Trastuzumab-emtansine: A very nice poison

Trastuzumab-emtansine: A very nice poison
T-DM1 (Trastuzumab – emtansine) Pre-Clinical and Early Clinical Development Mauricio Lema Medina – Clínica de Oncología Astorga / Clínica SOMA, Medellín Trastuzumab-emtansine: A very nice poison May, 2014 1 1

Overview Anti HER2 therapy in mBC – A bird’s eye view
Introduction to Antibody Drug Conjugates (ADCs) Modes of ADC toxicity T-DM1 pre-clinical development T-DM1 early clinical development Concluding remarks

DFS Human Breast Cancer: Correlation of Relapse and Survival with Amplification of the HER-2/neu Oncogene OS Slamon DJ, Science 235: , 1987

Mechanism of Action of Trastuzumab
1 1 2 2 1 Erb receptors Trastuzumab: Inhibits ligand-independent HER2 signaling Activates ADCC Prevents HER2 ECD shedding This slide shows the mechanism of action of lapatinib compared with trastuzumab. Trastuzumab targets the extracellular domain of HER2, whereas lapatinib targets both HER2 and HER1, as shown here. Both agents have been shown to block the downstream signaling pathways of HER2. Downstream signaling pathways Cell proliferation Cell survival

Trastuzumab with chemotherapy in HER2 postive MBC
Design and enrolment Metastatic breast cancer HER2 overexpression 2/3+ No prior CT for MBC Measurable disease KPS ³60% Eligible patients (n=469) No prior anthracyclines Prior anthracyclines Trastuzumab + AC (n=143) AC (n=138) Trastuzumab + paclitaxel (n=92) Paclitaxel (n=96) AC = doxorubicin/epirubicin + cyclophosphamide Slamon DJ et al. N Engl J Med 2001;344:783 5

Proportion of Progression-Free Survival
Trastuzumab Combination Pivotal Trial: Subgroup Analysis–TTP* AC Paclitaxel 1.0 0.8 0.6 0.4 0.2 0.0 1.0 0.8 0.6 0.4 0.2 0.0 Trastuzumab + Paclitaxel (N=92) Paclitaxel (N=96) Trastuzumab + AC (N=143) AC (N=138) P< 0.001 P< 0.001 Proportion of Progression-Free Survival Months Months * Median follow-up: 35 mo (range: 30–51). . Slamon. N Engl J Med. 2001;344:783.

Probability of survival
Overall survival 1.0 0.8 0.6 0.4 0.2 H + CT CT Probability of survival 25.1 months (20%) 20.3 months HR=0.80 p=0.046 Time (months) CT patients treated with trastuzumab after disease 24% 62% % 72% progression Slamon DJ et al. N Engl J Med 2001;344:783 7

QT vs QT+Trastuzumab Enrolled, n Response Rate, % (% Improvement)
Response Duration, Mos (% Improvement) Time to Progression, Mos (% Improvement) H + CT 235 49 (-53%) 9.3 (-58%) 7.6 (-65%) CT 234 32 5.9 4.6 H + AC 138 52 (-20%) 9.1 (-40%) 8.1 (-33%) AC 145 43 6.5 6.1 H + T 92 42 (-163%) 11.0 (-150%) 6.9 (-130%) T 96 16 4.4 3.0 This slide is a summary of the registrational trial that initially won trastuzumab its approval for use in breast cancer. That [registrational trial] study was in first‑line metastatic disease in which best available chemotherapy was compared with the best available chemotherapy plus trastuzumab. The data in the slide show that when trastuzumab was added, response rates improved by 53%, response duration improved by 58%, and the primary endpoint of the study—time to progression—had the largest improvement by 65%. H, trastuzumab; CT, chemotherapy; AC, doxorubicin + cyclophosphamide; T, paclitaxel Slamon et al. N Engl J Med. 2001;344:

Cardiac Dysfunction Associated With Trastuzumab
Alone + AC AC + Paclitaxel Paclitaxel Any 3% to 7% 27% 8% 13% 1% Class III-IV 2% to 4% 16% 4% 2% The cardiac dysfunction associated with trastuzumab was described over several years. For trastuzumab alone, the rate of any cardiac dysfunction was about 3‑7%. However, when trastuzumab was given in addition to previous anthracycline and cyclophosphamide (AC) therapy, cardiac dysfunction rose to 27% from 8% with AC therapy alone. Trastuzumab plus paclitaxel was associated with a rate of 13% of any cardiac dysfunction. Trastuzumab is clearly a factor to be reckoned with in cardiac dysfunction. AC, anthracycline/cyclophosphamide; T, paclitaxel Seidman A, et al. J Clin Oncol. 2002;20:

There is significant reversibility of LV dysfunction with trastuzumab-related cardiac toxicity
Ewer, et al J of Clinical Oncology 2005,23;p

Study M77001 was conducted to demonstrate the activity of Trastuzumab plus docetaxel
HER2-positive MBC (IHC 3+ and/or FISH+) n=188 Two patients did not receive study medication n=92 n=94 Docetaxel* 100mg/m2 q3w x 6 Docetaxel 100mg/m2 q3w x 6 + *Patients progressing on docetaxel alone could crossover to receive Herceptin® Trastuzumab® 4mg/kg loading,  2mg/kg weekly until PD Marty M, et al. J Clin Oncol 2005;23:4265 11

M77001: OS (IHC 3+ / FISH+) Patients alive (%) 100 Trastuzumab + docetaxel (n=92) Docetaxel alone (n=94) 80 p=0.0325* 60 40 20 p=0.0325 +37% References Extra J-M et al. J Clin Oncol (Meeting Abstracts) 2005; 23: 16S–17S. Marty M et al. J Clin Oncol 2005; 23: 4265–4274. 22.7 31.2 5 10 15 20 25 30 35 40 45 50 Months 8.5 months Twice as many patients receiving trastuzumab survived 3 years (33% vs 16%) *Statistically significant difference Extra 2005; Marty et al 2005 12

Mechanism of Action of Lapatinib Compared to Trastuzumab
1 1 2 2 1 2 Erb receptors Lapatinib L This slide shows the mechanism of action of lapatinib compared with trastuzumab. Trastuzumab targets the extracellular domain of HER2, whereas lapatinib targets both HER2 and HER1, as shown here. Both agents have been shown to block the downstream signaling pathways of HER2. Downstream signaling pathways Cell proliferation Cell survival

EGF study design1,2 Patients with ErbB2-positive locally advanced or metastatic breast cancer that progressed after prior anthracycline, taxane and trastuzumab (N=399) RANDOMISATION EGF study design1,2 This slide shows the design for the pivotal Phase III registration trial evaluating Tyverb in combination with capecitabine versus capecitabine alone (EGF100151) in women with advanced or metastatic breast cancer with documented ErbB2 overexpression (IHC3+ or IHC2+ with FISH confirmation) who had received prior therapy which included anthracyclines, taxanes and trastuzumab.1 Trastuzumab must have been administered for at least 6 weeks in the locally advanced/metastatic setting, but may also have been given in the adjuvant setting.2 Patients were randomised to receive the combination of Tyverb (1250mg/day) and capecitabine (2000 mg/m2/day on Days 1-14, every 3 weeks) or capecitabine alone (2500 mg/m2/day on Days 1-14, every 3 weeks). This dosage was based on the OTR established in a phase I dosage-ranging study (EGF10005).3 The 2500mg/m2 dosage in the capecitabine monotherapy arm was based on the recommended dosage in the capecitabine Summary of Product Characteristics.4 The clinical use of capecitabine in this setting is supported by a technology appraisal published by the National Institute for Health and Clinical Excellence.5 Eligible patients had to have an ECOG Performance Status of 0 or 1 and have a left ventricular ejection fraction (LVEF) within institutional range of normal as measured by echocardiogram (MUGA scan may have been performed if ECHO was not available). Patients could not have been treated with capecitabine prior to the trial.2 Patients continued to receive treatment until disease progression or unacceptable toxicity or withdrawal from the trial for other reasons (e.g. consent withdrawal). Patients were stratified by stage of disease and site of metastases (visceral / non-visceral / both). A recruitment target of 528 patients was planned for the study, based on a primary endpoint of independently-assessed time to progression (TTP) and overall survival as a secondary endpoint. A pre-planned interim analysis of the primary endpoint (independently-assessed Time to Progression, TTP) was to be conducted after approximately133 time-to-progression events. Cameron D, Casey M, Press M et al. A phase III randomized comparison of lapatinib plus capecitabine versus capecitabine alone in women with advanced breast cancer that has progressed on trastuzumab: updated efficacy and biomarker analyses. Breast Cancer Res Treat 2008; 11 Jan [e-publication ahead of print]. Geyer GE, Forster J, Lindquist D et al. Lapatinib plus capecitabine for HER2-positive advanced breast cancer. N Engl J Med 2006;355: Chu QSC, Schwartz G, de Bono J. Phase I and pharmacokinetic study of lapatinib in combination with capecitabine in patients with advanced solid malignancies. J Clin Oncol 2007;25: Xeloda®Summary of Product Characteristics. National institute for Clinical Excellence. Guidance on the use of capecitabine for the treatment of locally advanced or metastatic breast cancer. Technology Appraisal No.62, 2003. Lapatinib 1250 mg po od continuously + capecitabine 2000 mg/m2/d po days 1-14 q 3 wk Capecitabine 2500 mg/m2/day po days 1-14 q 3 wk Treatment continued until progression or unacceptable toxicity po = oral; od = once daily; q 3 wk = once every 3 weeks 1. Cameron D et al. Breast Cancer Res Treat 2008; 2. Geyer GE et al. N Engl J Med 2006 14

