) Tras + Cap (N=269) Median PFS, months 6.6 8.0 Hazard ra8o (95% CI) 1.30 (1.04, 1.64) HR: 1.70 (1.15-2.50) Stra8fied log- rank p- value 0.021 Pivot X, et al. JCO 2015

Dual HER2 blockade by Lapatinib and Trastuzumab

Lapatinib-induced HER2 stabilization Vazquez-Martin et al, J Cell Physiol 2010

Scaltriti, Verma, et al, Oncogene 2009

EGF104900 study: lapatinib ± trastuzumab in progressing mbc HER2-positive mbc (FISH+) (n=296) Lapatinib 1,000mg daily + trastuzumab 2mg/kg weekly (n=148) Lapatinib 1,500mg daily (n=148) Crossover if PD Blackwell, et al. JCO 2012

EGF104900: Significant Overall Survival (OS) Benefit With Trastuzumab + Lapatinib Following Disease Progression 100 80 80% L N = 148 L+T N = 148 Died, N (%) 113 (78) 105 (72) Survival, % 60 40 70% 6 Month OS 56% 41% Median, months 9.5 14.0 Hazard ratio (95% CI).74 (.57-.97) Log-rank P-value.026 20 12 Month OS 0 0 5 10 15 20 25 30 35 Time from Randomization, months Blackwell, et al. JCO 2012

Trial EGF117165: Primary Objective To evaluate changes in the expression of biomarkers associated with HER family, immunomodulation, apoptosis, and ABC transporters between pretreatment and disease progression biopsy within each treatment arm

HER2 is activated in two different ways Overexpression: o o Arising from gene amplification which results in constitutive activation of HER2 (ligand independent) Antagonized by trastuzumab Co-receptor: o For other HER family members especially HER3 (ligand dependent) o Antagonized by pertuzumab

HER2 is activated in two different ways Overexpression: o o Arising from gene amplification which results in constitutive activation of HER2 (ligand independent) Antagonized by trastuzumab Co-receptor: o For other HER family members especially HER3 (ligand dependent) o Antagonized by pertuzumab

Pertuzumab binds to a different epitopes on HER2 than trastuzumab Pertuzumab Dimerization domain of HER2 Ø Pertuzumab prevents ligand-induced HER2-HER3 dimerization Ø Pertuzumab flags cells for destruction by the immune system via ADCC

In contrast to trastuzumab, pertuzumab inhibits HRG-mediated HER2 signaling ErbB2/ErbB3 co-ip AKT Agus DBm et al. Cancer Cell 2002 18

Pertuzumab mediates ADCC (similar to Trastuzumab)

Hypotheses HER2 positive breast cancers are driven by both ligand-independent and ligand-dependent HER2-HER3 complexes Dual blockade of these complexes with trastuzumab and pertuzumab should result in greater clinical activity than treatment with trastuzumab alone

Pertuzumab + Trastuzumab combination therapy in a breast cancer xenograft model

CLEOPATRA Study OS (%) Patients with HER2-positive MBC centrally confirmed (N = 808) 10 900 80 70 60 50 40 30 20 10 0 Ptz + T + D Pla + T + D 1:1 HR 0.68 95% CI = 0.56, 0.84 p = 0.0002 n=406 n=402 40.8 months Placebo + trastuzumab Docetaxel* 6 cycles recommended Pertuzumab + trastuzumab Docetaxel* 6 cycles recommended 0 10 20 30 40 50 60 70 Time (months) Study Design Baselga J, et al. NEJM 2012 Overall survival Δ 15.7 56.5 months months Swain S, et al. NEJM 2015

CLEOPATRA Study: CNS metastases Time to CNS metastases as first site of DP OS in pts who develop CNS metastases as first site of DP Swain S, et al. Ann Oncol 2015

Trastuzumab Emtansine (T-DM1): Mechanism of Action HER2 Antibody: Trastuzumab Antibody drug conjugate: T-DM1 Emtansine release Stable linker: MCC Cytotoxic: DM1 Emtansine P P Inhibition of microtubule polymerization Lysosome P Internalization Nucleus Adapted from LoRusso PM, et al. Clin Cancer Res 2011

EMILIA Trial HER2-positive LABC or MBC (N=980) 1:1 T-DM1 Study Design Prior taxane and trastuzumab Capecitabine + Lapatinib 1.0 Proportion surviving 0.8 0.6 0.4 0.2 0.0 HR=0.68 p=0.0006 25.1 ms 30.9 ms 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 Time (months) Overall survival Verma S, et al. NEJM 2012

EMILIA OS Analysis for Patients with CNS mets at baseline Krop I, et al. Ann Oncol 2015

