Annual Scientific Report 2016

Heilung durch Innovation, Kompetenz und Partnerschaft

Annual Scientifi c Report 2016 Annual Scientific Report 2016

Heilung durch Innovation, Kompetenz und Partnerschaft Annual Scientific Report 2016 4 | Annual Scientific Report 2016 | 5

Index

Introduction 7

New Study Concepts 35 GBG 93: PADMA 36 GBG 94: PATINA 38 GBG 95: ULTIMATE 40

Recruiting Studies 43 GBG 89: GeparNuevo 44 GBG 88: GeparX 46 GBG 90: GeparOla 48 GBG 87: PALLAS 50 GBG 82: OLYMPIA 52 GBG 78: PenelopeB 54 GBG 68: GAIN-2 56 GBG 54: MALE 59 GBG 29: Breast Cancer in Pregnancy (BCP) 61 GBG 79: Brain Metastases in Breast Cancer (BMBC) 63 GBG 86: DESIREE 65 GBG 75: INSEMA 67

Studies in Follow-up 71 GBG 34: IBIS-II 72 GBG 53: PANTHER 74 GBG 59: SOLE 76 GBG 67: APHINITY 78 GBG 69: GeparSepto 80 GBG 77: Katherine 83 GBG 81: Brightness 85 GBG 84: GeparOcto 87

Completed Studies 91 GBG 74: GENEVIEVE 92 GBG 80: SenSzi 94

Follow-up Activities 97

Translational Research 101

GBG Studienfinder 106

Notes 108 6 | Annual Scientific Report 2016 | Introduction | 7

Introduction

1. About the German Breast Group 8 2. Infrastructure of the German Breast Group 9 3. Cooperations with other study groups 10 4. Publications in 2016 11 4.1. Peer-reviewed articles in 2016 11 4.2. Peer-reviewed reviews in 2016 13 4.3. Congress contributions in 2016 13 4.4. GBG-Publications Grading System 17 4.5. Standard Operating Procedure (SOP) for Authorship 18 4.6. Oral and poster presentations 19 8 | Annual Scientific Report 2016 | Introduction

Introduction

Headquarters: The main focus of the GBG is on the investigator initiated trials (IIT). These are clinical studies GBG Forschungs GmbH based on the work of doctors conducting Martin-Behaim-Strasse 12 research and are focused on the optimization of 63263 Neu-Isenburg therapy and the overall improvement of its GERMANY quality, unlike industrial studies which are Phone: +49 6102 7480-0 typically affected by approval and marketing Fax: +49 6102 7480-440 aspects. www.GBG.de The GBG currently manages over 40 clinical 1. About the German Breast Group trials. All services provided by GBG are to the highest standard of the International Conference The German Breast Group (GBG), a leading on Harmonisation of Good Clinical Practice cooperative study group in the field of breast (ICH-GCP1998) and if necessary regulatory cancer in Germany, provides the comprehensive requirements. We offer a comprehensive range management of clinical trials in all major of services, including: therapeutic categories: prevention, neoadjuvant, adjuvant, and palliative. The vision of the GBG is • Idea and Conception of Study Design best described as healing by innovation, • Clinical Project Management competence and partnership, from the protocol • Clinical Monitoring design and feasibility assessments to the final • Data Management study report. Through project management in • Biometric and Statistics combination with the expert data management • External Documentation and statistical analyses, the GBG delivers • Translational Research consistent high-quality results in order to • Biobanking improve treatment therapies of cancer patients • Pathological Central Laboratory and their quality of life. • Continuous Medical Education • Medical Writing • Sponsorship • Quality Control

Pharma Independent Data Operations: and Safety Monitoring GBG Committee (IDMC) Forschungs GmbH Staffmeetings

Scientific Board / Retreat

Five Subboards: neoadjuvant, adjuvant, palliative, operative and translational research

Participating Sites

Figure 1: Structure of the German Breast Group | 9

Introduction

2. Infrastructure of the Subboards German Breast Group Five subboards were active during the last year in the fields of neoadjuvant, adjuvant, palliative, Participating sites and operative therapy as well as in the field of Participating sites are actively recruiting sites. translational research. Members of the subboards An official membership is not required, however are all well-known professionals, experienced in any physician who takes part in our trials auto treating breast cancer patients and active in the matically becomes a member of the study field of breast cancer research and clinical studies. group. In the year 2016, the GBG counted around When a subboard decides to launch a new study, 600 German and 600 international Sites sites the GBG Forschungs GmbH plans, organizes and and 1000 investigators, who are currently manages the study, in line with the GBG’s belief working on facilitating their patient’s entry into that a clinical study must be directly related to the optimal GBG trial. Traditionally, most of our the potential improvement of the therapy and its investigators work in gynecological institutions benefits for the patient. Thus, a strict quality such as university clinics, general hospitals, monitoring is essential and is ensured by following specialist practices and general practices. For the GBG in-house standard operating procedures several years an increasing number of (SOP). The members of the subboards meet once gynecologic and medical oncologists have been a year face-to-face and 3 times via telephone taking part in our trials, thus enriching the trial conferences. Our subboards have been active conception with their knowledge. discussing current studies, research results and further innovative study designs. Recruitment of patients Patients are recruited through the participating The members of our subboards in 2016 are sites which provide detailed information on the shown below: GBG studies to the patient. This way, all existing uncertainties are clarified and an absolute Neoadjuvant transparency on the conduct of clinical trials can Prof. Dr. J. U. Blohmer, Berlin be ensured. Patients are treated according to the Prof. Dr. C. Denkert, Berlin latest scientific findings and are carefully Prof. Dr. P. Fasching, Erlangen controlled and monitored. Thanks to the clinical Dr. C. Hanusch, München trials, breast cancer therapies are nowadays Prof. Dr. J. Huober, Ulm carried out on the highest possible standard. Prof. Dr. Ch. Jackisch, Offenbach The annual patient recruitment is shown in PD Dr. S. Kummel, Essen figure 2. Prof. Dr. S. Loibl, Neu-Isenburg Prof. Dr. G. von Minckwitz, Neu-Isenburg PD Dr. K. Rhiem, Köln 6000 Number of patients Prof. Dr. A. Schneeweiss, Heidelberg

5000 Prof. Dr. M. Untch, Berlin

4000 Adjuvant Prof. Dr. W. Janni, Ulm Prof. Dr. S. Loibl, Neu-Isenburg 3000 Prof. Dr. F. Marme, Heidelberg 2000 Prof. Dr. G. von Minckwitz, Neu-Isenburg Prof. Dr. V. Möbus, Frankfurt am Main 1000 Prof. Dr. T. Reimer, Rostock 0 2011 2012 2013 2014 2015 2016 Dr. M. Reinisch, Essen Dr. S. Schmatloch, Kassel Figure 2: Annual recruitment of patients PD Dr. B. Sinn, Berlin Prof. Dr. E. Stickeler, Freiburg Prof. Dr. M. Untch, Berlin 10 | Annual Scientific Report 2016 | Introduction

Palliative Staff Meetings Dr. J. Barinoff, Frankfurt am Main Staff meetings are conducted on a regular basis, PD Dr. T. Decker, Ravensburg either at the GBG headquarters or via telephone Prof. Dr. S. Loibl, Neu-Isenburg conferences, to ensure sufficient information Dr. K. Lübbe, Hannover transfer between the responsible study project Prof. Dr. C. Mundhenke, Kiel managers, study chairs and representatives of Prof. Dr. V. Müller, Hamburg the supporting pharmaceutical companies. PD Dr. M. Schmidt, Mainz Prof. Dr. G. von Minckwitz, Neu-Isenburg Prof. Dr. M. Thill, Frankfurt am Main 3. Cooperations with Operativ other study groups Dr. B. Ataseven, Essen Prof. Dr. B. Gerber, Rostock The GBG maintains outstanding cooperative Prof. Dr. C. Denkert, Berlin relations with peer national and international Prof. Dr. M. Hahn, Tübingen study groups, including: PD Dr. J. Heil, Heidelberg Dr. D. Krug, Heidelberg ABCSG: Prof. Dr. T. Kühn, Esslingen Austrian Breast & PD Dr. S. Kümmel, Essen Colorectal Cancer Prof. Dr. S. Loibl, Neu-Isenburg Study Group Prof. Dr. G. von Minckwitz, Neu-Isenburg AGO: Translational Research Arbeitsgemeinschaft Prof. Dr. C. Denkert, Berlin Gynäkologische Onkologie Prof. Dr. P. Fasching, Erlangen PD Dr. T. Karn, Frankfurt am Main AGO-B: Prof. Dr. S. Loibl, Neu-Isenburg Breast Study Group Dr. M. van Mackelenbergh, Kiel Prof. Dr. F. Marme, Heidelberg AFT: Prof. Dr. V. Müller, Hamburg Alliance PD Dr. C. Schem, Kiel Foundation Trials Prof. Dr. E. Stickeler, Freiburg Prof. Dr. G. von Minckwitz, Neu-Isenburg BIG: Breast International The Independent Data and Safety Group Monitoring Committee (IDMC) As early as in 2006, the GBG established the BOOG: Independent Data and Safety Monitoring Borstkanker Committee (IDMC) to ensure continual Onderzoeksgroep improvement of working processes in clinical Nederland trials, in-house observation, monitoring and consultation. CECOG: Central European The IDMC reviews all GBG sponsored trials Cooperative Oncology regarding: Group

1. Objectives, the scientific impact of the CIRG: findings and adverse events (AE, SAE, non- Cancer breast cancer deaths) of ongoing trials, International Research 2. All major modifications to the trial protocol Group (including accrual goals), 3. The interim and final efficacy analysis of CRUK: trials, when the protocol-specified number of Cancer Research UK recruited patients or events has been reached. | 11

CTRU: 4. Publications in 2016 Clinical Trials Research Unit Timely publication of study results is a pre- requisite for all clinical trials. GBG is responsible DKG: for an unbiased and independent release of all Deutsche study results and the subsequent, related trans- Krebsgesellschaft lational research projects.

GEICAM: Our research reports were published in leading Grupo Español scientific journals like the New England Journal de Investigación of Medicine, The Lancet, Journal of Clinical del Cáncer de Mama Oncology, The Lancet Oncology, Journal of the National Cancer Institute, Annals of Oncology, IBCSG: European Journal of Cancer, Breast Cancer International Breast Research and Treatment and others. Cancer Study Group Our studies are constantly presented as oral pres ICCG: entations, poster discussions or posters at International Collaborative international congresses such as ASCO, SABCS, Cancer Group EBCC, ECCO-ESMO, St. Gallen and IMPAKT.

NOGGO: Peer-review articles, reviews and congress Nord-Ostdeutsche contributions in 2016 are listed in 4.1., 4.2. and Gesellschaft für 4.3. Gynäkologische Onkologie

NSABP: 4.1. Peer-reviewed articles in 2016 National Surgical Adjuvant Breast and Schwentner L, Helms G, Nekljudova V, Ataseven Bowel Project B, Bauerfeind I, Ditsch N, Fehm T, Fleige B, Hauschild M, Heil J, Kümmel S, Lebeau A, SBG: Schmatloch S, Schrenk P, Staebler A, Loibl S, Scandinavian Breast Untch M, Von Minckwitz G, Liedtke C, Kühn T: Cancer Group Using ultrasound and palpation for predicting axillary lymph node status following neoadjuvant SOLTI: chemotherapy - Results from the multi-center Grupo Español de SENTINA trial. Breast. 2016;31:202-207. Estudio Tratamiento y otras Estrategias Unger U, Denkert C, Braicu I, Sehouli J, Dietel M, Experimentales en Tumores Loibl S, Darb-Esfahani S: Prognostic impact of Solidos HER3 based on protein and mRNA expression in high-grade serous ovarian carcinoma. Virchows UZL: Arch. 2017;470(2):143-151. University Hospital of Leuven Foukakis T, von Minckwitz G, Bengtsson NO, Brandberg Y, Wallberg B, Fornander T, Mlineritsch WGS: B, Schmatloch S, Singer CF, Steger G, Egle D, Westdeutsche Karlsson E, Carlsson L, Loibl S, Untch M, Hell Studiengruppe ström M, Johansson H, Anderson H, Malmström P, Gnant M, Greil R, Möbus V, Bergh J; Swedish Breast Cancer Group (SweBCG), the German Breast Group (GBG), and the Austrian Breast & Colorectal Cancer Study Group (ABCSG): Effect of Tailored Dose-Dense Chemotherapy vs Standard 3-Weekly Adjuvant Chemotherapy on 12 | Annual Scientific Report 2016 | Introduction

Recurrence-Free Survival Among Women With BV, Sinn P, Sirtaine N, O'Toole SA, Viale G, Van de High-Risk Early Breast Cancer: A Randomized Vijver K, de Wind R, von Minckwitz G, Klauschen Clinical Trial. JAMA 2016;316(18):1888-1896. F, Untch M, Fasching PA, Reimer T, Willard-Gallo K, Michiels S, Loi S, Salgado R: Standardized Loibl S, Jackisch C, Schneeweiss A, Schmatloch S, evaluation of tumor-infiltrating lymphocytes in Aktas B, Denkert C, Wiebringhaus H, Kümmel S, breast cancer: results of the ring studies of the Warm M, Paepke S, Just M, Hanusch C, Hackmann international immuno-oncology biomarker work J, Blohmer JU, Clemens M, Costa SD, Gerber B, ing group. Mod Pathol. 2016;29(10):1155-64. Engels K, Nekljudova V, von Minckwitz G, Untch M; investigators of the German Breast Group von Minckwitz G, Rezai M, Tesch H, Huober J, (GBG) and the Arbeitsgemeinschaft Gynä Gerber B, Zahm DM, Hilfrich J, Costa SD, Dubsky kologische Onkologie—Breast (AGO-B) study P, Blohmer JU, Denkert C, Hanusch C, Jackisch C, groups: Dual HER2-blockade with pertuzumab Kümmel S, Fasching PA, Schneeweiss A, Paepke and trastuzumab in HER2-positive early breast S, Untch M, Burchardi N, Mehta K, Loibl S; cancer: a subanalysis of data from the randomized German Breast Group and Austrian Breast and phase III GeparSepto trial. Ann Oncol. 2016;Nov 9. Colon Cancer Study Group Investigators: Zole [Epub ahead of print] dronate for patients with invasive residual di sease after anthracyclines-taxane-based chemo- Laakmann E, Witzel I, Scriba V, Grzyska U, therapy for early breast cancer - The Phase III Zu Eulenburg C, Burchardi N, Hesse T, Würschmidt NeoAdjuvant Trial Add-oN (NaTaN) study (GBG F, Fehm T, Möbus V, von Minckwitz G, Loibl S, 36/ABCSG 29). Eur J Cancer. 2016;64:12-21. Park-Simon TW, Mueller V: Radiological Patterns of Brain Metastases in Breast Cancer Patients: A Fribbens C, O'Leary B, Kilburn L, Hrebien S, Subproject of the German Brain Metastases in Garcia-Murillas I, Beaney M, Cristofanilli M, Breast Cancer (BMBC) Registry. Int J Mol Sci. Andre F, Loi S, Loibl S, Jiang J, Bartlett CH, Koehler 2016;17(10). M, Dowsett M, Bliss JM, Johnston SR, Turner NC: Plasma ESR1 Mutations and the Treatment of Darb-Esfahani S, Denkert C, Stenzinger A, Salat Estrogen Receptor-Positive Advanced Breast C, Sinn B, Schem C, Endris V, Klare P, Schmitt W, Cancer. J Clin Oncol. 2016;34(25):2961-8. Blohmer JU, Weichert W, Möbs M, Tesch H, Kümmel S, Sinn P, Jackisch C, Dietel M, Reimer T, Harbeck N, Iyer S, Turner N, Cristofanilli M, Ro J, Loi S, Untch M, von Minckwitz G, Nekljudova V, André F, Loi S, Verma S, Iwata H, Bhattacharyya Loibl S: Role of TP53 mutations in triple negative H, Puyana Theall K, Bartlett CH, Loibl S: Quality and HER2-positive breast cancer treated of life with palbociclib plus fulvestrant in pre with neoadjuvant anthracycline/taxane-based viously treated hormone receptor-positive, chemotherapy. Oncotarget. 2016;7(42):67686- HER2-negative metastatic breast cancer: patient- 67698. reported outcomes from the PALOMA-3 trial. Ann Oncol. 2016;27(6):1047-54. Furlanetto J, Eiermann W, Marmé F, Reimer T, Reinisch M, Schmatloch S, Stickeler E, Thomssen Ingold Heppner B, Untch M, Denkert C, Pfitzner C, Untch M, Denkert C, von Minckwitz G, Lederer BM, Lederer B, Schmitt WD, Eidtmann H, B, Nekljudova V, Weber K, Loibl S, Möbus V: Fasching PA, Tesch H, Solbach C, Rezai M, Zahm Higher rate of severe toxicities in obese patients DM, Holms F, Glados M, Krabisch P, Heck E, Ober receiving dose-dense (dd) chemotherapy A, Lorenz P, Diebold K, Habeck JO, Loibl S: Tumor- according to unadjusted body surface area: infiltrating lymphocytes: a predictive and prog results of the prospectively randomized GAIN nostic biomarker in neoadjuvant treated HER2- study. Ann Oncol. 2016;27(11):2053-2059. positive breast cancer. Clin Cancer Res. 2016;22(23):5747-5754. Denkert C, Wienert S, Poterie A, Loibl S, Budczies J, Badve S, Bago-Horvath Z, Bane A, Bedri S, Loibl S, Majewski I, Guarneri V, Nekljudova V, Brock J, Chmielik E, Christgen M, Colpaert C, Holmes E, Bria E, Denkert C, Schem C, Sotiriou C, Demaria S, Van den Eynden G, Floris G, Fox SB, Loi S, Untch M, Conte P, Bernards R, Piccart M, von Gao D, Ingold Heppner B, Kim SR, Kos Z, Kreipe Minckwitz G, Baselga J: PIK3CA mutations are HH, Lakhani SR, Penault-Llorca F, Pruneri G, associated with reduced pathological complete Radosevic-Robin N, Rimm DL, Schnitt SJ, Sinn response rates in primary HER2-positive breast | 13

cancer - pooled analysis of 967 patients from five cancer-the road to new treatment strategies. prospective trials investigating lapatinib and Lancet. 2016;Dec 6. [Epub ahead of print] trastuzumab. Ann Oncol. 2016;27(8):1519-25. Paluch-Shimon S, Pagani O, Partridge AH, Bar- Berns K, Sonnenblick A, Gennissen A, Brohée S, Meir E, Fallowfield L, Fenlon D, Friedman E, Hijmans ME, Evers B, Fumagalli D, Desmedt C, Gelmon K, Gentilini O, Geraghty J, Harbeck N, Loibl S, Denkert C, Neven P, Guo W, Zhang F, Higgins S, Loibl S, Moser E, Peccatori F, Raanani Knijnenburg TA, Bosse T, van der Heijden MS, H, Kaufman B, Cardoso F: Second international Hindriksen S, Nijkamp W, Wessels LF, Joensuu H, consensus guidelines for breast cancer in young Mills GB, Beijersbergen RL, Sotiriou C, Bernards women (BCY2). Breast. 2016;26:87-99. R: Loss of ARID1A activates ANXA1, which serves as a predictive biomarker for trastuzumab resis Imgold Heppner B, Loibl S, Denkert C: Tumor- tance. Clin Cancer Res. 2016;22(21):5238-5248. Infiltrating Lymphocytes: A Promising Biomarker in Breast Cancer. Breast Care (Basel). 2016;11(2): Untch M, Jackisch C, Schneeweiss A, Conrad B, 96-100. Aktas B, Denkert C, Eidtmann H, Wiebringhaus H, Kümmel S, Hilfrich J, Warm M, Paepke S, Just Biganzoli L, Aapro M, Loibl S, Wildiers H, Brain E: M, Hanusch C, Hackmann J, Blohmer JU, Clemens Taxanes in the treatment of breast cancer: Have M, Darb-Esfahani S, Schmitt WD, Dan Costa S, we better defined their role in older patients? A Gerber B, Engels K, Nekljudova V, Loibl S, von position paper from a SIOG Task Force. Cancer Minckwitz G; German Breast Group (GBG); Treat Rev. 2016;43:19-26. Arbeitsgemeinschaft Gynäkologische Onkologie- Breast (AGO-B) Investigators: Nab-paclitaxel Ataseven B, von Minckwitz G: The Impact of Neo versus solvent-based paclitaxel in neoadjuvant adjuvant Treatment on Surgical Options and Out chemotherapy for early breast cancer comes. Ann Surg Oncol. 2016;23(10):3093-9. (GeparSepto-GBG 69): a randomised, phase 3 trial. Lancet Oncol. 2016;17(3):345-56. 4.3. Congress contributions in 2016 Budczies J, Pfarr N, Stenzinger A, Treue D, Endris V, Ismaeel F, Bangemann N, Blohmer JU, Dietel SABCS: San Antonio Breast Cancer M, Loibl S, Klauschen F, Weichert W, Denkert C: Symposium, 6-10 December 2016, Ioncopy: A novel method for calling copy San Antonio, Texas, USA number alterations in amplicon sequencing data Denkert C, von Minckwitz G, Darb-Esfahani S, including significance assessment. Oncotarget. Ingold Heppner B, Klauschen F, Furlanetto J, 2016; 7(11):13236-47. Pfitzner B, Huober J, Schmitt W, Blohmer J-U, Kümmel S, Engels K, Lederer B, Schneeweiss A, Hartmann A, Jakisch C, Untch M, Hanusch C, 4.2. Peer-reviewed reviews in 2016 Weber K, Loibl S: Evaluation of tumor-infiltrating lymphocytes (TILs) as predictive and prognostic bio Abramovitz M, Williams C, Loibl S, Leyland-Jones marker in different subtypes of breast cancer B: Dual Blockade of HER-2 Provides a Greater treated with neoadjuvant therapy - A metaanalysis Magnitude of Benefit in Patients With Hormone- of 3771 patients. Oral presentation S1-09, SABCS 2016 Negative Versus Hormone-Positive Breast Cancer. Clin Breast Cancer 2016;16(6):444-455. Bidard F-C, Michiels S, Mueller V, Riethdorf S, Esserman LJ, Lucci A, Naume B, Horiguchi J, Loibl S, Gianni L: HER2-positive breast cancer. Gisbert-Criado R, Sleijfer S, Toi M, Garcia-Saenz Lancet. 2016 Dec 6. [Epub ahead of print] JA, Hartkopf A, Generali D, Rothe F, Smerage J, Muinelo L, Stebbing J, Viens P, Magbanua M, Hall Turner NC, Neven P, Loibl S, Andre F: Advances in CS, Engebråtenm O, Takata D, Vidal-Martínez J, the treatment of advanced oestrogen-receptor- Onstenk W, Fujisawa N, Diaz-Rubio E, Taran F-A, positive breast cancer. Lancet. 2016;Dec 6. [Epub Cappelletti MR, Ignatiadis M, Proudhon C, Wolf ahead of print] D, Bowman Bauldry J, Borgen E, Nagaoka R, Carañana V, Kraan J, Maestro M, Brucker SY, Denkert C, Liedtke C, Tutt A, von Minckwitz G: Weber K, Reyal F, Amara D, Karhade MG, Ruud R, Molecular alterations in triple-negative breast Tokiniwa H, Llombart-Cussac A, D'Hollander K, 14 | Annual Scientific Report 2016 | Introduction

Cottu P, Park JW, Loibl S, Pierga J-Y, Pantel K: Loibl S, Jackisch J, Schneeweiss A, Schmatloch S, IMENEO: International MEta-analysis of circu Aktas B, Denkert C, Schem C, Wiebringhaus H, lating tumor cell detection in early breast cancer Kuemmel S, Luebbe K, Warm M, Just M, Hanusch patients treated by NEOadjuvant chemotherapy. C, Hackmann J, Blohmer J-U, Clemens M, Engels Oral presentation S3-01, SABCS 2016 K, Nekljudova V, von Minckwitz G, Untch M: Dual HER2-blockade with pertuzumab and von Minckwitz G, Timms K, Untch M, Elkin EP, trastuzumab in HER2-positive early breast Hahnen E, Fasching PA, Schneeweiss A, Salat CT, cancer: A subanalysis of data from the Rezai M, Blohmer J-U, Zahm D-M, Jackisch C, randomized phase III GeparSepto trial. Poster Gerber B, Klare P, Kümmel S, Paepke S, Schmutzler P4-21-06, 2016 R, Chau S, Reid J, Hartman A-R, Nekljudova V, Weber KE, Loibl S: Homologous repair deficiency Furlanetto J, von Minckwitz G, Jackisch C, (HRD) as measure to predict the effect of Schneeweiss A, Aktas B, Denkert C, Wiebringhaus carboplatin on survival in the neoadjuvant phase H, Kuemmel S, Warm M, Paepke S, Just M, II trial GeparSixto in triple-negative early breast Hanusch C, Hackmann J, Blohmer J-U, Clemens cancer. Poster P1-09-02, SABCS 2016 M, Costa SD, Gerber B, Nekljudova V, Untch M, Loibl S: Peripheral sensory neuropathy van Mackelenbergh M, Denkert C, Nekljudova V, occurrence and resolution: Results from the Karn T, Schem C, Marme F, Stickeler E, Jackisch C, neoadjuvant randomized GeparSepto study Hanusch C, Huober J, Fasching P, Blohmer J-U, (GBG 69). Poster P5-16-03, SABCS 2016 Kümmel S, Müller V, Schneeweiss A, Untch M, von Minckwitz G, Weber K, Loibl S: Outcome Reimer T, von Minckwitz G, Loibl S, Hildebrandt G after neoadjuvant chemotherapy in progesterone Denkert C, Nekljudova V, Kundt G, Schneider- receptor negative breast cancer patients - A Schranz C, Gerber B: Comparison of axillary pooled analysis of individual patient data from sentinel lymph node biopsy versus no axillary ten prospectively randomized controlled neo surgery in patients with early-stage invasive adjuvant trials. Poster P1-09-11, SABCS 2016 breast cancer and breast-conserving surgery: a randomized prospective surgical trial. The van Mackelenbergh M, Loibl S, Jackisch C, Lück Intergroup-Sentinel-Mamma (INSEMA)-Trial; H-J, Schneeweiss A, Tesch H, Harbeck N, Poster OT2-04-02, SABCS 2016 Schmatloch S, Fehm T, Huober J, Müller V, Bauerfeind I, Untch M, von Minckwitz G, ESMO: European Society for Medical Nekljudova V, Möbus V: Efficacy and safety of Oncology, 7-11. October 2016, Copenhagen, darbepoetin alfa or epoetin beta in 2994 high risk Denmark early breast cancer patients participating in the Loibl S, Schneeweiss A, Burchardi N, Blohmer German adjuvant intergroup node-positive study J.-U, Hanusch C, Costa S, Huober J, Jackisch C, (GAIN). Poster P2-11-04, SABCS 2016 von Minckwitz G, Kümmel S, Paepke S, Denkart C, Untch M: A randomized phase II study to Loibl S, Pfarr N, Weber K, Neunhöffer T, Villegas investigate the addition of PD-L1 antibody S, Stenzinger A, Furlanetto J, Aktas B, Budczies J, MEDI4673 (durvalumab) to a taxane anthracycline Marmé F, Kahmann L, Denkert C, Weichert W: containing chemotherapy in triple negative Comparison of the mutational landscape of breast cancer (GeparNuevo), ESMO 2016, 221TiP breast cancer during pregnancy and non- pregnant controls. Poster P2-03-09, SABCS 2016 Loibl S, Demichele A, Turner N, Cristofanilli M, Loi S, Verma S, Bhattacharyya H, Ke Z, Giorgetti Liedtke C, Kolberg H-C, Kerschke L, Goerlich D, C, Bartlett C, Iyer S, Colleoni M, Masuda N, Im Bauerfeind I, Fehm T, Fleige B, Hauschild M, S.-A, Harbeck N: Impact of palbociclib plus Helms G, Lebeau A, Schmatloch S, Schrenk P, fulvestrant on patient reported general health Schwentner L, Staebler A, von Minckwitz G, Loibl status compared with,fulvestrant alone in HR S, Untch M, Kuehn T: Development and validation + , HER2- metastatic breast cancer, ESMO of a nomogram predicting pathological axillary 2016, 260P status (ypN0 vs. ypN+) in a subgroup of patients converting from cN+ to ycN0 through Bischoff J, Barinoff J, Mundhenke C, Bauerschlag neoadjuvant therapy (NAT) - A transSENTINA D, Costa S, Herr D, Lübe K, Marmé F, Maass N, substudy. Poster P3-13-06, SABCS 2016 von Minckwitz G, Grischke E.-M, Müller V, | 15

