William J. Gradishar MD FACP FASCO

Transcription

1 Northwestern University Feinberg School of Medicine Highlights of Non-Endocrine Adjuvant Therapy SABCS 2018 William J. Gradishar MD FACP FASCO Betsy Bramsen Professor of Breast Oncology Chief- Hematology/Oncology Robert H. Lurie Comprehensive Cancer Center

2 Disclosures Roche/Genentech-Scientific Advisory Board, Independent Data Monitoring Committee Seattle Genetics- Data Monitoring Committee

3 Agenda Adjuvant Chemotherapy in the Elderly Sinha et al. Short course adjuvant HER2 therapy?- Pivot et al Residual disease post neoadjuvant HER2 therapy: Katherine- Geyer et al

4 Efficacy and Utilization trends of adjuvant chemotherapy for stage I, II, and III breast cancer in the elderly population: A National Cancer Database Analysis Shreya Sinha MD, Lauren Panebianco MD, Xiancheng Wu BS, Dongliang Wang PhD, Danning Huang PhD, Abirami Sivapiragasam MD SUNY Upstate Medical University Syracuse, New York

5 Background: Estimated new cases diagnosed in ,120 (15.3% of all cancers) Around 90% of them are diagnosed with local or regional disease Median age at diagnosis is 62 Standard of Care: Surgery with neoadjuvant or adjuvant chemotherapy. Followed by RT +/- hormonal therapy Data for elderly patient is limited in clinical trials

6 Background: Adjuvant chemo in the elderly

7 Background: CALGB Patients age >65 yo with stage I-III breast cancer (n = 600) Randomized to CMF or AC VS. Capecitabine Non-inferiority study Result: Standard adjuvant chemotherapy is superior to Capecitabine Muss HB, Berry DA et.al. Adjuvant Chemotherapy in Older Women with Early-Stage breast Cancer. N Engl J Med May 14; 360(20):

8 Background: ICE trial: Phase III trial presented at SABCS 2014 Patient population yo (n = 1,358) Capecitabine + Ibandronate VS. Ibandronate alone Result: No difference in DFS or OS 2017 Retrospective study by Yan-Shuang et.al. Patients >60 yo (n = 251) Adjuvant Capecitabine vs no chemotherapy Result: DFS was improved with adjuvant therapy No difference in OS 1. Von Monckwitz G et.al. S3-04; General Session 3 2. Li, Yan-Shuang et.al. Mol Clin Oncol Oct; 7(4)

9 Objectives:

10 Method: Source of Data: National Cancer Database (NCDB) Clinical oncology database Sourced from hospital registry data, collected from >1,500 Commissionon-cancer-accredited facilities. It is used to analyze and track patients with malignant neoplastic diseases, their treatment, and outcomes. US based database Inpatient and outpatient information Incorporates information for >70% of newly diagnosed cancers

11 Methods: Patient Selection -

12 Methods: Total patients analyzed: 2,445,730 Age <65: 1,468,499 Stage 0/ IV, missing info.: 977,231 Included: 160,676 With adjuvant Therapy: 97,128 (60.45%) With no adjuvant Therapy: 63,548 (39.55%)

13 Factors associated with utilization: Odds Ratio (adjusted) Estimates and Wald Confidence Intervals Variables Odds Ratio 95% confidence Limits P-value Grade 2 vs 1 / 3 vs / / /3.65 <0.0001/< ER-/PR- vs ER+/PR < ER+/PR- vs ER+/PR < Stage 1 vs 3 / 2 vs / / /0.59 <0.0001/< < Race: Black vs white Private vs Medicaid < Private vs. Medicare < Lumpectomy vs Mastectomy RT vs No RT < Academic vs Community program CDCC Score >=3 vs <0.0001

14 Average Age: Average age with chemotherapy: 70.7 yo Average age without chemotherapy: 75.5 yo Age Range Percent population yo 46.5% yo 24.6% yo 15.4% yo 8.8% 4.8%

15 Median Overall Survival Data: With Chemo (in months) Without Chemo (in months) P-value Overall: < ER+/PR < ER+/PR < ER-/PR < Stage < Stage < Stage < HR (adjusted)

16 Results: P-value: < HR: OS at 120 months: 59.5% 46.7%

17 OS based on Stage: Stage I P-value: < HR: Stage II P-value: < HR: Stage III P-value: < HR: 0.666

18 OS based on hormonal status: ER+/PR+ P: < HR: ER+/PR- P: < HR: ER-/PR- P: < HR: 0.547

19 Limitations: Hospital based data - retrospective study HER-2 population not well defined Data not available from before 2010 Study population does not represent everyone: COC-accredited programs only Genotype Prediction scores were not taken into account