EGF100151: independently assessed time to progression1
This slide shows the results for the primary endpoint, independently assessed TTP, for the updated analysis based on the 399 patients recruited as of the 3 April 2006 halt. The TTP results for the interim analysis were published in full in the New England Journal of Medicine (Dec 2006).1 The data shown here have been published in Breast Cancer Research and Treatment (online).2 They demonstrate that Tyverb plus capecitabine provides a highly statistically significant improvement in time to progression (TTP) versus capecitabine alone (6.2 months [27.1 weeks] vs. 4.3 months [18.6 weeks], HR=0.57, p = ).2 The curves separate early and stay separated. A hazard ratio (HR) of 0.57 indicates a 43% reduction in the risk of progression at any time point for the combination of Tyverb + capecitabine vs. capecitabine alone. A 2-month improvement in TTP is deemed to be a clinically meaningful benefit for patients at this late stage of their disease. Geyer GE et al. Lapatinib plus capecitabine for HER2-positive advanced breast cancer. N Engl J Med 2006;355: Cameron D, Casey M, Press M et al. A phase III randomized comparison of lapatinib plus capecitabine versus capecitabine alone in women with advanced breast cancer that has progressed on trastuzumab: updated efficacy and biomarker analyses. Breast Cancer Res Treat 2008, 11 Jan [e-publication ahead of print]. 15

Lapatinib + Capecitabine after Taxane + Trastuzumab: One approach on progression
1st-line 2nd--line Trastuzumab + Paclitaxel. 2001 Trastuzumab + Docetaxel 2005 Lapatinib + Capecitabina 2006 Phase II Phase III Phase III Slamon DJ, et al. (2001) N Engl J Med 344:783–792. Marti M, et al. J Clin Oncol 23:4265–4274. Geyer, et al. NEJM, 2006

Pertuzumab and Trastuzumab: Complementary Mechanisms of Action
HER2 Trastuzumab Pertuzumab HER1/3/4 Dimerization domain Subdomain IV Trastuzumab: Inhibits ligand-independent HER2 signaling Activates ADCC Prevents HER2 ECD shedding Pertuzumab: Inhibits ligand-dependent HER2 dimerization and signaling Activates ADCC ADCC = antibody-dependent cell-mediated cytotoxicity; ECD = extracellular domain

R CLEOPATRA Study Design Docetaxel (≥6 cycles recommended) N = 406
Centrally confirmed HER2-positive locally recurrent, unresectable or metastatic BC (mBC) ≤1 hormonal regimen for mBC Prior (neo)adjuvant systemic rx, incl trastuzumab and/or taxane allowed if followed by DFS ≥12 mo Baseline LVEF ≥ 50%; no CHF or LVEF < 50% during or after prior trastuzumab N = 406 Trastuzumab Placebo R 1:1 Docetaxel (≥6 cycles recommended) Trastuzumab N = 402 Pertuzumab Primary endpoint: Independently assessed progression-free survival Baselga J et al. N Engl J Med 2012;366(2):

CLEOPATRA: Progression-Free Survival
Independently assessed Pertuzumab (n = 402) Control (n = 406) HR p-value Median PFS 18.5 mo 12.4 mo 0.62 <0.001 Baselga J et al. N Engl J Med 2012;366(2):

Independently-assessed PFS (%)
3 CLEOPATRA: Significant improvement in median PFS1,2 (and OS)3 with pertuzumab 10 20 30 40 50 60 70 80 90 100 Ptz+T+D Pla+T+D Independently-assessed PFS (%) HR = % CI 0.51, 0.75 p < 0.001 References Baselga J, Kim S-B, Im S-A, et al. A Phase III, randomized, double-blind, placebo-controlled registration trial to evaluate the efficacy and safety of placebo + trastuzumab + docetaxel vs. pertuzumab + trastuzumab + docetaxel in patients with previously untreated her2-positive metastatic breast cancer (CLEOPATRA). Cancer Res 2011; 71 (15 Dec suppl.): Abstract S5-5 (and associated oral presentation). Baselga J, Cortés J, Kim SB, et al. Pertuzumab plus trastuzumab plus docetaxel for metastatic breast cancer. New Engl J Med 2012; 366: 109–119. Swain S, Kim S-B, Cortés J, et al. Confirmatory overall survival (OS) analysis of CLEOPATRA: a randomized, double-blind, placebo-controlled Phase III study with pertuzumab (P), trastuzumab (T), and docetaxel (D) in patients (pts) with HER2-positive first-line (1L) metastatic breast cancer (MBC). SABCS 2012: Poster P 12.4 18.5 5 10 15 20 25 30 35 40 PFS time (months) Number at risk Ptz+T+D 402 345 267 139 83 32 10 1. Baselga J, et al. SABCS 2011 (Abstract S5-5); 2. Baselga J, et al. N Engl J Med 2012; 366: 109–119; 3. Swain S, et al. SABCS 2012 (Poster P ). Pla+T+D 406 311 209 93 42 17 7 D, docetaxel; Ptz, pertuzumab; T, trastuzumab

CLEOPATRA: Overall Survival (Interim Analysis)
Pertuzumab (n = 402) Placebo (n = 406) HR p-value Deaths* 17.2% 23.6% 0.64 0.005 * Did not meet the O’Brien-Fleming stopping boundary of the Lan-DeMets alpha spending function for this interim analysis of overall survival and was therefore not significant. Baselga J et al. N Engl J Med 2012;366(2):

CLEOPATRA: Safety Results
Select adverse events (Grade ≥3) Pertuzumab (n = 407) Placebo (n = 397) Neutropenia 48.9% 45.8% Febrile neutropenia 13.8% 7.6% Leukopenia 12.3% 14.6% Diarrhea 7.9% 5.0% Peripheral neuropathy 2.7% 1.8% Left ventricular systolic dysfunction 1.2% 2.8% Baselga J et al. N Engl J Med 2012;366(2):

CLEOPATRA: Cardiac Tolerability of Pertuzumab plus Trastuzumab plus Docetaxel in Patients with HER2-Positive mBC Pertuzumab Placebo LVSD (any grade) (n = 407, 397) 4.4% 8.3% Symptomatic LVSD (Grade ≥3) 1.0% 1.8% LVEF decline to <50% and by ≥10% points from baseline (n = 393, 379) 3.8% 6.6% LVSD = left ventricular systolic dysfunction; LVEF = left ventricular ejection fraction Ewer M et al. Proc ASCO 2012;Abstract 533.

Lapatinib + Capecitabine after Taxane + Trastuzumab: One approach on progression
1st-line 2nd--line Trastuzumab + Paclitaxel. 2001 Trastuzumab + Docetaxel 2005 Lapatinib + Capecitabina 2006 Phase II Phase III Phase III Pertuzumab + Trastuzumab + Docetaxel 2012 T-DM1 2012 Phase III EMILIA Phase III CLEOPATRA Slamon DJ, et al. (2001) N Engl J Med 344:783–792. Marti M, et al. J Clin Oncol 23:4265–4274. Geyer, et al. NEJM, 2006 Baselga, et al. NEJM, 2012 Verma, et al. NEJM, 2012

Percentage of Patients with HER+ mBC Receiving nth Line of Chemotherapy (CT) and Duration of CT by Line of Treatment Retrospective Medical Record Review of 207 Women at DFCI Line of CT Total N % receiving CT Median duration of CT 1 58 100% 9.0 mo 2 44 76% 5.1 mo 3 40 69% 6.3 mo 4 30 52% 4.7 mo 5 24 40% 4.0 mo 6 19 33% 4.2 mo Seah DS et al. Proc ASCO 2012;Abstract 6089.

Anatomy of an Antibody-Drug Conjugate (ADC)
Linker stable in circulation Antibody targeted to tumor Linker biochemistry Acid labile (hydrazone) Enzyme dipeptides (cleavable) Thioether (uncleavable) Hindered disulfide (uncleavable) Site of conjugation Fc, HC, LC Humanized monoclonal Ab (IgG1) mAb with Fc modifications (modulate ADCC, CDC activity) Other mAb fragments Very potent chemotherapeutic drug Antibody drug conjugates are a unique type of therapeutic agent that is a hybrid of a large molecule and small molecule. ADCs consist of an antibody targeted to a tumor antigen attached to a highly potent cytotoxic drug by means of relatively stable linker. Antbody: The type of antibody used in ADCs are usually IgG1. This particular class of antibody can trigger antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC). In the context of an Adc, modulation of the Fc portion of the mAb may be desirable to alter the ability of the antibody to trigger ADCC or CDC immune responses, since this type of activity may not be desirable to the efficacy or safety of conjugate. In addition, modulation of the Fc portion may be desirable to inhibit triggering of a specific signaling cascade. An alternative to using a full length or Fc modulated Ab in ADCs, is to use a fragments of the mAb (such as Fabs). Cytotoxin: A number of different classes of cytotoxins are available for the treatment of cancer. These drugs target rapidly proliferating cells by disrupting different aspects of cellular proliferation, inluding DNA replication, repair, translation, and cell division. Because cancer cells have a higher rate of cell proliferation than normal cells, cancer cells are more susceptible to cytotoxic effects of these drugs; however it should be noted that the cytotoxins are not specific for cancer cells. The cytotoxins commonly used in ADCs fall into two categories: those that target microtubules (tubulin polymerization inhibitors) and DNA damaging agents. The maytansines (DM1) and the auristatins(MMAE) are compounds that inhibit microtubule assembly. The maytansines and their derivatives were discovered in the early 1970s and were shown to be 100 to 1000-fold more potent than other microtubule inhibitors such as vinblastine. Similarly, auristatins (MMAE) which are synthetic analogs of dolastatin, have been shown to be 50 to 200-fold more potent than doxorubicin. Aside from high potency, other desirable characteristics of a cytotoxin suitable for ADCs are good solubility and stability in aqueous soltions, low MW, accessibility of reactive groups for conjugation and low immunogenicity. Linker: the linker is a critical part of the multicomponent ADC. There are generally two classes of linkers based on their biochemistry; cleavable and uncleavable. Acid labile linkers undergo hydrolysis in the acidic environment (associated with non-specific release of the drug in clinical studies); peptide-based linkers which utilize a peptide bond to link the Ab to drug and undergo hydrolysis by intracellular lysosomal proteases. Hindered disulfide linkers are selectively cleaved in the cytosol due to high intracellular concentrations of glutathione The class of thioether linkers if noncleavable and drug is likely released by intracellular proteolytic degradation. So, what’s the benefit for using such as complex multicomponent therapeutic compound to treat cancer? Tubulin polymerization inhibitors Maytansines (DM1, DM4) Auristatins (MMAE, MMAF) DNA damaging agents Calicheamicins Duocarmycins Anthracyclines (doxorubicin) 26

Trastuzumab Emtansine (T-DM1): Mechanism of Action
Trastuzumab-specific MOA Antibody-dependent cellular cytotoxicity (ADCC) Inhibition of HER2 signaling Inhibition of HER2 shedding HER2 P Nucleus Adapted from LoRusso PM, et al. Clin Cancer Res 2011.