Schmidt M, Gerber B, Kümmel S, Schumacher C, Minckwitz G, Kronenwett R, Loibl S: Risk Krabisch P, Thill M, Nekljudova V, Loibl S: assessment after neoadjuvant chemotherapy in A randomized phase II study to determine the luminal breast cancer: A prospectively planned efficacy and tolerability of two doses of eribulin validation of gene expression based and clinical plus lapatinib in trastuzumab pre-treated prognostic scores in 428 residual tumor samples patients with Her2-positive metastatic breast from two neoadjuvant clinical trials. J Clin Oncol cancer - E-Vita, ESMO 2016, 273P 34, 2016 (suppl; abstr 522), ASCO 2016 Chicago

André F, Kaufman B, Juric D, Ciruelos E, Iwata H, Bergh J, Foukakis T, von Minckwitz G, Bengtsson Mayer I, Conte P, Rugo H.S, Loibl S, Rubovszky G, N.O, Wallberg B, Fornander T, Mlineritsch B, Tesch H, Inoue K, Lu Y.-S, Ryvo L, Longin A.-S, Schmatloch S, Singer C. F, Steger G.G, Karlsson E, Mills D, Wilke C, Germa C, Campone M: A phase Carlsson L, Loibl S, Untch M, Hellstrom M, III study of alpelisib and fulvestrant in men and Johansson H, Malmstrom P, Gnant M, Greil R, postmenopausal women with hormone receptor- Moebus V: PANTHER: Prospective randomized positive (HR+), human epidermal growth phase III trial of tailored and dose-dense versus factor receptor 2-negative (HER2–) advanced standard tri-weekly adjuvant chemotherapy for breast cancer (BC) progressing on or after high-risk breast cancer in the modern era of aromatase inhibitor (AI) therapy (SOLAR-1), endocrine and anti-HER2 therapy. J Clin Oncol ESMO 2016, 311TiP 34, 2016 (suppl; abstr 1002), ASCO 2016 Chicago

Curigliano G, Loibl S, Müller V, Pivot X, Wardley Fasching P.A, Blohmer J.U, Burchardi N, Costa A, Cameron D: A phase 2 randomized, double- S.D, ,Denkert C, Hanusch C, Huober J.B, von blinded, controlled study of ONT-380 vs. placebo Minckwitz G, Paepke S, Schneeweiss A, Kümmel in combination with capecitabine (C) and trastu S, Untch M, Loibl S, Jackisch C: A randomized zumab (T) in patients with pretreated HER2+ phase II trial to assess the efficacy of paclitaxel unresectable locally advanced or metastatic and olaparib in comparison to paclitaxel/carbo breast carcinoma (MBC), ESMO 2016, 312TiP platin followed by epirubicin/cyclophosphamide as neoadjuvant chemotherapy in patients with ASCO: American Society of Clinical HER2-negative early breast cancer and Oncology, Annual Meeting 2016, 3-7 June, homologous recombination deficiency (HRD): 2016, Chicago, Illinois, USA GeparOLA. J Clin Oncol 34, 2016 (suppl; abstr Kümmel S, von Minckwitz G, Nekljudova V, Costa TPS1096), ASCO 2016 Chicago SC, Denkert C, Hanusch C, Huober J, Jackisch C, Paepke S, Blohmer JU, Untch M, Schneeweiss A, Schwentner L, Helms G, Nekljudova V, Ataseven Loibl S, German Breast Group: Investigating B, Bauerfeind I, Ditsch N, Fehm T, Fleige B, denosumab as add-on neoadjuvant treatment Hauschild M, Heil J, Kümmel S, Lebeau A, Liedtke for hormone receptor-negative, RANK-positive C, Loibl S, Schmatloch S, Schrenk P, Staebler A, or RANK-negative primary breast cancer and two Untch M, von Minckwitz G, Kühn T; Using different nab-Paclitaxel schedules - 2x2 factorial ultrasound and palpation for predicting axillary design (GeparX). J Clin Oncol 34, 2016 (suppl; lymph node status following neoadjuvant abstr TPS635), ASCO 2016 Chicago chemotherapy: Results from the prospective multi-center SENTINA trial. J Clin Oncol 34, Andre F, Campone M, Ciruelos EM, Iwata H, Loibl 2016 (suppl; abstr 1054), ASCO 2016 Chicago S, Rugo H, Wilke C, Mills D, Chol M, Longin A.S, Juric D: SOLAR-1: A phase III study of alpelisib + Janning M, Müller V, Vettorazzi E, Cubas-Cordova fulvestrant in men and postmenopausal women M, Eulenburg C, Schem C, Fasching P. A, Karn T, with HR+/HER2– advanced breast cancer (BC) Fehm T, Kühn T, Overkamp F, Rack B, Denkert C, progressing on or after prior aromatase inhibitor Untch M, Pantel K, Bokemeyer C, Loibl S, von therapy. J Clin Oncol 34, 2016 (suppl; abstr Minckwitz, Loges S: Serum carbonic anhydrase IX TPS618), ASCO 2016 Chicago as predictive marker for efficacy of bevacizumab: A biomarker analysis from the GeparQuinto Denkert D, Weber K, Krappmann K, Huober J, phase III neoadjuvant breast cancer trial. J Clin Marmé F, Schem C, Engels K, Pfitzner B.M, Oncol 34, 2016 (suppl; abstr 11505), ASCO 2016 Kummel S, Furlanetto J, Hartmann A, Darb- Chicago Esfahani S, Mueller F, Staebler A, Mehta K, von 16 | Annual Scientific Report 2016 | Introduction

Fontanella C, Lederer B, Del Mastro L, Denkert C, Denkert C: Prediction of therapy resistance by De Placido S, Hanusch C, Cognetti F, Huober J, targeted massive-parallel sequencing in primary De Laurentiis M, Jackisch C, Bisagni G, Kummel S, HER2-positive breast cancer. Abstract Number: Garrone O, Schneeweiss A, Montemurro F, Untch 5010, Poster Session, AACR 2016 New Orleans M, Bighin C, von Minckwitz G, Loibl S,Puglisi F; Development and validation of a new prognostic EBCC-10: European Breast Cancer score on 4,646 patients with luminal-like breast Conference, 9-11 March, 2016, Amsterdam, cancer (BC) enrolled in 7 randomized prospective The Netherlands trials. J Clin Oncol 34, 2016 (suppl; abstr 529), Paepke S, Huober J, Sherko K, Eidtmann H, Untch ASCO 2016 Chicago M, Dan Costa S, Blohmer J.U, Eiermann W, Gerber B, Hanusch C, Hilfrich J, Jackisch C, Schneeweiss Loibl S, Turner N. C, Ro J, Cristofanilli M, Iwata H, A, Denkert C, Engels K, Klare P, Fasching P, Im S. A, Masuda N, Loi S, Andre F, Harbeck N, Burchardi N, Loibl S, von Minckwitz G: Random Verma S, Folkerd E, Theall K.P, Zhang K, Huang ized, open-label, phase II study comparing the Bartlett C, Dowsett N: Palbociclib (PAL) in com- efficacy and the safety of cabazitaxel versus bination with fulvestrant (F) in pre-/peri-meno- weekly paclitaxel given as neoadjuvant treatment pausal (PreM) women with metastatic breast in patients with operable triple negative or cancer (MBC) and prior progression on endocrine luminal B/HER2 normal breast cancer therapy – results from Paloma-3. J Clin Oncol 34, (GENEVIEVE). European Journal of Cancer 57, 2016 (suppl; abstr 524), ASCO 2016 Chicago suppl. 2 (2016) S19–S153, Abstract number: 350, EBCC 2016 Amsterdam Turner N. C, Jiang Y, O'Leary B, Hrebien S, Cristofanilli M, Andre F, Loibl S, English P. A, DKK: Deutscher Krebs Kongress (German Zhang K, Randolph S, Huang Bartlett C, Koehler Cancer Congress), 24-27 February 2016 M, Loi S: ; Efficacy of palbociclib plus fulvestrant Ingold Heppner B., Untch M, Denkert C, Pfitzner (P+F) in patients (pts) with metastatic breast B. M, Lederer B, Schmidt W, Eidtmann H, Fasching cancer (MBC) and ESR1 mutations (mus) in P, Tesch H, Solbach C, Rezai M, Zahm D.- M, circulating tumor DNA (ctDNA). J Clin Oncol 34, Holms F, Glados M, Krabisch P, Heck E, Ober A, 2016 (suppl; abstr 512), ASCO 2016 Chicago Lorenz P, Diebold K, Habeck J, Loibl S: Tumor-infil trating lymphocytes and response prediction to Nekljudova V, Loibl S, von Minckwitz G, Gluck S, neoadjuvant chemotherapy in HER2-positive Crane R, Li H, Luo H: A model for trial level pre- breast cancer. Oncol Res Treat 2016;39 (suppl 1): diction of long term outcome based on patho- 1–175, Abstract Nr. 512, Seite 5, DKK 2016 Berlin logical complete response (pCR) after neoad juvant chemotherapy for early-stage breast cancer Laakmann E, Witzel I, Fehm T, Hesse T, von (EBC). J Clin Oncol 34, 2016 (suppl; abstr 1031) Minckwitz G, Möbus V, Park-Simon T.-W, Neun höffer T, Schmidt M, Loibl S, Müller V: Brain Mueller V, Loibl S, Laakmann E, Augustin D, Flock Metastases in Breast Cancer Network Germany F, Dohmen H, Strittmatter H-J, Frank M, Hesse T, (BMBC, GBG 79): The introduction of the Ignatov A, Kuehn T, Park-Simon T, Schmidt M, multicenter register and analysis of patient data. Wuerschmidt F, Fehm T, Moebus V, von Minckwitz Oncol Res Treat 2016;39(suppl 1):1–175, Abstract G, Burchardi N, Witzel W; Brain Metastases in Nr. 236, Seite 50, DKK 2016 Berlin Breast Cancer Network Germany (BMBC, GBG 79): Treatment patterns and clinical outcome of Pfitzner B, Lederer B, Lindner J, Solbach C, Engels more than 1000 patients with brain metastases K, Rezai M, Dohnal K, Tesch H, Hansmann M.- L, from breast cancer. J Clin Oncol 34, 2016 (suppl; Salat C, Beer M, Schneeweiss A, Sinn P, von abstr 2070) Minckwitz G, Denkert C, Loibl S: HEr2 testing in neoadjuvant breast cancer trials – improved AACR: American Association for Cancer concordance of HER2 status in different patho Research, Annual Meeting 2016, 16-20 April logy laboratories. Oncol Res Treat 2016;39(suppl 2016, New Orleans, Lousiana USA 1):1–175, Abstract Nr. 496, Seite 56, DKK 2016 Loibl S, Budczies J, Weichert W, Furlanetto J, Berlin Stenzinger A, Pfarr N, von Minckwitz G, Jackisch C, Schneeweiss A, Fasching P, Schmatloch S, Möbus V, Lück H-J, Forstbauer H, Wachsmann G, Aktas B, Nekljudova V, Weber K, Untch U, Ober A, Schneeweiss A, Christensen B, von Abel | 17

E, Grischke E-M, Höffkes H-G, Klare P, Ko Y- D, Prospectively Randomized GAIN study. Oncol Schmatloch S, Furlanetto J, Burchardi N, von Res Treat 2016;39(suppl 1):1–175, Abstract Nr. Minckwitz G, Loibl S: GAIN2: Adjuvant phase III 562, Seite 58, DKK 2016 Berlin trial comparing an intensified dose-dense (idd) adjuvant therapy with EnPC compared to a dose- Budczies J, Pfarr N, Stenzinger A, Treue D, Endris dense (dd), dose-adapted therapy with dtEC-dtD V, Ismaeel F, Bangemann N, Blohmer J- U, Dietel in patients with primary high risk breast cancer: M, Loibl S, Weichert W, Denkert C: Ioncopy: A Results of the second safety interim analysis. novel method for calling copy number Oncol Res Treat 2016;39(suppl 1):1–175, Abstract alterationsin amplicon sequencing data including Nr. 560, Seite 57, DKK 2016 Berlin significance assessment. Oncol Res Treat 2016;39(suppl 1):1–175, Abstract Nr. 460, Seite Untch M, von Minckwitz G, Jackisch C, Schnee 123, DKK 2016 Berlin weiss A, Conrad B, Aktas B, Denkert C, Eidtmann H, Wiebringhaus H, Kümmel S, Hilfrich J, Warm M, Paepke S, Just M, Hanusch C, Hackmann J, 4.4. GBG-Publications Grading Blohmer J-U, Clemens M, Dan Costa S, Gerber B, System Nekljudova V, Loibl S: nab-Paclitaxel at a dose of 125 mg/m² weekly is more efficacious but less To set internal publication goals and to measure toxic than at 150 mg/m². Results from the our own success, we established our GBG in- neoadjuvant randomized GeparSepto study house grading system as follows: (GBG 69). Oncol Res Treat 2016;39(suppl 1):1– • 7 GBG points for preparation or final pub 175, Abstract Nr. 561, Seite 58, DKK 2016 Berlin lication in a high quality peer-reviewed journal with an impact factor greater than 5, Furlanetto J, Eiermann W, Marmé F, Reimer T, • 5 GBG points for publication preparation or Reinisch M, Schmatloch S, Stickeler E, Thomssen final publication in a journalwith an impact C, Untch M, Denkert C, von Minckwitz G, factor of less than 5, Nekljudova V, Möbus V, Loibl S: Higher Rate of • 3 GBG points for an oral presentation or Severe Toxicities in Obese Patients Receiving poster discussion, dose-dense (dd) Chemotherapy according to • and 2 GBG points for a poster presentation at Unadjusted Body Mass Index – Results of the an international congress.

Peer reviewed Reviews (8) Oral presentation or Poster Total publications (16) poster disscusion (5) presentation (31) 250 231

200

150 106 100 In-house grading 62 48 50 15 0 2016 Figure 3: Overview of GBG’s in-house grading for publications in 2016

GBG IF Official IF 350 321 300 250 200 154 150 100 50 0 2016 Figure 4: GBG and official Impact Factor (IF) in 2016 18 | Annual Scientific Report 2016 | Introduction

4.5. Standard Operating Procedure (SOP) for Authorship

In order to guarantee a maximum of transparency when assigning the co-authorship we have established internal GBG guidelines: Standard Operating Procedure (SOP) for Authorship. Details are listed below:

Score for Authors General Rules (will be used to select and rank co-authors)

§ Important posi,ons: 1st author, senior author, corresponding 1 point for every fulﬁlled criteria: author § Regular par,cipa,ng in TCs and mee,ngs of Subboard and/or § Shared authorship for 1st and 2nd author, if applicable Protocol board § Separate rules for: § Protocol wri,ng § Main publica,on on primary endpoint § Recruitment among best 3rd of par,cipa,ng sites § Publica,ons on secondary endpoints § Sta,s,cal Analysis Plan development § Transla,onal research publica,ons § Manuscript prepara,on § No honorary authorships § In ,me response to emails concerning the trial and the § Author posi,ons can be transferred to a junior person, if also manuscript (within 4 weeks) involved in the study § In ,me response for COI (within 2 weeks) (nega,ve point for subsequent publica,ons)

What to do before submission Publica=on on primary endpoint

§ Select journal § 1st author: PI (or Co-PI group 1) § Ask poten,al authors for their interest to become co-author § Subboard / protocol board members according to Score* § Present proposed list of authors to subboard / protocol board § Best recruiters § Circulate manuscript amongst authors § Biometrician, § Collect COI § Senior author (Co-PI group 2, or group chairman) § Addendum with study team, subboard / protocol board member, and all other recruiters with 3+ pa,ents as „on behalf of the study groups“

* Subboard and protocol board members will share in general authorships with best recruiters on a 2:1 basis

Publica=on on secondary endpoints / retrospec=ve analyses Publica=ons on transla=onal research project

§ 1st author: „project“ leader § Project leader (should prepare manuscript) § Subboard / protocol board members according to score for § Involved team member of this TRAFO project this sub-project* § TRAFO board / protocol board members* § Best recruiters for this sub-project § Biomaterial provider § Biometrician § 1-2 local pathologists providing most tumor ,ssue § PI or group chairman (if involved in sub-project) § Biometrician § PI (if involved in TRAFO project)

* Subboard and protocol board members will share in general authorships * Subboard and protocol board members will share in general authorships with best recruiters on a 2:1 basis with best biomaterial providers on a 2:1 basis | 19

4.6. Oral and poster presentations

San Antonio Breast Cancer Symposium, December 6 -10, 2016 San Antonio Breast Cancer Symposium, December 6 -10, 2016 DistribuOon of TILs in different breast Evalua&on of tumor-infiltra&ng lymphocytes (TILs) as cancer subtypes TILs meta-analysis predic&ve and prognos&c biomarker in different SABCS 2016 subtypes of breast cancer treated with neoadjuvant all 45 36 19 therapy - a metaanalysis of 3771 pa&ents TNBC 29 41 30 low (0-10%)

intermed. (11-59%) Carsten Denkert, Gunter von Minckwitz, Silvia Darb-Esfahani, Barbara Ingold- Heppner, Frederick Klauschen, Jenny FurlaneFo, Berit Pﬁtzner, Jens Huober, Wolfgang HER2+ 44 37 19 high (≥60%) SchmiF, Jens-Uwe Blohmer, Sherko Kümmel, Knut Engels, Bianca Lederer, Andreas Schneeweiss, Arndt Hartmann, ChrisOan Jackisch, Michael Untch, Claus Hanusch, Karsten Weber, Sibylle Loibl lum/HER2- 56 32 13

0% 20% 40% 60% 80% 100% % of tumors

This presentaOon is the intellectual property of the author/presenter. Contact them at This presentaOon is the intellectual property of the author/presenter. Contact them at [email protected] for permission to reprint and/or distribute. [email protected] for permission to reprint and/or distribute.

San Antonio Breast Cancer Symposium, December 6 -10, 2016 San Antonio Breast Cancer Symposium, December 6 -10, 2016 TILs as conOnous parameter: increased pCR and survival pCR rates in all subtypes pCR • Summary aim 1: TILs and increased pCR rate (OR per 10% increase in TILs) response to neoadjuvant 9 OR p- value pCR rate non-pCR all tumors 8 1.20 <0.0005 therapy

TNBC 7 - TILs are linked to increased univariate 1.16 <0.0005 37%

6 pCR rate in all subtypes and p<0.0005 analysis HER2+ 1.12 <0.0005 TNBC

5 most clinical subgroups lum/HER2- 1.34 <0.0005 pCR

4 all tumors 1.17 <0.0005 • Aim 2: How does this 3 mul9variate TNBC 1.17 <0.0005 increased pCR rate translate non-pCR

2 analysis HER2+ 1.12 <0.0005 into prognosis? pCR rate

1 lum/HER2- 1.27 <0.0005 10% 0 • Background: 0,80 1,00 1,20 1,40 - TNBC: high pCR rate and large Lum/HER2- BC p=0.001

odds raOo per 10% increase in stromal TILs survival eﬀect of pCR - Luminal: low pCR rate and limited survival eﬀect of pCR

San Antonio Breast Cancer Symposium, December 6 -10, 2016 San Antonio Breast Cancer Symposium, December 6 -10, 2016 TILs and prognosis in diﬀerent subtypes (Cox- TILs and prognosis – diﬀerences between regression) TNBC and luminal tumors Disease-free survival Overall survival

high TILs = beFer surv. low TILs = beFer surv. high TILs = beFer surv. low TILs = beFer surv.

13 TILs low

13 HR per 10% p-value HR per 10% p-value all tumors 1.02 ns all tumors 1.01 ns TILs high

11 univariate TNBC 11 0.93 0.011 TNBC 0.92 0.032 analysis TILs high HER2+ 0.94 0.017 HER2+ 0.94 ns

9 lum/HER2- 9 1.02 ns lum/HER2- 1.10 0.011 TILs intermed. all tumors 0.94 0.001 all tumors 0.96 ns

7 mul9variate TNBC 7 0.91 0.003 TNBC 0.91 0.020 analysis TILs intermediate all baseline HER2+ 0.92 0.004 HER2+ 0.92 0.061 TILs low Survival rate (%) Survival rate (%) 5 parameters lum/HER2- 5 0.99 ns lum/HER2- 1.09 0.044

mul9variate all tumors 0.96 0.037 all tumors 0.99 ns

analysis 3 TNBC 3 0.95 ns TNBC 0.95 ns all baseline HER2+ 0.96 ns TNBC: DFS p=0.02; n=906 lum/Her2- : OS p<0.0005; n=1366 parameters HER2+ 0.94 0.041 OS: p=0.071 DFS:p=0.036

1 and pCR lum/HER2- 1 1.01 ns lum/HER2- 1.11 0.014

Disease-free survival Ome (months) Overall survival Ome (months)

-1 0,80 1,00 -1 1,20 0,80 1,00 1,20

San Antonio Breast Cancer Symposium, December 6 -10, 2016 San Antonio Breast Cancer Symposium, December 6 -10, 2016 TILs in TNBC vs. luminal BC – main prognosOc Summary and conclusions difference in non-pCR group 1. TILs are a strong predicOve marker for pCR in all subtypes (TNBC; HER2+; luminal) and in nearly all clinical subgroups. pCR rate 2. This pCR effect is translated into a survival benefit in HER2+ and no pCR+low TILs: TNBC. poor prognosis in TILs 37% p=ns p=ns TNBC (n.s.) low 3. In contrast, no survival benefit is observed in luminal tumors. good TNBC, no pCR, n=573 TNBC, pCR, n=333 prognosis of 4. In luminal breast cancer pCR pts.: - increased TILs are linked to increased pCR rate TILs regardless of low pCR rate TILs; - paOents with pCR have a good prognosis no pCR+low similar in TILs: 10% TNBC and - paOents with low TIL-tumors have am improved prognosis p=ns good prognosis p<0.0005 lum/HER2- (overall survival) despite non-pCR in lum BC highly lum/HER2-, no pCR, n=1224 lum/HER2-, pCR, n=142 5. The effects in luminal tumors may suggest a contribuOon of significant subsequent therapies, but should be interpreted with cauOon. overall survival (months) overall survival (months) 6. The role of TILs in resistance to different types of endocrine therapy should be invesOgated in further studies.

PANTHER PANTHER PANTHER trial design - data ASCO 2016 Prospective randomized phase III trial of tailored and dose-dense tdd EC versus standard tri-weekly adjuvant chemotherapy for high-risk q 2 w x 4 courses breast cancer in the modern era of endocrine and anti-HER2 therapy n=1006 q 3 weeks 2017 patients INT=1002 A tdd Docetaxel • Adjuvant RT were • ET for HR + for at randomized q2w x 4 courses least 5 yrs R Duration 15 weeks in both arms •Trastuzumab 1 y February 2007- Control arm September for HER2 (+) Jonas Bergh, Theodoros Foukakis, Gunter von Minckwitz, Nils-Olof Bengtsson, Birgi?a Wallberg, FEC100 Tommy Fornander, Brigi?e Mlineritsch, Sabine Schmatloch, ChrisFan F. Singer, Guenther G. Steger, 2011 n=1011 Eva Karlsson, Lena Carlsson, Sibylle Loibl, Michael Untch, Mats Hellström, Hemming Johansson, INT= 1004 B q 3w x 3 courses Per Malmstrom, Michael Gnant, Richard Greil, Volker Moebus on behalf of the Swedish Breast Cancer Group (SweBCG), the German Breast Group (GBG) Docetaxel100 and the Austrian Breast & Colorectal Cancer Study Group (ABCSG) q 3w x 3 courses JB ASCO 4.6.2016 2

Summary: Primary and secondary efficacy endpoints Primary endpoint: 17-23% improvements, statistically significant for EFS, Breast cancer relapse free survival (BCRFS) and secondary cancers (Loco-regional recurrence, distant recurrence or death due to breast cancer) Events 5-year BCRFS: 88.7% for tddEC-D 5-year B BCRFS: 88.7% for tddEC-D Study end points tddEC-D FEC-D HR † (95% CI) P‡ 85% for FEC 3.7% gain Primary: Breast Cancer Recurrence 118 151 0.79 (0.61-1.01) 0.0620

Secondary: Distant disease-free survival1 110 134 0.83 (0.64-1.08) 0.1660 FEC-D Event-free survival 139 182 0.79 (0.63-0.99) 0.0420 tddEC-D Overall survival 82 104 0.77 (0.57-1.05) 0.0930 Number of patients 1001 1002 Median Follow-up: 5.3 y 37 secondary non-breast cancer malignancies, 18 tdd EC-D and 19 FEC-D, including

5 myelodysplas=c syndrome / acute myeloid leukemia. , 3 tdd EC-D and 2FEC-D arm 2 Failure defined as distant metastases or death due to breast cancer. 3† Hazard ratio estimated using proportional hazards regression. ‡Log-rank p-value. JB ASCO 4.6.2016 3 JB ASCO 4.6.2016 4

BCRFS in relation to actually delivered doses in tddEC-D Grade 3-4 hematological toxicity

ddtEC-D (%) FEC-D (%) tddEC tddD FEC D Type of toxicity Cycle 1-4 Cycle 5-8 Cycle 1-3 Cycle 4-6

Ø Anemia 37 (4)( 12 (1) 7 (1) 6 ()(1) Similar outcomes irrespective Leukopenia 896 (90) 145 (15) 680 (68) 468 (48) of the dose levels, supporting Neutropenia 878 (88) 166 (18) 854 (86) 518 (53) the tailored dosing strategy Thrombocytopenia 41 (4) 7 (1) 16 (2) 3 (0)

Any hematological toxicity 935 (93) 198 (21) 890 (89) 551 (56)

Numbers of patients starting cycles 1001 948 998 980

Dose level groups EC Dose level groups D

The dose level group is the sum of dose levels a patient received during the four therapy courses.