20 Tailoring your therapy: Chemotherapy toxicity prediction calculator Cancer and Aging Research Group (CARG) Moffitt Cancer Center Senior Adult Oncology Program Tools Genetic-based assays OncoType Dx Mammaprint

21 Conclusions: Factors predicting utilization of chemotherapy included a higher grade, CDCC score, stage, administration of radiation, hormonal status, facility type, having private insurance Race had no bearing on it Getting adjuvant chemotherapy is associated with increased overall survival in patients with stage I III breast cancer Across all stage and hormonal status Geriatric Assessment tools have shown accuracy in predicting the risk of chemotherapy

22 How long should trastuzumab be given in the adjuvant setting?

23 HERA TRIAL DESIGN Accrual (n=5102) Women with locally determined HER2-positive invasive early breast cancer Surgery + (neo)adjuvant CT ± RT Centrally confirmed IHC 3+ or FISH+ and LVEF 55% Randomization OBSERVATION n=1698 After ASCO 2005, option of switch to Trastuzumab CT, chemotherapy; RT, radiotherapy 1 year Trastuzumab 8 mg/kg 6 mg/kg 3 weekly schedule n= years Trastuzumab 8 mg/kg 6 mg/kg 3 weekly schedule n=1701 M Piccart SABCS 2012

24 OS FOR 2 YEARS VS. 1 YEAR TRASTUZUMAB AT 8 YRS MFU Overall Survival (%) % 92.6% 86.4% 96.5% 91.4% 87.6% Trastuzumab 2 years Trastuzumab 1 year Pts Events HR (2 vs 1) 95% CI p-value 2 years ( ) year Years from randomization No. at risk Trastuzumab 2 years Trastuzumab 1 year

25 Trials Exploring Shorter Duration Trastuzumab Trial Duration (months) Chemotherapy Start Accrual Status PHARE 6 vs 12 Investigator Choice 5/ Completed 5/2011 Hellenic Oncology 6 vs 12 ddfec/d 10/ Completed 5/2012 Short-HER 3 vs 12 A=T vs T+FEC 12/ Completed 9/2013 SOLD 2 vs 12 T+FEC 1/ Completed 11/2014 Persephone 6 vs 12 Investigator Choice 10/ Completed 7/2015

26 Optimal Duration of Trastuzumab PHARE Study Non-inferiority study; pre-specified HR 1.15 n = 3,380 HR = 1.28 (p = 0.29) 2 yr DFS: Tras 12 mos = 93.8% Tras 6 mos = 91.1% 6 months not non-inferior to 12 months of trastuzumab Short-HER Non-inferiority study, pre-specified HR yrs f follow-up N = HR = 1.15 ( ); 0.78 probability 5yr DFS (87.5% LONG vs. 85.4% SHORT) Months from randomization Number at risk 627 B short A long B short 9 weeks not non-inferior to 12 months of trastuzumab

27 Persephone Study Design: 6 vs 12 months trastuzumab Earl H et al, ASCO 2018

28 Persephone: Disease-Free Survival Disease-free survival 6 months is non-inferior to 12 months of trastuzumab Earl H et al, ASCO 2018

29 Persephone: Cardiotoxicity Stopped trastuzumab due t cardiotoxicity: 8% in 12 mo arm 4% in 6 mo arm 6 mo patients had a more rapid recovery of EF (p=0.0 Earl H et al, ASCO 2018

30 Treatment approach different than in the modern era (sequential tx, non-taxane regimens, no dual-her2 treatment) makes it challenging to use these data to change practice Does however provide comfort if we need to discontinue therapy early due to cardiac toxicity or other reasons Persephone: Subgroup Analysis DFS:<br /><br />Pre-defined subgroup analysis

31 Protocol of Herceptin Adjuvant with Reduced Exposure PHARE* randomized trial final results comparing 6 to 12 months of trastuzumab in adjuvant early breast cancer Xavier Pivot, Gilles Romieu, Marc Debled, Jean-Yves Pierga, Pierre Kerbrat, Thomas Bachelot, Alain Lortholary, Marc Espié, Pierre Fumoleau, Daniel Serin, Jean-Philippe Jacquin, Christelle Jouannaud, Maria Rios, Sophie Abadie-Lacourtoisie, Laurence Venat-Bouvet, Laurent Cany, Stéphanie Catala, David Khayat, Laetitia Gambotti, Iris Pauporté, Celine Faure- Mercier, Sophie Paget, Julie Henriquez, Jean-Marie Grouin. *lighthouse in French