Trastuzumab-specific MOA Antibody-dependent cellular cytotoxicity (ADCC) Inhibition of HER2 signaling Inhibition of HER2 shedding HER2 T-DM1 P Emtansine release Inhibition of microtubule polymerization Lysosome Internalization Nucleus Adapted from LoRusso PM, et al. Clin Cancer Res 2011.

Improving the Therapeutic Window
ADCs can selectively deliver a potent cytotoxic drug to tumor cells via tumor-specific and/or over-expressed antigens Increase drug delivery to tumor Reduce normal tissue drug exposure Chemotherapy ADC TOXIC DOSE (MTD) Therapeutic Window Combining a cytotoxic drug with a mAbs can provide the best of both worlds in terms of efficacy and safety. One is able to selectively deliver a highly potent cytotoxic drug to a tumor cell using the exquisitive specificity of a monoclonal ab for a particular Ag- an Ag that is either overexpressed or, preferentially, only expressed, in tumor cells. When compared to a standard chemotherapy, selectively delivering a potent cytotoxin to a tumor using an ADC allow you to deliver a much higher dose of drug to your cell of interest while effectively increasing the window b/w your toxic dose and minimally efficacious dose. So, let’s look at an example of the advantages that ADCs have over standard “chemo”: TOXIC DOSE (MTD) DRUG DOSE Therapeutic Window EFFICACIOUS DOSE (MED) EFFICACIOUS DOSE (MED) MTD: Maximum tolerated dose; MED: Minimum Efficacious Dose

ADC Better Tolerated than Free Cytotoxin in Rats
Single IV dose; rats T-DM1 (2040 µg DM1/m²) (% change from baseline) Body Weight Besides efficacy, tolerability of the drugs are a major consideration when comparing two therapeutic agents. Tolerability in tox studies are often crudely assessed by monitoring changes in body weights over time. This data speaks to the point that the ADC is better tolerated than the free cytotoxin in rats. Again, using the same T-DM1 ADC in a single IV injection, we can see that there is a dramatic decline in the body weights of animals and early mortality that were dosed with the free cytotoxin (DM1) in comparison to little- no impact on the body weights of rats dosed with the ADC. In fact, TDM1 dosed rats continue to gain weight after the initial dose. (does this bind to rat?) Free DM1 (2400 µg DM1/m²) Early mortality (100%) Time (Day) 30 T-DM1: Trastuzumab emtansine 30

Modes of Anti-tumor Activity of ADCs
Tumor Cells Tumor Cell Tumor cytotoxicity is target-enhanced (bystander effect) ADC-Ag binding → extracellular cleavage of toxin → release of toxin in local tumor environment → diffusion of toxin intracellularly to neighboring tumor cells → tumor cell death The mechanism of anti-tumor activity of an ADC can be two fold: target-directed and target-enhanced (or secondary to the bystander effect). An ADC can selectively bind to an antigen on the cell surface of a tumor cell, internalized in the lyzosome, degraded and the cytoxin released from the Ab intracellularly. The other mechnism that an ADC can bind to the Ag, extracellular cleavage of the toxin may occur and release of toxin in the local environment may diffuse into the neighboring cell causing cytotoxicity. A highly charged metabolite may be hindered from free difuse through a lipid membrane- and judicious use of a linkers may be valuable in manipulating the bystander killing effect. The antigen is highly important in the specificity of the target-directed tumor cytotoxicity and bears a bit of attention before moving on to the modes of toxicity of ADCs. Tumor cytotoxicity is target-directed ADC-Ag binding → internalization in lysosomes → ADC degradation → release of toxin intracellularly → tumor cell death 31

Tissue Antigen Characteristics Are Key in ADCs
Careful selection of target antigens are an important criterion for both the safety and efficacy of an ADC The ‘ideal’ tissue antigen should have: High level of target expression in cancer cells Little to no expression in normal cells Expressed on the cell surface Readily internalized No shedding into the blood by cleavage of the antigen from cancer cell surface The number of antigen molecules and antibody binding affinity for the antigen may affect the potency of the ADC Careful selection of target antigens are an important criterion for both the safety and efficacy of an ADC. The ideal tissue antigen should be be expressed in high levels on cancer cells (both the primary tumor and the metastatic tumor) and ideally not on normal cells; but in the real world, may have decreased of limited expression on normal cells. The ag should be expressed on the cell surface allowing for Ab binding and should be readily internalized- since the MOA of ADCs relies on internalization of the ADC after binding. Ideally, the ag should not be shed by cleavage from cancer cells b/c one would want to avoid ADC binding and systemic cytotox. The binding affinity of the Ab and number of AG molecules on the cell surface also likely impact the ADC efficacy. However, a clear correlation between antigen expression and sensitivity to the therapeutic mAb has not been shown.

Modes of Toxicity of ADCs
+ Normal Cell Systemic release of toxin Instability of linker Catabolism of ADC Unwanted ADC-mediated cytotoxicity Targeted binding to normal tissues expressing antigen Off-target (cross reactive) binding to normal tissues Non-antigen-mediated ADC uptake (e.g., Fc-mediated uptake, pinocytosis) So, let’s talk in more depth about the modes of toxicity of ADCs. ADCs can cause toxicity by either systemic release of the toxin or by unwanted ADC-mediated cytotxicity. (Speak each type of sub bullet) We will consider each of these subcategories in more details now: 33

+ Normal Cell Systemic release of toxin Instability of linker Catabolism of ADC Unwanted ADC-mediated cytotoxicity Targeted binding to normal tissues expressing antigen Off-target (cross reactive) binding to normal tissues Non-antigen-mediated ADC uptake (e.g., Fc-mediated uptake, pinocytosis) I mentioned before when introducing you to the anatomy of ADCs that the linker is a critical component to the stability of the overall structure. Instability of the linker can result in release of the cytotoxin at an undesirable time or in an undesirable environment. The impact of linker stability of the toxicity profile was investigated in greater depth with Genentech’s CD22 ADC program. 34 34

Slower Drug Deconjugation With Uncleavable Linker
Single IV dose 20 mg/kg ADC Days post dose Concentration (µg/ml)  Total Ab  Uncleavable linker To investigate the impact of linker stability on the release of drug and systemic toxicity, various linker-drug combinations that all result in the release of cytotoxic metabolites , were dosed to rats and the PK and tolerability/toxicity was assessed. In this study, the investigators used maytansinoid linker-drugs consisting of the maytansine DM1 with one of two different linkers: a disulfide linker SPP(which will release drug through the reduction of the sulfide bond) or the thioether linker MCC (which is uncleavable such that the Ab must be degraded to release the drug. (Two other ADCs used in these experiements and I’ll speak to those on the next slide: They are based on the auristatins: the cytotoxin MMAE linked to Ab cysteines by MC-vc-PAB and cleaved internally by cathepsin B In contrast MC-MMAF are uncleavable and must be internalized and degraded within the cell. In this study, female SD rats were given a single IV dose of one of the 4 linker-drug combinations conjugated to CD22 (matched at 20 mg/kg ADC- highest tolerated dose for DM1).The CD22 mAb does not bind rats and so the results should reflect systemic toxicity of the ADC in general rather than the effects of targeting the ADC to specific tissues. In examining the PK curves from animals treated with the cleavable vs. noncleavable linker drug conjugates; total Ab clearance (measured by ELISA) was similar for both cleavable and uncleavable conjugates suggesting that the type of linker has a minimal effect on the metabolism of the Ab. However: whereas the drug-loaded uncleavable linker conjugates (measured by ???) cleared with similar kinetics as the total ab, the cleavable linker conjugates lost drug more quickly (SPP-DM1). This suggests that the cleavable linker ADCs release more free drug into the circulation. Given that there was no detectable differences in the clearance rates of the total Abs and that there it appears that the ADC cleavable linker releases more drug(deconjugates) faster than the noncleavable llinker drug conjugate in systemic circulation, the authors hypothesized that the ADCs with uncleavable linkers (MCC-DM1 and MC-MMAF, may be better tolerated because they released smalerl amounts of free drug or small molecule metabolites into systemic circulation:  Cleavable linker Polson, et al., Cancer Res., 69(6), 2009

More Stable Linker Reduces Systemic Toxicity of ADC in Rats
Single IV dose given on Day 1 : Change in bodyweight (grams) Days post dose   CD22-DM1 with cleavable linker Looking first the the body weight data as a measure of tolerability: Only the CD22-SPP-DM1 had a negative effect on body weight. Polson, et al., Cancer Res., 69(6), 2009

More Stable Linker Reduces Systemic Toxicity of ADC in Rats
Single IV dose given on Day 1 : Explain the hem and chem changes in the cleavable linkers over uncleavable linkers vs. control. In these studies, ADCs with uncleavable linkers showed no significant hematologic effects, minimal impact on body weights and slower deconjugation or loss of drug from the ADC: suggesting that they are much better tolerated than the ADCs with cleavable linkers due to their relatively stability of the drug on Ab, limiting the release into systemic circulation. . Polson, et al., Cancer Res., 69(6), 2009