TH3RESA Trial HER2-positive (central) advanced BC a 2 prior HER2- directed therapies for advanced BC Prior treatment with trastuzumab, lapatinib, and a taxane 2 1 T-DM1 3.6 mg/kg q3w IV (n=400) Treatment of physician s choice (TPC) b (n=200) PD PD T-DM1 c (optional crossover) Study Design 1.0 Proportion surviving 0.8 0.6 0.4 0.2 HR=0.55 p=0.0034 Efficacy stopping boundary HR<0.363 or P<0.0000013 14.9 ms NE Overall survival 0.0 0 2 4 6 8 10 12 14 Time (months) 16 Krop I, et al. Lancet Oncol 2014

MARIANNE Study Design HER2-positive (central) LABC a or MBC No prior chemotherapy for LABC/MBC >6 months from prior neo-/adjuvant vinca alkaloid or taxane chemotherapy N = 1095 Trastuzumab + docetaxel (8 mg/kg LD then 6 mg/kg + 100 or 75 mg/m 2 q3w) OR Trastuzumab + paclitaxel (4 mg/kg LD then 2 mg/kg + 80 mg/m 2 qw) T-DM1 + placebo b (3.6 mg/kg + 840 mg LD then 420 mg q3w) T-DM1 + pertuzumab (3.6 mg/kg + 840 mg LD then 420 mg q3w) Stratification factors: World region, Prior neo-/adjuvant therapy (if Yes: prior trastuzumab/lapatinib), Visceral disease Primary end point: PFS by independent review facility (IRF), non-inferiority and superiority assessed Key secondary end points: OS, PFS by investigator, ORR, Safety, Patient-reported outcomes LD, Loading dose. a Locally progressive or recurrent and not amenable to resection with curative intent; b Pertuzumab placebo. Ellis P, et al. ASCO 2015

Statistical Considerations Statistical analyses were conducted independently for T-DM1 vs HT and for T-DM1+P vs HT Hierarchical statistical testing was performed in pre-specified sequential order Two-sided alpha = 2.5% Two-sided alpha = 2.5% T-DM1 vs HT 1. PFS non-inferiority 2. PFS superiority 3. OS superiority 4. Other secondary end points T-DM1+P vs HT 1. PFS non-inferiority 2. PFS superiority T-DM1+P vs T-DM1 3. PFS superiority 4. OS superiority 5. Other secondary end points % power for target HR=0.75 (T- DM1/T- DM1+P vs. HT) and target HR=0.73 (T- DM1+P vs. T- DM1 Primary end point of PFS by IRF: Trial was powered at 80% for both non-inferiority and superiority analyses of PFS Non-inferiority: Established if the upper limit of the 97.5% CI for the HR is below 1.1765 (non-inferiority margin) Superiority: Target HR = 0.75 (T-DM1/T-DM1+P vs HT) and target HR = 0.73 (T-DM1+P vs T-DM1). Established if P 0.025 Ellis P, et al. ASCO 2015

Progression-Free Survival by IRF HT T-DM1 T-DM1+P 100 Median PFS (mo) 13.7 14.1 15.2 Events (no) 231 236 217 Progression-Free Survival (%) 80 60 40 20 HT T-DM1 T-DM1+P Stratified HR (97.5% CI) vs HT Stratified HR (97.5% CI) vs T- DM1 0.91 (0.73 1.13) P=0.31 0.87 (0.69 1.08) P=0.14 0.91 (0.73 1.13) No. at Risk 0 0 6 12 18 24 30 36 42 48 54 Time (mo.) HT T-DM1 T-DM1+P 365 367 363 265 257 261 163 176 177 107 133 135 75 104 109 50 67 75 21 28 25 5 3 5 1 Non-inferiority: Established if the upper limit of the 97.5% CI for the HR is below 1.1765 (non-inferiority margin). Ellis P, et al. ASCO 2015

Objective Response and Duration of Response by IRF Patients, % 100 90 80 70 60 50 40 30 Objective Response Rate 67.9% 64.2% 59.7% Patients without progression, % 100 80 60 40 20 Duration of Response Median, mo. (95% CI) HT T-DM1 T-DM1+P HT 12.5 (10.5 16.6) T-DM1 20.7 (14.8 25.0) T- DM1+P 21.2 (15.8 29.3) 20 0 10 0 195/287 181/303 192/299 HT T-DM1 T-DM1+P No. at Risk HT T-DM1 T-DM1+P 0 6 12 18 24 30 36 42 48 54 195 181 192 150 150 162 81 110 118 55 85 92 Time (mo.) 37 65 75 24 35 44 5 14 11 1 1 Error bars depict 95% confidence intervals. Non-inferiority: Established if the upper limit of the 97.5% CI for the HR is below 1.1765 (non-inferiority margin). Ellis P, et al. ASCO 2015