32 PHARE* Study design trastuzumab up to 12 months 1690 patients trastuzumab 6 months Stratification 1. ER pos / neg 2. Chemo: concurrent/sequential R stop trastuzumab 1690 patients Clinical exam LVEF mos Mammography Up to 60 mos * Protocol of Herceptin Adjuvant with Reduced Exposure Pivot X, et al, Lancet Oncol 14: , 2013

33 Statistical Methods Non inferiority randomized trial 2% variation in terms of absolute difference of recurrence The 95% CI HR margins should not cross the 1.15 boundary HR were estimated from the stratified Cox model

34 Primary endpoint scenarii A Equivalent B 6 month Superior C PHARE trial D 6 month Non Inferior E HR 6 month Inferior Pivot X, et al, Lancet Oncol 14: , 2013

35 Statistical Methods First analysis Database locked: 31/07/2012 based on the recommendations of the IDMC: to continue the patients follow-up till July 2012, which corresponds to 4 years of enrolment and a minimum of 2 years of follow-up, before releasing the data. The margin for non-inferiority is not changed.

36 Statistical Methods Final analysis Database locked: 14/12/2017 Based on an amended statistical plan Preservation of the 1.15 margin for non-inferiority, 680 DFS events are required at the 5% level of significance to conclude to non-inferiority.

37 Patient Characteristics 12 months n= months n=1690 Age, median (range) 54 (21 86) 55 (23 85) Tumor: Nodes: 0 2 cm 2 5 cm 5 negative 1 3 nodes 4 nodes 44.9% 44.4% 10.8% 55.4% 30.0% 14.6% 42.5% 45.6% 11.9% 54.7% 30.2% 15.1% Positive Estrogen receptor 57.7% 58.9% Inflammatory disease 3.5% 3.4% SBR: I II III 3.1% 41.0% 55.9% 3.3% 40.9% 55.8%

38 Treatment Characteristics Type of Chemotherapy No Anthracyclines Anthracyclines no taxanes Anthracyclines and Taxanes Concomitant Chemotherapy Sequential Chemotherapy 12 months n= % 15.9% 73.9% 57.5% 42.5% 6 months n= % 15.5% 72.7% 56.9% 43.1% Radiotherapy 87.7% 88.2% Hormonotherapy 50.6% 50.2% Trastuzumab duration, mean (sd) 11.8 (2.03) 6.3 (1.46)

39 DFS Events 7.5 years median Follow-up 12 mos (n=1690) 6 mos (n=1690) DFS Events (n=704) 20.4% 21.2% Local Recurrence Regional Recurrence Distant Recurrence Controlateral Breast Cancer 2 nd Primary Malignancy Death 2.1% 1% 9.6% 1.6% 4.6% 1.4% 2.5% 1.1% 11.1% 2% 3.6% 1.1%

40 1 : Adjusted for stratification factors : Oestrogen receptor status (+ vs -) and chemotherapy (concomitant vs sequential) Disease Free Survival HR: 1.08 (95%CI: ) Non-inferiority margin = 1.15 p=0.39

41 Subgroup effects

42 Disease Free Survival

43 Disease Free Survival Per protocol analysis 1 : Adjusted for stratification factors : Oestrogen receptor status (+ vs -) and chemotherapy (concomitant vs sequential) 2 : adjusted for Oestrogen-receptor status (+ vs -), chemotherapy (concomitant vs sequential), Age ( < 50 vs 50), Tumour size ( < 2 vs 2 cm), progesterone-receptor status (+ vs -), nodal status (+ vs -).

44 Overall Survival 1 : Adjusted for stratification factors : Oestrogen receptor status (+ vs -) and chemotherapy (concomitant vs sequential) 2 : adjusted for Oestrogen-receptor status (+ vs -), chemotherapy (concomitant vs sequential), Age ( < 50 vs 50), Tumour size ( < 2 vs 2 cm), progesterone-receptor status (+ vs -), nodal status (+ vs -).