+ Normal Cell Systemic release of toxin Instability of linker Catabolism of ADC DAR Unwanted ADC-mediated cytotoxicity Targeted binding to normal tissues expressing antigen Off-target (cross reactive) binding to normal tissues Non-antigen-mediated ADC uptake (e.g., Fc-mediated uptake, pinocytosis) The next mode that ADC can cause toxicity is through catabolism or breakdown of the ADC.A Breakdown the ADC itself can result in the generation of various fragments/components of the ADC; including the liberation of free drug from the Ab. One of the important factors that will impact an ADCs toxicity is the number of drugs per antibody or the DAR. 38

Early Observation: Highly Drugged ADCs More Toxic
DAR: Drug-to-Antibody Ratio DAR 2 DAR 4 DAR 6 DAR 2 DAR 4 DAR 6 An early observation in ADCs was that conventional Ab drug conjugation processes resulted in a heteorgenous species with variable numbers of drugs conjugated to the Ab- DAR ranging from Its important to remember that even when you have a lot of ADC with an average DAR of 3, there can still be substantial variability in the species of ADC with DARs from 0-6 or more. Bioanalytical assays are an important component of assessing the DAR of these ADC species and at Genentech, the standard assay paradigm involves a total AB assay (capturing the Ab by ELISA) and a LC-MS assay that measures the mass differences in these complex ADC species: this way one can get a sense of the average DAR in a lot with accuracy. Having a high proportion of these highly drugged Abs in your mixture was concerning in terms of toxicity higher DAR species were suspected to be potentially more toxic. To investigate this, a single dose IV toxicity studies using Herceptin-MMAF with a cleavable linker to evaluate the effects of variable DAR on rats showed decreased tolerabilities based on body weight effects and elevations in AST in rats treated with increasing DAR ratios. (Does herceptin bind in rats?????) Explain the graph on tolerabiltity and the graph on increasing DAR and increased AST elevations at matched ug/m2 MMAE. The higher DAR species were associated with faster in vivo clearance and increased toxicity.

ThioMAb Technology: Controlling Heterogeneity
Proportion ADC DAR Proportion TDC In an effort to retain the in vivo efficacy of a conventional ADC and control the variability of the DAR, thiomab engineering technology was developed at genentech. This allowed for more homogenous drug load and elimination of a highly drugged ADC (High DAR) that may be more toxic. Talk about the engineering differences between an ADC vs TDC (engineered cysteines on the TDC that allow for controlled stoichemetry) and illustration of the DAR differences- in particular make the point about elimination of the highly drugged species. Efficacy studies demonstrated that a TDC was of equivalent efficacy as the ADC when you compare a mg/kg basis of ADC/TDC. However, looking at efficacy from a matched cytotoxin dose (MMAE), the TDC is actually twice as efficacious as the ADC because you are able to deliver half the DAR compared to the ADC. In rat tox studies in which the dose of matched MMAE ug/m2 ADC vs. 16TDC at (different DAR ratio), the TDC was better tolerated than the ADC in rats (less changes of BW and Hem/chem parameters). Looking at the PK in more detail from these studies gives us a sense of how these two types of ADCs vary. DAR Engineered ThioMAb backbone allows more homogenous drug load (MMAE) Efficacy of TDC  ADC (mg/kg basis) and  2 x ADC (ug MMAE/m2 basis) Junutula, et al., Nat. Biotech., 26(8), 2008

Catabolism and Deconjugation of TDC is Slower than ADC in Rats
Single dose I.V. PK study: ADC or TDC with matched cytotoxin (MMAE) doses A thorough kinetic analysis using chimeric antibodies (Fig. 5b–d) showed that the total TDC is cleared somewhat more slowly than the ADC (9.5 ± 2.9 versus 16.1 ± 3.5 ml/day/kg), and the proportion of TDC still bearing at least one drug decreased substantially more slowly than the corresponding proportion of ADC (14.1 ± 3.0 versus 41.6 ± 4.8 ml/day/kg). Therefore, we conclude that despite bearing fewer drugs per antibody on average, the TDC variants retain the conjugated drugs more effectively in rats than their ADC counterparts. (((Insert rat body weight data to speak about tolerability of TDCs over ADCs))) Deconjugation of the Antibody Catabolism of the Antibody Junutula, et al., Nat. Biotech., 26(8), 2008

MMAE TDC is Better Tolerated Than ADC in Monkeys
Repeat IV doses of ADC or TDC, Days 1 and 23: Examining TDC vs ADC tolerability in NHPs was performed in this example. In a multidose study (Ag-dependent??) where the ADC and TDC doses were delivered to the NHP at matched MMAE ug/m2 doses (again, because there are half as many Drugs per ab on the TDC, we delivered 2x the dose on a mg/kg basis in comparison to the ADC), we monitored the impact of the neutrophils. Looking at the day 8 and day 32 neutrophil counts (~7 days post dose, when we expect that the neutrophils, if impacted by the toxin, would be depleted), we can see there is a a 2 fold difference(decrease) in the neutrophils on the ADC vs the TDC. Again, we see similar results on day 32. Examining the impact of increasing TDC doses (increased ug/m2 dose of MMAE), there are corresponding decreases in the neutrophil count. What this data tells us, is that there is little to no impact on neutrophils, a sensitive marker of MMAE-mediated toxicity, in NHPS given equivalent MMAE doses as that of the ADC in which there is a demonstrated neutropenia. No neutrophil decreases with TDC compared to equivalent ug/m2 dose of ADC Junutula, et al., Nat. Biotech., 26(8), 2008 42

+ Normal Cell Systemic release of toxin Instability of linker Catabolism of ADC DAR Site of conjugation Unwanted ADC-mediated cytotoxicity Targeted binding to normal tissues expressing antigen Off-target (cross reactive) binding to normal tissues Non-antigen-mediated ADC uptake (e.g., Fc-mediated uptake, pinocytosis) Not only does the DAR impact of the toxicity of ADCs, but also the site of drug conjugation on the Ab. 43

+ Normal Cell Systemic release of toxin Instability of linker Catabolism of ADC Unwanted ADC-mediated cytotoxicity Targeted binding to normal tissues expressing antigen Off-target (cross reactive) binding to normal tissues Non-antigen-mediated ADC uptake (e.g., Fc-mediated uptake, pinocytosis) ADCs can cause toxicity from targeted binding to normal tissues expressing antigen 44 44

+ Normal Cell Systemic release of toxin Instability of linker Catabolism of ADC Unwanted ADC-mediated cytotoxicity Targeted binding to normal tissues expressing antigen Off-target (cross reactive) binding to normal tissues Non-antigen-mediated ADC uptake (e.g., Fc-mediated uptake, pinocytosis) Unwanted-mediated cytotox can also occur because of off-target (cross-reactive) binding to normal tissues and non-Ag mediated ADC uptake (pinocytosis,etc….) Talk a little bit more about these theoretical concerns? 45 45

Summary An ADC is both a “large molecule” and a “small molecule”.
ADCs hold great promise for improving current oncology therapies. Highly potent cytotoxic agents are delivered directly to cancer cells, sparing normal tissues. ADCs tend to be better tolerated than standard chemotherapy. Increased therapeutic window allows for better balance between safety/efficacy. There is a fine balance between efficacy and toxicity. Choice of linker, cytotoxic drug and mAb are all important determinants of safety, PK, and efficacy. Toxicity is usually antigen-independent, ADC/drug-dependent. Linker stability, DAR, and site of drug conjugation impacts toxicity.

Constructing a highly effective ADC for HER2-positive breast cancer
Least stable Five trastuzumab–maytansinoid conjugates were constructed with various linkers to assess impact on drug activity X DM R is: Maytansine: CH3 DM1:CH2CH2SSMe DM3:CH2CH2CH(CH3)SSMe DM4: CH2CH2C(CH3)2SSMe T-SPDP-DM1 X T-SPP-DM1 X T-SSNPP-DM3 X T-SSNPP-DM4  Script Five trastuzumab–maytansinoid conjugates were constructed with various linkers to assess impact on drug activity These were: Trastuzumab-SPDP-DM1 (no methyl substitutions around disulphide bridge) Trastuzumab-SPP-DM1 (one methyl group) Trastuzumab-SSNPP-DM3 (two methyl groups) Trastuzumab-SSNPP-DM4 (three methyl groups) Trastuzumab-MCC-DM1 (thioether linker) Increase in methyl groups around the disulphide bridge reflects increasing resistance to cleavage DM3 and DM4 nomenclature signifies the addition of methyl groups to DM1 (one or two, respectively) Trastuzumab linked to DM1 via MCC (thioether linker) displayed superior activity compared with unconjugated trastuzumab or the other trastuzumab–maytansinoid conjugates Potent activity was observed on all HER2-overexpressing tumour cells whereas normal cells and tumour cells not overexpressing HER2 were unaffected Activity was also seen on all trastuzumab-refractory cell lines References Lewis Phillips GD, et al. Cancer Res 2008; 68: 9280–9290 T-MCC-DM1 Most stable Discovery of antibodies Hormonal therapy and chemotherapies used to treat haematological & solid cancers mAbs Linker and ADC advances Ehrlich’s vision of targeted treatment Maytansine HER2 Trastuzumab 1900s 1900s 1930–60s 1970s 1980s 1990s 2000s Key milestones in the development of Kadcyla Lewis Phillips GD, et al. Cancer Res 2008; 68: 9280–9290