JB ASCO 4.6.2016 5 JB ASCO 4.6.2016 6

Grade 3–4 other toxicities, mostly non-hematological Conclusions according CTC v 3 • Tailored-dose-dense EC-D improved all efficacy endpoints with 17-23%, ddtEC-D (%) FEC-D (%) compared with the three-weekly FE100C-D100 schedule, being statistically tddEC tddD FEC D significant only for EFS, at median follow-up of 5.3 years Type of toxicity Cycle 1-4 Cycle 5-8 Cycle 1-3 Cycle 4-6 1. Diarrhea 14 (1) 33 (3) 8 (1) 32 (3) • BCRFS within the tailored-dose-dense group was similar irrespective of the 2. Nausea 42 (4) 10 (1) 15 (2) 7 (1) 3. Vomiting 32 (3) 4 (0) 23 (2) 3 (0) achieved level of dose escalation, supporting the tailoring strategy aiming at 4. Mucositis (oral) 30 (3) 53 (6) 9 (1) 24 (2) similar toxicity - outcome for each patient, tailoring worked best for the EC part 5. Neutropenic event 98 (10) 23 (2) 79 (8) 87 (9) 6. Infection (normal ANC) 25 (2) 74 (8) 30 (3) 37 (4) • Tailored-dose-dense EC-D resulted in higher doses of cyclophosphamide 7. Pain 26 (3) 109 (11) 12 (1) 112 (11) (120%) and epirubicin (35%) and grade 3-4 toxicity was increased for the non- 8. Fatigue 95 (9) 162 (17) 27 (3) 89 (9) 9. Hand-foot skin reaction 3 (0) 91 (10) 1 (0) 17 (2) hematological items but vice versa was seen for the hematological toxicity for 10. Neuropathy (motor) 1 (0) 17 (2) 0 (0) 6 (1) standard D 11. Neuropathy (sensory) 2 (0) 31 (3) 2 (0) 11 (1) 12. Nail changes 0 (0)) 23 (2)) 0 (0) 6 (1)) • The “superiority” of tailored-dose-dense EC-D was consistent in all studied Any [1-12] 269 (27) 403 (43) 164 (16) 277 (28) subgroups, including ER positive disease and HER2 positive disease treated

Numbers of patients starting cycles 1001 948 998 980 with trastuzumab JB ASCO 4.6.20164 6 2016 7 JB ASCO 4.6.2016 8 | 21

San Antonio Breast Cancer Symposium – December 6-10, 2016 San Antonio Breast Cancer Symposium – December 6-10, 2016 CTC detec'on Interna'onal MEta-analysis N pa'ents ≥1 CTC ≥2 CTC ≥5 CTC con'nuous IMENEO study of circula'ng tumor cell detec'on Before NCT 1574 25.2% 12.6% 5.9% in early breast cancer pts SABCS 2016 treated by NEOadjuvant chemotherapy (IMENEO study) cT size p<.0001 p<.0001 p<.0001 p<.0001 cT1 122 (7.9%) 18.9% 8.2% 3.3% cT2 770 (49.8%) 22.3% 10.3% 3.5% cT3 343 (22.2%) 24.2% 12.2% 6.1% FC Bidard*, S Michiels, V Mueller, S Riethdorf, LJ Esserman, A Lucci, B Naume, J Horiguchi, cT4a-c 108 (7.0%) 28.7% 16.7% 8.3% R Gisbert-Criado, S Sleijfer, M Toi, JA Garcia-Saenz, A Hartkopf, D Generali, F Rothé, cT4d 204 (13.2%) 41.2% 24.5% 15.7% J Smerage, L Muinelo, J Stebbing, P Viens, M Magbanua, CS Hall, O Engebråtenm, D Takata, J Vidal-MarQnez, W Onstenk, N Fujisawa, E Diaz-Rubio, FA Taran, MR CappelleT, cN status p=.051 p=.021 p=.009 p=.024 cN0 656 (41.9%) 22.7% 10.4% 4.1% M IgnaUadis, C Proudhon, D Wolf, J Bowman Bauldry, E Borgen, R Nagaoka, V Carañana, cN+ 911 (58.1%) 27.1% 14.4% 7.2% J Kraan, M Maestro, SY Brucker, K Weber, F Reyal, D Amara, MG Karhade, RR Mathiesen, p=.23 p=0.54 p=.12 H Tokiniwa, A Llombart-Cussac, K d'Hollander, P CoZu, JW Park, S Loibl, JY Pierga, K Pantel Subgroup p=.028 HR+ 365 (23.2%) 24.1% 11.0% 4.7% HER2+ 800 (51.0%) 24.1% 11.5% 5.3% * Medical Oncology, Ins'tut Curie, Paris, France Triple Neg. 405 (25.8%) 28.4% 16.5% 8.4%

This presentaUon is the intellectual property of the presenter. contact [email protected] for permission to reprint/ distribute. This presentaUon is the intellectual property of the presenter. contact [email protected] for permission to reprint/ distribute.

San Antonio Breast Cancer Symposium – December 6-10, 2016 San Antonio Breast Cancer Symposium – December 6-10, 2016 CTC detec'on: associa'on with pCR CTC before NCT & Overall Survival

pCR was deﬁned as ypT0/isN0 in 92.5% of pa'ents (N=1916/2072) pCR was observed in 24.3% of pa'ents (N=503/2072)

N pa'ents ≥1 CTC ≥2 CTC ≥5 CTC con'nuous Stra'ﬁed p value <.0001 Before NCT p=.076 p=.65 p=.90 p=.10 pCR 374 (24.0%) 21.7% 12.0% 6.1% months No pCR 1183 (76.0%) 26.3% 13.0% 5.9% N pts % events Hazard Ra'o Before surgery p=.45 p=.13 p=.53 p=.52 0 CTC 1175 9.8% 1 pCR 300 (26.3%) 13.7% 7.0% 1.3% 1 CTC 199 10.6% 1.09 [0.65-1.69] No pCR 841 (73.7%) 15.7% 4.6% 1.0% 2 CTC 59 23.7% 2.63 [1.42-4.54]

3-4 CTC 47 29.8% 3.84 [2.08-6.66]

≥ 5 CTC 93 46.2% 6.25 [4.34-9.09]

San Antonio Breast Cancer Symposium – December 6-10, 2016 San Antonio Breast Cancer Symposium – December 6-10, 2016 Overall clinical validity (single point; ≥2 CTC) Changes during therapy; ≥2 CTC

Mul'variate analyses N HR <2 / <2 CTC 609 1 Time point OS DDFS LRFI <2 / >2 CTC 21 3.57 HR p HR p HR p Overall Survival >2 / <2 CTC 103 4.29 CTC at baseline 4.19 <.0001 3.79 <.0001 3.20 <.0001 Landmark analysis >2 / >2 CTC 11 9.68 [2.97-5.88] [2.84-5.03] [1.93-5.19] (landmark analysis) Stra'ﬁed p value <.0001

CTC [-5;0]w 2.56 .0020 2.69 <.0001 1.05 .92 N HR before surgery [1.45-4.23] [1.67-4.12] [0.32-2.55] <2 / <2 CTC 606 1 (landmark analysis) <2 / >2 CTC 21 3.89

Distant Disease-Free Survival >2 / <2 CTC 103 4.18 Landmark analysis >2 / >2 CTC 10 4.94 Stra'ﬁed p value <.0001

months This presentaUon is the intellectual property of the presenter. contact [email protected] for permission to reprint/ distribute. This presentaUon is the intellectual property of the presenter. contact [email protected] for permission to reprint/ distribute.

San Antonio Breast Cancer Symposium – December 6-10, 2016 San Antonio Breast Cancer Symposium – December 6-10, 2016 Added value to comprehensive prognos'c models Conclusion Likelihood ra'o tests For the ﬁrst 'me, we showed: Model (1) Model (2) Endpoint Average increase P value in Chi2 [95% CI] Independent prognos'c factor for death, distant metastasis Baseline Baseline OS 28.9 [13.3-44.1] <.0001 & locoregional relapses + CTC baseline DDFS 38.7 [21.9-58.5] (resampling <.0001 procedure) CTC number-dependent impact, signiﬁcant above ≥2 CTC/7.5ml LRFI 8.7 [2.4-19.2] .003

Baseline Baseline OS 3.1 [0.03-11.3] .07 + CTC + CTC baseline Baseline CTC count improves predic've models of survival DDFS 3.1 [0.02-9.7] .07 baseline + pCR + pCR + CTC before LRFI 0.9 [0.00-3.5] .34 CTC changes during therapy were associated with survival (resampling surgery & landmark analysis)

57 BC . the

CA .

291 72,0

in of one, treated 0.039

PIK3 38,7 = p=0.001

HER 2+ G3 without

in in and 2. pCR rates for ). ). A total mutations were 60,9 1 interaction 148 tumors CA TP53 P

cohort

pertuzumab p=0.690

/ 55,9 Abstract # 5010 # 5010 Abstract Table Table , wt no, Table Table 5 . in PIK3

) BC .

had 61,3

mut vs 0.412 HER 2+

= Figure Figure p=0.052 43,9

CA trastuzumab in addition, from G3 ( G3 71 from In

PIK3 Buch; Germany Buch; 151 tumors - 80,0

interaction wildtype tumors. The results show that ). P to p=0.029 54,2 , Peter A. Fasching , Peter 3

able able 2 T from G7 and significant difference was observed between the 2 blockade HER 2 blockade 66,7

no group, when anti-HER2 therapy was given . This effect 293 p=0.009 wt ), anti- 47,7 Helios Kliniken Kliniken Berlin Helios group ( 8 anti-HER2 therapy. G7 compared

87 % in dual

difference difference between the

XRAS nab -paclitaxel cohort, pCR rates were significantly lower mut vs

11,1 to

no 0.126 Conclusions Conclusions CA the p=0.094 the =

33,3 tumors ( to In 2

taxane paclitaxel cohort, mutation in the untreated (G3) and untreated in the mutation

PIK3 417 30,8 the , Andreas Schneeweiss , interaction

the -paclitaxel nab -paclitaxel 4

in P in ). ). There was PIK3CA wt PIK3CA cohort. This presentation is the intellectual property of the author/presenter. the author/presenter. of property intellectual the is presentation This 364/ - p=0.660 to PIK3CA PIK3CA 20,0 Universitätsklinikum Essen; Universitätsklinikum in more mutated genes. Mutation frequencies are displayed 7 HR or

BC predicts resistance Contact them at [email protected] for permission to reprint and/or distribute. reprint and/or to for permission them at [email protected] Contact were combined ) cohorts were were cohorts ) evaluated separately two trials ( forthe and Figure Figure 2 16,3 wt

+ , Carsten Denkert p=0.339 HRAS 27,3 HR 8

HER2+

group (

group, whereas in wt the wt PIK3CA mut PIK3CA in on FFPE core biopsies reliably identified the most common genomic alterations 57,8 rates according to rates according NRAS,

, Christian Jackisch wildtype and wildtype ( and 1 and 8 tumors three alterations are relevant for response are response for alterations relevant NGS was successful p=0.014 Results 42,5 pCR by predict lower response mut vs seen mut KRAS, (mut) to Elisabeth Krankenhaus, Kassel; Krankenhaus, Elisabeth 6 0 CA mutation CA CA 90 80 70 60 50 40 30 20 10 2 HER 2 treatment.

, Michael Untch

1 (%) pCR tumors two, mutant mutational Analysis NGS Targeted found EGFR, EGFR, non -synonymously mutated genes were detected, the most commonly mutated genes were The pCR rate was significantly lower PIK3 was was also PIK3 PIK3 anti- (G7) HER2+ cohort overall and by and HR by HER2+ cohort overall and (G7) Figure 2: Figure

62,6

N=293) 0.113 0.113 0.167 0.063 0.063 0.466 0.466 p-value 75) 21,2 HER2

, Gunter Gunter , von Minckwitz

, Karsten Weber 50,0 1 49 (16.8) 29 (10.1) Universitätsklinikum Erlangen; Universitätsklinikum 131 131 (45.6) 158 158 (53.9) 204 204 (69.6) 260 260 (88.7) 49 (22- N (valid N (valid %) 5 dual dual anti- 0,0

(30.4)

blockade blockade (G7; 4 (3.6) with or without anti withoutor with 72 (64.3) 60 (53.6) 34

N (valid N (valid %)

60,6 HR - (N=112) status

25,0 , Nicole Pfarr

N=71)

genes 3 HR

dual anti-HER2 blockadedual (G7) anti-HER2 78) HER2 by (G3;

64,7 27 - ( and the German Cancer Consortium (DKTK). Consortium Cancer German the and anti- and 5 (2.0) Sana Klinikum Offenbach; Offenbach; Klinikum Sana 45 (64.3) 45 17 (23.9) 20 (32.3) 27 (38.6) 48 (67.6) 63 (88.7) 63

4 15,4 47 53 (21.0) N (valid N (valid %) 112 (44.4) 112 141 141 (56.0) N (valid N (valid %)

no HR+ (N=252) HR+ , Valentina Nekljudova , Valentina 7

treatment

66,7 overall

16,3 and

47,7 The project has partly been funded within the EU-FP7 project RESPONSIFY No 278659 No RESPONSIFY EU-FP7project within the funded been partly has project The

, Albrecht Stenzinger , 9 ( 2.5) 1 87 (23.9) N (valid N (valid %) 213 213 (58.5) 172 172 (47.3) patient frequencies

27,3

Overall (N=364) , Bahriye Aktas , Bahriye

no anti-HER2 therapy (G3) no therapy anti-HER2 6 -lobular -lobular invasive

62,5 Universitätsklinikum Heidelberg; Universitätsklinikum 3 PgR PgR positive

; ; Baseline Mutation

19,7 mut characteristics mut

mut

mutation

0 , Berlin , -4

+ (%) pCR 80 70 60 50 40 30 20 10

Figure 1: pCR rates for mutant and wildtype wildtype pCR 1: rates formutant and Figure HER2 therapy Parameter cN and/or ER Grade 3 Ductal/ductal >60% TILs LPBC, cT3 Age, median median Age, (range), years Any PIK3CA TP53 XRAS Table 2: Table tumor Table Table 1: Charité , Jenny , Jenny Furlanetto

of of to no . CA CA CA 3 40 part Sabine Schmatloch defined PIK3 panel PIK

PIK3 as 133 - PIK3CA, 1 to , mutations,

re included. re included. 2+ tumors the 53 variant allele BC . 124 : 3) received in CA we Lancet Oncol. at TP . (G FGFR2, HRAS, investigate key HER al Institute of Pathology, Pathology, of Institute somatic 2 somatic

1, to 2 therapy. HER 2 therapy. PIK3 2010 ; of

et of , Wilko Weichert , Wilko 2 . 3 genes as ESR 21 500 t anti- t clinical and clinical endpoints.

Treat. Treat. Untch Untch of GeparTrio GeparTrio HER 2+ primary in 2.

ithou BC received trastuzumab and in w BC solvent-based paclitaxel 20 .

2+ and hormone receptor (HR) status. or assign these mutations 364 pretherapeutic core biopsies promising tool

or 2+ mutations to 22 genes (including HER2+ breast cancer (BC). a in of % tumor cell cell 20 % tumor content

the additional in , Jan Budczies HER is

1 of of HER the coding region called in 32 :3212- neoadjuvant neoadjuvant therapy. taxane taxane tions, Prediction of therapy resistance by targeted massive-parallel sequencing in primary HER2-positivebreast cancer massive-parallelin by targeted resistancesequencing of therapy Prediction to of determine the relation between individual Cancer Breast Res . deep targeted massive parallel sequencing. to References 1, ATM, BRAF, EGFR, Objectives Objectives

al therapy

muta Background Background German Breast Group, Neu-Isenburg; Group, Breast German by to 1 et nab -paclitaxel assess the predictive value of

assess the predictive value AKT and predict lower pathological complete response (pCR) to to addition to to ) with minimum a coverage Patients with in spot regions Sibylle Sibylle Loibl 2 ), patients with subgroups subgroups Patients and Methods and Methods Patients cancer. Genomic alterations, such in Huober Huober hot 3B1 and only ≥ cases with in (G 7 SF addition

in possible, . J Clin Oncol. 2014; 07 . 3. 10 % if al and pathway-related groups the frequency

alin-fixed paraffin alin-fixedembedded paraffin (FFPE) samples from s for pCR pCR s for blockade with trastuzumab/lapatinib et s were shown NRAS, FBXW7, PTEN,

non -synonymous mutations ency ency ≥ 2 HER 2 treatment ; 75 ;S2- G7 were and were G7 analysed interrogated primary aim was investigated genomic alterations GeparSepto double . Loibl In 417 form G3 We CDH1, KRAS, frequ pertuzumab neoadjuvant therapy. anti- mutations mutation determine pathways, Massive Massive parallel sequencing (NGS) mutation We trialswith clinicalprospective fromtwo The ypN0). pCR (ypT0 for mutations •Secondary aims were 1 2016 molecular events are important for response •Exploratory aims included analyses to • • • • Only Presented at: AACR Annual Meeting 2016 in New Orleans, Louisiana, USA – April 16-20, 2016 16-20, 2016 April – USA Orleans, Louisiana, 2016 in New Meeting Annual AACR Presented at: | 23

rate pCR + SURGERY in or be

III triple-

to to surgery surgery either II 2-positive 2-positive early as stratified stage HER in in paclitaxel weeks weeks and patients with to 12 12

in neoadjuvant once-per-week Homologous Recombination planned to or or by be conducted

8 8 is chemotherapy chemotherapy , 3 2014 . neoadjuvant breast

for for 56 . imaging tests breast in ** EC 90mg/m²/600mg/m² 90mg/m²/600mg/m² EC 90mg/m²/600mg/m² EC

addition 747 - : SABCS in 15 and EC

2014 ; afterPO/PCb by patients with triple-negative patients with TNBC,Results from GeparSixto. ASCO in in carboplatin (C) compared with docetaxel (D) for patients 21 . of with/without tBRCA mutation; 13 - on pathologic complete response rates positive. positive. HRD vs 33 : 1/2 breast cancer. 1/2 breast deficient tumors defined carboplatin and/or bevacizumab III trial assess the pCR (ypT0/is ypN0) rate rate pCR ypN0) the assess (ypT0/is

known gBRCA mutation. mutation. gBRCA known

of email: [email protected] a higher pCR rate 12 2015 ; to BRCA or HR

in , Gunter , Gunter von Minckwitz 7 or pathological complete response (pCR) -Klinikum, Offenbach, Germany ; Germany Offenbach, Sana -Klinikum, BRCA of HER2neg/HR+ HER2neg/HR+ Ki67 t HRD positive positive HRD weeks weeks 12 high Oncol. ; ; ypT(any) ypN0); pCR rate Germany. Enrollment 12 12 • • • • and Clin the addition Results chemotherapy chemotherapy /+ safety. safety. and in of Objectives Objectives . . Neoadjuvant carboplatin PwO followed References and cyclophosphamide al HRD biopsies (both mandatory) mandatory) (both biopsies Conclusions positive positive BC

Centrally in confirmedCentrally initial coretumor

Technische Universitäts-Frauenklinik, München, Germany; Germany; München, Universitäts-Frauenklinik, Technische Carboplatin AUC CarboplatinAUC for2 * biopsy: of QII/2016. The study will 8 . . Prediction et

Universitäts-Frauenklinik, Magdeburg, Germany ; Germany Magdeburg, Universitäts-Frauenklinik, + al Olaparib 100mg bid for for bid 100mg + Olaparib blood collection (mandatory) (mandatory) collection blood pCR rate according other definitions (ypT0 ypN0; 4

Paclitaxel weekly 80mg/m² 80mg/m² weekly Paclitaxel 80mg/m² weekly Paclitaxel core in pCR. and et . Impact , Jens Huober al on physical examination HRD 6

sites Loibl S et please contact: please contact: German Breast Group, Neu-Isenburg, Germany;

): a randomised phase 2 trial. Lancet Oncol. Lancet trial. 2 phase randomised a 66 ): A, GeparOLA aims GeparOLA , Christian Jackisch months. months. or (Alliance). J (Alliance). 40603 3 50 CM -

Untch M * 24 N=102 N=102

. al . The TNT trial: A randomized phase

K, For For further information and for permission to reprint and/or distribute et 40 65PO:37PCb 65PO:37PCb Perou dose-dense doxorubicin ; ; ypT0/is ypN0 by DA, screening HR -) /+ recurrent locally advanced triple negative negative triple locally advanced recurrent

: >= 40 >= or G, Timms G, Schneeweiss Kilburn L L Kilburn vs vs P,

, Claus Hanusch , Sibylle Loibl 5 11 Ellis

HELIOS Klinikum Berlin Buch, Germany; Germany; Buch, Berlin HELIOS Klinikum 11 omologous omologous ecombination (HRD) eficient negative breast cancer: CALGB CALGB cancer: breast negative von Minckwitz 2015 . Tutt metastatic with paclitaxel paclitaxel followed Deficiency (HRD) after carboplatin-containing neoadjuvant chemotherapy Sikov WM, Berry breast cancer (GeparSixto; GBG (GeparSixto; cancer breast von Minckwitz neoadjuvant treatment H R D HER2- Stratification Hormone-receptor (HR+ status Age years (<40 years) 3. 4. 2. olaparib olaparib with paclitaxel will result with cancerbreast early HER2-negative Primary objective: objective: Primary of cancer patients with early mutation tBRCA1/2 design trial GeparOLA 2: Figure Recruitment will start will GeparOLA assess whether a platinum agent 1. Secondary Secondary Objectives: ypT0 ypN0 subgroups; subgroups; pCR rate response rate based conservation rate, tolerability tolerability rate, conservation Exploratory analyses will relate germline sequencing)generation and tumor mutations (by next approximately approximately within completed

Universitäts-Frauenklinik, Ulm, Germany; Germany; Ulm, Universitäts-Frauenklinik, - 7 or an on be +/

, Michael Untch -neg , Carsten Denkert 4 10 score be

German Breast Group, Neu-Isenburg, Germany; Germany; Neu-Isenburg, Group, Breast German tumor triple- 3 weeks

PR HER2- or in to in 12 HRD

Myocet - and %). %). Based pCR (ypT0/is and 55 and ER 1,2

). %) or olaparib yrs) (Figure 2). (Figure yrs)40 , Sherko Kümmel 9 65 of , Serban Dan Costa ≥ 3 HRD score and BRCA support a subsequent vs 40603 to including response rate for patients 40

positive). in 4 AUC 2 weekly for pNSLN+ not (< or 37 patients will provide a point Cb - a germline (gBRCA) Brustzentrum Kliniken Essen-Mitte, Essen, Germany; Essen, Essen-Mitte, Kliniken Brustzentrum HRD 10 +/ age Klinikum zum Roten Kreuz, München, Germany; Germany; München, Kreuz, Roten zum Klinikum 6 Brustzentrum, Charité, Berlin; Germany; Berlin; Charité, Brustzentrum, 2 and mg/m² 80 mg/m² weekly (Pw) plus olaparib shown how plus GeparSixto study. study. GeparSixto , Nicole Burchardi gBRCA mutation. Moreover among % 80 % power) 2 and % in docetaxel. Pw patients are needed for PwO, the study , Andreas Andreas Schneeweiss ,

8 60 been vs and Financial and Drug Supply was provided by Astra Financial Astra and ZenecaDrug Supply wasprovided by neoadjuvant Paclitaxel 65 or HR -) to deficiency has been recognized 50 -

the central test results available; 3. Male to vs has paclitaxel

HR be it 70 %, to

1c with either cN+ of

cT (HR+

Stefan Stefan Paepke Background well described. data Toxicity for the combination with or , Jens-Uwe Blohmer TNBC 1 not mutations BRCA patients with gBRCA 1/2 mutations ( 1/2 mutations gBRCA with patients -003509-4) -003509-4) will assess the efficacy In breast cancer patients with Homologous Recombination

% 90 % confidence interval (CI) in be excluded (α=0.1; is in 4a-d 4a-d 12 weeks (PwO) epirubicin/cyclophosphamide (EC). Randomization will (BRCA 1/2 mutations and/or mutations 1/2 (BRCA Patients and Methods cT 2015 102 patients and high pCR rates be enrolled prior 65 % ( by : : June June 2016 A randomized phase II trial to assess the efficacy of paclitaxel and olaparib in comparison to comparison olaparib in assess and efficacy of paclitaxel the to trial II phase randomized A of th 2 - ≥ HRD No HRD is

cT – 7 % width). width). 27 % the GeparSixto trial rd % % should

of and Peter A. Fasching Peter in

of 55 with HER2-negative early breast cancer and Homologous Recombination Deficiency (HRD) Deficiency GeparOLA HER2-negative Recombination – Homologous early breast cancer and with

also scarce. scarce.also pCR Frauenklinik, Universitätsklinikum Erlangen, Germany; Germany; Erlangen, Universitätsklinikum Frauenklinik, paclitaxel/carboplatin followed by epirubicin/cyclophosphamide as neoadjuvant chemotherapy in patients chemotherapy neoadjuvant as epirubicin/cyclophosphamide by followed paclitaxel/carboplatin olaparib (O) 1 is National Center for Tumor Diseases, Heidelberg, Germany; Germany; Heidelberg, Diseases, Tumor for Center National if BC