45 Conclusion PHARE failed to show that 6 months of trastuzumab is non inferior to 12 months HR: 1.08 (95%CI: ) p= PERSEPHONE Showed that 6 months of trastuzumab is non inferior to 12 months HR: 1.07 (90%CI: ) p=

46 Phase III Study of Trastuzumab Emtansine (T-DM1) vs Trastuzumab as Adjuvant Therapy in Patients with HER2-Positive Early Breast Cancer with Residual Invasive Disease after Neoadjuvant Chemotherapy and HER2-Targeted Therapy Including Trastuzumab: Primary Results from KATHERINE (NSABP B-50-I, GBG 77 and Roche BO27938) ct1-4/n0-3/m0 at presentation (ct1a-b/n0 excluded) Centrally confirmed HER2-positive breast cancer Neoadjuvant therapy must have consisted of Minimum of 6 cycles of chemotherapy Minimum of 9 weeks of taxane Anthracyclines and alkylating agents allowed All chemotherapy prior to surgery Minimum of 9 weeks of trastuzumab Second HER2-targeted agent allowed Residual invasive tumor in breast or axillary nodes Randomization within 12 weeks of surgery R 1:1 N=1486 T-DM1 3.6 mg/kg IV Q3W 14 cycles Trastuzumab 6 mg/kg IV Q3W 14 cycles Radiation and endocrine therapy per protocol and local guidelines Stratification factors: Clinical presentation: Inoperable (stage ct4 or cn2 3) vs operable (stages ct1-3n0-1) Hormone receptor: ER or PR positive vs ER negative and PR negative/unknown Preoperative therapy: Trastuzumab vs trastuzumab plus other HER2-targeted therapy Pathological nodal status after neoadjuvant therapy: Positive vs negative/not done

47 San Antonio Breast Cancer Symposium December 4 8, 2018 Statistical Considerations Primary Endpoint: IDFS Final IDFS IDFS analyses Adjuvant trastuzumab assumed to provide 3-year IDFS of 70% and improvement with T-DM1 to 76.5% (HR=0.75) would be clinically meaningful To provide 80% power with 2-sided alpha 5%, 384 events required Single Pre-specified Interim Analysis of IDFS When 67% of events (~257) occurred Early reporting boundary: HR < or P < First Interim OS analysis Performed at time of interim IDFS analysis IF boundary is crossed The overall type I error controlled at 0.05 using the Lan DeMets alpha spending function with O Brien Fleming boundary

48 San Antonio Breast Cancer Symposium December 4 8, 2018 Study Population Trastuzumab (n=743) T-DM1 (n=743) Randomized ITT (efficacy analyses) Treated (safety analyses) Median duration of follow up (months) First patient in: Last patient in: Clinical data cutoff: IDMC review of interim analysis: 03 April, December, July, September, 2018

49 San Antonio Breast Cancer Symposium December 4 8, 2018 Baseline Characteristics of ITT Population (1) Baseline Characteristics of ITT Population (1) Trastuzumab (n=743) T-DM1 (n=743) Median age (range), years 49 (23 80) 49 (24 79) <40 years, n (%) 153 (20.6) 143 (19.2) years, n (%) 522 (70.3) 542 (72.9) 65 years, n (%) 68 (9.2) 58 (7.8) Race, n (%) White 531 (71.5) 551 (74.2) Asian 64 (8.6) 65 (8.7) American Indian^ or Alaska Native 50 (6.7) 36 (4.8) Black or African American 19 (2.6) 21 (2.8) Multiple/Unknown/Other 79 (10.6) 70 (9.4) Region, n (%) North America 164 (22.1) 170 (22.9) Western Europe 403 (54.2) 403 (54.2) Rest of world 176 (23.7) 170 (22.9) Prior anthracycline, n (%) 564 (75.9) 579 (77.9) ^Includes North, Central, and South American Indians.

50 San Antonio Breast Cancer Symposium December 4 8, 2018 Characteristics of ITT Population (2): Stratification Factors Trastuzumab (n=743) T-DM1 (n=743) Clinical stage at presentation, n (%) Operable (stages ct1-3n0 1M0) 553 (74.4) 558 (75.1) Inoperable (stage ct4nxm0 or ctxn2 3M0) 190 (25.6) 185 (24.9) Hormone receptor status, n (%) ER and/or PgR positive 540 (72.7) 534 (71.9) ER negative and PgR negative/unknown 203 (27.3) 209 (28.1) Preoperative HER2-targeted therapy, n (%) Trastuzumab alone 596 (80.2) 600 (80.8) Trastuzumab plus additional HER2-targeted agent(s)^ - Trastuzumab plus pertuzumab * Pathologic nodal status after preoperative therapy, n (%) 147 (19.8) 139 (18.7) 143 (19.2) 133 (17.9) Node positive 346 (46.6) 343 (46.2) Node negative/not done 397 (53.4) 400 (53.8) ^Non-pertuzumab HER2-targeted agents included: neratinib, dacomitinib, afatinib, lapatinib. * Not a stratification factor, included for informational purposes.