Another HER-2 Targeted Therapy in Development Trastuzumab-DM1 (T-DM1)
The trastuzumab-mertansine drug conjugate efficiently delivers a chemotherapeutic drug to HER2 positive breast cancer cells by encompassing three components (drug, linker, and antibody). The unique chemical linker, MCC, is conjugated via lysine side chains to the cytotoxic agent, mertansine, and is designed to provide a stable antibody-drug bond that preserves the binding activity of trastuzumab to the extracellular domain of HER2. Conjugation of the highly cytotoxic mertansine to a HER2 specific monoclonal antibody is an important approach to confer selectivity to mertansine and potentially increases the therapeutic index. Trastuzumab Mertansine: anti-tubulin

T-DM1 Selectively Delivers a Highly Toxic Payload to HER2-Positive Tumor Cells
Trastuzumab-like activity by binding to HER2 Targeted intracellular delivery of a potent antimicrotubule agent, DM1 T-DM1 binds to the HER2 protein on cancer cells Trastuzumab and DM1 are covalently linked via a thioether linker DM1 is a highly potent derivative of antimicrotubule agent T-DM1 specifically targets HER2 + tumor cells by antibody-dependent cellular cytotoxicity and inhibiting HER2 signaling. A phase I study demonstrated that MTD was 3.6 mg/kg every 3 wks, and systemic DM1 exposure was low (5 ng/mL maximum plasma levels). Receptor-T-DM1 complex is internalized into HER2-positive cancer cell Potent antimicrotubule agent is released once inside the HER2-positive tumor cell 49

ADC More Efficacious than Free Cytotoxin in Mice
DM1 Free DM1 (cytotoxin) In this example, we look at the demonstrated increased efficacy in mice of the adc over the free cytotocin (mimicking the ADC vs standard chemotherapy comparison) by evaluating the mean tumor volume over time in mouse tumor models (efficacy models). This study shows that dosing mice harboring a HER2 positive breast cancer xenograft treated with standard chemo/free cytotoxin, DM1 (cytotoxcin alone) vs DM1 conjugated to the a HER Ab (traztuzumab-DM1) ADC: the mean tumor volumes remain smaller for a longer period of time in the ADC treated animals vs the free drug. This implies that the ADC is more efficacious at inhibiting tumor growth in this model than in the free drug arm alone. There is much better efficacy, of course, of ADC over Ab vs. control as well. (question about traz insensitive?) T-DM1 (ADC) Parsons et al, AACR (2007); Modified from S. Spencer 50

T-DM1 q3w – Phase I Trial: T-DM1 DOT
Krop IE, et al. JCO, 2010

Contra Pro T-DM1 q3w Phase I Trial
Heavily pre-treated HER2+ BC patients Dose/Efficacy Significant AEs at MTD 3.6 mg/kg – q3w 73% Clinical-Benefit 44% Objective responses DLT at 4.8 mg/kg: Thrombocytopenia Thrombocytopenia Elevetad aminotrasferases Fatigue Nausea Contra Pro “T-DM1 was associated with mild, reversible toxicity, and substantial clinical activity in a heavily treated (HER2+) population” Krop IE, et al. JCO, 2010 52 52

T-DM1 weekly – Phase I Study: T-DM1 DOT
Beeram M, et al. Cancer, 2012

Contra Pro T-DM1 weekly Phase I Trial
Heavily pre-treated HER2+ BC patients Dose/Efficacy Significant G>=3 AEs 2.4 mg/kg – every week 57% Clinical-Benefit 46% Objective responses 18.6m median DOT Anemia (14%) Thrombocytopenia (11%) Pneumonia (11%) Increased AST (11%) Contra Pro “weekly dose of T-DM1 2.4 mg/kg has antitumor activity and is well tolerated in patients with HER2+ mBC ” Beeram M, et al. Cancer, 2012 54 54

Phase II Trial of Trastuzumab-DM1 for the Treatment of HER2+ mBC After Prior anti-HER2 therapy
Burris HA, et al. J Clin Oncol, 2010

TDM1 Versus Trastuzumab + Docetaxel 1st line
HER2-positive, recurrent locally advanced BC or MBC (n=137) T-DM1 3.6 mg/kg Q3W until PD 1:1 Trastuzumab 8 mg/kg dose; 6 mg/kg Q3W + Docetaxel 75 or 100 mg/m2 Q3W PD Crossover T-DM1 Randomized, phase II, international, open-label study HER2-positive, measurable disease required Stratification factors World region, prior adjuvant trastuzumab therapy, disease-free interval Primary endpoints: PFS by INV, safety Key secondary endpoints: ORR, clinical benefit, OS, QOL, symptom control Perez EA, et al. ESMO Abstract LBA3. 57

T-DM1 Versus Trastuzumab (T) + Docetaxel (D) in HER2-Positive MBC With No Prior Chemotherapy for MBC
Efficacy Summary Overall response rate (ORR) 47.8% 41.4% Safety Summary Grade ≥3 adverse event (AE)† 37.3% 75.0% † Most common AEs, any grade, T + D: alopecia: 66.2%, neutropenia: 57.4%, diarrhea: 45.6% — these were 1.5%, 7.5%, and 10.4% in pts receiving T-DM1. Most common AEs, any grade, T-DM1: nausea: 47.8%, fatigue: 46.3%, pyrexia: 35.8% — these were 39.7%, 46.2%, and 20.6% in pts receiving T + D. Perez EA, et al. Proc ESMO Abstract LBA3.

T-DM1 Activity: Improved PFS
Median Progression-Free Survival (months) trastuzumab + docetaxel T-DM1 HER2+ locally advanced or metastatic 10 15 5 14.2 9.2 P=0.035 Treatment with T-DM1 reduced the probability of disease progression or death by 41% compared with Trastuzumab + Docetaxel HR=0.59, P=0.035 Hurvitz S, et al. ECCO-ESMO Abstract 5001. 59

From Ehrlich’s vision to Kadcyla: Over 100 years in the making
Maytansine isolated4,5 mAbs6 HER2 oncoprotein7 Herceptin (trastuzumab) in trials and approval7,9–12 1897 ADC & linker advances13 1970s Five trastuzumab-DM1 conjugates with various linkers were evaluated13 Trastuzumab-MCC-DM1 selected 2000’s Greater activity compared with non-conjugated trastuzumab Good PK profile Best safety profile Taken into clinical studies Pivotal EMILIA* study reports successful survival and QoL benefits compared with lapatinib plus capecitabine and thus the final proof-of-concept of an ADC14 Script Ehrlich’s vision of targeted therapy began in 1897 and more discoveries and advances followed such as antibodies and chemotherapies1–3 Maytansine was isolated in the 1970s and some time later was followed by its analogues which were highly toxic in vitro and were not usable in the clinic as they lacked tumour selectivity4,5 The 1970s also saw the discovery of monoclonal antibodies (continuous cultures of fused cells secreting antibody of pre-defined specificity)6 In the early 1980s HER2 was identified as an oncoprotein and was then found to be overexpressed in breast tumours and then overexpression was found to be associated with a more aggressive phenotype7 The late 80s/early 90s saw the development of the mouse anti-HER2 mAbs and subsequently the humanised mAb resulting in trastuzumab and thus Herceptin8 In 1993, Herceptin clinical programme was initiated and subsequently Herceptin became the SoC in HER2-positive eBC and HER2-positive 1st L mBC7,9–12 Further advances in molecular targets led to the ADC concept which was the inspiration for the development of Kadcyla – following the experiments with the five conjugates, one – that is Kadcyla – was found to have greater activity compared with non-conjugated trastuzumab, having the best PK and safety profile, and thus was taken into clinical studies13 The pivotal EMILIA study of Kadcyla as a single agent in the treatment of patients with HER2-positive metastatic breast cancer, previously treated with Herceptin plus a taxane, reported significant progression-free survival and overall survival benefits over Xeloda plus lapatinib and was thus the final proof-of-concept of an ADC and resulted in the FDA and EMA approval of Kadcyla in this patient population14 References Strebhardt K & Ullrich A. Nat Rev Cancer 2008; 8: 473– Yarden Y. The Oncologist 2011; 16 suppl 1: 23–29 Kauffman S. Nat Rev Cancer 2008; 7: Gianni L, et al. Lancet 2010; 375: 377–384 DeVita VT Jr & Chu E. Cancer Res 2008; 68:8643– Marty M, et al. J Clin Oncol 2005; 23: 4265–4274 Remillard S, et al. Science 1975; 189: 1002– Perez EA, et al. J Clin Oncol 2011; 29: 3366–3373 Chari RV. Acc Chem Res 2008; 41: 98– Slamon D, et al. N Engl J Med 2011; 365: 1273–1283. Köhler G & Milstein C. Nature 1975; 256: 495– Lewis Phillips GD, et al. Cancer Res 2008; 68: 9280–9290 Slamon D, et al. N Engl J Med 2001; 344: 783– Verma S, et al. N Engl J Med : 2012 FDA & EMA approval 2013 * EMILIA study population: Patients with HER2-positive locally advanced or metastatic breast cancer, who had received prior Herceptin and taxane therapy. Patients had either progressed within 6 months of completing adjuvant therapy or during metastatic treatment

Trastuzumab-emtamsine
Concluding remarks And the rest of the story is about to come… EMILIA Hay be studied in combination Effective, with high response rate even in heavily pretreated HER2+ BC patients Safe, with most cytotoxic activity inside the HER2+ cell ADC (Trastuzumab + anti microtubule

Updated Overall Survival Results From EMILIA,
a Phase 3 Study of Trastuzumab Emtansine (T-DM1) vs Capecitabine and Lapatinib in HER2-Positive Locally Advanced or Metastatic Breast Cancer S Verma,1 D Miles,2 L Gianni,3 IE Krop,4 M Welslau,5 J Baselga,6 M Pegram,7 D-Y Oh,8 V Diéras,9 E Guardino,10 L Fang,10 MW Lu,10 S Olsen,10 K Blackwell11 1Sunnybrook Odette Cancer Center, Toronto, Canada; 2Mount Vernon Cancer Center, Northwood, UK; 3San Raffaele Hospital, Milan, Italy; 4Dana-Farber Cancer Institute, Boston, MA, USA; 5Medical Office Hematology, Aschaffenburg, Germany; 6Massachusetts General Hospital, Boston, MA, USA; 7University of Miami Sylvester Comprehensive Cancer Center, Miami, FL, USA; 8Seoul National University College of Medicine, Seoul, Korea; 9Institut Curie, Paris, France; 10Genentech, Inc, South San Francisco, CA, USA; 11Duke Cancer Institute, Durham, NC, USA