CI 9 ≤ twice daily for of Furthermore the TNT study showed a doubling of 3 3 hormone-receptor status hormone-receptor %; %; 2. Centrally confirmed tumor Homologous Recombinant Deficiency score mg 90 % a high correlation between tBRCA by 20 Charité University Hospital, Institute of Pathology, Berlin, Germany; Berlin, Pathology, of Institute Hospital, University Charité is 5 rates were even were higher even rates

63 %). 67 > Ki

BRCA BRCA 1/2 (tBRCA) mutation chemotherapy Carboplatin (Cb) increased the pathological complete response (pCR) rate The efficacy Deficiency (HRD) positive (former score high) tumors negative breast cancer (TNBC) patients (GeparSixto, CALBG high ( carboplatinreceiving mutations 1/2 gBRCA with There patients involved mutations overlap (Figure 1). independent predictor Carboplatin. GeparOLA (EudraCT early negative ypN0) Figure 1: Overlap Overlap 1: Figure previous previous results, pCR rate for PwCb will GeparOLA GeparOLA will randomize tablets 100 (PwCb), each followed criteria:inclusion Main 1. Untreated primary and/or tBRCA status can being positive, ( estimate female patients. patients. female considerations: Statistical A pCR rate phase III Assuming study. a pCR rate (tBRCA (tBRCA positive/mutated and/or HRD positive) (mySeq) . Patients with known gBRCA stratified stratified or TPS1096 TPS1096 Presented at: ASCO Annual Annual Meeting 2016 , ASCO Presented at: Chicago, Illinois, USA, 3 24 | Annual Scientific Report 2016 | Introduction

to 22

trial SABCS

18 : 15 . 1,8 q metastasis. Clin other 307 - and addition vs low; other day in 145 : the ABCSG- ): ): a multicentre, randomised, . . positive HER2- positive HER2- of 2014 ; 18 of of accoding BC pCR (ypT0 (ypT0 pCR ypN0) Germany. Germany. Primary Primary Endpoint: tumorigenesis mg/m² mg/m² in definitions (LRRFS) survival (DDFS) survival (IDFS) pCR in 2 125mg/m² day 1,8 q 22 Treatment backbone Treatment Pertuzumab Trastuzumab+ HER2+: 2 AUC weekly Carboplatin TNBC: 125 case case RANK high • Localrelapse free survival • Distant disease free • Invasivedisease free •• Event free survival (EFS) Overall survival (OS) Secondary Secondary Endpoint: •

nP in in

in ,

Surgery primary nP or in

sites 60 breast cancer (ABCSG-

postmenopausal patients in or in

EC 425 

without denosumab EC , Sibylle LoiblSibylle , prefer

email: [email protected] 11 or EC EC  to ; ; pCR rates EC EC - - breast cancer. Breast Cancer Res Treat.Res Cancer Breast cancer. breast in the RANKL/RANK/OPG pathway BC standard neoadjuvant chemotherapy will of 12x 12x , Hanusch Claus months 18 4 to predictive marker and predictive please contact: please contact: German Breast Group, Neu-Isenburg, Germany; Charité University Hospital, Institute of Pathology, Berlin, of Institute Pathology, Hospital, University Charité disease-free survival: Results from 3, 4 Technische Universitäts-Frauenklinik, München, München, Universitäts-Frauenklinik, Technische on 8 planned for planned Objectives Objectives . . e. trastuzumab/pertuzumab . . e. trastuzumab/pertuzumab denosumab treatment with RANK expression; pCR rates Results References denosumab is (i (i Conclusions For For further information and for permission to reprint and/or distribute of 125mg/m² weekly (Cb) osteoclast-independent roles of , Andreas Andreas Schneeweiss , a prognostic a , Carsten Denkert , Carsten 43 . 3 10 and as TNBC and HER2+ nP Colorectal Cancer Study Group. Adjuvant denosumab without Denosumab without Denosumab and

-paclitaxel schedule nab -paclitaxel in 433 - 125mg/m² weekly (Cb) and of . . . also planned. planned. also

386 : nP -Paclitaxel 125mg/m² d 1,8 q22 - -Paclitaxel 125mg/m² d 1,8 q22 - is -Paclitaxel 125mg/m² weekly -Paclitaxel 125mg/m² weekly

BC BC 2016 Denosumab Denosumab 120 mg q4w24 weeks adjuvant denosumab neoadjuvant treatment with Denosumab Denosumab 120 mg q4w 24 weeks expression expression 2015 ; of of Nab Nab of Nab Nab RANK National Center for Tumor Diseases, Heidelberg, Germany; Heidelberg, Diseases, for Tumor Center National . QIII/ 11 the addition al which and

. . Austrian Breast if in al et

The impact et , Michael Untch , Michael

9 PC, PC Sana-Klinikum, Offenbach, Germany ; Germany Offenbach, Sana-Klinikum, Loibl Loibl S 35 . 7 , Serban Dan Dan Costa Serban , D, 2 326 - survival. survival. and 18 : Universitäts-Frauenklinik, Magdeburg, Germany; Germany; Magdeburg, Universitäts-Frauenklinik, 3 G, Dubsky G, Dubsky TNBC vs GeparX trial design trial GeparX plus anti-HER2 treatment plus anti-HER2 treatment 2012 ; patients with early with patients early with patients . . Molecular pathways: osteoclast-dependent vs no

in in Pfeiler Pfeiler WC EC EC

M, M, vs q3 /HR+ yes R definitions definitions for both randomizations; response rates; breast conservation rates; toxicity and -   N=778 arm for both randomizations randomizations both arm for -primary objectives: objectives: -primary Dougall Dougall Res. Cancer Gnant Pfitzner BM, Branstetter BM, Pfitzner Gnant 2015 double-blind, placebo-controlled trial. Lancet. trial. placebo-controlled double-blind, *second core before EC Stratification Stratification HER2 HER2+ LPBC EC q2 (Cb) Co pCR (ypT0 ypN0) rates backbone treatment consisting (Cb) compliance; program translationallarge A diagnosticcore* will start Recruitment GeparX investigates 1. 4. 2. 3. status) status) rate pCR the increase pCR (ypT0 ypN0) rates Objectives: Secondary Secondary objectives are interaction per pCR , Jens-Uwe , Jens-Uwe Blohmer Figure 1: Figure 8 HELIOS Klinikum Berlin Buch, Germany; Germany; Buch,Berlin HELIOS Klinikum

10 % as EC , Valentina Nekljudova Valentina , and and the 2 45

Universitäts-Frauenklinik, Ulm, Germany; Ulm, Universitäts-Frauenklinik, 6 to in a 2x2 factorial design (GeparX) a 2x2 design factorial in to

2 patients requiring

the first in ), ; 58 ; 2-sided defined 36 % by neoadjuvant

is , Stefan Paepke , Stefan 20 administered 7 . the GeparTrio German Breast Group, Neu-Isenburg, Germany; Germany; Neu-Isenburg, Breast Group,German to 2 0 be in from α=

randomized randomized be tested % % (OR=1. be Financial and Drug Supply was provided by Amgen Financial Amgen and Drug and CelgeneSupply was provided by 46 Brustzentrum, Germany; Berlin; Charité, Brustzentrum, patients overall. patients 9

overall

, Gunter von Minckwitz von, Gunter of . 1

to schedule s.c. every 4 weeks for 6 . . ; hormone receptor negative 778 patients

in %

mg nP enrolled. LPBC BC Germany; Germany; could demonstrate that elevated 35 , Christian Jackisch , Christian be power) 14 .5% 6 power

Klinikum zum Germany; München,Kreuz, zum Roten Klinikum

5 we , , tumor infiltrating lymphocytes and in RANK/L-negative primary breast cancer and two different nab-Paclitaxel different schedules twoRANK/L-negative breast cancer and primary 601 patients treated 4a-d 4a-d BC, centrally confirmed hormone 67 neoadjuvant chemotherapy. chemotherapy. neoadjuvant 80 % -

HER2+ 80 % different 4 of

cT -paclitaxel -paclitaxel (nP) 125 mg/m² weekly + in Ki triple negative breast cancer (TNBC) Germany; Germany; associated Sherko Kümmel Sherko 10 ; by . June June 2016 1,8 q 22 + EC, stratified HER2+), and epirubicin/cyclophosphamide in is

Moreover denosumab improves disease- Background Investigating Denosumab as an add-on to neoadjuvant chemotherapy RANK/L-positive or chemotherapy in to neoadjuvant add-on as an Denosumab Investigating 10 ; 1c- th 1 . 3 nab vs

0 Jens Huober Jens cT day α= 0 and – 7 Brustzentrum Kliniken Essen-Mitte, Essen, Germany; Germany; Essen-Mitte,Essen, Kliniken Brustzentrum of

lymphocyte predominant breast cancer (LPBC; α= rd addition

. . Secondarily patients will by - denosumab (120 in study showed that adjuvant denosumab reduces antibody against RANKL, will Patients and Methods postmenopausal woman with hormone receptor

denosumab from be given TNBC sided +/ % % stromal tumor infiltrating lymphocytes),subtype in - 18 an on core biopsy can RANK found was by power) 2 sided BC vs % stromal infiltrating lymphocyte. lymphocyte. infiltrating stromal %

( - 50 of

HER2 status, 2 50 vs q3 w)

> an 92 % 125 mg/m²

vs 45 ; .

1 10 ; patients with higher pathological complete response (pCR) rates. (pCR) response completepathological higherwith .

nP a retrospective analysis 0

50 % Denosumab, Denosumab, TPS635 TPS635 RANK ligand (RANKL) pharmacologic inhibition attenuates breast cancer development functions via progression. metastatic its receptor RANK and In The ABCSG- clinical fractures, improves bone health, and can without added toxicity. High RANK expression primarywith positive breast cancer (BC). cancerbreast positive free free survival BC study with chemotherapy (TAC) study with chemotherapy (TAC) expression RANK status chemotherapy chemotherapy Trial design: Trial GeparX (NCT02682693) will randomize or 1). (Figure randomization Carboplatin will criteria:inclusion Main Patients with primary receptor than more considerations: Statistical Sample size improvement (primary endpoint) planning (OR= assumed a 778 pCR ≤ backbone backbone treatment trastuzumab + pertuzumab + trastuzumab (EC, q2w (HER2-/HR+ (HER2-/HR+ cycles), stratified α= Presented at: ASCO Annual Annual Meeting 2016 , ASCO Presented at: Chicago, Illinois, USA, 3

Poster # 273P Abstract # 2935 A Randomized Phase II Study to Determine the Efficacy and Tolerability of two Doses of Eribulin plus Lapatinib in Trastuzumab pre-treated Patients With Her2-Positive Metastatic Breast Cancer (E-VITA)

Bischoff J1, Barinoff J2, Mundhenke C3, Bauerschlag D4, Costa SD1, Herr D5, Lübbe K6, Marmé F7, Maass N4, von Minckwitz G8, Grischke EM9, Müller V10, Schmidt M11, Gerber B12, Kümmel S2, Schumacher C13, Krabisch P14, Thill M15, Nekljudova V8, Loibl S8

1University Women’s Hospital Magdeburg, Germany, 2 Breast Center Clinics Essen-Mitte, Essen, Germany, 3Department of Obstetrics and Gynecology, Kiel University Hospital, Germany, 4Department of Obstetrics and Gynecology, Aachen University Hospital, Germany5University Women’s Hospital Würzburg, Germany, 6Department of Obstetrics and Gynecology, Hannover- Kirchrode Hospital, Germany, 7Department of Obstetrics and Gynecology, Heidelberg University Hospital, Germany, 8German Breast Group, Neu-Isenburg, Germany, 9University Women’s Hospital Tübingen, Germany, 10University Women’s Hospital Hamburg-Eppendorf, Germany, 11University Women’s Hospital Mainz, Germany, 12University Women’s Hospital Rostock, Germany, 13St. Elisabeth-Hospital, Köln, Germany, 14Chemnitz Hospital, Germany, 15Department of Obstetrics and Gynecology, Agaplesion Markus Hospital, Frankfurt am Main, Germany

Background Results

Lapatinib (L) in combination with capecitabine has been approved for the treatment of HER2-positive Figure 1 E-VITA study design Table 1 Baseline characteristic of patients advanced breast cancer (ABC) progressed after anthracycline-, taxane-, and trastuzumab (T)-containing E1.23+L E1.76+L E1.23+L E1.76+L therapies1, 2. The use of this combination is limited by overlapping toxicities such as diarrhea and Lapatinib (1000 mg/d p.o. daily d1-21) Parameter Category (n=21) (n=20) Parameter Category (n=21) (n=20) cutaneous side effects. Therefore, other combinations of L with less toxic agents are needed. N (%) N (%) N (%) N (%) Eribulin mesylate (E7389) is a synthetic analog of halichondrin B, a large polyether macrolide isolated from Eligibility 2 3 Eribulin (1.23 mg/m i.v.; days 1+8, q d21) median 50.0 60.0 G1 0 ( 0.0) 0 ( 0.0) a marine sponge . It is a mechanistically unique antagonist of microtubule dynamics among tubulin- Trastuzumab pre-treated patients with Age, years targeted agents, leading to inhibition of microtubule growth, but not shortening, in association with HER2-positive MBC 1:1 range 32.0-84.0 39.0-69.0 Grading, G2 10 (50.0) 8 (40.0) 4 R sequestration of tubulin into aggregates . Eribulin mesylate has shown better overall survival of 2.5 months 0 13 (61.9) 15 (75.0) primary tumor G3 10 (50.0) 12 (60.0) at a dose of 1.4 mg/m² given on day 1 and 8 every 3 weeks as compared with treatment of physician’s Stratification N = 43 Lapatinib (1000 mg/d p.o. daily d1-21) choice in patients with locally advanced or metastatic breast cancer who were previously treated for 2-5 • previous line of chemotherapy for ECOG 1 8 (38.1) 4 (20.0) missing 1 0 lines with anthracyclines, taxanes, and capecitabine (EMBRACE study)5. metastatic disease (0-1 vs. 2-3) Eribulin (1.76 mg/m2 i.v.; day 1, q d21) performance ductal or In the E-VITA study the efficacy and tolerability of two schedules of Eribulin (E) in association with L have • estimate of tumor burden by LDH status 2 0 ( 0.0) 1 ( 5.0) ductal-lobular 20 (95.2) 19 (95.0) level (≤1.5 x ULN vs. >1.5 ULN) been compared in patients with metastatic trastuzumab (T) pre-treated HER2+ breast cancer. Primary endpoints: invasive • Time to progression (TTP) Histological • Safety and compliance M0 16 (76.2) 11 (57.9) tumor type lobular invasive 1 ( 4.8) 0 ( 0.0)

Patients and Methods M at first M1 5 (23.8) 8 (42.1) other 0 ( 0.0) 1 ( 5.0) diagnosis Trial design: Figure 2 TTP and OS in the two treatment arms Number of E-VITA (NCT01534455; GBG64) is a multicentre, randomized, open-labelled phase II study to determine missing 0 1 metastatic ≥ 3 13 (61.9) 9 (45.0) sites the efficacy and tolerability of two doses of E+L in T pre-treated patients with HER2+ ABC (Figure 1). 100% 100% Patients were randomized in 1:1 ratio to receive E1.23 mg/m2 i.v. days 1+8, q d21 (E1.23) or E1.76 mg/m2 both ER, PgR 90% + CensoredE1.23+L E1.76+L + CensoredE1.23+L E1.76+L 8 (38.1) 5 (25.0) Anti-HER2 no 8 (38.1) 9 (45.0) 90% ER/PgR ,primary negative i.v. day 1, q d21 (E1.76) plus L 1000 mg p.o. once daily from d1-21 (3-week cycles). Stratification factors Median TTP, months 8.1 6.5 Median OS, 23.1 23.2 E 1.23 mg 14/21 events treatment 80% E 1.23 mg 19/21 events 80% months for randomization were: i) previous line of chemotherapy for metastatic disease (0-1 vs. 2-3) and ii) 95% CI 4.8-9.4 4.6-13.4 tumor ER and/or PgR palliative 95% CI 12.5-35.0 13.7-30.1 13 (61.9) 15 (75.0) yes 13 (61.9) 11 (55.0) E 1.76 mg 17/20 events E 1.76 mg 12/20 events estimate of tumor burden by LDH level (≤1.5 x ULN vs. >1.5 ULN). Treatment was given until disease 70% 70% positive progression or unacceptable toxicity of the study drug, or withdrawal of consent of the patient. Main inclusion criteria: 60% 60% Table 3 Haematological and non-haematological toxicities

• Histological confirmed carcinoma of the breast with over-expression of HER2 (IHC3+ or FISH positive); 50% 50% • ABC not suitable for surgery or radiotherapy alone; E1.23+L E1.76+L Overall AE Any Grade Grade 3-4 Any Grade Grade 3-4 Any Grade Grade 3-4 • The following previous systemic treatments are eligible: 40% 40% N (%) N (%) N (%) N (%) N (%) N (%)

adjuvant and up to 3 chemotherapy regimen for ABC; Proportion alive o 30% 30% Most frequently haematological AEs o previous treatment with trastuzumab either as (neo)adjuvant treatment for early breast cancer Anemia 18 (85.7) 0 ( 0.0) 15 (75.0) 0 ( 0.0) 33 (80.5) 0 ( 0.0) 20% 20% and/or first and/or second line treatment for metastatic breast cancer; Leukopenia 19 (90.5) 5 (23.8) 17 (85.0) 11 (55.0) 36 (87.8) 16 (39.0) Proportion progression-free o if previous chemotherapy regimen were anthracycline based, the maximum cumulative dose of prior 10% 10% Neutropenia 16 (76.2) 10 (47.6) 16 (80.0) 14 (70.0) 32 (78.0) 24 (58.5) anthracycline therapy must not exceed 360 mg/m² for doxorubicin and 720 mg/m² for epirubicin; Most frequently non-haematological AEs 0% 36 0% 24 27 30 33 36 Alopecia 20 (95.2) n.a. 18 (90.0) n.a. 38 (92.7) n.a. • Patients must have either measurable or non-measurable target lesions according to RECIST criteria. 12 15 18 21 0 6 12 18 24 30 0 3 6 9 Fatigue 16 (76.2) 4 (19.0) 16 (80.0) 0 (0.0) 32 (78.0) 4 (9.8) • ECOG performance status 0-2. 2 1 2 2 2 2 3 1 Increased ASAT 13 (61.9) 1 (4.8) 15 (75.0) 0 (0.0) 28 (68.3) 1 (2.4) • Normal cardiac ejection function as determined by cardiac ultrasound (LVEF above institutional normal 1 8 2 2 1 1 8 4 4 3 1 2 0 2 6 1 9 7 E 1.23 mg 1 0 E 1.23 mg Increased ALAT 15 (71.4) 0 (0.0) 12 (60.0) 0 (0.0) 27 (65.9) 0 (0.0) 2 2 1 1 2 0 range). 7 3 3 0 2 8 1 5 1 9 5 13 (61.9) 0 (0.0) 14 (70.0) 0 (0.0) 27 (65.9) 0 (0.0) E 1.76 mg 0 9 2 1 1 1 9 E 1.76 mg Sensory neuropathy Statistical considerations: Diarrhea 12 (57.1) 2 (9.5) 12 (60.0) 1 (5.0) 24 (58.5) 3 (7.3) • As this was a feasibility study, no formal sample size calculation was performed. It was planned to recruit TTP, months OS, months Nausea 10 (47.6) 0 (0.0) 9 (45.0) 0 (0.0) 19 (46.3) 0 (0.0) 80 patients in 35 sites of Germany with enrolment time of 24 months (Q-III 2011 – Q-II 2013) and follow- HFS 7 (33.3) 0 (0) 5 (25.0) 0 (0) 12 (29.3) 0 (0) up period of maximal 12 months after the last patient entered the study. The recruitment in the study was Table 2 ORR and CBR in the two treatment arms stopped on July 7, 2014 in agreement with the study IDMC due to the slow recruitment, 43 patients were E1.23+L E1.76+L enrolled. Response • One interim safety analysis was performed after 20 patients had received at least 2 cycles with eribulin N % 95% CI N % 95% CI Conclusions and lapatinib. No modification of study protocol was necessary based on this analysis. CR 1 4.8 0 0.0 • Kaplan-Meier estimates with the 95% confidence interval (CI) were performed for TTP and OS. ORR PR 10 47.6 9 45.0 The combination of E+L shows an acceptable safety profile. E1.23+L is the preferred SD 8 38.1 8 40.0 and CBR were reported as number and percentage of patients in each treatment arm including 95% CI. 1 PD 2 9.5 1 5.0 regimen due to its lower toxicity profile. Compared to capecitabine/lapatinib regimen , E+L The AEs and compliances were reported as number and percentage of patients in each treatment arm. combination has higher sensory neuropathy and neutropenia and lesser HFS adverse Missing (considered as no response) 0 0.0 2 10.0 events. Due to premature study termination, no definitive conclusion on efficacy can be drawn. Objectives ORR (CR or PR) 11 52.4 (31.0%, 73.7%) 9 45.0 (23.2%, 66.8%) CBR (CR or PR or SD) ≥ 24 weeks 15 71.4 (52.1%, 90.8%) 15 75.0 (56.0%, 94.0%) Primary objective: References • To assess the time to progression (TTP) defined as the time period between randomization and documented disease progression or disease related death Between February 2012 and July 2014, 43 patients were randomized (41 started treatment). The study was 1.Geyer CE, Forster J, Lindquist D, et al. Lapatinib plus capecitabine for HER 2- positive advanced brast cancer N • To assess the safety and toxicity by NCI-CTCAE version 4.0 and compliance by the number and stopped in July 2014 due to slow accrual. Median follow-up was 28.7 months. At the time of analyses 19 TTP Engl. Med 2006; 355: 2733- 2743 reasons of patients whose treatment had to be reduced, delayed or permanently stopped eribulin 2.Cameron D, Casey M, Press M, et al. A phase III randomized comparison of lapatinib plus capecitabine versus events and 14 deaths occurred in E1.23+L arm (N=21) and 17 TTP events and 12 deaths in the E1.76+L arm and/or lapatinib capecitabine alone in women with advanced breast cancer that has progressed on trastuzumab: updated efficacy (N=20). Baseline patient’s characteristics are described in Table 1. The median number of erubilin cycles was 8 in and biomarker analyses. Breast Cancer Res Treat 2008; 112:533–543 Secondary objectives: both arms. There was no difference in TTP and OS between the two eribulin doses arms (Figure 2). Also ORR 3.Towle MJ, Salvato KA, Budrow J, et al. In vitro and in vivo anticancer activities of synthetic macrocyclic ketone • To determine the objective response rate (ORR) defined as complete or partial response (CR or PR) analogues of Halichondrin B. Cancer Res. 2001; 61:1013-1021. • To determine the overall clinical benefit rate (CBR) defined as CR or PR or stable disease (SD) for ≥ 24 and CBR were similar (Table 2). High grade (grade 3-4) haematological and non-haematological adverse events 4.Jordan MA, Kamath K, Manna T, Okouneva T, Miller HP, Davis C, Littlefield BA, Wilson L. The primary antimitotic weeks (AEs) were more common in patients with E1.76+L (Table 3). The most frequently reported grade 3-4 AE was mechanism of action of the synthetic halichondrin E7389 is suppression of microtubule growth. Mol Cancer Ther. • To determine the overall survival (OS) defined as the time period between randomization and death due neutropenia (47.6% with E1.23+L vs. 70.0% with E1.76+L), fatigue (19.0% vs. 0.0%) and diarrhea (9.5% vs. 2005;4(7):1086-95. to any cause. 5.0%). Hand-food syndrome (HFS) was observed in 29.3% of patients. Overall 5 (12.1%) patients discontinued 5.Cortes J, O'Shaughnessy J, Loesch D, et al. EMBRACE (Eisai Metastatic Breast Cancer Study Assessing Physician's Choice Versus E7389) investigators. Eribulin monotherapy versus treatment of physician's choice in • To assess biomarkers like PI3K mutation, PTEN expression, c-myc on the primary tumor and correlate therapy due to AEs, 2 (9.6%) in E1.23+L arm and 3 (15%) in E1.76+L arm. Two deaths occurred during treatment, patients with metastatic breast cancer (EMBRACE): a phase 3 open-label randomised study. Lancet. 2011; them with TTP (here not presented) but only one (E1.23+L arm) was related to the study treatment. 377(9769):914-23.

Presented at the ESMO Copenhagen, October 7-11, 2016 For further information and for permission to reprint and/or distribute please contact: German Breast Group, Neu-Isenburg, Germany, e-mail: publications@germanbreastgroup

|25 26 | Annual Scientifi c Report 2016 | Introduction

. ; ; III in

2012 2013 based

- .

better .

phase

GBG 7

; -

)

standard and + two

Nature months

Nature

and

. 1421

: 9

)

7 9 to 18 ( from

35 - nanoparticle

assessed vi genes : response

112 rate 6 ab - Paclitaxel GeparSepto ) . cancers breast ( i Epirubicin Cyclophosphamide

be definitions

surgeries weekly cancers study

2015 ;

Suppl .

will pCR all : breast research cancer with

the associated

other -

) breast

in Cancer 0

safety :

to 2012 ;

from J breast

removed

Durvalumab .

approximately cancer

negative Br

Randomisation higher -

. ypN .