Hyperactivation of the PI3K pathway regulates trastuzumab and lapatinib sensitivity in HER2 breast cancer cells HER2 Untreated Trastuzumab Lapatinib control PI3K PTEN PTEN TORC2 Ptenkd PIP3 Akt Tuberin Tuberin Tuberin PDK1 p110α E545K Rheb H1047R TORC1 Eichhorn et al. Cancer Res 2008

PI3K signaling pathway alteration results in reduced response to trastuzumab n = 55 patients (T +/- CT) PTEN low (IRS 3) 22% PI3K mutation 25% Berns, et al. Cancer Cell 2007

BOLERO-3: Study Design N = 572* Locally advanced or metastatic HER2 + breast cancer Prior taxane required TRAS resistance Adjuvant: progression on or within 12 months of TRAS Metastatic: progression within 4 weeks of TRAS Measurable disease only Randomize 1:1 Stratification by prior lapatinib use (yes/no) Everolimus (5 mg PO daily) + Vinorelbine (25 mg/m 2 weekly) + TRAS (2 mg/kg week ) (n = 284) Placebo (PO daily) + Vinorelbine (25 mg/m 2 weekly) + TRAS (2 mg/kg weekly*) (n = 285) Therapy until PD or intolerable toxicity Endpoints: Primary: PFS Secondary: OS, ORR, time to deterioration of ECOG PS, safety, DoR, CBR, and QoL * Actual enrollment was 569. Following a 4-mg/kg loading dose on day 1, cycle 1 (1 cycle = every 21 days). Abbreviations: CBR, clinical benefit rate; DoR, Duration of response; ECOG, Eastern Cooperative Oncology Group; HER2, human epidermal growth factor receptor 2; ORR, overall response rate; OS overall survival; PD, progressive disease; PFS, progressive-free survival; PO, oral; PS, performance status; QoL, quality of life. http://www.clinicaltrials.gov/ct2/show/nct01007942?term=bolero3&rank=1 Presented by: Ruth M. O Regan, MD Non-inferiority: Established if the upper limit of the 97.5% CI for the HR is below 1.1765 (non-inferiority margin). Andre F, et al. Lancet Oncol 2014

BOLERO-3: PFS Subgroup Analyses Subgroup N All 569 Age < 65 years 472 65 97 Region Europe 223 North America 123 Asia 166 LaZn America 36 Other 21 Prior lapaznib* Yes 161 No 408 Prior adj/neo trastuzumab** Yes 251 No 318 Baseline ECOG PS 0 382 1 or 2 186 Hormonal status ER /PgR 250 ER + /PgR + 317 Visceral involvement Yes 439 No 130 Hazard RaZo [95% CI] 0.78 [0.65-0.95] 0.77 [0.62-0.95] 0.93 [0.56-1.57] 0.72 [0.53-0.99] 0.86 [0.55-1.32] 0.83 [0.59-1.18] 0.61 [0.27-1.38] 1.28 [0.48-3.45] 0.79 [0.56-1.11] 0.78 [0.62-0.99] 0.65 [0.48-0.87] 0.92 [0.71-1.18] 0.79 [0.63-1.00] 0.75 [0.53-1.05] 0.65 [0.48-0.87] 0.93 [0.72-1.20] 0.89 [0.72-1.10] 0.48 [0.30-0.76] Abbreviations: CI, confidence interval; ECOG PS, Eastern Cooperative Oncology Group performance status. Favors EVE Favors PBO Presented by: Ruth M. O Regan, MD Non-inferiority: Established if the upper limit of the 97.5% CI for the HR is below 1.1765 (non-inferiority margin). Andre F, et al. Lancet Oncol 2014

BOLERO-1/TRIO 019: Study Design N = 719 Locally advanced or metastatic HER2+ breast cancer No prior therapy for advanced or metastatic disease (except endocrine therapy) Prior (neo)adjuvant TRAS and/or chemotherapy allowed 1 Measurable disease or presence of bone lesions (lytic or mixed) Endpoints Randomized 2:1 Stratification factors: Prior neo/adjuvant TRAS Visceral metastases Everolimus (10 mg PO daily) + Paclitaxel 2 + Trastuzumab 3 Placebo + Paclitaxel 2 + Trastuzumab 3 Therapy until disease progression or intolerable toxicity 4 Primary: PFS (investigator-assessed) Overall population and HR - subpopulation Secondary: OS, ORR, CBR, Time to response, Safety, Duration of response 1 Discontinued > 12 mo before randomization; 2 Paclitaxel: 80 mg/m 2 weekly; 3 Trastuzumab: 4 mg/kg loading dose on day 1 at cycle 1 followed by 2 mg/kg weekly doses 4 Patients could discontinue any study treatment due to AEs; other study treatments continued until disease progression or intolerable toxicity Non-inferiority: Established if the upper limit of the 97.5% CI for the HR is below 1.1765 (non-inferiority margin). ABC, advanced breast cancer; CBR, clnical benefit rate; ORR, overall response rate; OS, overall survival; PFS, progression free survival. Hurvitz S, et al. Lancet Oncol 2015 36