Disclosure Slide Verma: Compensated consultant/advisory relationship with Roche/GSK; honoraria from GSK/Roche; research funding from Genentech/Roche Miles: Compensated consultant/advisory relationship with Genentech/Roche; honoraria from Genentech/Roche Gianni: Compensated consultant/advisory relationship with Genentech/Roche, GSK, Pfizer Krop: Uncompensated consultant/advisory relationship with Novartis; research funding from Genentech/Roche Welslau: None Baselga: Compensated consultant/advisory relationship with Genentech/Roche Pegram: Compensated consultant/advisory relationship with Genentech/Roche; honoraria from Genentech/Roche Oh: None Dieras: Compensated consultant/advisory relationship with Genentech/Roche, Novartis, Sanofi, Amgen, Clovis, Pfizer, GSK; honoraria from Genentech/Roche, Novartis, Sanofi, Amgen, Clovis, Pfizer, GSK Guardino: Genentech employee; owns Roche stock Fang: Genentech employee; owns Roche stock Lu: Genentech employee; owns Roche stock Olsen: Genentech employee; owns Roche and Sanofi stock Blackwell: None

HER2 Antibody: Trastuzumab P Stable linker: MCC Emtansine Cytotoxic: DM1 Nucleus Adapted from LoRusso PM, et al. Clin Cancer Res 2011.

Trastuzumab-specific MOA Antibody-dependent cellular cytotoxicity (ADCC) Inhibition of HER2 signaling Inhibition of HER2 shedding HER2 P Nucleus Adapted from LoRusso PM, et al. Clin Cancer Res 2011.

Trastuzumab-specific MOA Antibody-dependent cellular cytotoxicity (ADCC) Inhibition of HER2 signaling Inhibition of HER2 shedding HER2 T-DM1 P Emtansine release Inhibition of microtubule polymerization Lysosome Internalization Nucleus Adapted from LoRusso PM, et al. Clin Cancer Res 2011.

HER2-positive LABC or MBC (N=980)
EMILIA Study Design 1:1 HER2-positive LABC or MBC (N=980) Prior taxane and trastuzumab Progression on metastatic treatment or within 6 months of adjuvant treatment PD T-DM1 3.6 mg/kg q3w IV Capecitabine 1000 mg/m2 PO bid, days 1–14, q3w + Lapatinib 1250 mg/day PO qd Stratification factors: World region, number of prior chemo regimens for MBC or unresectable LABC, presence of visceral disease Primary endpoints: PFS by independent review, OS, and safety Key secondary endpoints: PFS by investigator, ORR, DOR Statistical considerations: Hierarchical statistical analysis was performed in pre-specified sequential order: PFS by independent review → OS → secondary endpoints PFS analysis: 90% power to detect HR=0.75; 2-sided alpha 5% OS analyses: 80% power to detect HR=0.80; 2-sided alpha 5%

EMILIA Analyses Final PFS analysis: Targeted number: 508 events Actual number: 569 events 1st Interim OS analysis: Preplanned at time of final PFS analysis Did not cross efficacy stopping boundary Safety and Secondary Endpoint analysis Final PFS and 1st Interim OS Analysis Data cut-off Jan 14, 2012 Presented at ASCO 2012 2nd Interim OS Analysis Data cut-off July 31, 2012 Following health authority interactions 50% of targeted number of OS events (n=316) Actual number of OS events: 331 events Final OS Analysis Expected 2014 Targeted number of events: 632

Patient Demographics and Baseline Characteristics (1)
Cap + Lap (n=496) T-DM1 (n=495) Median age, years (range) 53 (24–83) 53 (25–84) Race, n (%) White Asian Black/African American Other Not available 374 (75) 86 (17) 21 (4) 10 (2) 5 (1) 358 (72) 94 (19) 29 (6) 7 (1) World region, n (%) United States Western Europe Asia 136 (27) 160 (32) 76 (15) 124 (25) 134 (27) 157 (32) 82 (17) 122 (25) ECOG PS, n (%) 1 312 (64) 176 (36) 299 (61) 194 (39)

Patient Demographics and Baseline Characteristics (2)
Cap + Lap (n=496) T-DM1 (n=495) Measurable disease by independent review, n (%) 389 (78) 397 (80) Site of disease involvement, n (%) Visceral Non-visceral 335 (68) 161 (32) 334 (67) 161 (33) Metastatic sites, n (%) <3 ≥3 Unknown 307 (62) 175 (35) 14 (3) 298 (60) 189 (38) 8 (2) ER/PR status, n (%) ER+ and/or PR+ ER− and PR− 263 (53) 224 (45) 9 (2) 282 (57) 202 (41) 11 (2)

Prior Systemic Treatment
Cap + Lap (n=496) T-DM1 (n=495) Prior treatment type, n (%) Taxanes Anthracyclines Endocrine agents 494 (100) 302 (61) 204 (41) 493 (100) 303 (61) 205 (41) Prior therapy for MBC, n (%) Yes No 438 (88) 58 (12) 435 (88) 60 (12) Prior trastuzumab treatment, n (%) Early breast cancer only 495 (100) 77 (16) 78 (16) Duration of trastuzumab treatment, n (%) <1 year ≥1 year 212 (43) 284 (57) 210 (42) 285 (58) Median time since last trastuzumab, months (range) 1.5 (0–98) 1.5 (0–63)

Patient Disposition Primary analysis data cut-off Jan 14, 2012
Updated OS July 31, 2012 Cap + Lap T-DM1 Randomized, n 496 495 Treated, n 488 490 Median follow-up, months 12.4 12.9 18.6 19.1 On study at data cut-off date, n 316 366 262 308 On treatment, n 125 182 55 106 Deaths, n 129 94 149

Drug Exposure Cap (n=487) Lap (n=488) T-DM1 (n=490)
Median dose intensity, % 77.2 93.4 99.9 Pts with dose reduction, n (%) 260 (53.4) 133 (27.3) 80 (16.3) T-DM1 decreased to 3.0 mg/kg, n (%) — 58 (11.8) T-DM1 decreased to 2.4 mg/kg, n (%) 22 (4.5)

Progression-Free Survival by Independent Review
Median (months) No. of events Cap + Lap 6.4 304 T-DM1 9.6 265 Stratified HR=0.650 (95% CI, 0.55, 0.77) P<0.0001 0.0 0.2 0.4 0.6 0.8 1.0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Proportion progression-free Time (months) 496 404 310 176 129 73 53 35 25 14 9 8 5 1 495 419 341 236 183 130 101 72 54 44 30 18 3 Cap + Lap T-DM1 No. at risk by independent review: Unstratified HR=0.66 (P<0.0001).

Progression-Free Survival Subgroup Analyses Pre-specified Stratification Factors
Cap + Lap T-DM1 Baseline characteristic T-DM1 better Cap + Lap better Total n Median, mos Median, mos HR (95% CI) All patients 991 6.4 9.6 0.66 (0.56, 0.78) World region US 270 5.7 8.5 0.70 (0.51, 0.98) Western Europe 317 6.4 10.9 0.56 (0.41, 0.74) Other 404 6.9 9.6 0.73 (0.56, 0.94) Number prior chemo regimens for MBC or unresectable LABC 0–1 609 6.7 10.3 0.68 (0.55, 0.85) >1 382 5.7 8.5 0.63 (0.49, 0.82) Disease involvement Visceral 669 5.7 9.6 0.55 (0.45, 0.67) Nonvisceral 322 10.2 8.5 0.96 (0.71, 1.30) Hazard ratio 0.2 0.5 1 2 5

Overall Survival: First Interim Analysis
Median (months) No. of events Cap + Lap 23.3 129 T-DM1 NR 94 Stratified HR=0.621 (95% CI, 0.48, 0.81); P=0.0005 Efficacy stopping boundary P= or HR=0.617 Proportion surviving 0.0 0.2 0.4 0.6 0.8 1.0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 77.0% 65.4% 47.5% 84.7% Time (months) 496 469 438 364 296 242 195 155 129 97 74 52 31 17 7 3 2 1 495 484 461 390 331 277 220 182 149 123 96 67 46 29 16 5 Cap + Lap T-DM1 No. at risk: Unstratified HR=0.63 (P=0.0005). NR, not reached.

Overall Survival: Confirmatory Analysis
Median (months) No. of events Cap + Lap 25.1 182 T-DM1 30.9 149 Stratified HR=0.682 (95% CI, 0.55, 0.85); P=0.0006 Efficacy stopping boundary P= or HR=0.727 1.0 85.2% 0.8 78.4% 64.7% 0.6 Proportion surviving 51.8% 0.4 0.2 0.0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 Time (months) 496 471 453 435 403 368 297 240 204 159 133 110 86 63 45 27 17 7 4 495 485 474 457 439 418 349 293 242 197 164 136 111 62 38 28 13 5 Cap + Lap T-DM1 No. at risk: Data cut-off July 31, 2012; Unstratified HR=0.70 (P=0.0012).