51 San Antonio Breast Cancer Symposium December 4 8, 2018 Baseline Characteristics of ITT Population (3) Baseline Characteristics of ITT Population (3) Trastuzumab (n=743) T-DM1 (n=743) Primary tumor stage (at definitive surgery)^, n (%) ypt0, ypt1a, ypt1b, ypt1mic, yptis 306 (41.2) 331 (44.5) ypt1/ypt1c 184 (24.8) 175 (23.6) ypt2 185 (24.9) 174 (23.4) ypt3, ypt4 67 ( 9.0) 63 ( 8.5) Regional lymph node stage (at definitive surgery), n (%) ypn0 335 (45.1) 344 (46.3) ypn1 213 (28.7) 220 (29.6) ypn2, ypn3 133 (17.9) 123 (16.6) ypnx 62 ( 8.3) 56 ( 7.5) Residual invasive disease 1 cm or less AND negative axillary nodes (ypt1a, ypt1b or ypt1mic and ypn0) ^One patient in the trastuzumab arm was reported as yptx; Five patients had ypt1 disease without further subspecification. 161 (21.7) 170 (22.9)

52 San Antonio Breast Cancer Symposium December 4 8, 2018 Invasive Disease-Free Survival 100 Invasive Disease-Free Survival Rate (%) Trastuzumab T-DM1 Trastuzumab T-DM1 (n=743) (n=743) IDFS Events, no. (%) 165 (22.2) 91 (12.2) Unstratified HR=0.50 (95% CI, ) P< year IDFS 77.0% 88.3% 0 No. at Risk Trastuzumab T-DM Time (months)

53 San Antonio Breast Cancer Symposium December 4 8, 2018 IDFS Subgroup Analysis (1) Trastuzumab (n=743) T-DM1 (n=743) Group Total N 3-Year IDFS 3-Year IDFS Hazard Ratio 95% CI T-DM1 Better Trastuzumab Better All Clinical stage at presentation Operable Inoperable Hormone receptor status Negative (ER negative and PgR negative/unknown) Positive (ER and/or PgR positive) Preoperative HER2-directed therapy Trastuzumab alone Trastuzumab plus additional HER2-directed agent(s) Pathological nodal status after preoperative therapy Node positive Node negative/not done Age group (years) < Race^ White Asian American Indian or Alaska Native Black or African American ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ^149 were of multiple races or unknown race

54 San Antonio Breast Cancer Symposium December 4 8, 2018 IDFS Subgroup Analysis (2) IDFS Subgroup Analysis (2) Group Total N Trastuzumab (n=743) 3-Year IDFS T-DM1 (n=743) 3-Year IDFS Hazard Ratio 95% CI T-DM1 Better Trastuzumab Better All Primary tumor stage (at definitive surgery) ypt0, ypt1a, ypt1b, ypt1mic, yptis ypt1, ypt1c ypt2 ypt3 ypt4^ Regional lymph node stage (at definitive surgery) ypn0 ypn1 ypn2 ypn3 ypnx Residual disease 1 cm with negative axillary lymph nodes ypt1a, ypt1b or ypt1mic and ypn ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) Central HER2 status by IHC * 0/ < (0.00 NE) ( ) ( ) ^Includes all ypt4 and 1 patient with yptx; * Three patients had unknown HER2 IHC status

55 San Antonio Breast Cancer Symposium December 4 8, 2018 First IDFS Events Trastuzumab T-DM Patients (%) Total patients with IDFS event^ 11.6 CNS* (4.3) CNS* (5.9) Distant recurrence Locoregional recurrence Contralateral breast cancer Death without prior event ^Patients who experience additional IDFS event(s) within 61 days of their first IDFS event are reported in the category according to the following hierarchy: [1] Distant recurrence; [2] Locoregional recurrence; [3] Contralateral breast cancer; [4] Death without prior event. *CNS metastases as component of distant recurrence (isolated or with other sites). Trastuzumab T-DM1