BOLERO-1/TRIO 019: PFS Full Population (Investigator-assessment) 100% Hazard Ratio = 0.89; 95% CI [0.73, 1.08] Log rank p value = 0.1166 80% Median PFS Everolimus: 14.95 months; 95% CI [14.55, 17.91] (n/n = 271/480) Placebo: 14.49 months; 95% CI [12.29, 17.08] (n/n = 154/239) Probability (%) 60% 40% Censoring times 20% 0% No. of patients still at risk Everolimus Placebo 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 Time (months) 480 416 365 324 289 260 217 178 151 130 122 107 94 80 72 63 58 48 42 35 26 21 17 13 10 5 3 3 0 239 221 199 166 144 123 106 91 80 69 53 47 43 38 36 36 31 24 17 15 12 9 7 6 4 3 1 1 0 - One-sided p-value is obtained from the log-rank test stratified by prior use of trastuzumab (Y/N) and Visceral metastasis (Y/N) from IWRS. ABC, advanced breast cancer; CBR, clnical benefit rate; ORR, overall response rate; OS, overall survival; PFS, progression free survival. Non-inferiority: Established if the upper limit of the 97.5% CI for the HR is below 1.1765 (non-inferiority margin). 37 Hurvitz S, et al. Lancet Oncol 2015 37

BOLERO-1/TRIO 019: PFS HR Subpopulation (Investigator Assessment) 100% Hazard Ratio = 0.66; 95% CI [0.48, 0.91] Log rank p value = 0.0049 Probability (%) 80% 60% 40% Median PFS Everolimus: 20.27 months; 95% CI [14.95,24.08] (n/n = 97/208) Placebo: 13.08 months; 95% CI [10.05,16.56] (n/n = 66/103) Censoring times 20% 0% 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 Time (months) No. of patients still at risk Everolimus 208 183 166 151 138 125 100 84 73 64 62 55 49 40 35 32 30 24 21 19 15 11 10 7 5 2 1 1 0 Placebo 103 96 83 68 58 49 43 34 32 28 24 21 20 19 19 19 17 13 7 6 5 4 2 1 1 0 0 0 0 - One-sided p-value is obtained from the log-rank test stratified by prior use of trastuzumab (Y/N) and Visceral metastasis (Y/N) from IWRS. ABC, advanced breast cancer; CBR, clnical benefit rate; ORR, overall response rate; OS, overall survival; PFS, progression free survival. Non-inferiority: Established if the upper limit of the 97.5% CI for the HR is below 1.1765 (non-inferiority margin). 38 Hurvitz S, et al. Lancet Oncol 2015 38

Margetuximab To evaluate the safety of margetuximab using two dosing regimens V H Fab C H1 VL C L C H2 Fc C H3 Tumor Cell A D C C A D CC A D C C

FIH phase 1 Study with Margetuximab All evaluable pts Evaluable MBC 39/60 prior antiher2 therapy 22/23 prior antiher2 therapy Burris HA, et al. ASCO 2015

New opportunities with Margetuximab Scaltriti M, et al. Oncogene 2009

Margetuximab in HER2 1+/2+ (FISH neg)? HER2 1+ Unpublished data

Conclusion MBC1 MBC1 Pertuzumab + Trastuzumab-based Previous Rx* De novo Rx Therapy <12 mo; >12 mo HR- Pos HR- Neg MBC1 Previous Rx* TDM1 <6 mo MBC2 Short DFI HR- Pos MBC2 TDM1 Long DFI HR- Neg TH3RESA MBC3 Trastuzumab and/or Lapatinib-based MBC4

CONCLUSIONS 100 HER2 positive, Herceptin (n = 191) HER2 negative (n = 1782) HER2 positive, no Herceptin (n = 118) HER2 positive, pertuzumab, TDM1? Probability of Survival, % 80 60 40 20 0 0 12 24 36 48 60 Time From Diagnosis, months

HER2+ Early Breast Cancer

Adjuvant Trastuzumab Trials: >13,000 Patients Treated HERA (ex-usa) BCIRG 006 (global) IHC/FISH (n=5,090) Observation 1 year 2 years FISH (n=3,222) 1 year 1 year NCCTG N9831 (USA) NSABP B-31 (USA) IHC/FISH (n=3,505) 1 year IHC / FISH (n=2,030) 1 year 1 year Standard chemotherapy Doxorubicin + cyclophosphamide Docetaxel Docetaxel + carboplatin Trastuzumab Paclitaxel FISH = fluorescence in situ hybridization Romond et al. N Engl J Med 2005; Piccart-Gebhart et al. N Engl J Med 2005; Slamon et al. SABCS 2006