Overall Survival Subgroup Analyses
Cap + Lap T-DM1 Baseline characteristic Total n Median (mos) HR (95% CI) Better All patients 991 25.1 30.9 0.70 (0.56, 0.87) World region United States 270 23.7 NR 0.62 (0.41, 0.96) Western Europe 317 28.6 27.8 0.95 (0.65, 1.39) Asia 158 22.7 34.3 0.48 (0.27, 0.85) Other 246 26.1 0.68 (0.45, 1.04) Number of prior chemotherapeutic regimens for LABC or MBC 0–1 609 28.0 29.8 0.80 (0.61, 1.07) >1 382 31.9 0.58 (0.41, 0.81) Disease involvement Visceral 669 21.9 28.4 0.59 (0.46, 0.76) Nonvisceral 322 33.9 1.05 (0.69, 1.61) Hazard ratio 0.2 0.5 1 2 5 NR, not reached. From confirmatory OS analysis; data cut-off July 31, 2012.

Overall Survival Subgroup Analyses
Cap + Lap T-DM1 Baseline characteristic Total n Median (mos) HR (95% CI) Better All patients 991 25.1 30.9 0.70 (0.56, 0.87) Age group <65 years 853 24.6 0.66 (0.52, 0.83) 65–74 years 113 27.1 NR 0.74 (0.37, 1.47) ≥75 years 25 11.1 3.45 (0.94, 12.65) ER and PR status ER+ and/or PR+ 545 25.3 31.9 0.62 (0.46, 0.85) ER– and PR– 426 23.7 0.75 (0.54, 1.03) Line of therapya First-line 118 27.9 0.61 (0.32, 1.16) Second-line 361 0.88 (0.61, 1.27) Third- and later-line 512 23.3 33.9 0.62 (0.46, 0.84) Hazard ratio 0.2 0.5 1 2 5 aDefined as any systemic therapy including endocrine and chemotherapy. NR, not reached. From confirmatory OS analysis; data cut-off July 31, 2012.

ORR and DOR in Patients with Measurable Disease
Objective response rate (ORR) Duration of response (DOR) Difference: 12.7% (95% CI, 6.0, 19.4) P=0.0002 Median, months (95% CI) Cap + Lap 6.5 (5.5, 7.2) T-DM1 12.6 (8.4, 20.8) 1.0 43.6% 50 0.8 40 30.8% 0.6 30 Proportion progression-free Patients, % 0.4 20 0.2 10 120/389 173/397 0.0 Cap + Lap T-DM1 Time (months) No. at risk Cap + Lap 120 105 77 48 32 14 9 8 3 3 1 1 T-DM1 173 159 126 84 65 47 42 33 27 19 12 8 2

Overview of Adverse Events
Cap + Lap (n=488) T-DM1 (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 within 30 days of last dose of study drug, n (%)a 4 (0.8) 1 (0.2) aCap + Lap: coronary artery disease, multi-organ failure, coma, and hydrocephalus; T-DM1: metabolic aencephalopathy.

Adverse Events Grade ≥3 AEs With Incidence ≥2%
Cap + Lap (n=488) T-DM1 (n=490) Adverse Event All Grades, % Grade ≥3, % Diarrhea 79.7 20.7 23.3 1.6 Hand-foot syndrome 58.0 16.4 1.2 0.0 Vomiting 29.3 4.5 19.0 0.8 Neutropenia 8.6 4.3 5.9 2.0 Hypokalemia 4.1 2.2 Fatigue 27.9 3.5 35.1 2.4 Nausea 44.7 2.5 39.2 Mucosal inflammation 19.1 2.3 6.7 0.2 Thrombocytopenia 28.0 12.9 Increased AST 9.4 22.4 Increased ALT 8.8 1.4 16.9 2.9 Anemia 8.0 10.4 2.7 ALT, alanine aminotransferase; AST, aspartate aminotransferase.

Cardiac Dysfunction Cap + Lap T-DM1 Cardiac dysfunction AEs,a n (%)
All grades Grade 3 (n=488) 15 (3.1) 2 (0.4) (n=490) 9 (1.8) 1 (0.2) Lowest post-baseline LVEF value, n (%) ≥45% ≥40 to <45% <40% (n=461) 454 (98.5) 4 (0.9) 3 (0.7) (n=482) 476 (98.8) 3 (0.6) LVEF <50% and ≥15-point decrease from baseline, n (%) (n=445) 7 (1.6) (n=481) 8 (1.7) aIncludes preferred terms ‘decreased ejection fraction’ and ‘left ventricular dysfunction’; Does not include cardiac AEs (e.g. myocardial infarction, atrial fibrillation).

Conclusions In the EMILIA study, T-DM1 achieved:
Significant improvement in PFS Median PFS: Cap + Lap 6.4 mos; T-DM1 9.6 mos HR=0.650; P<0.0001 Significant improvement in OS Median OS: Cap + Lap 25.1 mos; T-DM mos HR=0.682; P=0.0006 Key secondary efficacy endpoints including time to symptom progression1 were also significantly improved with T-DM1 The safety profile of T-DM1 was favorable to that of Cap + Lap T-DM1 should offer an important therapeutic option in the treatment of HER2-positive metastatic breast cancer 1Welslau et al. ESMO 2012, Poster 329P.

Thanks To the scientists To the investigators, clinicians and research staff at the 213 sites in 26 countries To all of the patients who participated in the trial and their families

Krop I, et al. EMILIA investigators, SABCS 2013

EMILIA Biomarker Analysis: PFS by HER2 mRNA Level and Treatment Arm
Baselga J, et al. EMILIA Investigators

EMILIA Biomarker Analysis: PFS by EGFR, HER3 and PTEN Expression

Cytotoxic Activity of T-DM1 in HER2-Positive Breast Cancer Cell Lines With PIK3CA Mutations

EMILIA Biomarker Analysis: PFS by PIK3CA Mutation Status Level and Treatment Arm

EMILIA Biomarker Analysis

T-DM1 for HER2-Positive MBC: Primary Results From TH3RESA, a Phase 3 Study of T-DM1 vs Treatment of Physician’s Choice H Wildiers,1 S-B Kim,2 A Gonzalez-Martin,3 PM LoRusso,4 J-M Ferrero,5 M Smitt,6 R Yu,6 A Leung,6 IE Krop7 1University Hospitals Leuven, Leuven, Belgium; 2Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea; 3Centro Oncológico MD Anderson International España, Madrid, Spain; 4Karmanos Cancer Institute, Wayne State University, Detroit, MI, USA; 5Department of Medical Oncology, Centre Antoine Lacassagne, Nice, France; 6Genentech, Inc, South San Francisco, CA, USA; 7Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA, USA 2013

Disclosure Slide Wildiers: advisory board for Roche
Kim: advisory board for Novartis, research grant from Novartis and Ildong Gonzalez-Martin: advisory board for Roche LoRusso: advisory board for Genentech, research grant from Genentech, speakers bureau for Genentech Ferrero: research grant from Roche and Novartis Smitt, Yu, Leung: Genentech employee, Roche stock Krop: research grant from Genentech

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

T-DM1: Background Phase 3 EMILIA study in patients with HER2-positive LABC and MBC previously treated with a taxane and trastuzumab1: Median PFS and OS were longer in the T-DM1 arm vs the capecitabine + lapatinib arm PFS: HR=0.65 (95% CI, 0.55, 0.77); P<0.001 OS: HR=0.68 (95% CI, 0.55, 0.85); P<0.001 Fewer grade ≥3 AEs were reported with T-DM1 vs capecitabine + lapatinib (41% vs 57%) Approved in several countries worldwide T-DM1 has not previously been studied in a randomized trial in patients who received prior treatment with both trastuzumab and lapatinib for advanced disease 1 Verma S, et al. N Engl J Med 2012. AE, adverse events; CI, confidence interval; HER2, human epidermal growth factor receptor 2; HR, hazard ratio; LABC, unresectable locally advanced breast cancer; MBC, metastatic breast cancer; PFS, progression-free survival; OS, overall survival.

TH3RESA Study Schema PD 2 Treatment of physician’s choice (TPC)b 1 PD
HER2-positive (central) advanced BCa (N=600) ≥2 prior HER2-directed therapies for advanced BC Prior treatment with trastuzumab, lapatinib, and a taxane T-DM1 3.6 mg/kg q3w IV (n=400) PD 2 Treatment of physician’s choice (TPC)b (n=200) 1 T-DM1c (optional crossover) PD Stratification factors: World region, number of prior regimens for advanced BC,d presence of visceral disease Co-primary endpoints: PFS by investigator and OS Key secondary endpoints: ORR by investigator and safety a Advanced BC includes MBC and unresectable locally advanced/recurrent BC. b TPC could have been single-agent chemotherapy, hormonal therapy, or HER2-directed therapy, or a combination of a HER2-directed therapy with a chemotherapy, hormonal therapy, or other HER2-directed therapy. c First patient in: Sep Study amended Sep 2012 (following EMILIA 2nd interim OS results) to allow patients in the TPC arm to receive T-DM1 after documented PD. d Excluding single-agent hormonal therapy. BC, breast cancer; IV, intravenous; ORR, objective response rate; PD, progressive disease; q3w, every 3 weeks.

Statistical Considerations
Final PFS analysisa (by investigator assessment) Targeted number of events: 324 80% power to detect HR=0.65; 2-sided alpha 0.5% 1st interim OS analysisa (prespecified to occur with final PFS analysis) Number of observed events: 105 Efficacy crossing boundary HR<0.363; P< Final OS analysis Targeted number of events: 492 80% power to detect HR=0.76; 2-sided alpha 4.5% a Clinical data cutoff: February 11, 2013.

Baseline Characteristics (1)
TPC (n=198) T-DM1 (n=404) Age, % <65 years 65–74 years ≥75 years 82.8 14.1 3.0 85.4 11.4 3.2 World region, % United States Western Europe Other 24.2 42.9 32.8 24.5 42.3 33.2 Race, % White Asian Othera 81.3 12.1 6.6 80.4 5.4 ECOG PS,b % 1 2 41.4 51.0 7.6 44.8 49.8 5.5 a Multi-racial patients are included in the Other category. b Two patients in the T-DM1 arm had missing ECOG PS scores: TPC, n=198; T-DM1, n=402. ECOG PS, Eastern Cooperative Oncology Group performance status.