Durvalumab Placebo (MEDI4736) early negative for all response

- 66 0 Oncol a of new

R = endpoints

triple chemotherapy

enrolled

with for tissues

to Ann

triple cancers

endpoint ypT

. 2959 -

( ) according Germany

)

32 :

primary

patients lead of

(TNBC) Frauenklinik, Ulm, Germany; different Quality of life Correlative science Tolerability and Tolerability pCR Clinical (c) and imaging ( pCR

2014 ; - neoadjuvant addition conducted

the patients

for . in

publications@germanbreastgroup considering lymphatic Pathological • • • ● • Primary Secondary •

will

be and 15 : ) 11

cancer lymphocytes

Oncol

the spectrum to M therapy

are comparing

1 Clin

collection - cancer

breast J

- mail

. in trial e

Universitäts infiltrating Surgery

6 III there evolution /ligand

1199

1 cyclophosphamide - planned

/ Untch Tissue+Blood negative

( whether - tumor ,

is phase Results

chemotherapy of

Biomaterial Biomaterial ECOG 10 it

2016 , triple protein

heterogeneity C mutational

and for

References

value Germany, Germany, Conclusions

and ) anthracyline ) death

Durvalumab and , HELIOS Klinikum Berlin Buch, assess

2197 by i.v.

i.v. 11 randomized GeparNuevo

cell

Mutational .

2016 . clonal treatments

al TNBC

September ECOG al Prognostic

will

neoadjuvant : mg/m . et mg/m followed Denkert

th et

QII/

The

, .

trials

9 et 07 . in

3 al early

Cyclophosphamide programmed (600

Cyclophosphamide of - Isenburg

(600 Weeks ) - 2 S

+ ) study i.v. ) Neoadjuvant Polak based . Anti

paclitaxel As .

- .

. cancer with Part

Neu i.v M A i.v

2 2 Schneeweiß started

Demaria

Goya based C, , Berlin, Germany;

breast

Ribas Martin

RJ, presentation mg/m

mg/m has Placebo (1500 mg Durvalumab Stojanov B, Paepke

Frauenklinik, Magdeburg, Germany;

patients -

9 8 , Germany solvent - -

Epirubicin (90 Oral 8 Roth Epirubicin (90 Gray

MS, in

Jackisch

adjuvant

395 214 S, anthracycline : : with

Moreno

/ S P,

GeparNuevo )

2014

Pathology

Minckwitz

7403 ) 7457 ) of sites Clinical and imaging response Shah Adams Lawrence Untch von Homet collection

...... 30 outcome Recruitment The 1 SABCS taxane 2 randomized 3 486 ( 4 499 ( 5 6 paclitaxel Universitäts 5

further information and for permission to reprint and/or distribute please contact: contact: please distribute and/or reprint to permission for and information further Kümmel Tissue+Blood

, ( 1 of as

Biomaterial Biomaterial For Group, Breast German

G and with

phase overall

such (DDFS),

and

eeks

l treatment

W window

Part 2

) survival examination

)

2 Minckwitz )

the signatures 2 i.v.

- L1 antibody MEDI4673 (

free definition) TNBC Paclitaxe -

pharmacogenetics -

Paclitaxe after ) -

, von physical neoadjuvant 7 nab (125 mg/m FDA Charité University Hospital, Institute

and early Durvalumab (1500 mg

nab ; (125 mg/m

Placebo C collection

on disease ; expression

per

0 )

with TILs

, )

i.v. ; rate 1 zum Roten Kreuz, München, Germany; ypN Tissue+Blood gene

(

L ) - (EFS based distant

Biomaterial Biomaterial chemotherapy, (any)

and

patients

Jackisch ypN in subpopulations , ypT opportunity

eeks 6 Klinikum

; nodes

0 ; 4 of survival

dose dose mg 750 W Part 1 /+ before

0 (LRRFS), 2

markers (ypT

subpopulations MRI) antigens,

free

ypN

stratified

;

window or axillary

s tarting (

Durvalumab ( Placebo /is

pCR

) 0 Durvalumab biopsies

tumor survival

Huober event

and

stratified +/ 5 ypT molecular

the

collection ; core

free

treatment .

the /+ ; EC

Frauenklinik, München, Germany; Objectives 0 - on tumor

(IDFS), according

by for .

Tissue+Blood ( ypN 59%) vs.

mammography, expression, - etc specified taxane

and N = 174

0 -

, Taxane compare

, Biomaterial Biomaterial

; breast

; cN 67 study design pre recurrence

to ypT survival -

, Costa SD

after

followed ; 4

arms

rate chemotherapy 0 100%) prognosis .

FACT the

- separately

C any ,

free

of 1 rate (N=174) 10%) vs. :

aims of d,

L - ypN both : - biomarkers -

term arm sonography

( invasive using in compare

end /is

cTa 0 compliance

: - Paclitaxel , per rates Brustzentrum, Charité, Berlin; Germany;

long study

1 life 3 the

TILs (0 and disease

tests

Technische Universitäts Technische (OS) ypT

and 9 o response of

conservation

nab -

TNBC cT1b Eligibility

high TILs (60 high Stratification n intermediateTILs (11 Hanusch

objectives PD of and

molecular rates

objectives regional ,

after

- 3 objective

rates pCR toxicity loco breast clinical imaging invasive response examine and TILs, quality survival JU

● ● ● ● ● ●

● ● ●

sequential correlate GeparNuevo response Figure 1 GeparNuevo Figure Secondary Exploratory ● ● Primary containing chemotherapy in triple negative breast cancer ( Blohmer

. ) - . II 4 , Heidelberg, Germany; a , . is to

v 1 so an

are cell cell i Mitte, Essen, Germany; and had It

plus -

TILs N be nab

(part

T sided nodal .

better TNBC - of

. mg 2 system every

surgery, )

phase clinically since of

standard

test

of have

plus

- binding the mutational

a Diseases assessed

and might

events

cycles 750

randomized

the 1

( lymphocytes .

) L until

negative, be - /Placebo

a .

must

2 high are addition on

for (TNBC) (Figure Placebo PD immune Programmed pCR ) rate

intermediate Tumor Tumor

( a

ypN controlled

4 /Placebo

given , or ; would , Burchardi 0 3 who for adverse the using to vs . irrespective

2 the

and ) pCR

v conducted

infiltrating

. power be d A i standard weeks

rate patients blocks

- prospectively patient cancer %

this

load

(ypT Durvalumab

arm

2 be Durvalumab the to

10 placebo

be All DFS

- will out

mg the

response

that .

related 0

cT patients 80 %

without will tumor ( rate : -

) negative/HER

will was

b ; mediated Durvalumab breast

6 every - all

1 of reactivate

Brustzentrum Kliniken Essen drop - receive

inhibition/suppression

or )

(no 8

because better

1500 .

mAb cT pCR v inhibitor

blinded, - anthracycline ( for

. a

mutational

and - i

National Center will

study

which 2

10 % . help treatment

analysis achieve tumor

TILs complete followed Durvalumab consent , the 66 levels with

into

immune

) a negative high

done ; to of stage

assumptions Schneeweiss be

biomarkers

will double randomization a the negative/PR

mg/m²

(PP) 48 %

biopsy of 11, 2016 - 11,

TNBC

mediated High

antibody 1 to arms, 7 Methods patients -

of a

. triple checkpoint with

1 conducted be stromal

ER 1 Durvalumab

days)

(part

600

( target 1

have - compare arm) . in translate

by to

early L

be GeparSepto

to assuming

21 - the

prior

study following study PD pathological

protocol

taxane

withdrawal will

(≤

PD and

will be the breast

predefined with ) -

Loibl S Loibl

weeks expected status

chemotherapy the each or especially the

October

randomized, option per - study in

, Offenbach, Germany;, Offenbach,

monoclonal , both superiority is will confirmed of

a anti Isenburg, Germany;

the

followed

12 - subtypes, randomized phase II study to investigate the addition of PD patients 1 expected important

months higher

investigation

other

eventually

TILs

this In

immunogenicity

of an the

3 is

. arm for 79

analysis IgG patients

to a ( an like toxicity, Background

))

A and ultrasound 02685059 and into

ratio

with

in ; focusing that based Klinikum

might paclitaxel - arm

treatment 1 weeks,

centrally 1

addition, -

clinical :

show

within

Cyclophosphamide 0

1 2 prospective, NCT

neoadjuvant 100 % basal mITT )

- ,

( weekly value opportunity) human ; Group, Neu to

nab - .

breast

only

: Sana in up

with

- for

patients which randomization addition 60 Copenhagen

carcinoma and is 2 markers of 89 provides

( subjects

. 67

Patients

under

- )

0 . . : Durvalumab status Placebo ) the with treat analyses

v for

- Ki 1 the .

associated overcoming/preventing

i work

unacceptable based α= Breast 158 L

is mg/m² 174

Durvalumab

- to therapy are

(GBG

calculation specially - TILs the

the

4736 )

benefit

demonstrated

disease

ESMO

multicenter tested

are ;

primary in immune in

(window set

factor (PD

criteria

with

125 - a

treated level (CT)

start

randomize size

interim

; 1

mg/m² be

high intent )

recruit

TiP 1 study

was

is staging

CD 80

the

- (MEDI

agents :

rate rate TNBC, German with monotherapy

to confirmed years

Therefore

nP mammography,

90 will

considerations 1 It

mITT (

( . vs part at

TNBC Performance

status,

negative

progression, ) 1 and

In ≥ 18 safety sample anthracycline and actually pCR pCR meaningful patients in ligand

/ L

that . the

inclusion

design option

modified

• •

1 59 % targeted Stratification - - Placebo planned Six Patients planned PD Triple ECOG involvement centrally Complete bilateral significance A Age In The and

. ) 11

valid

● ● or 3 Main PD study ● ●

● death Chemotherapy (TILs) outcome burden Durvalumab activation GeparNuevo taxane immunotherapy Trial GeparNuevo weeks Paclitaxel Epirubicin disease Statistical ● ( ● ● ● ● far Abstract # 3769 Abstract Poster # 221 # 221 Poster

Presented | 27

. . of in

or in of

vs vs HR The - 9. Genes and

the

non - %

. gene

29 %) in ) . % 0

1/2 . 3 4 detection

diagnosed diagnosed

different 22 . (positive vs (G 502 ) and diagnosed

mutations . - 03 09 .

( 52 relevance Within be

% HR reliable vs sensitivity cancer samples) both cohorts

CA Figure mutated P1 11 ( p= 0 in

) . to differences BCP

; sample) 2 clinical per

or for

45 ), .

of 53 enables grading per

PIK 3 significantly less -pregnant controls CA (

41 - TP patients subgroup . 2013 10;31(20):2532 2013 Oncol . .

Cancer

and control

had

non vs Figure PIK3 - 61 Clin for frequently

information

not seem sequencing to Breast

differences

- 40 mutations whereas % no -pregnant controls TCGA with

mutations

36 015 ; 03 most pregnant 37 . ancillary

for does 0

. of women with primary breast cancer breast primary with women of . negative) patients - generation ] Epub % HR - negative the the vs of breast cancer during pregnancy using genomic genomic using pregnancy during cancer breast of and and TCGA cohorts was performed difference status vs ( p=

0 .

next Diagnosed

of in

provides 54

[ 22431 BCP to detect mutations

85 . compared . 2012; ;51(5):653 2012; Oncol . and

was Prognosis biological features and prognosis of breastof prognosis and features biological % . 31 - 35 vs. 65 %

BCP and /gcc . 4 as

. rate Acta vs

Targeted The The gene expression patterns are currently

53 2

et al. et in . vs References

mutational rate after matching might 1002 53 ( al Conclusions Conclusions % controls %; Figure18 %; 2).

Pregnant FA et al. Biology Biology al. et FA cancer

TP 8 significant 10 .

et .

mutational

30 :

CA TP 204 - 209 only : V

vs patients from both datasets (Table 2). 41 patients from both datasets (Table doi 23 . ; Being . 195 (average 1. no

. 26 - breast in 7 ( 3 ) , F ;

tumors

HER 2 (positive

control study. study. - control Peccatori vs PIK3 in 000 ) of the mutational patterns between 2016 Endris .

- 2014;21(4):545 Cancer. . mutational overall

most frequent somatic mutations S,

(6% 1 N+ 2012

R, . BCP ; in

Amant a remaining bias caused

) with status Relat ,

of methods used

The (p= 102 mutations (average 1. by Cancer

4 not used for matching. Further comparisons using other SN,

for Brohée Botteri E, Renne G et al.et G Renne E, Botteri

2 ) age ( age on

, , (Basel) 3 detected Penzel

046 Jr Jr non - pregnant higher

dataset and yielded % GATA3 Han

Endocr N, 0

Berlin, Berlin Berlin,

S, Care (ERBB pregnancy: results from an international collaborative study. J study. collaborative international an from results pregnancy: . negative),

GATA In the matched cohorts (p=0. were frequently G3) 75 . Exact matching (1:1) based vs BCP Comparison pregnant controls (TCGA cohort) before any overall matching showed with the subgroups, TCGA. were detected and Pfarr Loibl

2.Amant F, von Minckwitz G, Han SN, SN, Han G, Minckwitz von F, 2.Amant during pregnancy: a casea pregnancy: during during 3.Azim HA HA 3.Azim Breast 1.Azim HA HA 1.Azim profiling. 4 . 5 . Chromosomes HER 2 conducted. Overall the mutational landscape imbalances between pregnant patients explained variables datasets, looking into specificity • • • Charité 100 80 This presentation is the intellectual property of the author/presenter. author/presenter. the of property intellectual the is presentation This Parsberg, Germany Parsberg, , Jan Budczies,

, ,

TCGA-SET BCP-SET 60 Contact them at [email protected] for permission to reprint and/or distribute. and/or reprint to permission for [email protected] at them Contact

TCGA-SET HR-TCGA-SET HR-BCP-SET . cohorts

50 40 cohorts

Institute of Pathology, Pathology, of Institute 4 , , 20 % of samples with with mutationssamples% of BCP % of samples with with mutationssamples% of matched

-, matched Bahriye Aktas Bahriye in

0 , HR B) 0 1 RB1 ATM MYC D) TP53 AKT1 ESR1 PTEN BRAF CBFB CTCF RB1 CDH1 KRAS Clinic Landkreis Neumarkt ClinicLandkreis EGFR ATM MDM2 MYC TP53 AKT1 GATA3 FGFR2 ERBB2 MED12 8 ESR1 RUNX1 PTEN BRAF CTCF CBFB CCND1 PIK3R1 KRAS CDH1 EGFR PIK3CA MDM2 MAP3K1 MAP2K4 GATA3 ERBB2 FGFR2 CDKN2A MED12 RUNX1 CCND1 PIK3R1 PIK3CA MAP3K1 MAP2K4 CDKN2A

overall

80 60

Results

cohorts

TCGA-SET BCP-SET 60

patterns TCGA-SET HR+TCGA-SET HR+BCP-SET

40 trols cohorts 20 regnancy and non-pregnant and non-pregnant controls regnancy

matched , Jenny Furlanetto 5 % of samples with with mutationssamples% of con 20 -pregnant controls controls non -pregnant , Helios Clinics Wiesbaden, Germany Wiesbaden,Clinics Helios , 0 HR+, Mutation % of samples with with mutationssamples% of C) RB1 matched ATM MYC , WilkoWeichert² TP53 AKT1 ESR1 PTEN BRAF CTCF CBFB CDH1 KRAS EGFR MDM2 4 0 FGFR2 GATA3 ERBB2 MED12 RUNX1 CCND1 PIK3R1 PIK3CA MAP3K1 MAP2K4 CDKN2A non - RB1

ATM MYC

TP53 AKT1 ESR1 BRAF PTEN CTCF CBFB CDH1 KRAS EGFR MDM2 FGFR2 GATA3 ERBB2 MED12 RUNX1 CCND1 PIK3R1 pregnant PIK3CA MAP3K1 MAP2K4 CDKN2A A)

Figure 2. Figure 100

Gynaecology non -

cohorts TCGA-SET HR-TCGA-SET HR-BCP-SET vs .

45 cohort 60 - 38

26 - , Albrecht Stenzinger ,

4 7 (17.1%) BCP 34 (82.9%) 14 (34.1%) 27 (65.9%) Carsten Denkert

TCGA Departement of Departement

40 3 ,

in matched

1 (2.4%)

matched matched 21 (51.2%) 40 (97.6%) 20 (48.8%) 12 (29.3%) 29 (70.7%) % of samples with with mutationssamples% of 43 non - 20 37 vs . -cohort

28 - the matching the NCT, Section Translational Gynaecologic Oncology, Heidelberg, Oncology, Gynaecologic Translational Section NCT, 4 (9.8%) 7 status of 37 (90.2%) 23 (57.5%) 17 (42.5%) , , BCP HR -, 0

BCP

B) HR RB1

ATM MYC TP53 AKT1 ESR1 CTCF BRAF PTEN CBFB CDH1 KRAS EGFR MDM2 in FGFR2 GATA3 ERBB2 MED12 RUNX1 CCND1 PIK3R1 definition PIK3CA MAP3K1 MAP2K4 45 CDKN2A cohort

by - by 40

26 - 60 40 (27.8%) 24 (16.2%) 52 (49.5%) 23 (15.0%) 61 (39.6%) 93 (60.4%) 53 (50.5%) 104 (72.2%) 124 (83.8%) 130 (85.0%) TCGA

, Laura Kahmann

7 cohorts

s Hospital, Essen Hospital, s controls 㼿 patterns -matched non -matched 43 40

TCGA-SET HR+TCGA-SET HR+BCP-SET 34 -cohort 26 - 43 (43.4%) 56 (56.6%) 13 (13.1%) 86 (86.9%) 30 (30.3%) 82 (82.8%) 17 (17.2%) 48 (49.5%) 49 (50.5%) 69 (69.7%) BCP characteristics matched

pregnant -2 -2 -4 Mutation non - % of samples with with mutationssamples% of

G3 Tanja Neunhöffer³ Sonia Villegas Tanja T1 T3 G1 median positive positive positive positive positive positive , negative negative negative min-max Category linical 1 non -

University Women University C 0 HR+, 6

Frederik Marmé

, ,

A) RB1 ATM MYC TP53 AKT1 ESR1 CTCF BRAF PTEN CBFB CDH1 KRAS EGFR MDM2

FGFR2 GATA3 ERBB2 MED12 RUNX1 CCND1 PIK3R1 PIK3CA The project has partly been funded by the German Cancer Consortium (DKTK) and the BANSS Foundation. BANSS the (DKTK)and Consortium Cancer German the by funded been partly has project The Figure 3. Figure MAP3K1 MAP2K4 CDKN2A

Table 2. Table

Parameter Age, years size Tumor Nodal status Grading* HR* HER2* *Numbers in matched BCP-set vs.TCGA-set are identical are vs.TCGA-set BCP-set matched in *Numbers Institute of Pathology, Technical University Munich, Germany, Germany, Munich, University Technical Pathology, of Institute 2

. of by not

BCP (N=848) (N=848) Excluded using

Breast covers

therapy

were

classical

patients

have that . pregnant

and pregnancy

Proton/PGM on

Analysed Analysed Analysed tests that

before

diagnosis diagnosis

performed 10, 2016 10, designed :

(N=41/154, 26.6%) 26.6%) (N=41/154, - pools

associated with believed

(N=154/1002; 15.4%) (N=154/1002; 6 database

on between pregnant

is outside

mainly TCGA-Cohort (N=1002) TCGA-Cohort about the impact

taken

is ;

Torrent

were , Nicole Pfarr², KarstenWeber Eligibility: Eligibility: • M0 at • 25

of it

1 ;

statistical

mutations 4

primer , and custom

TCGA 3 Ion

All a

status .

known two biopsies

an ge ; is mutations diagnosed on December December from HR

analyses statement

dataset

dataset Matching based based Matching into •a • • HER2 status • grading

a core

Institute of Pathology, University Hospital Heidelberg Hospital University Pathology, of Institute Before Before matching further

data using . Little cancer Sibylle Loibl (N=42) (N=42) split

Excluded small

somatic 1,2 assayed

obtained using Consort Raw 45

(FFPE)

non - synonymous

German Breast Group; Neu-Isenburg, Germany, Germany, Neu-Isenburg, Group; Breast German of

one ) . 1. diagnosis diagnosis

-Cohort (N=141) -Cohort (N=141)

breast 1 Comparison of the mutational landscape of breast cancer during p cancerof breast during landscape of the mutational Comparison

processed 05 . in were .

Analysed Analysed Analysed Analysed 0 Only mutations a rare coexistence BCP

= amplicons cancer

is from ) . controls 

prognosis

pattern (N=41/99, 41.4%) 41.4%) (N=41/99, (N=99/141; 70.2%) 70.2%) (N=99/141;

Eligibility: Eligibility: • M0 at • 25

(Table the genomic level. Based Figure .

to a multicenter observational study for breast cancer embedded 4

at 236

and is

analysis the samples

set

somatic breast origin Background Background different genes

02 )

The was paraffin 9 3 8 3 3 2 5 3 1 3 3 3 6 9 4 2 5 for

11 10 15 43 25 29 12 19 25

03 not (version

with ) . Amplicons

1 of Amplicons comprises compare

non - pregnant

236 mutational level

Materials and Methods Materials germline - fixed to

22 that of and

is and

29 ; BIG exons 5

21 ) analysed Software about its biology (Figure patients

biologically

3 2 138 is Exons Exons 37 Formalin 2,3 1,2 3,4 being 5-9 2-4 3-5 4-6 3-9 1-9 study 3- 10 2- 20 2- 16 2- 16 1,4,8 1,3,4 panel 11,15 11,15 18 - 21

4,7,11 4,7,11 3,7,12

Exons Exons study

Suite of significance 138 BCPv2 Panel BCPv2 San Antonio Breast Cancer Symposium Breast Cancer Antonio Symposium San

as breast cancer biology ( BCPv 2,5,8,10,14,21 2,5,8,10,14,21 , 7,8,14,17,19-

the platform BCP on

2,3,8,13,14,16-18,20-

Gene pregnant of in - study (GBG

- sided

Torrent Panel regions

pregnancy

retrospectively non reported aim

BCP

RB1 ATM MYC TP53 TP53 Ion AKT1 AKT1 ESR1 ESR1 CTCF PTEN PTEN CDH1 BRA EGFR EGFR CBFB KRAS KRAS

MDM2 Genes Genes GATA3 GATA3 FGFR2 MED12 RUNX1 ERBB2 CCND1 PIK3R1 PIK3R1 PIK3CA PIK3CA

MAP2K4 MAP2K4 MAP3K1 CDKN2A CDKN2A 25 Genes Genes 25 Table 1. Table Breast cancer during pregnancy (BCP) pregnancy immunohistochemistry during The and enrolled contradicting results hotspot been default during pregnancy. during pregnancy. The were sequencing Cancer the Presented Presented at: 28 | Annual Scientific Report 2016 | Introduction

the

mg/m² 2015

150 150 but grade 1

from - 16 03

125 to

continuing. of P5

nP125 to SABCS is

6.4 [4.1-10.0] [4.1-10.0] 6.4 Results at

32.0 [6.0-122.7] 32.0

dose up .