Trastuzumab Provides Consistent DFS Benefits DFS Benefit Median Follow-up (Years) HERA CTg H 1 year 2 B-31/N9831 ACg PH 3 BCIRG 006 ACg DH 3 BCIRG 006 DCarboH 3 0 Favours Favours No Trastuzumab Trastuzumab 2 Hazard Ratio Size of square represents sample size; horizontal bars indicate 95% confidence intervals CT = chemotherapy; H = trastuzumab; AC = doxorubicin, cyclophosphamide; P = paclitaxel; D = docetaxel; Carbo = carboplatin Joensuu et al. N Engl J Med 2006; Perez et al. ASCO 2007, Abstract #512; Slamon et al. SABCS 2006; Smith et al. Lancet 2007; Spielmann et al 2007

Trastuzumab Consistently Reduces the Risk of Death by One Third Overall Survival Benefit Median Follow-up (Years) HERA CTg H 1 year 2 B-31/N9831 ACg PH 3 BCIRG 006 ACg DH 3 BCIRG 006 DCarboH 3 0 Favours Favours No Trastuzumab 1 Trastuzumab Hazard Ratio 2 Size of square represents sample size; horizontal bars indicate 95% confidence intervals Gianni et. al Breast 2009: 4 yr OS HR = 0.85 (NS) CT = chemotherapy; H = trastuzumab; AC = doxorubicin, cyclophosphamide; P = paclitaxel; D = docetaxel; Carbo = carboplatin Perez et al. ASCO 2007, Abstract #512; Slamon et al. SABCS 2006; Smith et al. Lancet 2007

Summary of OS ITT Analyses for 1 Year Trastuzumab vs. Observation Across Analysis Time Points Median follow-up (% follow-up time after selective crossover) OS benefit No. of deaths 1 year trastuzumab vs observation 2005 (0%) 1 yr MFU 0.76 29 vs 37 P=0.26 2006 (4.1%) 2 yrs MFU 0.66 59 vs 90 P=0.0115 2008 4 yrs MFU (30.9%) 2012 8 yrs MFU (45.5%) 0.76 0.85 182 vs 213 P=0.1087 278 vs 350 P=0.0005 0 Favours 1 year trastuzumab 1 Favours observation 2 HR (95% CI) Extended from Gianni et al. Lancet Oncol 2011;12:236-44 Goldhirsch et al. (presented by Piccart-Gebhart), SABCS 2012, Abstract #S5-2, oral presentation

B-31/N9831 Joint Analysis Demonstrates Disease- Free Survival Benefit after 8.4-y Median Follow-up 100 80 AC Ú P+H 81.4% 76.8% 73.7% % Event-Free 60 40 AC Ú P 69.5% 64.9% 62.2% 11.5% 20 N Events AC P 2018 680 AC P+H 2028 473 HR=0.60 (95% CI: 0.53-0.68) P<0.0001 0 0 2 4 6 8 10 Years from Randomization No. at risk 2028 1959 1848 1747 1675 1611 1514 1293 910 619 350 2018 1887 1689 1529 1423 1329 1232 1027 705 449 255

B-31/N9831 Joint Analysis Demonstrates Overall Survival Benefit after 8.4-y Median Follow-up 100 80 AC Ú P 93.2% 90.3% 89.8% 84.3% AC Ú P+H 87.0% 84.0% 79.4% 75.2% 8.8% % Survival 60 40 20 N Events AC P 2018 418 AC P+H 2028 286 =2.9% =5.5% =7.6% =8.8% HR=0.63 (95% CI 0.54-0.73) P<0.0001 No. at risk 0 0 2 4 6 8 10 Years from Randomization 2028 1995 1959 1897 1843 1785 1709 1506 1085 735 439 2018 1962 1883 1806 1730 1640 1534 1336 944 604 353 Romond et al. SABCS 2012, Abstract #S5-5