Baseline Characteristics (2)
TPC (n=198) T-DM1 (n=404) ER and/or PR-positive, % 52.0 51.5 Visceral involvement, % 75.8 74.8 Disease extent at study entry, % Metastatic Unresectable locally advanced/recurrent BC 94.4 5.6 96.8 3.2 Number of prior regimens for advanced BC,a median (range) ≤3, % 4–5, % >5, % 4 (1–19) 39.4 32.8 27.8 4 (1–14) 32.6 37.1 30.3 Brain metastasis at baseline, % 13.6 9.9 a Two patients in the T-DM1 arm had missing information for prior treatment in the advanced BC setting: TPC, n=198; T-DM1, n=402. ER, estrogen receptor; PR, progesterone receptor.

TPC Treatment Category
(n=184a) Combination with HER2-directed agent, % Chemotherapyb + trastuzumab Lapatinib + trastuzumab Hormonal therapy + trastuzumab Chemotherapyb + lapatinib 83.2 68.5 10.3 1.6 2.7 Single-agent chemotherapy,b % 16.8 T-containing 80.4 a Includes patients who received study treatment. b The most common chemotherapy agents used were vinorelbine, gemcitabine, eribulin, paclitaxel, and docetaxel.

Study Discontinuation
TPC (n=198) T-DM1 (n=404) Discontinued study, % 36.9 21.0 Reasons for study discontinuation, % Death Withdrawal by patient Physician’s decision Other 22.2 13.1 1.0 0.5 15.1 4.7 0.7

PFS by Investigator Assessment
TPC (n=198) T-DM1 (n=404) Median (months) 3.3 6.2 No. of events 129 219 Stratified HR=0.528 (95% CI, 0.422, 0.661) P<0.0001 14 12 10 8 6 4 2 0.0 0.2 0.4 0.6 0.8 1.0 Proportion progression-free Time (months) 198 120 62 28 13 6 1 404 334 241 114 66 27 12 TPC T-DM1 No. at risk: Median follow-up: TPC, 6.5 months; T-DM1, 7.2 months. Unstratified HR=0.521 (P<0.0001).

PFS Subgroup Analyses (1)
By Investigator Assessment TPC T-DM1 T-DM1 Better TPC Better Baseline characteristic Total n Median (months) Median (months) n Event n Event HRa (95% CI) All patients 602 198 129 3.3 404 219 6.2 0.52 (0.42, 0.65) Age group <65 years 509 164 108 3.4 345 191 5.8 0.55 (0.44, 0.70) 65–74 years 74 28 17 3.2 46 25 6.9 0.42 (0.22, 0.80) ≥75 years 19 6 4 3.0 13 3 NE 0.14 (0.02, 0.79) World region United States 147 48 24 4.1 99 58 5.8 0.71 (0.44, 1.14) Western Europe 256 85 61 3.2 171 91 6.9 0.44 (0.32, 0.61) Other 199 65 44 3.1 134 70 5.8 0.53 (0.36, 0.78) Race White 488 161 104 3.4 325 177 6.3 0.50 (0.39, 0.64) Asian 81 24 17 2.8 57 30 5.4 0.63 (0.35, 1.14) Other 35 13 8 3.3 22 12 6.6 0.57 (0.23, 1.41) Baseline ECOG PS 262 82 48 3.6 180 84 7.0 0.44 (0.31, 0.64) 1 301 101 68 3.1 200 120 5.4 0.63 (0.47, 0.85) 2 37 15 13 1.6 22 13 6.9 0.41 (0.19, 0.92) 0.2 0.5 1 2 5 a Unstratified HR. NE, not estimable.

PFS Subgroup Analyses (2)
By Investigator Assessment TPC T-DM1 T-DM1 Better TPC Better Baseline characteristic Total n Median (months) Median (months) n Event n Event HRa (95% CI) All patients 602 198 129 3.3 404 219 6.2 0.52 (0.42, 0.65) ER and PR status ER+ and/or PR+ 311 103 66 3.9 208 109 5.9 0.56 (0.41, 0.76) ER– and PR– 270 85 58 2.9 185 105 6.0 0.51 (0.37, 0.71) Unknown 21 10 5 3.9 11 5 8.3 0.17 (0.03, 0.93) Disease involvement Visceral 452 150 95 3.4 302 168 6.2 0.56 (0.44, 0.72) Nonvisceral 150 48 34 3.1 102 51 6.7 0.41 (0.26, 0.64) Number of prior regimens for advanced BC ≤3 209 78 49 3.3 131 60 6.9 0.48 (0.32, 0.70) 4–5 214 65 45 3.7 149 83 6.2 0.58 (0.40, 0.83) >5 177 55 35 2.9 122 75 5.8 0.48 (0.32, 0.73) Brain metastasis at baseline Yes 67 27 16 2.9 40 24 5.8 0.47 (0.24, 0.89) No 535 171 113 3.6 364 195 6.2 0.53 (0.42, 0.66) 0.2 0.5 1 2 5 a Unstratified HR.

PFS for Patients Treated With Trastuzumab-Containing Regimens
TPC (T-containing) (n=149) T-DM1 (n=404) Median (months) 3.2 6.2 No. of events 101 219 Stratified HR=0.558 (95% CI, 0.437, 0.711) P<0.0001 14 12 10 8 6 4 2 0.0 0.2 0.4 0.6 0.8 1.0 Proportion progression-free Time (months) 149 99 50 20 12 5 1 404 334 241 114 66 27 TPC T-DM1 No. at risk: Unstratified HR=0.54 (P<0.0001).

First Interim OS Analysis
0.0 0.2 0.4 0.6 0.8 1.0 Proportion surviving Observed 21% of targeted events TPC (n=198) T-DM1 (n=404) Median (months) 14.9 NE No. of events 44 61 Stratified HR=0.552 (95% CI, 0.369, 0.826); P=0.0034 Efficacy stopping boundary HR<0.363 or P< 2 4 6 8 10 12 14 16 Time (months) No. at risk: TPC 198 404 169 381 125 316 80 207 51 127 30 65 9 30 3 7 T-DM1 44 patients in the TPC arm received crossover T-DM1 treatment after documented progression. Unstratified HR=0.57 (P=0.004).

ORR in Patients With Measurable Disease
By Investigator Assessment Difference: 22.7% (95% CI, 16.2, 29.2) P<0.0001 T-DM1 TPC 31.3% 8.6% 108/345 14/163 Patients, %

Overview of AEs TPC (n=184a) T-DM1 (n=403a) All-grade AEs, % 88.6 93.5
Grade ≥3 AEs,b % 43.5 32.3 AEs leading to treatment discontinuation,c % 10.9 6.7 AEs leading to dose reduction, % 19.6 9.4 LVEF <50% and ≥15% decrease from baseline,d % 1.1 1.5 a One patient randomized to the TPC arm received 2 cycles of T-DM1 by mistake; this patient was included in the T-DM1 group for safety analyses. b Grade 5 AEs: TPC, 1.6% (n=3); T-DM1, 1.2% (n=5). Three were considered related to T-DM1: hepatic encephalopathy, subarachnoid hemorrhage, and pneumonitis. One was considered related to TPC: noncardiogenic pulmonary edema. c For any study drug. d No patient experienced an LVEF <40%. LVEF, left ventricular ejection fraction.

Grade ≥3 AEs With Incidence ≥2% in Either Arma
TPC (n=184) T-DM1 (n=403) Any grade Grade ≥3 Nonhematologic AEs, % Diarrhea 21.7 4.3 9.9 0.7 Abdominal pain 12.5 2.7 6.5 1.2 AST increased 5.4 2.2 8.4 Fatigue 25.0 27.0 2.0 Asthenia 15.8 15.6 1.0 Cellulitis 3.3 0.5 Pulmonary embolism Dyspnea 9.2 1.6 Hematologic AEs, % Neutropenia 5.5 2.5 Febrile neutropenia 3.8 0.2 Anemia 10.3 8.9 Leukopenia 6.0 Thrombocytopenia 15.1 4.7b a Medical Dictionary for Regulatory Activities (MedDRA) preferred term. b Grade 5 subarachnoid hemorrhage was reported for 1 patient with grade 4 thrombocytopenia; grade 4 tumor hemorrhage was reported for 1 patient with grade 3 thrombocytopenia. The incidence of grade ≥3 hemorrhage of any type was 2.2% (T-DM1) and 0.5% (TPC). AST, aspartate aminotransferase. Highlighting indicates grade ≥3 AEs with >3% difference between the TPC and T-DM1 arms.

Conclusions T-DM1 demonstrated improved efficacy and safety compared with TPC Significant improvement in PFS HR=0.528; P<0.0001 A clear and consistent treatment effect across subgroups Interim OS favored T-DM1 but efficacy stopping boundary not crossed HR=0.552; P=0.0034 Safety and ORR favored T-DM1 Fewer grade ≥3 AEs with T-DM1 vs TPC: 32.3% vs 43.5% Fewer discontinuations and dose reductions due to AEs with T-DM1 ORR 31.3% vs 8.6%, P<0.0001 These data reaffirm the results from the EMILIA study, demonstrating a consistent benefit with T-DM1 in patients with previously treated HER2-positive advanced BC

Thanks To all of the patients who participated in the trial and their families, as well as the participating study sites

Trastuzumab-emtansine: A very nice poison

Similar presentations

Presentation on theme: "Trastuzumab-emtansine: A very nice poison"— Presentation transcript:

Similar presentations

About project

Feedback

Log in

Auth with social network:

Trastuzumab-emtansine: A very nice poison

Similar presentations

Presentation on theme: "Trastuzumab-emtansine: A very nice poison"— Presentation transcript:

Similar presentations

About project

Feedback