125 125 was associated 56 .

mg/m² nP 9.0 [6.6-11.9] [6.6-11.9] 9.0 grade 1 345 - compared 83.4 [11.6-157.3] 83.4

Presented

. 17 : 150 and grade 3-4 (B)

PSN -paclitaxel -paclitaxel versus solvent-based 69 ) 150. 150. Follow-

170.4] [11.6-

at Paclitaxel nP150 - PSN to

of 2016 ; nP Nab

12.7 [8.9-16.0] [8.9-16.0] 12.7 of nab . 157.3 157.3

than to (GBG .

al ,

9 al

Median TTR n (weeks); [95% CI] TTR[95% Median n (weeks); et

occurred after

et toxic study A,

P PSN risk are needed. needed. are risk less up

PSN grade 2-4 (A) 7 [6.0-9.1] at

References 8.6 [4.0-123.3] 8.6

Conclusions Conclusions

of but München; München;

Jackisch C resolution (TTR) GeparSepto M,

to ,Stefan Paepke ,Stefan

8 Schneeweiss

C, , Sibylle Loibl , Sibylle resolve 17 Untch efficacious to Trier; , Trier;

randomized

neoadjuvant chemotherapy for early breast cancer (GeparSepto-GBG more in Jackisch

associated with a lower frequency is

M, nP125 4* 4*

uvant is

vs comparison groupscomparison , Mathias Warm , Mathias 7 nP Borromäerinnen

125 vs Untch Untch von von Minckwitz Table 1: Median time Table Figure 3: Time paclitaxel paclitaxel Oncol. Lancet 3 trial. phase 69 ): a randomised, weekly neoadj grade 3- grade grade 2-4 grade grade 2-4 grade 3- grade , Michael Untch , Michael nP150 nP150 P *Probable overestimation due to the non continuous follow- continuous non the to due overestimation *Probable 1. 2. higher higher frequency than P 80 .The with with a more rapid resolution compared patients selecting for Markers nP und Poliklinik für Frauenheilkunde, Frauenheilkunde, für und Poliklinik 1 A

in in in at Klinik Klinik This presentation is the intellectual property of the author/presenter. the author/presenter. of property intellectual the is presentation This 125 Mutterhaus Mutterhaus der PSN .5% nP of EOT

14 at Contact them at [email protected] for permission to reprint and/or distribute. reprint and/or for permission to Contact them at [email protected]

resolve Sherko Sherko Kümmel grade 1 14Klinik 14Klinik , FU TTR 6 significant no significant EOT EOT to not PSN G G PSN 3-4 grade 3-4 was group, resolved to G1 n= 8/18 (44.4%) n=18/31 (58.1%) Berlin; n= 31/385 (8.1%) treatment treatment phase not resolved to G1 39 .2% with Results . 0.001); PSN significant difference , Valentina Nekljudova Valentina , day 1. PSN grade 2-4 vs n=385 nP125

16 no ( on the P 80

in 150 - Buch rechts der Isar der TU München, TUMünchen, Isar der der rechts 125

grade 2-4 did Paclitaxel Epirubicin Epirubicin nab-Paclitaxel nab-Paclitaxel Cyclophosphamide Cyclophosphamide 125 Pertuzumab (HER2 positive) Pertuzumab (HER2 positive) Trastuzumab (HER2 positive) Trastuzumab nP . A) . Similarly, the FU nP EOT EOT

nP Berlin

PSN G G PSN 2-4

PSN University , - University Hospital

Surgery resolved to G1 Klinikum Klinikum 0.043) 0.043) but and n=23/45 (51.1%) ). treatment treatment phase 9 n=45/151 (29.8%) patients not resolved to G1

125, 125, whereas n=151/385 (39.2%)

of Clinics p= grade 3-4 did not resolve nP grade 2-4 was resolved 019 ( 150 .8% with , Charité , Bernd Gerber , Bernd Hermann , Hermann Wiebringhaus Köln; Köln; 0. 5 15 41 nP PSN 125

p= PSN 2.7% ; 17Helios 17Helios in 385 patients

in nP

B) .

s Hospital Essen, Essen; 5Institute of Pathology and German Cancer Consortium (DKTK), Charité-University (DKTK), Consortium Cancer and Charité-University German 㼿 s of Pathology 5Institute Essen, Essen; Hospital FU 125 EOT EOT 125: .1% .1% with and , - Holweide and Brustzentrum and PSN G G PSN 3-4

nP 51 nP

resolved to G1

n= 8/20 (40.0%) n= 20/32 (62.5%) treatment treatment phase 80 not resolved to G1 n= 32/220 (14.5%) , Rostock; of 150 150 und

vs 150 150 0.372) 0.372) (Table 1; Figure 3 nP nP n=220 nP150 p= Carsten Carsten Denkert ( was reported , 29 .8% 150 150 and 4

- Frauenklinik Serban Costa Dan , Serban

125 nP patients respectively. respectively. patients Gynäkologie PSN

14 in Krankenhaus Köln Krankenhaus in and FU of EOT EOT nP

PSN G G PSN 2-4 resolved to G1 150 n=35/92 (38.0%) n= 16/35 (45.7%) treatment treatment phase 0.535) (Table 1; Figure 3 Figure 1; (Table 0.535) n=92/220 (41.8%) not resolved to G1 and 44 .4% nP , 80 , 220 patients p= 125 group (nP150

of nP .7% 45 .7% with the patients respectively (Figure 2). 2). (Figure respectively patients the Bahriye Aktas Bahriye and patients treated with P

, of 3 of P,

.0% the

( nP125

. 162 weeks after EOT, the remaining Mitte; 8Brustzentrum - Mitte; 8Brustzentrum in FU 38 0.531). 0.531). Grade 3-4 EOT EOT .0% Magdeburg; 16Universitäts Magdeburg;

40 .0% of vs , , Michael Clemens , Michael

58 .1% p= PSN G G PSN 3-4 P, 19 n=5/5 (100%) 13

: resolved to G1 n=5/16 (31.3%)

Essen grade 2-4 was significantly different between between 2-4 grade was different significantly n=16/601 (2.7%) treatment treatment phase FU not resolved to G1 of Arm A A Arm Arm Arm B in R and

125 design P80 PSN 100 .0%,

n=601 nP

of patients with - Frauenklinik in The trial is financially supported by Roche Celgene by and supported is financially trial The 29 .8% vs 62 .5% of

150 150 and 8.1% in

nP of opportunity FU EOT EOT .8% Andreas Andreas Schneeweiss PSN G G PSN 2-4 Overall Overall 601 patients received P the the EOT EOT was observed (nP150 (nP150 31 .3%, After After a median 61 resolved The The TTR difference was seen between P grade grade 3-4 was significantly different between was observed for P P for observed was , R resolved to G1 2 randomized GeparSepto study (GBG 69) study (GBG GeparSepto randomized n= 21/34 (61.8%) treatment treatment phase 15Universitäts n=34/114 (29.8%) n=34/114 not resolved to G1 n= 114/601 (19.0%) n= 114/601 • • • Figure 2: Consort diagram Window Figure 1: Study , Blohmer UweJens 12 Brustzentrum an den Kliniken an Kliniken den Brustzentrum

2 of by %, and 150 150 inäres inäres PSN 29

. 3%, nP nP

case on vs In paclitaxel 125 mg/m . P peripheral and % to Interdiszipl

mg/m² mg/m² every vs as

Christian Jackisch Christian nP , 1

% 600 and comparable , John Hackmann , John to 38 patients with early 11 well P

in PSN as

10, 2016 10, nP and of subsequently continued -paclitaxel -paclitaxel 150/ grade 1 within 3 weeks, 125mg/m² 125mg/m² weekly (nP125) was improved over to grade 2-4 and 3-4 resolved 3-4 and resolved 2-4 grade 3-4 and resolved 2-4 grade and to nab

125 nP125 collect long-term data or PSN PSN to nP reported according treatment

a sequential taxane followed cyclophosphamide cyclophosphamide of be of December December 6 - , Claus Hanusch , Claus

- , nP150 10 one week P with P with P resolution. resolution. and 30 days after the last chemotherapy order weekly Gunter von Minckwitzvon Gunter ) ) increases the pathological complete 60 patients completed study treatment, will 3-4 for P compared 2 , 10Onkologische Schwerpunktpraxis, Bielefeld; 11Klinikum zum Roten Kreuz München; 12Marien 13 Witten; 12Marien Hospital München; zum Roten Kreuz 11Klinikum Bielefeld; Schwerpunktpraxis, 10Onkologische by solvent-based paclitaxel ( 10 in 1 part and and to Peripheral Sensory Neuropathy Occurrence and Resolution: Results from the neoadjuvant neoadjuvant the Results from and Resolution: Occurrence Peripheral Sensory Neuropathy and as PSN mg/m Berlin; 6St. Barbara Klinik Heessen, 7 Hamm; Heessen, Klinik Berlin; 6St. Barbara 125 125 grade 3-4 taxane treatment was discontinued. mg/m² mg/m² plus 80 nP nP GeparSepto showed that PSN grade 3–4 was 90 occurrence occurrence PSN of -paclitaxel -paclitaxel was reduced

Hospital, Hospital, Marianne Just Marianne of 150, 150, 1German Breast Group, Neu-Isenburg; 2Sana Klinikum Offenbach; 3National Center for Tumor Disease, University Hospital Heidelberg; 4University Women Hospital Heidelberg; 4University University Disease, Tumor Center for 3National Klinikum 2Sana Offenbach; Neu-Isenburg; 1German Group, Breast -paclitaxel -paclitaxel ( nP nab PSN grade 2-4 Objectives Objectives

nP nP If symptoms did not resolved Background Background of nab on patients treated with with treated patients -paclitaxel -paclitaxel 150mg/m² weekly (nP150) compared paclitaxel (P) Jenny Furlanetto Jenny overestimated. overestimated. PSN in case a phase 3 randomised neoadjuvant trial epirubicin to

), In mg) were every administered three weeks (Figure 1). nab of of relative dose lower intensity, frequency treatment (EOT), i.e. -paclitaxel -paclitaxel compared been of Materials and Methods Materials 840 weekly).

and 2 received paclitaxel dose reductions, treatment discontinuations nab 3-4 3-4 and (FU) data (FU) of end BC

(LD

treatment modalities. Since the data were retrospectively analyzed, time 1. mg FU on -paclitaxel -paclitaxel treatment was delayed

four cycles the day as patients treated with with treated patients with treated patients by nab at nP125 the study the protocol was amended patients with HER2+ tumors trastuzumab 6mg/kg (loading (LD) dose 8mg/kg) grade 2-4 grade

of of on of In to

San Antonio Breast Cancer Symposium Breast Cancer Antonio Symposium San well resolution (mTTR) PSN A safety interim analysis, conducted after

report follow- up report 1 end as to weekly. Therefore the Therefore dose weekly. of 2 resolution could have grade 2, We 2 grade 1 during1 grade Rate Percentage Percentage administration Percentage compared to to Time Time PSN 0.0001). A 0.0001). sub-analysis A showed that the risk-benefit ratio 0.001). mg/m PSN a lower dose (100mg/m

Primary objectives: Primary • • • • a outcome outcome to received dose Patients Patients with untreated weekly followed three weeks. taxane treatment was stopped. After the and pertuzumab and 420 pertuzumab at 80 The The GeparSepto (NCT01583426 breast cancer (BC), showed response (pCR) that rate compared epirubicin/cyclophosphamide (EC) neoadjuvant p< chemotherapy (pCR indicated a higher rate during the trial. The significantly final higher safety after results with better drug adherence pCR. sensory sensory neuropathy (PSN) with p< Presented Presented at: | 29

vs and 41 ;

has and the HR - than rash 10 % 034 ), pCR -3.

of Oncol . HER 2- of .4% diarrhea GBG 69): a 69): - GBG based - based 85 positive breast - positive pertuzumab

1. 23 part nab -Pac - papular CI with as Pac followed and treatment HER 2- patients 025 ), the the HER+/ . GeparSepto or

. %

3 neoadjuvant in

2+ patients was the pertuzumab 95

2- group ( ( p= 0 maculo ) . of - 356. of the

label, phase 2 trial. Lancet trial. 2 phase - label, HER patients

05 [ HER paclitaxel versus solvent versus - paclitaxel 003 of .

in , open , 0 043 ), 50 % along with a ≥ . in , toxicities 1 0

. 0.4%

< (p=

to Efficacy and safety of safety and Efficacy (p= vs

multicentre of pCR . 2016; 17: 345 17: 2016; Oncol .

, References Conclusions Conclusions 11 dyspnea anthracycline-containing and anthracycline-free chemotherapy chemotherapy anthracycline-free and anthracycline-containing chemotherapy for early breast cancer ( cancer breast early for chemotherapy HER 2+

HER 2+ than 0. 003 ) included anaemia (p=0. randomised together with nab-Pac anorexia and cyclophosphamide

and ): a ): p= in

/stomatitis/esophagitis toxicity, this treatment regime ),

neoadjuvant of the

in women with locally advanced, inflammatory, or early HER2 early or inflammatory, advanced, locally with women in - 32.

047 ) 001 , phase 3 trial. Lancet trial.3 phase , the largest cohort HER 2+ cohort, particularly . . 18 rubicin rubicin is 0 0 NeoSphere reported previously itive itive early breast cancer receiving a dual HER2- decrease from baseline. Multivariate regression analysis showed that the only significant predictor discontinuations were more frequent with with Pac. LVEF decreases from baseline were uncommon, with 2.0% showing decreases HR -negative status (OR 2. As (p= 78 .0%; p= 0. 006 ). High-grade toxicities common that were significantly more mucositis (p< epi the , Warm Mathias 1. Untch M, Jackisch C, Schneeweiss A et al. Nab et A Schneeweiss C, Jackisch M, Untch 1. in neoadjuvant paclitaxel randomised al. etYH Im T, Pienkowski L, Gianni 2. 2012; 13: 25 13: 2012; trastuzumab ( cancer 3. Schneeweiss A, Chia S, Hickish T et al. Pertuzumab plus trastuzumab in combination combination in trastuzumab plus Pertuzumab al. et T HickishS, Chia A, Schneeweiss 3. standard with cardiac II phase randomized a cancer: breast early HER2-positive with patientsin regimens 2278-2284. 24: 2013; Oncol. Ann (TRYPHAENA). study safety • This by pos targeted neoadjuvant therapy trastuzumab, in negative subgroup, supporting its use neoadjuvant treatment. noteworthy acceptable toxicity and achieved higher rates Although the HER2+ patients experienced more • • • 10 be -

, Michael , Michael Untch 1 + than

vs. HER2-) 001 001 001 001 001 034 451 550 025 043 047 003 008 023 410 00 003 513 120 001 012 00 005 038 004 685 p-value (HER2+ (HER2+ p-value 0. 0. 0. 0. 0...... 112 67 .8%) with

HER 2 0 0 0 0 < 0 0 0 0 0 0 1 0 0 0 0 0 1 0 0 0 < < < < 0 0

of responses , Kristina , Kristina Lüb 9 in

Rotkreuzklinikum Munich, Germany, Rotkreuzklinikum Germany, Munich,

). 13 University Hospital Schleswig-Holstein Hospital (UKSH), Schleswig-Holstein Germany University 52 .5%, 7 +/ hormone receptor 2 This presentation is the intellectual property of the author/presenter. the author/presenter. of property intellectual the is presentation This Overall N (%) N Overall patients 99 .9) 78 .0) 97 .8) 69 .5) 88 .5) 21 .8) 99 .9) 30 .1) 20 .3) 53 .3) 43 .1) 31 .5) 15 .2) 22 .2) 16 .9) 22 .1) HER HER2- safety set safety HER2- ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( ( - 60 .1% ( (3.3) (2.0)

(0.9) (0.5) (0.1) (0.5) (0.9) (0.1) (0.5) (0.5) (0.4) 809 632 792 563 715 7 2 27 176 4 2 809 244 164 1 2 431 16 2 349 4 255 7 123 1 180 4 137 4 3 179 were achieved , von Minckwitz Gunter of the 1 Sherko Sherko Kümmel Contact them at [email protected] for permission to reprint and/or distribute. reprint and/or for permission to Contact them at [email protected]

Helios Clinics Berlin-Buch Berlin-Buch Clinics Helios , in HER 2 8

18 & N (%)

) ) + 100 85 .4) 98 .5) 74 .0) 94 .2) 100 38 .4) 25 .8) 62 .1) 62 .1) 80 .6) 41 .7) 26 .3) 28 .5) Charite Berlin, Germany Berlin, Germany Charite 23 .2) 17 .9) ( ( ( ( ( ( ( ( ( ( ( ( ( ( Overall 6 (2.8) HER 2+ cohort excluding the window- of HER2+ safety set safety HER2+ (6.3) 25 92 ( 71 ( (2.0) 8 (7.6) 30 (1.0) 4 (1.8) 7 (2.5) 10 (1.0) 4 (0.3) 1 11 (2.5) 10 0 0 0 0 (2.0) 8 396 338 390 293 373 113 396 152 102 246 246 319 165 104 HER 2

the

Grade in

3 -4 3 -4 3 -4 3 -4 3 -4 3 -4 3 -4 3 -4 3 -4 3 -4 3 -4 3 -4 Any Any Any Missing Any Any Missing Any Any Missing Any Missing Any Any Any Any Any Any

(AE) of pCR (ypT0 ypN0) Results

Practice of Oncology, Germany Bielefeld, Oncology, of Practice 12

, Valentina Nekljudova , Valentina 17 events 52 .7% ( 45 .1%,60 .2%) with Pac (p=0. 212

, Hermann Wiebringhaus 7

Pathology Center Neuss, Germany Neuss, Germany Center Pathology 17 AE

AEs

ial interest ial

Adverse %* AP ALAT /stomatitis/esophagitis

HER 2- tumors, with the highest rate AE

-papular rash -papular

University Hospital Essen, Germany Germany Hospital Essen, University neutropenia spec 2. , Knut Engels Knut , 5 of Overall, higher rates nab -Pac and in negative (HER2+/HR-) cohort (Figure 2). A similar pattern was seen using the ypT0/is ypN0 definitionfor pCR (Figure 2) A sensitivity analysis opportunity patients showed a pCR rate AEs -hematological -hematological

16 y AE * •

• An Hematological Anemia Febrile Thrombopenia Non Increased Increased Mucositis Diarrhea Anorexia Maculo Dyspnea baseline from decrease LVEF≥10% <50and to Serious Table

, Christian , Christian Schem Cologne Clinics GmbH, Germany, GmbH, Germany, Clinics Cologne 6 11

48,9%

, Bernd Gerber 48,2%

2 15 0 (0.0) 0 N=810 Overall Overall HR+ ypT0 HR+ ypT0 ypN0 HR+ ypT0/is ypN0 HR- ypT0 ypN0 HR- ypT0/is ypN0 37 (4.6) 37 25 (3.1) 25 20 (2.5) 20 21 (2.6) 21 27 (3.3) 27 112 (13.8) 112 (14.7) 119 243 (30.0) 243 567 (70.0) 567 534 (65.9) 534 298 (36.8) 298 (27.0) 219 276 (34.1) 276 363 (44.8) 363 (19.4) 157 291 (35.9) 291 653 (80.6) 653 332 (41.0) 332 156 (19.3) 156 760 (94.1) 760 458 (56.5) 458 19,8%

nab-Pac HER2- nab-Pac Klinikum Südstadt Rostock, Germany Germany Südstadt Rostock, Klinikum HER2-negative 16 N=810/1206 (67.2%) N=810/1206 16,0% Carsten Carsten Denkert

, molecular subtypes Elisabeth-Hospital Kassel, Germany Germany Kassel, Elisabeth-Hospital 5 4

29,2%

26,3%

Pac HER2- Pac 2 (0.5) 2 5 (1.3) 5 N=396 16,2% Overall Overall

14 (3.5) 14 12 (3.0) 12 34 (8.6) 34 13 (3.3) 13 15 (3.8) 15 61 (15.4) 61 (18.7) 74 90 (22.7) 90 130 (32.8) 130 266 (67.2) 266 289 (73.0) 289 133 (33.6) 133 (25.8) 102 107 (27.0) 107 172 (43.4) 172 132 (33.3) 132 362 (91.4) 362 192 (48.5) 192 368 (92.9) 368 199 (50.3) 199 , Michael Clemens HER 2+ HER 2- & patients HER2-positive 6 12,0%

/is ypN0) N=396/1206 (32.8%) N=396/1206 of The trial was financially supported by Celgene and Roche. Roche. by and Celgene supported financially trial was The

Bahriye Aktas , Bahriye

81,4% 4

ypT0 Diakovere Henriettenstift Hannover, Germany Germany Henriettenstift Hannover, Diakovere 10

or Blohmer 74,6%

12 ) positive negative

64,3% PgR

PgR nab-Pac HER2+ nab-Pac Uwe

(IRS 0-5) (IRS

Category

(IRS 6-

56,4%

40 50 60 70 ER,

detected and/or

1-3 4a-c 4d 20 % 30 20 %

≤ > < 40 -< 70 + 30 -< 50 -< 60 -< G 2 G 1 Missing None Negative Positive Not both negative positive ER cT G 3 cT cT , Jens- 72,9%

University Hospital Heidelberg, Germany Germany Hospital Heidelberg, University Mutterhaus der Borromäerinnen Trier, Germany, Germany, Trier, Borromäerinnen der Mutterhaus 14

3 15 66,7%

biopsy , Sabine Schmatloch Pac HER2+ Pac 59,7%

node

49,7% Parameter

pathology) pathology) grading years PgR

Table 1. Baseline characteristics Table Figure 2. pCR (ypT0 ypN0 Age, Sentinel ER/ (central Tumor (central SPARC

Ki Clinics Essen-Mitte, Essen, Germany Germany Essen, Essen-Mitte, Clinics 9

, John Hackmann Surgery

. - II 13

to of ;

subanalysis of data from the randomized phase III GeparSepto trial phase III GeparSepto the randomized subanalysis of data from

(CI) was

were nab - phase

a 10, 2016 10,

regimen -

- positive phase

SANA Clinic, Offenbach, Germany Germany Offenbach, Clinic, SANA Marien Hospital Witten, Germany Hospital Germany Witten, Marien group 6 data

label, regimen pertuzumab Multivariable intervals -

, Andreas Andreas Schneeweiss , -

2 two . 12 weeks 12 neoadjuvant comparison HER 2

other definitions definitions other as in open in test loading dose) loading

negative loading dose) ( loading by

- 2 the 2 response , Claus Claus Hanusch , HER2+ HER2+ subanalysis: the 2 χ neoadjuvant

targeted

12 of patients receiving a pCR

a of December December confidence mg/kg

mg/m

840 mg mg 840

trastuzumab in - HER

of % of .

by 420 mg mg 420 by by 6 by in

blockade 95

vs - missing

data

HER 2 , randomised + part

corrected ,

2 -

followed followed Pertuzumab Pertuzumab followed followed pCR rates the HER2+ vs. HER2-negative cohort vs. HER2+ the HER2-negative

Epirubicin 90 Epirubicin Cyclophosphamide 600 mg/m 600 Cyclophosphamide ( mg/kg 8 Trastuzumab

as .

with

sided in

3 HER 2 - Dual HER2-blockade with pertuzumab and trastuzumab in HER2-positive early breast cancer: in HER2-positive trastuzumab and pertuzumab with Dual HER2-blockade dual

showed that substituting 12 weeks 12 , Christian Jackisch St. Barbara-Clinic Barbara-Clinic Hamm-Heessen, Germany St. HER for

safety

1 8 1 A fety fety two assess the pCR pCR the rates assess Marianne Marianne Just of

dual To

continuity and a

prospective with

Secondary objectives objectives Secondary • • Sa Patients weekly weekly and cyclophosphamide

paclitaxel 2 factors

impressive

analysis with using weekly weekly

2 test

TRYPHAENA

Sibylle Sibylle Loibl 1:1 efficacy -Paclitaxel -Paclitaxel German Breast Group, Neu-Isenburg, Germany Germany Neu-Isenburg, Group, Breast German 1

standard . subgroups and arms Safety

predictive

nab mg/m (125) 150 Paclitaxel Paclitaxel mg/m 80 multicentre,

1 N=1206 .

treated

R in

achieved a

Background Background Fisher’s and

All All tumor types

design

before study entry) entry) study before

ypT0 ypN0

) III GeparSepto study treatment anti-HER2 (after cancer Figure study

identify as .

biopsy Core in sided

treatment - reported study to focuses

taxane, epirubicin

response Patients and Methods Patients and Methods

NeoSphere paclitaxel of

two

breast - no the HER2+ subanalysis subanalysis the HER2+ cohort

were shown

trastuzumab of

weeks nab the 6

between San Antonio Breast Cancer Symposium Breast Cancer Antonio Symposium San analysis

GeparSepto having and positive performed studies,

opportunity -

the pCR defined pCR the defined

2 of as using consideration

- negative

design

from R was

parameters

HER comparing HER2+ tumors

window assess assess present entry) study (before compared

21 -

study

neoadjuvant phase HER 2 biopsy Core To

trial

Stratification factors: factors: Stratification • vs. HER2-/HR-vs. HER2-/HR+ HER2+/HR-vs.HER2+/HR+ •• >20%) vs. (≤20% Ki67 SPARC • Primary objectives Primary P 4- the considered analysis The The paclitaxel (nab-Pac) improved for the pathological standard complete response solvent-based (pCR) rate paclitaxel treatment (Pac) for high-risk significantly primary breast pertuzumab cancer. neoadjuvant patients design Trial GeparSepto (NCT01583426) was Statistical treated performed Figure 1: GeparSepto sequential regimen Efficacy III The and Presented Presented at: 30 | Annual Scientifi c Report 2016 | Introduction

of s of to

ESF arm event TX

en prior

- 04 84 .7%

t/prevent .2% 83 .2% vs.

significant TX EC - - 11 trea : D betwe No

the to P2 TX - in the trial ns warranted EC In

is .8%). 89 .8%).

C optio ETC ETC arm anemia rates, thromboembolic difference is 2%;

of no 87 . the other

In analys

as and a similar long term outcome . Epo number vs showed

90 .5% vs. Epo

benefit the

86 .1% risk especially Conclusions Conclusions

analyses B ESF

: D

ESF, However,

DFS careful

hand . an

grade

a patients, respectively. and 1512 received Epoetin beta. at of

and of

any n beta. , on and are , thromboembolic events

chemotherapy schedules have comparable incidences , OS 7

chen

high Epoeti Mün

they receive a preventive treatment with either Darbepoetin alfa .5%). is applicati anemia if or Figure 1: Disease free (DFS) and overall survival (OS) in (OS)Disease overall survival free 1: (DFS) and Figure C: EC- B: ETC, A: overall, ESF. to patients according High risk breast cancer patients treated with dense two different dose anemia

A 13 y

Schmatloch 3.6% vs 3.1% Universität ). Interestingly there were more thromboembolic events in applied (anemia

der der 355 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. ESF

n.s. n.s. n.s. n.s. n.s. n.s. p-value p-value This presentation is the intellectual property of the author/presenter. the author/presenter. of property intellectual the is presentation This , Sabine , Sabine to 6 p-value .3% vs. Epo Frankfurt-Höchst Klinikum ( GAIN) 13

no significant difference between the , Volker Möbus , Volker

chemotherap 1

Brustzentrum 6 Contact them at [email protected] for permission to reprint and/or distribute. reprint and/or for permission to Contact them at [email protected] beta to

was was Harbeck arm. PX : D 57 (7.7) 57 34 (4.6) 34 95 (6.3) 95 61 (7.9) 61 96 (12.5) 96 153 (10.1) 153 TX N (valid %) N (valid EC - Epoetin 4 (3.4) 5 (2.0) 5 (3.4) 3 (1.2)

there Nekljudova 96 (81.4) 96 200 (79.4) 200 128 (88.3) 128 205 (82.3) 205 EC - N (valid %) %) N (valid , Nadia , Nadia

Epoetin beta Epoetin 5 1482 patients were given Darbepoetin alfa but alfa according

Praxis Bethanien, Bethanien, Praxis 5 -TX -TX

the EC of Tesch Helios Kliniken Berlin-Buch, Kliniken Helios Results (N=1496) alfa

12 these 53 (7.0) 53 38 (5.0) 38 96 (6.5) 96 58 (8.0) 58 82 (11.3) (11.3) 82 135 (9.1) 135 events Valentina , Valentina . Ep o 7.7%; of

N (valid %) N (valid 1 grade 3/4 anemia was observed vs Darbepoetin Heidelberg, Heidelberg, 60 + years, positive - positive Study

, Hans 1 (1.0) 4 (1.5) 2 (1.7) 9 (3.9)

4 and 86 (88.7) 86 (83.5) 96 and patients treated with Epoetin (p=0. 221 (81.5)221 190 (82.6)190 N (valid %) %) N (valid of

age groups

Darbepoetin aged

3-4 3-4 3-4 any any any Klinikum Landshut, Landshut, Klinikum 11 CTC Grade 10 .1%

various in Schneeweiss Universitätsklinikum in Thromboembolic patients

4 n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s.