Various trials investigated the optimal duration of trastuzumab in EBC Reported Ongoing Trastuzumab for <1 year vs. trastuzumab for 1 year Trastuzumab for 2 years vs. trastuzumab for 1 year 6 months 9 weeks 2 years 1 year (standard of care) PHARE 6 months vs. 1 year 1 SOLD 4 9 weeks vs. 1 year HERA 2 years vs. 1 year 7 HERA 1 year vs. observation 8 10 HORG 6 months vs. 1 year 2 SHORT-HER 9 weeks vs. 1 year 5 NCCTG N9831 11 NSABP B-31 12 PERSEPHONE 6 months vs. 1 year 3 FinHer 9 weeks vs. chemo 6 BCIRG 006 12 Trastuzumab for 1 year remains the standard of care in EBC, as recommended by international guidelines 13 15 1. Pivot X, et al. Lancet Oncol 2013; 14:741 748; 2. http://clinicaltrials.gov/ct2/show/nct00615602; 3. Earl HM, et al. ASCO 2013 (Abstract TPS667); 4. http://clinicaltrials.gov/ct2/show/nct00593697; 5. http://clinicaltrials.gov/ct2/show/nct00629278; 6. Joensuu H, et al. J Clin Oncol 2009; 34:5685 5692; 7. Goldhirsch A, et al. Lancet 2013; 382:1021 1028; 8. Piccart-Gebhart MJ, et al. N Engl J Med 2005; 353:1659 1672; 9. Smith I, et al. Lancet 2007; 369:29-36; 10. Gianni L, et al. Lancet Oncol 2011; 12:236-244; 11. Perez EA, et al. J Clin Oncol 2011; 29:3366 3373; 12. Slamon D, et al. N Engl J Med 2011; 365:1273-1283; 13. NCCN Clinical Practice Guidelines in Oncology; Breast Cancer v3.2013; 14. Senkus E, et al. Ann Oncol 2013; epub ahead of print; 15. Goldhirsch A, et al. Ann Oncol 2013; epub ahead of print.

PHARE Trial: Non-inferiority of 6 months vs. 1 year of trastuzumab DFS trastuzumab 6 ms trastuzumab up to 12 ms 100 80 Clinical exam LVEF 0 3 stop trastuzumab 6 9 12 15 18 21 24 30 ms DFS (%) 60 40 20 0 Trastuzumab 6 months Trastuzumab 1 year 0 12 24 36 48 60 Months from randomisation Mammo R: Randomization after informed consent Up to 60 ms Patients Events HR (6 months vs. 1 year) 95% CI p value 6 months 1690 93 1.46 (1.06, 2.01) 0.03 1 year 1690 66 HR (95% CI): 1.46 (1.06, 2.01) (above the prespecified non-inferiority CI of 1.15) Pivot X, et al. Lancet Oncol 2013

Targeted therapy trials for HER2-positive Breast Cancer EBC MBC Neo Surgery Adjuvant Relapse First line Progression Second + lines NOAH MDACC GeparQuattro GeparQuinto NeoALLTO Neosphere TRYPHAENA HannaH NeoALLTO HERA NSABP-B31 NCCTG N9831 BCIRG 006 ALLTO Aphinity HO648g M77001 US Oncology TanDEM BCIRG 007 CHAT HERNATA RHEA AVEREL CLEOPATRA MARIANNE GBG-26 EGF104900 Numerous Phase II Studies BO17929 PHEREXA TDM4374g EMILIA Th3resa EBC, early breast cancer; MBC, metastatic breast cancer; Neo, neoadjuvant

NeoALTTO Study Design Eligibility Invasive operable HER2+ breast cancer Tumor >2 cm (inflammatory breast cancer excluded) LVEF 50% Stratification Tumor 5 cm vs Tumor >5 cm ER or PgR+ vs ER and PgR- N0-1 vs N 2 Conservative surgery or not R A N D O M I Z E Lapatinib Paclitaxel Trastuzumab Paclitaxel Lapatinib Trastuzumab Paclitaxel 6 weeks +12 weeks S U R G E R Y F E C X 3 Lapatinib Trastuzumab Lapatinib Trastuzumab 52 weeks of anti-her2 therapy LVEF, left ventricular ejection fraction Baselga J, et al. Lancet. 2012;379(9816):633-640.

NeoALTTO: pcr and tpcr P =.0001 P =.0007 P =.34 51.3 P =.13 46.8 29.5 27.6 L N = 154 T N =149 pcr L+T N = 152 L N = 150* T N = 145* total pcr L+T N = 145* *Excludes 15 patients with nonevaluable nodal status L: lapatinib; T: trastuzumab Baselga J, et al. Lancet. 2012;379(9816):633-640.

NeoALTTO: DFS ER+ ER- De Azambuja E, et al. Lancet Oncol 2014

First results from the phase III ALTTO trial (BIG 02-06; NCCTG 063D) comparing one year of anti- HER2 therapy with lapatinib alone (L), trastuzumab alone (T), their sequence (T" L) or their combination (L + T) in the adjuvant treatment of HER2-positive early breast cancer (EBC) Martine Piccart-Gebhart, Andrew P. Holmes, José Baselga, Evandro de Azambuja, Amylou Dueck, Giuseppe Viale, Jo Anne Zujewski, Aron Goldhirsch, Sergio Santillana, Kathleen Pritchard, Antonio C. Wolff, Christian Jackisch, Istvan Lang, Michael Untch, Ian Smith, Frances Boyle, Binghe Xu, Henry Gomez, Richard D. Gelber and Edith A. Perez On behalf of the ALTTO Study Team