56 San Antonio Breast Cancer Symposium December 4 8, 2018 Distant Recurrence 100 Distant Recurrence Distant Recurrence-Free Rate (%) Trastuzumab T-DM1 Trastuzumab T-DM1 (n=743) (n=743) Events, no. (%) 121 (16.3) 78 (10.5) Unstratified HR=0.60 (95% CI, ) 3-year event-free rate 83.0% 89.7% 0 No. at Risk Trastuzumab T-DM Time (months)

57 San Antonio Breast Cancer Symposium December 4 8, 2018 Overall Survival 100 Overall Survival 80 Survival (%) Trastuzumab T-DM1 Trastuzumab T-DM1 (n=743) (n=743) Events, no. (%) 56 (7.5) 42 (5.7) Unstratified HR=0.70 (95% CI, ) P= No. at Risk Trastuzumab T-DM Time (months) Boundary for statistical significance HR<0.43 or P<

58 San Antonio Breast Cancer Symposium December 4 8, 2018 Exposure to Study Treatment Exposure to Study Treatment Cycles of trastuzumab/t-dm1 completed, n (%) Trastuzumab (n=720) T-DM1 (n=740) 7 cycles 664 (92.2) 637 (86.1) 14 cycles 583 (81.0) 528 (71.4) Patients with a dose reduction, n (%) No dose reduction N/A 634 (85.7) One dose level reduction (3.0 mg/kg) N/A 77 (10.4) Two dose level reductions (2.4 mg/kg) N/A 29 (3.9)

59 San Antonio Breast Cancer Symposium December 4 8, 2018 Safety Overview Safety Overview Number of patients with at least one, n (%) Trastuzumab n=720 T-DM1 n=740 Grade 3 AEs 111 (15.4) 190 (25.7) Serious AEs 58 (8.1) 94 (12.7) AE leading to treatment discontinuation 15 (2.1) 133 (18.0) AE with fatal outcome^ 0 1 (0.1) ^Fatal AE was intracranial hemorrhage diagnosed after a fall with platelet count of 55,000.

60 San Antonio Breast Cancer Symposium December 4 8, 2018 AEs Leading to Treatment Discontinuation ( 1% Incidence Either Arm) AEs Leading to Treatment Discontinuation ( 1% Incidence Either Arm) Trastuzumab n=720 T-DM1 n=740 Patients discontinuing due to adverse events 15 (2.1%) 133 (18.0%) Platelet count decreased 0 31 ( 4.2%) Blood bilirubin increased 0 19 ( 2.6%) Aspartate aminotransferase (AST) increased 0 12 (1.6%) Alanine aminotransferase (ALT) increased 0 11 (1.5%) Peripheral sensory neuropathy 0 11 (1.5%) Ejection fraction decreased 10 (1.4%) 9 ( 1.2%)

61 San Antonio Breast Cancer Symposium December 4 8, 2018 All Grade AEs 15% Incidence in Either Arm All Grade AEs 15% Incidence in Either Arm Trastuzumab grade 1 60 Trastuzumab grade 2 Trastuzumab grade 3 50 T-DM1 grade 1 T-DM1 grade 2 T-DM1 grade Patients (%) Fatigue Nausea Platelet count decreased 6 5 AST increased Headache Arthralgia Radiation skin injury ALT increased Epistaxis Sensory neuropathy Constipation Myalgia

62 San Antonio Breast Cancer Symposium December 4 8, 2018 KATHERINE Summary and Conclusions KATHERINE Summary and Conclusions Adjuvant T-DM1 demonstrated both a statistically significant and clinically meaningful improvement in IDFS compared with trastuzumab Unstratified HR=0.50; 95% CI ; P< year IDFS rate improved from 77.0% to 88.3% (difference=11.3%) Benefit of T-DM1 was consistent across all key subgroups including HR status, extent of residual invasive disease, and single or dual HER2-targeted neoadjuvant therapy The safety data were consistent with the known manageable toxicities of T-DM1, with expected increases in AEs associated with T-DM1 compared to trastuzumab Additional follow-up will be necessary to evaluate the effect of T-DM1 on OS The KATHERINE data will likely form the foundation of a new standard of care in this population and increase the use of neoadjuvant therapy in HER2-positive EBC

63 San Antonio Breast Cancer Symposium December 4 8, 2018

64 San Antonio Breast Cancer Symposium December 4 8, 2018 The Standard of Care Has Changed: T-DM1 should be recommended to the vast majority of patients with residual disease after a taxane-based neoadjuvant regimen