0.027 0.014 , in p-value p-value

TX Gunter von Minckwitz Gunter von - type (EPC: D 7.0% ESF ,

ETC EC overall 12 the Epoetin group

Hannover . in

Table Table 4: and

, Andreas Andreas ,

ESF 3

as Untch

l the

no significant difference within the treatment arms according Lueck of regimen wel n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. n.s. 4 (3.9) 5 (2.2) 10 (3.7) 10 (7.8) 10 Universitätsklinikum Hamburg, Hamburg, Universitätsklinikum 91 (88.3) 91 ESF 115 (89.8) 115 224 (83.0) 224 205 (90.3) 205 p-value N (valid %) N (valid 10 as Epoetin beta Epoetin the dose dense chemotherapies

, Michael

Darbepoetin group and

of Schwerpunktpraxis 11

and

ETC ETC Joachim Joachim beta treatment were were treatment observed

there was the (N=1498) alfa one

irrespective irrespective in

but ESF 17 (2.2) 17 29 (4.0) 29 46 (3.1) 46

629 (82.3)629 635 (87.2)635 chemotherapy N (valid %) N (valid 1264 (84.7) 1264

, Hans but Epoetin by 2

4 (3.5) Bauerfeind 10 (4.0) 10 (7.3) 10 13 (5.3) 13 by

9.1% - Onkologischen 213 (85.2) 213 101 (88.6) 101 (90.5) 124 206 (83.4) 206 N (valid %) N (valid

receive alfa The trial was financially supported by Amgen and Roche. Roche. and Amgen by supported financially trial was The in

mia the Darbepoetin to Darbepoetin 001 ), Universitätsklinikum Ulm, Universitätsklinikum Ingo 9 chemotherapy in

ane Jackisch 10 to r

stratified of 16 (2.2) 16 37 (4.9) 37 53 (3.6) 53

3-4 3-4 3-4 3-4 any any any any CTC 593 (83.2) 593 644 (86.1) 644 Grade N (valid %) N (valid 1237 (84.7) 1237 German Adjuvant Adjuvant Intergroup Node German Gynäkologisch as 3

Mülle Darbepoetin

well

, Christian

1 according as

Offenbach, Offenbach, 3-4 3-4 3-4 any any any

the ETC arm anemia was most frequently observed Volkmar rall In the incidence the incidence , Loibl the ETC arm (p<0.

9 bolic events occurred CTC Grade in

to ove

Klinikum Anemia suffered suffered from anemia d 3%). .

Universitätsklinikum Düsseldorf, Universitätsklinikum

8 ally grade 3/4 were slightly higher, Sibylle group

82 , Sana Sana TX

49 59 2 - 1 , EC ETC overall <40 >60 40 - 50 - Age Table 2: Table various age groups in ESF and to chemotherapy Anemia acccording 3: Table compare Epo Thromboem especi 2994 patients were randomized patients differences treatment Jens Huober

, 8

, - Isenburg of nt of

a and with

g/kg Neu

were as Fehm

10, 2016 5 events participating in the , - analyse

date adjuva

4 6 age .

(

Group Group

(N=1512) 169 (11.2) anja 707 707 (46.9) 568 568 (37.6) 538 (35.6) 406 (26.9) 288 288 (20.3) 465 465 (30.9) 846 (56.2) to

epirubicin 1162 (76.9) 1162 1169 (77.3) 1169 N (valid %) N (valid

72) (23- 49.9 T Epoetin beta Epoetinbeta 26 (16.5-52.7) (16.5-52.7) 26 by fatigue fatigue of

regimen (D)

the GAIN trial

breast cancer Breast

during

Elisabethenkrankenhaus Kassel, Elisabethenkrankenhaus In

by prophylaxis alfa

alfa defined December December Marion Marion Mackelenbergh van

ESF

and German

due – 1 breast cancer patients 136 136 (9.2) thromboembolic (N=1482) 490(33.2) 490(33.2) 678 678 (45.8) 563 563 (38.0) 520 (35.1) 399 (26.9) 329 329 (23.7) 823 823 (55.7)

1152 (77.7) 1152 1132 (76.4) 1132 N (valid %) N (valid alternately . 71) (20- 49.7 2-week intervals i.v. (idd In high-risk primary (16.9-51.4) 26.1 at Darbepoetin

in Efficacy safety and of Darbepoetin alfa or Epoetin beta in2994 risk high early breast cancer patients often overall

and different is

and patients received either subgroups been reported.

cancer and rising during the course

Darbepoetin

three courses each this subanalysis was either in two

of (OS)), %

or of anemia happened 50 and

breast overall overall

of (N=2994) 805 805 (26.9) 617 617 (22.0) 955 955 (32.0) 305 (10.2) 1131 (37.8) 1131 1058 (35.3) 1058 1385 (46.4) 1385 2314 (77.3) 2314 N (valid %) N (valid 1669 (55.9) 1669 2301 (76.9) 2301

72) (20- 49.8 receive paclitaxel weekly (Tw) plus capecitabine be beneficial 26.04 (16.5-52.7) 26.04 (16.5-52.7) received survival

ESF to

The aim

by applied te stimulating factors (ESF) serves (rate weekly) anemia have to

py-induced anemia.

positive

of be above application

Background Background therapy delays. Despite red blood cell transfusion type overall

each to

the ), patients node

to of outcome ESF

450 IE/kg

erythrocy All ( hemothera

well ty (DFS to Materials and Methods Materials

life during chemotherapy c ) comparison with conventional chemotherapy intense dose Epoetin beta. patient as g of primary In or by

ddEC followed safe estimated Epo Patient or ( for

. San Antonio Breast Cancer Symposium Breast Cancer Antonio Symposium San

survival and higher incidences

characteristics Allocation

regimen). accordin tion - (range)

Baseline Baseline option. .

but beta PgR positive

imen)

free TX

regimens have proven (range) (range) -invasive -invasive tumor paclitaxel paclitaxel (T), cyclophosphamide (C) all given efficacy

and Table 1: Table Parameter Age, years, median BMI, median pT1 pT2 pT3 pN1 pN2 pN3 ER and/or HER2 positive 3 grade Tumor Ductal Reduced quality treatment patients, Darbepoetin alfa Patients were randomly assigned ETC-reg (X)(EC- dense two dose dense regimens were evaluated (E), Epoetin biweekly) randomiza disease compared dyspnea caused the incidence treatment resulting the administration the chemotherapy Presented Presented at: | 31

2 CI

no ive HR PgR term

status HER 58 ,

cancer

negat . its long achieving

the

- 09 11

HR 1. in for tivity

PgR P1 breast an independent

worse treatment HER 2+ (OS: and

a pCR overall posi study Interestingly, ic and factor therefore

also the .

and 001 ) ve ve . 308 ; DFS: in primary HER 2 0 and but

system p< e in

001 ) whereas there was positi

achieving on 406 -2. predicti 0.

of 755 ; ER p< CI 1. ding respons subset chemotherapy demonstrate

deci fic

801 , to chemotherapy

1. higher

independent

Conclusions Conclusions speci when , djuvant HR

.2% v 5.8%; likely 9

an with

a 11

neoa more baseline, histological type type histological baseline,

trials neoadjuvant at

). 106 ). This was also observed 1

after considered served were

influence influence long-term outcome. associated Jens Huober Jens represent

after be ,

subgroups after adjustment for age, tumor stage and

255 -2. 8 stage stage not and

in 1. pCR Offenbach, Offenbach, CI did did

negativity patient outcome tumors Figure Figure 3: Forrest Plot for model multivariable demonstrating the logistic odds regression and nodal This analysis demonstrates that a should Sibylle , Sibylle Loibl 1

2 2 negative subgroup ( 625 , Klinikum 1.

and 2 2 status and HER2- subgroups (OS: involvement

HER ) neoadjuvant HR Sana Sana 7 , Claus , Claus Hanusch HER 7 patients 769 - Buch the nodal in

in This presentation is the intellectual property of the author/presenter. the author/presenter. of property intellectual the is presentation This l

Aachen, Aachen, Karsten Weber Karsten

as , 077 -2. 1 clinica Contact them at [email protected] for permission to reprint and/or distribute. reprint and/or for permission to Contact them at [email protected] an independent predictive factor for pCR overall (OR 1.

709 ; LRFS: well CI 1. as

26 -1. respectively). Multivariable Cox regression analysis revealed that PgR was Helios Kliniken Kliniken Berlin Helios 727 ,

1. 13 001 ) , Jackisch Christian 1. advanced 6 CI Universitätsklinikum 0. 6 001 , HR an B D p<

the PgR the PgR negative tumors 467 ,

. Results in have

Heidelberg, Heidelberg,

to HR 1 and overall survival (OS)

Elmar Stickeler Elmar , 5 and PgR negative breast cancer with tended 13 .8% v 7.5%;

, Universitätsklinikum Hamburg, Hamburg, Universitätsklinikum 353 -2.606; LRFS: 68 ; DDFS: 12 Michael von Minckwitz Michael Untch, Gunter free (LRFS) grade Universitätsklinikum CI 1. 5 13 ). Interestingly, 151 ).Interestingly, Mitte , , 5 261 -1. 878 , 1. 07 -2. Essen Essen ER positive higher

Frederik Frederik Marmé CI 1. 1. , DFS, C: DDFS, D: LRFS D: DDFS, C: DFS, , A C 4 HR Holstein - Holstein Kiel, the entire cohort ( Figure Figure 2: Disease free (DFS), distant disease free (DDFS), local recurrence with without pCR treated with neoadjuvant chemotherapy. A: OS, B: Kliniken 456 in 11 0.117). 0.117). 517 , CI .

p= HR 1. significantly Schleswig

of HR 1

18 %; 279 ; DDFS: , Andreas Andreas Schneeweiss , were , Christian , Christian Schem 12 3 the multivariable logistic regression analysis PgR negativity was 895 ; DFS: 248 -2. in .1% v v 22 .1% negative patients ; LRFS: 95 ; LRFS: 1. R CI Universitätsklinikums 4 309 -1. , Pg Universitätsklinikum Erlangen, Erlangen, Universitätsklinikum 10 299 -1. 1. in 687 , 001 ). CI 1.

, Thomas Karn , Thomas 0. B D 1 Frankfurt Volkmar Müller , Volkmar

p< 11 HR 1. 575 , 2 HER 2 positive ( 592 , CI 992 ; the HR 1. 1. in 373 ; DFS: (OR Universitätsklinikum Ulm, Universitätsklinikum 9 Universitätsklinikum 3 Sherko Kümmel , Sherko 424 -3. factor factor (OS: 2 difference difference Berlin, Berlin,

1. Valentina Nekljudova , Valentina ; DDFS: HR 1. 912 ; DDFS: 2 CI adjusting adjusting for known predictive factors Charite 2 , Blohmer 306 -1. 192 , Figure 1: Disease free (DFS), distant disease free (DDFS), free (DDFS), Disease 1: free disease (DFS), distant Figure in (OS)recurrence overall survival and local free (LRFS) negative or positive PgR and positive ER patients with A: chemotherapy. breast cancer neoadjuvant treated with D: LRFS C: DDFS, B: DFS, OS, pCR rates were significantly higher significant After prognostic Tumors lacking PgR expression (1172 patients) lacking PgR expression (1172 Tumors negative patients Patients with PgR negative disease had a significantly worse DFS, OS, DDFS and LRFS (p<0. 2. 1. A C

Uwe

- Isenburg . ER this Universitätsklinikum München, Universitätsklinikum ity data 8 Neu Carsten Denkert Carsten of , Jens- 10, 2016 , overall the - breast 1 10

HER 2 6 ival

in of Group Group compared

01 by 4

expression ; ER activ <0.0 <0.001 p-value p-value between PgR groups 0.304 0.00 <0.001 <0.001 initially involve The aim

Breast ER

(DDFS), were PgR expression

action not of defined

of December December non genomic effects

do – German a predictive factor for 1 long term surv as survival ), positive Peter A Fasching A Peter (LRFS))

exert (5.7) 245 (68) 2916 (26.3) 1126 (77.7) 2703 (22.3)776 ER+ PgR+ N (valid%) (7.2) 316 (63.9) 2815 (16.8)738 (7.4) 325 (4.7) 209 (50.2) 219 2 (45.1) 1967 (3.8) 167 (0.9)38 (77.2) 340 1 (17.2)759 (5.6) 247

ng-term outcome up, Outcome Outcome after neoadjuvant chemotherapy in progesterone receptor negative breast cancer patients - to subgroups free

in which the progesterone receptor a pooled a pooled analysis of individual patient data from ten prospectively randomized controlled Marion Mackelenbergh Marion van survival

and to

and

l growth factor signalling. disease (n=9785) German neoadjuvant trials free

(6.2) (1.6)

10 (3.4)38 (58.2) 654 (38.4)431 (63.8) 598 (36.2)340 ER+ – PgR N (valid%) (7.5)87 (63.8) 744 (15.5) 181 (7.0)82 72 (45.7) 528 (47.6)550 (5.2)60 18 (81.1)950 (12.1) 142 (6.7)79 s that utilize the non-genomic genes absent. Therefore lack overal a nuclear transcription factor alters the

ane ane based chemotherapy were included. been described or and proliferation progesterone receptor (PgR) as distant altered

tumor of Background Background ), in PgR PgR expression may act also

recurrence and tax if

has (DFS

expression,

Materials and Methods Materials (5.2) 283 (66) 3570 (28.8) 1557 (74.7) 3301 (25.3) 1116 All All patientsER+ N (valid%) (7.2) 403 (63.9) 3559 (16.5)919 (7.3) 407 (5.0) 281 (49.3) 2720 (45.6) 2517 (4.1) 227 (1.0)56 (78) 435 1 (16.2)901 (5.8) 326 local ER be decreased PgR

estrogen estrogen sensitive The

and survival San Antonio Breast Cancer Symposium Breast Cancer Antonio Symposium San

investigate neoadjuvant chemotherapy gene transcription. These different patterns their

unknown from overall to to in to Baseline or free

anthracycline an anthracycline (OS)

stimulate tumorigenesis -

the assumption that be a surrogate marker to +/ belongs. to ases ases pathologic complete response (pCR)(ypT0, ypN0

interacting with several cell signalling pathways that Table 1: Table characteristics grade Tumor 1 2 3 Negative positive HER2 statusHER2 Tumor stage Tumor cT1 cT2 cT3 cT4a-c cT4d Nodal stage Nodal cN0 cN1 cN2 cN3 Histological type Histological invasive Ductal invasive Lob ular others The estrogen receptor (ER) gene lead order expression would could study was response transcription cancer cancer patients. by incre 5613 patients with primary breast cancer, follow- HER2 The receiving (disease survival according Presented Presented at: 32 | Annual Scientifi c Report 2016 | Introduction

II 02 of on ): ): a be with

had

66 09 - Results stage patients : response ) in

not in TNBC P 1- patients in dose-dense general 43 events

compared in by in (TNBC

the therapy as carboplatin and/or complete

of mproved EFS. i cancer of carboplatin

intact tumors deficiency (either an mutation. Nodal -

56 . chemotherapy breast

Pathological HR the addition BRCA and , . 747 - ation) was 40603 (Alliance). J Clin Oncol. HRD (p=0.1662) were al understand the role of 2223 , Figure 4). 13 . Neoadjuvant carboplatin

BRCA (p=0. 0526 , Figure 3). al to et the HRD score

this underpowered study.

and negative

et - of

status before treatment in PA, not S,

neoadjuvant ones -week paclitaxel followed . Impact

34 .3 months for EFS, 2014 ; 15 (7): triple al

per nodal et of Loibl on pathologic complete response rates and

HR -deficient tumors showed a better A, adding carboplatin could BRCA mut score Fasching

of , Peter Peter Klare , containing . not predict the effect E, - References high score

or 12 Conclusions Conclusions Please contact them at [email protected]. [email protected]. at them contact Please mutation HRD

2014 ) )

HRD and the value - non -deficient and lymphocyte predominant breast cancer Hahnen

0. 0231 ) but

and HER2-positive early breast cancer (GeparSixto; GBG patients without

clinical Schneeweiss carboplatin the G, ), ),

on EFS HR Women’s Hospital Erlangen, Germany, Germany, Erlangen, Hospital Women’s G, p= in gBRCA (ASCO

(

by to neoadjuvant once- 21 . after , the results can help but not statistically significant EFS rates and cyclophosphamide multivariate analysis revealed that

and LPBC remained the strongest prognostic 13 - been reported. 0048 0202 BRCA

Minckwitz HR - non -deficient patients (p=0. rates

, Gerber Bernd

University ) GeparSixto 5 -deficiency 11 have ( p= 0. (p=0. The After median follow- up Overall, patients with than EFS Patients with better to HR -deficiency did EFS (Figure 5). (LPBC; 1). independent significantprognostic factors for EFS (Table von pCR germline ( • from • • • • Within the GeparSixto study HRD the score high associated with a higher pCR rate 1. status along factorswith Carboplatin therapy. especially However HR The effect predicted 2. von Minckwitz 3. Sikov WM, Berry DA, Perou CM, to III 2015 ; 33 (1): triple-negative breast cancer: CALGB randomised phase 2 trial. Lancet Oncol. with with triple-negative bevacizumab doxorubicin This presentation is the intellectual property of the author/presenter. author/presenter. the of property intellectual the is presentation This

Women’s Hospital, SRH Wald-Clinic Gera, Germany, Germany, Gera, Wald-Clinic Hospital, SRH Women’s

Bethanien-Hospital, Germany Frankfurt am Main, Bethanien-Hospital, , value - .893 .862 .917 .546 .602 .708 .597 .521 .879 .486 .452 p 0 . 1662 0 . 0231 0 . 8151 0 .6116 0 . 0202 0 . 1052 0 . 0538 0 . 0048 Test for

Interaction 967 214 523 905 139 144 740 1 . 201 in subgroups , Christian , Christian Jackisch 33 - 95 % 95 % CI 10 .442 (.166, 1.18) .398 (.170, .931) .495 (.185, 1.32) .433 (.191, .980) .430 (.210, .883) .458 (.102, 2.06) .407 (.194, .856) .627 (.181, 2.17) .347 (.121, .999) .496 (.227, 1.08) .369 (.108, 1.26) .472 (.227, .978) .353 (.103, 1.21) .502 (.242, 1.04) .468 (.242, .905) .233 (.026, 2.09) .408 (.182, .915) .475 (.139, 1.62) .379 (.172, .833) .596 (.207, 1.72) .188 (.020, 1.82) .506 (.263, .973) .441 (.236, .826) Hazard Ratio (95% CI) 0 . 146 -0. 0 .3 0 . 535 -2. 0 . 578 -2. 0 . 253 -0. 0 . 292 -1. 0 . 880 -7. 1 . 301 -4. 6 21 21 21 21 21 21 21 21 21 21 21 21 21 21 21 21 HR 0 . 37 0 . 632 1 . 088 1 . 208 0 . 478 0 . 576 2 . 507 2 . 483

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

for the GBG/AGO-B study groups groups study the GBG/AGO-B for

hazard ratio

1,16 , Dirk , Zahm Dirk M. PMCb better PMCb better PM 9 Value Center for Hematology and Oncology Center for Hematology N+ vs. N0 yes yes vs. no yes yes vs. no Charité, Breast Center, Berlin, Germany, Germany, Berlin, Center, Breast Charité, 16 G3 G3 vs. G1-2 ≥ 50 vs. <50 9 cT2-4 vs. cT1 PMCb vs. PM 0.1 0.2 0.5 0.10.1 0.2 0.2 0.2 0.2 0.5 0.5 0.5 0.10.10.1 0.2 0.2 0.2 0.5 0.5 0.5 0.10.10.10.10.10.1 0.2 0.2 0.2 0.2 0.2 0.2 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.10.10.1 0.2 0.2 0.2 0.5 0.5 0.5 0.1 0.2 0.5 ≥ 60% vs. <60%

60 57 35 54 76 50 63 48 67 76 43 129 136 142 139 109 143 130 145 109 117 150 193

N patients

, Sibylle , Sibylle Loibl

Predictor Table 1. Multivariate model for EFS Table HRD Arm age Grading LPBC Ki 67 cT cN , Jens U. Blohmer 8 ich, Munich, Germany, ich, Munich, Germany, low high no yes no yes Negative/Uncertain Positive negative positive G1+2 G3 <60% >=60% no yes no yes <50 >=50 cT1 cT2-4 gBRCA tmBRCA score HRD HRdeficiency Grading Ki67 LPBC Family risk Subgroup Overall Age cT cN Mun Results N=56 TU Figure 5. Univariate model for EFS overall & 15 , Karsten E. Weber 1 (N=57, 29.5%) 29.5%) (N=57,

N=3 , Mahdi Rezai

48 48 48 48 N=4 HR non-deficient 7 mutations

42 42 42 42 arm 4 Censored Censored 5 N=1 + + 36 36 36 36 9 6 2 6 -rank p=0.2377 -rank -rank p=0.3708 -rank t BRCA mutant mutant N=29 BRCA 30 30 30 30 Log Log 2 2 9 2 (N=54, 28.7%) 28.7%) (N=54, N=1 treatment 24 24 24 24 BRCA mutant (N=35, 19.8%) 19.8%) (N=35, g 2 2 2 2 to EFS, months EFS, EFS, months EFS, N=18 18 18 18 18 8 4 4 3 HRD & 12 12 PMCb in HRD 12 12 HR=1.846 (95% CI 0.657-5.188)(95% HR=1.846 HR=1.808 (95% CI 0.485-6.735)(95% HR=1.808 deficient deficient 1 6 6 5 of 6 6 6 6 HR Christoph T. Salat T. Christoph 4 7 (N=136, 70.5%) 70.5%) (N=136, (N=193) (N=193) 8 6 C) by 0 0 0 0 ,

BRCA intact tumors+PMCb

0% 0% percent event free event percent 0% 0% 20% 40% 60% 80% 20% 40% 60% 80% percent event free event 40% percent 20% 60% 40% 80% 20% 60% 80% 100% 100% 100% 100% Samples analyzed analyzed Samples high N=81 HRD Score C) tm (N=129, 66.8%) 66.8%) (N=129, high HRD score HRD high low HRD score HRD low HR deficientHR HR non-deficientHR

Department of Gynecology, Luisenkrankenhaus GmbH, Düsseldorf, Germany, Germany, GmbH, Düsseldorf, Luisenkrankenhaus of Gynecology, Department 8 48 48 48 48 Center for Hereditary Breast and Ovarian Cancer, Center for Integrated Oncology (CIO), University Hospital Cologne, Cologne, Germany, Cologne, Hospital University Cologne, (CIO), Oncology Germany, Center for Integrated and Ovarian Cancer, Breast for Hereditary Center Breast Center Clinics Essen-Mitte, Essen, Germany, Germany, Essen-Mitte, Essen, Center Clinics Breast 4 Figure 2. Overlap 42 42 42 42 0 0 14 0 0 8 Censored Censored 4 + + 36 36 36 36 2 1 3 1 5 7 2 7 -rank p=0.1780 -rank -rank p=0.3773 -rank 30 30 30 30 6 dreas Schneeweiss 1 6 Log Log 2 7 8 9 8 24 24 24 24 2 1 An 5 1 8 9 EFS, months EFS, D in PM arm PM arm D in 0 9 EFS, months EFS, , 18 18 BRCA intact tumors & according 18 18 5 3 2 treatment 5 2 3 3 3 3 HR 12 12 The study was financially supported by Myriad and Teva Teva and Myriad by supportedfinancially was study The tm 12 12 HR=1.664 (95% CI 0.787-3.518)(95% HR=1.664 HR=1.454 (95% CI 0.630-3.355)(95% HR=1.454 to 3 2 6 2 8 7 8 6 6 6 in 6 6 , Suzanna Chau², Julia Reid², Valentina Nekljudova , Suzanna Chau², Julia Reid², Valentina 3 2 6 2 2 0 BRCA intact tumors+PM ) B) by 4 1 9 0 0 0 0

3 2

7 2 0% 0%

percent event free event percent 20% 60% 40% 80% 20% 60% 40% 80% 100% 100% 0% 0% percent event free event 40% percent 20% 60% 80% 40% 20% 60% 80% 100% 100% low B) tm 3 2 7 2 vs . Praxisklinik, Berlin; Germany, Berlin; Germany, Praxisklinik, 13 HR deficientHR HR non-deficientHR high HRD score HRD high low HRD score HRD low 48 48 48 48 , Peter A. Fasching A. , Peter 4 42 42 42 42 Censored 8 4 0 0 0 0 3 4 Censored + Rita , Rita Schmutzler 36 36 for the TNBC subgroup + 36 36 1 3 4 2 -rank p=0.2223 -rank 3 2 15 7 3 -rank p=0.0526 -rank 30 30 30 30 Log 8 1 0 8 Log 1 1 0 24 24 06 status status 24 24 -deficiency HR -deficiency & according HRD score (high 2 1 EFS, months EFS, 3 1 epke 2 1

design 1 EFS, months EFS, 18 18 10 18 18 7 6 4 1 Eric Hahnen by by Pa 2 1 HRD 7 12 12 21 HR=1.546 (95% CI 0.764-3.127)(95% HR=1.546 , 12 12 HR=1.805 (95% CI 0.985-3.309)(95% HR=1.805 4 1 3 4 1 3 1 3 6 6 29

6 6 BRCA intact tumors 9 5 5 2 A) by 3 2 7 0 0 36 tm 0 0

0% 0% 5 2

percent event free event percent 20% 40% 60% 80% 20% 40% 60% 80%

100% 100% 3 2 0% 0% 80% free event percent 40% 60% 20% 20% 40% 60% 80% 100% 100% , Stefan A) 6 3 4 2 14 high HRD score HRD high low HRD score HRD low Figure 1. Study Figure 3. EFS Figure 4. EFS HR deficientHR HR non-deficientHR Medical Center for Hematology and Oncology Munich MVZ GmbH, Munich, Germany, Munich, Germany, MVZ Munich GmbH, Oncology for and Hematology Center Medical Department of Obstetrics and Gynecology, Helios Clinics Berlin-Buch, Germany, Berlin-Buch, Clinics Germany, Helios and Gynecology, of Obstetrics Department 7 in the neoadjuvant phase II GeparSixto trial in triple-negative early breast cancer early breast GeparSixto trial in triple-negative phase II neoadjuvant the in 3

University Women’s Hospital Rostock, Germany, Germany, Hospital Rostock, Women’s University 2 a 0 in in to / 12

as we 0 0 the and with HR -

as LOH

HRD HRD were

been ypT 0 0

5 3 allelic 1

- fixed, Myriad, Myriad, BRCA BRCA study 2 had hazard

studies

7 3 HRD Fisher’s

Patients

2 1 a

, ) . the trial,

platinum - negative II BRCA 1 has . Sherko Sherko Kümmel high The

The and 4 1

carboplatin study 5 3

tumor as the A

or BRCA ( pCR ; of 05 .

defined . patients of .