DESIGN 1: SEQUENTIAL ANTI- HER2 THERAPY AFTER ALL CHEMOTHERAPY (N= 4,613) 3-weekly Trastuzumab Lapatinib* All (neo)adjuvant chemo prior to anti-her2 therapy Weekly Trastuzumab wash out Lapatinib 12 weeks 6 wks 34 weeks 12 weeks Lapatinib + 3-weekly Trastuzumab 6 wks 34 weeks 52 weeks All patients: radiotherapy, if indicated (concomitant with targeted therapy). Hormone receptor-positive patients: endocrine therapy for at least 5 years. *The L alone arm was closed on 18 Aug 2011 following IDMC recommendation Tras alone: 8 mg/kg " 6 mg/kg iv, q21 days Lap alone: 1500 mg po qd Tras " Lap: T 4 mg/kg " 2 mg/kg iv q7 days; L 1500 mg po qd Tras + Lap: T 8 mg/kg " 6 mg/kg iv, q21 days; L 1000 mg po qd

DESIGN 2: CONCURRENT ANTI- HER2 THERAPY AFTER ANTHRACYCLINE- BASED CHEMOTHERAPY (N= 3,337) w-p or 3-w D 3-weekly Trastuzumab w-p or 3-w D Lapatinib* Anthracyclinebased chemo first Weekly w-p or 3-w D Trastuzumab wash out Lapatinib 12 weeks 6 wks 34 weeks w-p 12 or weeks 3-w D Lapatinib + 3-weekly Trastuzumab 6 wks 34 weeks 52 weeks w-p: weekly paclitaxel (80 mg/m 2 ); 3-w D: q3 weeks docetaxel (75-100 mg/m 2 ) All patients: radiotherapy, if indicated (concomitant with targeted therapy). Hormone receptor-positive patients: endocrine therapy for at least 5 years. *The L alone arm was closed on 18 Aug 2011 following IDMC recommendation Tras alone: 4 mg/kg " 2 mg/kg iv, q7 days " 6 mg/kg iv, q21 days Lap alone: 750 mg po qd " 1500 mg qd Tras " Lap: T 4 mg/kg " 2 mg/kg iv q7 days; L 1500 mg po qd Tras + Lap: T 4 mg/kg " 2 mg/kg iv, q7 days " 6 mg/kg iv, q21 days; L 750 mg po qd " 1000 mg qd

DESIGN 2B: CONCURRENT ANTI- HER2 THERAPY WITH A NON- ANTHRACYCLINE CHEMOTHERAPY (N= 431) 3-w D + carbo 3-weekly Trastuzumab 3-w D + carbo Lapatinib* Non-anthracyclinebased chemo with anti-her2 therapy Weekly 3-w D + carbo Trastuzumab wash out Lapatinib 18 weeks 6 wks 28 weeks 3-w 18 weeks D + carbo Lapatinib + 3-weekly Trastuzumab 6 wks 28 weeks 52 weeks 3-w D: q3 weeks docetaxel (75 mg/m 2 ); carbo: carboplatin (AUC 6) All patients: radiotherapy, if indicated (concomitant with targeted therapy). Hormone receptor-positive patients: endocrine therapy for at least 5 years. *The L alone arm was closed on 18 Aug 2011 following IDMC recommendation Tras alone: 4 mg/kg " 2 mg/kg iv, q7 days " 6 mg/kg iv, q21 days Lap alone: 750 mg po qd " 1500 mg qd Tras " Lap: T 4 mg/kg " 2 mg/kg iv q7 days; L 1500 mg po qd Tras + Lap: T 4 mg/kg " 2 mg/kg iv, q7 days " 6 mg/kg iv, q21 days; L 750 mg po qd " 1000 mg qd

STATISTICAL CONSIDERATIONS Target enrolment of at least 8,000 pa8ents. Timing of primary analysis: 850 DFS events are required for the comparison of L + T vs. T (80% power using a 2- sided alpha error = 0.0167); OR 4.5 years median follow- up, whichever comes first. Event rate lower than anzcipated: the current analysis is based on 555 DFS events at a median follow- up of 4.49 years (range: 1 day to 6.40 years). First interim efficacy analysis (IDMC on 18 th August 2011): comparison of lapa8nib alone versus trastuzumab crossed the fu8lity boundary; pa8ents switched to trastuzumab. Lapa8nib alone arm is not reported here.

CURRENT ANALYSIS PLAN Sta8s8cal procedures for the two remaining pairwise comparisons are: Comparison L + T vs. T AssumpZons Test superiority in ITT popula8on at alpha = 0.025 T L vs. T Test non- inferiority in per protocol popula8on (PPP) at alpha = 0.025 Primary (inferen,al) analysis: 97.5% CI; Descrip,ve analyses: 95% CI 63

DISEASE- FREE SURVIVAL (DFS) ANALYSIS MFU = 4.5 yrs * ** * 97.5% CI **p- value 0.025 required for stazszcal significance

OVERALL SURVIVAL (OS) ANALYSIS MFU = 4.5 yrs * * 95% CI

Summary of OS ITT Analyses for 1 Year Trastuzumab vs. Observation Across Analysis Time Points Median follow-up (% follow-up time after selective crossover) OS benefit No. of deaths 1 year trastuzumab vs observation 2005 (0%) 1 yr MFU 0.76 29 vs 37 P=0.26 2006 (4.1%) 2 yrs MFU 0.66 59 vs 90 P=0.0115 2008 4 yrs MFU (30.9%) 2012 8 yrs MFU (45.5%) 0.76 0.85 182 vs 213 P=0.1087 278 vs 350 P=0.0005 0 Favours 1 year trastuzumab 1 Favours observation 2 HR (95% CI) Extended from Gianni et al. Lancet Oncol 2011;12:236-44 Goldhirsch et al. (presented by Piccart-Gebhart), SABCS 2012, Abstract #S5-2, oral presentation

DFS NON- INFERIORITY ANALYSIS Note: Null hypothesis hazard razo is 1.11 MFU = 4.5 yrs * * 97.5% CI ** **p- value 0.025 required for stazszcal significance

Neoadjuvant trastuzumab and pertuzumab Study WO20697 / NEOSPHERE HER2+ LABC and large stage II breast cancer (n=400) Trastuzumab + docetaxel q3w x 4 Trastuzumab + pertuzumab + docetaxel q3w x 4 Trastuzumab + pertuzumab q3w x 4 Pertuzumab + docetaxel q3w x 4 Surgery Surgery Surgery Surgery FEC q3w x 3 trastuzumab q3w until Week 52 FEC q3w x 3 trastuzumab q3w until Week 52 Docetaxel + trastuzumab q3w x 4 FEC q3w x 3 trastuzumab q3w until Week 52 End points: l pcr l biomarker analysis FEC q3w x 3 trastuzumab q3w until Week 52

NeoSphere pcr rates: ITT population summary p = 0.0198 50 p = 0.0141 p = 0.003 pcr, % ± 95% CI 40 30 20 10 H, trastuzumab; P, pertuzumab; T, docetaxel 0 29.0 45.8 16.8 24.0 TH THP HP TP

TRYPHAENA Study diagnosis epirubi/cyclo x 3 // docetaxel x 3 plus trastuzumab /pertuzumab docetaxel x 3 plus Epirubi/cyclo X3 trastuzu/ pertuzu SURGERY / XRT Trastuzumab + pertuzumab to 1 year Trastuzumab + pertuzumab to 1 year Docetaxel / carboplatin Trastuzumab /pertuzumab (TCH) X 6 Trastuzumab + pertuzumab to 1 year Additional chemorx if required Schneeweiss, et al. SABCS 2011

Pathological complete response ypt0/is ypt0 ypn0 Pathological complete response (%) 61.6 [49.5 72.8] 50.7 57.3 [45.4 68.7] 45.3 66.2 [54.6 76.6] 51.9 FEC+H+P x3 T+H+P x3 (n = 73) FEC x3 T+H+P x3 (n = 75) TCH+P x6 (n = 77) FEC, 5-fluorouracil, epirubicin, cyclophosphamide; H, trastuzumab; P, pertuzumab; T, docetaxel; TCH, docetaxel/ carboplatin/trastuzumab Schneeweiss, et al. SABCS 2011

APHINITY: Phase III Trial of Pertuzumab, trastuzumab and Chemo in the Adjuvant Setting 6-8 cycles Arm 1 Chemotherapy* Surgery Central confirmation of HER2 status R Arm 2 Trastuzumab Pertuzumab 6-8 cycles Chemotherapy* 10-year followup Trastuzumab Randomisation within 7 weeks of surgery Start treatment within 1 week Placebo Anti-HER2 therapy for a total of 1 year (52 weeks) * Investigator s choice of: FEC (or FAC) TH or AC (or EC) TH or TCH A, doxorubicin; C, cyclophosphamide; E, epirubicin; F, 5-fluorouracil; H, trastuzumab; T, docetaxel or paclitaxel

APT Trial HER2+ ER+ or ER Node Neg 3 cm N=400 Tolaney SM, et al. NEJM 2015

CONCLUSIONS 1. 1-year Trastuzumab is the standard of care in patients treated a. In adjuvant b. In neoadjuvant 2. Pertuzumab in combination with Trastuzumab seems to be a new SOC in the neoadjuvant setting 3. SC Trastuzumab might be a more convenient strategy to administer trastuzumab