8 3

of 6 4 (EFS 0 triple formalin

BRCA = 42 the of g

test 4 carboplatin g phase 1

6 4 α

tumor

( telomeric , Kirsten Timms², Michael Michael Untch Timms², , Kirsten 315 was

≥ 1

irrespective study. in a 4

6 4 assess

II proportional effect

the

carboplatin

of 4 HRD to 1 7 5

GeparSixto large response survival

from

GeparSixto in

breaks, resulting

transitions)

the with chemotherapy mutations 5 score

7 5 1 score Cox

assessment

. two

66 ) was a multicenter,

free mutation 5 the

1 used

DNA

mutation from without in

and in prognosis, TAI

Pretherapeutic

HRD

Myriad’s those BRCA EFS

predict

was biopsies

complete ) . Women’s Hospital Heidelberg; Germany, Germany, Heidelberg; Hospital Women’s to a two-sided

event

segment Addition ), 3 Group, Neu-Isenburg, Germany, Germany, Neu-Isenburg, Group,

Homologous recombination deficiency (HRD) score as a measure to predict the effect of carboplatin on survival on survival of carboplatin effect score as a measure to predict the (HRD) deficiency recombination Homologous and

somatic or BRCA

the

Women’s Hospital Offenbach, Germany, Germany, Hospital Offenbach, Women’s

1 and 11 TNBC can the

better tm - Meier either .

neoadjuvant core using algorithmic

and a high frequency 1

detected for

with

enrolled open -label phase (long a to

DNA -damaging drugs such

40603

subgroup

42 . as University with an

Guntervon Minckwitz 6 to

BRCA

HRD

≥

rate regression

analyze

Figure 1.

Kaplan tumors of score as - based

as pathological

(FFPE) TNBC in ( TNBC) status

to as

German Breast Breast German

ients

1 ) . general heterozygosity CALGB TNBC pCR

defined LST HRD - score Background & Aim & Background

pat

in pCR .

HRD logistic used

HRD

whether

of . in taxane

and

Materials and Methods Materials of

defined 2

defined mutation and defined retrospectively and was

the had

increase

and higher

shown

cancer

repair double-stranded

with ) were (Figure confirmed mutations are commonly associated with

embedded is a

status to to

(loss

rate was was

was

test

) pCR survival

0 significance level was set determine BRCA

predictor BRCA BRCA

GeparSixto with Homologous recombination–deficient (HRD) tumors have lost the ability agents. Genomic instability breast Patients general anthracycline/ shown g increased susceptibility ypN To g on score correlated a centrally ( The prospective, randomized, deleterious intact deficiency mutation Statistical consideration GeparSixto (NCT01426880; GBG design paraffin assessed score score imbalance) methods ( tm exact Presented at the 2016 San Antonio Breast Cancer Symposium, December 6-10, 2016. December Symposium, Cancer Breast Antonio San 2016 the at Presented Randomized, open-label, phase II study comparing the efficacy and the safety of cabazitaxel versus weekly paclitaxel given as neoadjuvant treatment in patients with operable triple negative or luminal B/HER2 normal breast cancer (GENEVIEVE)

Stefan Paepke1, Jens Huober2, Sherko Kümmel3, Holger Eidtmann4, Michael Untch5, Serban Dan Costa6, Jens Uwe Blohmer7, Wolfgang Eiermann8, Bernd Gerber9, Claus Hanusch10, Jörn Hilfrich11, Christian Jackisch12, Andreas Schneeweiss13, Carsten Denkert14, Knut Engels15, Peter Klare16, Christian Bayer17, Nicole Burchardi18, Sibylle Loibl18, Gunter von Minckwitz18 Abstract 350 1TU München, Germany; 2Frauenklinik Ulm, Germany; 3Klinikum Essen Mitte, Germany; 4Universitätsklinikum Schleswig-Holstein, Kiel, Germany; 5HELIOS Klinikum Berlin-Buch, Berlin, Germany; 6Klinikum der Otto-v.-Guericke-Universität Magdeburg, Germany; 7Charité, Universitätsmedizin Berlin, Germany; 8Interdisziplinäres Onkologisches Zentrum München , Germany; 9Klinikum Südstadt Rostock , Germany; 10Rotkreuzklinikum München, Germany; 11Eilenriede Klinik Hannover ,Germany; 12Klinikum Offenbach, Germany; 13Universitätsklinikum Heidelberg, Germany; 14Charité, Institute of Pathology, Berlin, Germany; 15Zentrum für Pathologie, Zytologie und Molekularpathologie Neuss, Germany; 16Praxisklinik Berlin, Germany; 17Universitätsklinikum Erlangen, Germany; 18German Breast Group, Neu-Isenburg, Germany;

Background Results

• Cabazitaxel is a new taxoid promoting the tubulin assembly in vitro and stabilizing Between 4/2013 and 6/2015, a total of 333 patients have been randomized and started treatment with 74.7 and 83.2% of patients completing treatment in the cabazitaxel and paclitaxel arm, respectively. microtubules against cold-induced depolymerization as efficiently as docetaxel. 1,2,3 Among patients who discontinued taxane treatment due to progression, 50% in the cabazitaxel arm vs 55% in the paclitaxel arm had TNBC, and 50% vs 18% were cN0, respectively (Figure 2). Baseline • It has shown superior survival against mitoxantrone plus prednisone in docetaxel pre- characteristics were well balanced (Table 1). pCR was lower in the cabazitaxel arm compared to the paclitaxel arm (Table 2). A total of 48 (29%) patients in the cabazitaxel arm and 20 (12%) in the treated hormone refractory metastatic prostate cancer patients leading to its registration. paclitaxel arm had at least one serious adverse event (SAE). Serious hematological and non-hematological toxicities occuring with cabazitaxel and paclitaxel are displayed in Table 3. A favorable toxicity profile with a low rate of alopecia was shown. 4,5,6,7 The median time of treatment exposure was 12 (range 3-15) weeks in the cabazitaxel arm corresponding to a median of 4 cycles (range 1-4) compared to 12 (1-17) weeks in the paclitaxel arm (p=0.019). The GENEVIEVE study compares cabazitaxel to weekly paclitaxel, which is currently The relative total dose intensity did not differ between the cabazitaxel and paclitaxel arm, with 100% (range 1.6-101.2) and 100% (8.3-102.4), respectively (p=0.822). There were significantly less treatment most widely used in breast cancer patients. delays in the cabazitaxel arm (19.9%) compared to the paclitaxel arm (44.3%, p<0.001), which were mostly due to organizational reasons (11.4 vs 24.6%, p=0.003). Dose reductions were observed in 9.6% patients in the cabazitaxel arm compared to 11.4% in the paclitaxel arm (p=0.721). Main reason for dose reductions were non-hematological toxicities in 3.0 vs 7.8% (p=0.087), respectively. Figure 2: Consort diagram Table 2: pCR rates Materials and Methods Cabazitaxel Paclitaxel Overall Patients screened (N=166) (N=167) (N=333) p-value* GENEVIEVE (NCT01779479) is a prospective, multicenter, randomized, open label, N=407 study comparing the efficacy and the safety of four 3-weekly cycles cabazitaxel pCR (primary) N (%; 95% CI) N (%; 95% CI) N (%; 95% CI) versus 12 weekly paclitaxel given as neoadjuvant treatment in patients with operable Randomized and Randomized and ypT0/is ypN0/+ 2 (1.2; 0.0 – 2.9) 18 (10.8; 6.1 – 15.5) 20 (6.0; 3.5 – 8.6) 0.001 triple negative (TNBC) or luminal B/ HER2-normal breast cancer (Figure 1). started Cabazitaxel started Paclitaxel Evaluable LN Evaluable LN Evaluable LN pCR (secondary) Randomization was conducted in a 1:1 ratio with stratification factors as follows: treatment treatment N=154 N=151 N=305 • nodal stage status (cN0 vs. c(p)N+) N=166 N=167 ypT0 ypN0 1 (0.6; 0.0 – 1.9) 10 (6.6; 2.7 – 10.6) 11 (3.6; 1.5 – 5.7) 0.013 • subtype (TNBC vs. luminal B/HER2-normal) N=42 discontinuations N=28 discontinuations ypT/is ypN0 2 (1.3; 0.0 – 3.1) 13 (8.6; 4.1 – 13.1) 15 (4.9; 2.5 – 7.3) 0.007 13 adverse event 7 adverse event ypN0 47 (30.5; 23.2 – 37.8) 42 (27.8; 20.7 – 35.0) 89 (29.2; 24.1 – 34.3) 0.694 Objectives: 6 investigator`s decision 6 investigator`s decision *two-sided Fisher's exact test between arms Primary objective was to compare the pathologic complete response (pCR) rate in the 1 patient`s wish 4 patient`s wish breast (ypT0/is ypN0/+) between the treatment arms. 20 progression 11 progression Table 3: Serious hematological and non-hematological toxicities per patient 2 death (unknown reason) pCR according to different definitions, as well as toxicity (NCI-CTC-AE V4.03) and Cabazitaxel Paclitaxel Overall p-value* compliance in both arms were defined as secondary objectives. Completed Completed Cabazitaxel Paclitaxel SAE N (%) N (%) N (%) Sample size calculations: N=124 N=139 Anemia 2 (1.2) 0 (0.0) 2 (0.6) 0.248 Assuming a 15% pCR rate, defined as ypT0/is ypN0/+, in controls (GBG database) and N=83 surgery after cabazitaxel N=88 surgery after paclitaxel Leukopenia 2 (1.2) 0 (0.0) 2 (0.6) 0.248 targeting a smallest clinical improvement of 10% (i.e. pCR = 25% in experimental arm), a Additional Additional total of 326 patients (163/arm) were required for the one-sided Fisher’s exact test (α=0.1). Neutropenia 23 (13.9) 1 (0.6) 24 (7.2) <0.001 EC/Taxan EC/Taxan Accounting for 2% of patients randomized but not treated, a total of 332 randomized N=78 N=77 Febrile neutropenia 17 (10.2) 1 (0.6) 18 (5.4) <0.001 patients were needed. Diarrhea 5 (3.0) 0 (0.0) 5 (1.5) 0.030 N=78 surgery N=75 surgery Figure 1: GENEVIEVE study design. Other gastrointestinal disorders 2 (1.2) 1 (0.6) 3 (0.9) 0.623 Table 1: Baseline patient and tumor characteristics Infection 6 (3.6) 8 (4.8) 14 (4.2) 0.786 Baseline Cabazitaxel Paclitaxel Overall p- Nervous system disorders 2 (1.2) 0 (0.0) 2 (0.6) 0.248 Treatment until surgery, disease progression, unacceptable parameter (N=166) (N=167) (N=333) value* toxicity, or withdrawal of consent Renal and urinary disorders 6 (3.6) 0 (0.0) 6 (1.8) N (valid %) N (valid %) N (valid %) 0.015 Age years 51 (23-78) 53 (25-77) 52 (23-78) 0.222 *two-sided Fisher's exact test between arms Immediately before surgery: (median, range) Presurgical core biopsy (optional) Tumor stage by cT1 39 (24.1) 37 (22.3) 76 (23.2) 0.594 Conclusions Cabazitaxel yes sonography cT2 120 (74.1) 123 (74.1) 243 (74.1) 25 mg/m², q3w If no cT3 3 (1.9) 6 (3.6) 9 (2.7) The GENEVIEVE study showed no short-term effect of cabazitaxel in TNBC or luminal B/HER2- primary R N=326 pCR missing 4 1 5 breast cancer, while there seemed to be no differences in drug exposure and patient compliance between cCR in the two arms. and no Nodal status cN0 68 (41.0) 77 (46.1) 145 (43.5) 0.377 breast no EC

Surgery cN+ 98 (59.0) 90 (53.9) 188 (56.5) and nodes before: References Paclitaxel Subtype TNBC 65 (39.2) 66 (39.5) 131 (39.3) 0.946 Core neg e.g. 80 mg/m², q1w e.g. Luminal B/HER2- 101 (60.8) 101 (60.5) 202 (60.7) Biopsy 4xEC 1. Bouchet BP et al. Cabazitaxel, a new taxane with favorable properties. Drugs Today (Barc). 2010; 46:735-742 pos 4xEC Grading Grade 1 3 (1.8) 2 (1.2) 5 (1.5) 0.629 2. Attard G et al. Update on tubulin binding agents. Pathol Biol (Paris) 2006; 54: 72–84. 3. Bissery MC et al. Preclinical evaluation of TXD258, a new taxoid. Proc Am Assoc Cancer Res (AACR) 2000, 41: Abst 1364. Grade 2 53 (31.9) 61 (36.5) 114 (34.2) 4. Villanueva C et al. A multicentre dose-escalating study of cabazitaxel (XRP6258) in combination with capecitabine in patients with before after EOT at surgery Grade 3 110 (66.3) 104 (62.3) 214 (64.3) metastatic breast cancer progressing after anthracycline and taxane treatment: a phase I/II study. Eur J Cancer 2011; 47:1037-1045. rando 2nd cycle if pCR/cCR 5. de Bono JS et al. Prednisone plus cabazitaxel or mitoxantrone for metastatic castration-resistant prostate cancer progressing after Histological Ductal invasive 141 (84.9) 144 (86.2) 285 (85.6) 0.942 docetaxel treatment: a randomised open-label trial. Lancet 2010; 376:1147-1154 Inclusion criteria: tumor type Lobular invasive 9 (5.4) 8 (4.8) 17 (5.1) 6. Mita AC et al. Phase I and pharmacokinetic study of XRP6258 (RPR 116258A), a novel taxane, administered as a 1-hour infusion every 3 Core biopsy/ weeks in patients with advanced solid tumors. Clin Cancer Res 2009; 15: 723–30. TNBC or HR+/HER2-, cT3 or cT2 or other 16 (9.6) 15 (9.0) 31 (9.3) blood collections 7. Pivot X et al. A multicenter phase II study of XRP6258 administered as a 1-h.i.v. infusion every 3 weeks in taxane-resistant metastatic cT1c and cN+ or cT1c and pNSLN+ *Chi2 test of baseline parameters between arms breast cancer patients. Annals of Oncology 2008; 19: 1547-1552.

Presented at: 10th European Breast Cancer Conference (EBCC-10), The study was financially supported by the Sanofi-Aventis Group For further information and for permission to reprint and/or distribute please contact: Amsterdam, The Netherlands, 9-11 March 2016 [email protected] |33 34 | Annual Scientific Report 2016 | New Study Concepts | 35

New Study Concepts

GBG 93: PADMA Interview with Prof. Dr. Sibylle Loibl 36

GBG 94: PATINA Interview with Dr. Jana Barinoff 38

GBG 95: ULTIMATE Interview with Prof. Dr. Fabrice André 40 36 | Annual Scientific Report 2016 | New Study Concepts

Interview with Prof. Dr. Sibylle Loibl, International Principal Investigator of the PADMA trial

A randomised, open-label, multicenter phase IV study evaluating palbociclib plus endocrine treatment versus a chemotherapy-based treatment strategy in patients with hormone receptor positive/HER2 negative metastatic breast cancer in a real world setting (PADMA)

Prof. Dr. Sibylle Loibl breast cancer population. In addition to that the Chair of the assessments done in clinical trials are usually stricter than in every-day clinical practice. The German Breast Group concept in the PADMA trial is seeking to mirror Executive Vice President GBG every-day clinical practice and to investigate in Forschungs GmbH a more realistic scenario whether first-line ther- Associate Professor at apy with palbociclib plus endocrine treatment University Frankfurt am Main (ET) is superior to chemotherapy +/- endocrine maintenance therapy. This trial will evaluate two approved approaches for the treatment of HR positive, HER2 negative MBC. The aim of the trial is to establish a superior but also patient-preferred treatment approach for this phase of breast cancer. This trial evaluates a guideline-driven approach and compares with a regimen that is acceptable by guidelines and used by physicians. PADMA is a prospective, randomized, open- label, multicenter controlled phase IV low in 2. What are the advantages of using a combi- tervention trial to test whether endocrine nation of endocrine therapy with CDK4/6 in- treatment with palbociclib is better than mo- hibitors in metastatic breast cancer? no-chemotherapy +/- endocrine maintenance therapy as per treating physician's choice as An endocrine therapy plus palbociclib is a new first line therapy in advanced/metastatic breast treatment option for women with treat- cancer (MBC). Patients will be randomized in a ment-naïve HR positive MBC as first-line thera- 1:1 ratio to receive either endocrine treatment py, because progression-free survival (PFS) was with palbociclib or chemotherapy with or with- improved compared with letrozole alone based out endocrine maintenance therapy. on the data from the PALOMA-1 and -2 trial. Also other CDK4/6 inhibitors have demonstrated Primary objective: To compare the Time- similar treatment effects in this setting. The to-Treatment Failure (TTF) for patients ran- PALOMA-3 study in the setting of women with domised to receive palbociclib and endocrine endocrine resistant MBC shows that the addition therapy versus those randomised to receive a of palbociclib can improve PFS compared to pre-defined chemotherapy treatment strategy. fulvestrant alone. This study also included premenopausal women. However, data comparing endocrine therapy alone with chemotherapy are 1. With regards to the concept of PADMA trial, scarce and less convincing. Since palbociclib im- what is the role of “real world” and low inter- proves the efficacy of endocrine therapy alone ventional studies for breast cancer? by about 50%, the hypothesis is that palbociclib + ET is superior to mono-chemotherapy of physi- The term “real world” study suggests that all cian´s choice followed by ET maintenance thera- other studies are not real world, which is not py in time-to-treatment failure. However, due to completely true. However, the classical rand- rigid inclusion and exclusion criteria, limited omized controlled phase III studies are charac- number of treatment options, and strictly pre- terized by very specific and sometimes tight in- scribed monitoring intervals the majority of cli and exclusion criteria, not reflecting the general nical trials are done in an “artificial environment” | 37

Interview with Prof. Dr. Sibylle Loibl, International Principal Investigator of the PADMA trial

PADMA Study Design

PATIENT POPULATION: PRIMARY ENDPOINT: N=360 Arm A Endocrine Therapy* + Time-to-treatment • HR+ / HER2- MBC Palbociclib Failure • No prior systemic an9cancer treatment for advanced/ KEY SECONDARY metasta9c disease 1:1 ENDPOINTS: • Male or female R PFS, TFST, TSST, • Pre-, peri-, post-menopausal Chemotherapy OS at 36 months, • Metastases (≥ 1 liver or ≥ 2 Arm B of physician´s paRent-reported HR metasta9c sites) choice** ± ET QoL (FACT-B), DMTI, • CT deemed op9on by maintenance therapy safety, tolerability, inves9gator compliance.

PFS = Progression-free survival * AI, fulvestrant +/- GnRH analogue TFST = Rme to ﬁrst subsequent therapy ** As deﬁned in the study protocol TSST = Rme to second subsequent therapy DMTI = Daily Monitoring Treatment Impact

STRATIFICATION FACTORS: - hormone resistant (relapse on or within 12 months of end of adjuvant ET) versus hormone sensi9ve (relapse beyond 12 months aKer end of ET or de novo metasta9c HR+ / HER2- breast cancer) - symptoma9c (deﬁned as per inves9gator) versus asymptoma9c (as deﬁned by inves9gator)

and often do not mirror real world situation. and will deliver important information on the dif- Therefore, this trial is planned as so called low ferences between endocrine-based and chemo- intervention real world trial. therapy-based treatment.

3. Why do you want to compare the Time-to- 4. What are the future perspectives for treat- Treatment Failure between patients with HR ment of metastatic breast cancer? positive metastatic breast cancer receiving a pre-defined chemotherapy regime and those Patients should be encouraged to consider receiving palbociclib and endocrine therapy? enrollment into clinical trials, including those receiving treatment in the first-line setting. Mul- The Time-to-Treatment Failure (TTF) is defined as tiple clinical trials are ongoing or are currently the time from randomization to treatment dis- planned, with a focus on improving response to continuation for one of the following reasons, endocrine therapy in metastatic disease. Deter- disease progression, treatment toxicity, patient mining biomarkers to predict response to specific preference, or death. TTF may also include fail- targeted agents is a critical investigative path. ure to meet the planned therapy objective/plan While the CDK4/6 inhibitors have already en- as specified for each patient in the treatment tered the treatment reality, other novel agents arm as defined by the protocol. Thus, this end- are currently under investigation including point introduces the idea of real-world accepta- PI3Kinase inhibitors, new SERDs, AKT and mTOR bility of treatment, its tolerance (including bur- inhibitors. The more precise characterization of den of care with visits in clinic, etc.), and overall the disease at relapse will become more and acceptability of treatment. Patient-reported more important and will shed further light on the outcome is one of the main secondary objectives disease. 38 | Annual Scientific Report 2016 | New Study Concepts

Interview with Dr. Jana Barinoff, PATINA Principal Investigator of the PATINA trial

A randomized, open label, phase III trial to evaluate the efficacy and safety of palbociclib + anti-HER2 therapy + endocrine therapy vs anti-HER2 therapy + endocrine therapy after induction treatment for hormone receptor positive (HR+)/HER2-positive metastatic breast cancer.

Dr. Jana Barinoff know that triple positive breast cancer exhibits a unique clinical behavior. For example, these pa- Department of Gynaecology tients show less benefit from endocrine therapy and Oncology than patients with HER2-negative, HR+ disease. Charité University Hospital, On the other hand constellation with low disease Berlin burden and high HR expression resemble luminal tumors. There is a preclinical evidence to confirm that cross-talk between HER2 and HR signaling pathways may contribute to the resistance of endocrine therapy. Adding palbociclib to trastuzumab may be very effective in endocrine- und trastuzumab resistant HR+ disease.

2. Regarding the different clinical trials that are currently conducted with CDK4/6 inhibitors, where would you position the PATINA trial? PATINA is an international, multicenter, ran domized, open label, phase III trial testing the First I would position the PATINA trial as a main- efficacy and safety of palbociclib + anti-HER2 tenance concept after response to first-line the therapy + endocrine therapy vs. anti-HER2 ther- rapy, which consists of chemotherapeutic drugs, apy + endocrine therapy after induction treat- trastuzumab and pertuzumab, where available in ment for hormone receptor positive (HR+)/ triple positive patients. This is the planned con- HER2-positive metastatic breast cancer. The cept. But I find it´s very important to investigate study also includes a comprehensive molecular the role of that combination in low disease bur- characterisation of the disease at study entry den as first-line therapy and as a chemotherapy- which will allow investigation of the benefits of free option. Regarding the other metastatic palbociclib in subsets of HER2+ disease such breast cancer trials such as PATRICIA (Spain, (e.g. PIK3CA mutant). SOLTI), a phase II clinical trial, PATINA has a potential to provide a new standard of cure to Primary objective: The primary objective of patients with HR+/HER2+ metastatic breast this study is to demonstrate that the addition of cancer. palbociclib to anti-HER2 and endocrine therapy is superior to anti-HER2 and endocrine therapy 3. What are the scientific expectations of the in improving the Progression-Free-Survival (PFS) translational research programme imple- of participants with HR+/HER2+ metastatic mented in the study? breast cancer. Activating mutations in the PIK3CA gene occur 1. What is the rational of combining CDK4/6 in about 30 % of breast cancers. The translation- inhibitors such as palbociclib with anti-HER2 al programme of PATINA investigates whether therapy in metastatic breast cancer? an activated PI3K pathway, e.g. HER2 amplifica- tion or oncogenic mutation in the PIK3CA gene Combining CDK 4/6 inhibitors with anti-HER2 are involved in the efficacy of palbociclib. It therapy has been considered for triple positive seems, synergy exists between CDK 4/6 inhibi- metastatic breast cancer: approximately 50 % of tors and PI3K pathway activation in PIK3CA mu- the patients with HER2-positive breast cancer tant tumors. The role of this synergy will be ex- are also hormone receptor-positive (HR+). We plored in the triple positive patients in PATINA. | 39

Interview with Dr. Jana Barinoff, Principal Investigator of the PATINA trial

PATINA Study Design

n=496 Arm A Palbociclib 125mg (D1-21 of 28d cycle) An,-HER2 Therapy (every 3w)* Endocrine Therapy**

Un,l disease progression HER2+ HR+ 1:1 ***Clinical FU Survival FU Metasta,c Breast q12w un,l tumor q6m un,l 5y Cancer progression

An,-HER2 Therapy (every 3w)* Endocrine Therapy**

Un,l disease progression Arm B

*An,-HER2 treatment op,ons are Trastuzumab *** for pa,ents who plus Pertuzumab or Trastuzumab only (limited discon,nue treatment prior to 20% of the study popula,on). The same an,- to disease progression HER2-regimen should be used before and post randomiza,on. **Endocrine therapy op,ons are either an Aromatase Inhibitor or Fulvestrant 40 | Annual Scientific Report 2016 | New Study Concepts

Interview with Prof. Dr. Fabrice André, ULTIMATE Coordinating Investigator of the ULTIMATE trial

A phase II trial testing durvalumab combined with endocrine therapy in patients with ER+/Her2- breast cancer eligible to neoadjuvant endocrine therapy and who present CD8+ T Cell Infiltration after 4-6 weeks exposure to immune-attractant

Prof. Dr. Fabrice André most of them present a significant mutational Professor of Medical Oncology load. It’s also very likely that anti-PD1 single agent will not be enough. This is why new strate- at Institute Gustave Roussy, gies that aim to use IT sequentially or in combi- Villejuif, France nation are needed. BIG Executive Board Member 2. Checkpoint inhibitors have been recently tested in the treatment of triple negative breast cancer (TNBC). Why do you believe check-point inhibitors could work in hormone receptor-positive (HR+) breast cancer?

Phase II trial has shown that ORR is not dramati- cally different between TNBC (18 %) and HR+ / HER2-negative (12 %) tumors. What is needed now is to amplify these response rates by propos- ing strategies that aims to sensitize tumors to ULTIMATE is an open-label, multicenter, inter- anti-PD1. national, phase II trial testing an aromatase inhibitor in combination with durvalumab, an an- 3. Are there any biomarkers identified that can ti-PD-L1 antibody in patients with CD8+ T cell be used to predict response to checkpoint in- infiltration (> 10 % CD8+ T cells in the tumor). hibitors in breast cancer? The trial includes two sequences: the first part of the treatment will consist of 4-6 weeks treat- It is expected that biomarker will be the same as ment with an immune-attractant; in the second in other tumors; i.e. mutational load, TILs and part, patients with CD8+ tumours will receive 6 PD-L1 expression. months of durvalumab combined with exemestane. 4. What needs special attention in the study design of ULTIMATE? Primary objective: To assess the efficacy of durvalumab combined with exemestane in pa- The most important point will be the biopsy du tients with CD8+ T cells on pathological re- ring the 1st part of the treatment; this should sponse at surgery after a lymphocyte attraction tell us which cancer is immunogenic and which phase. one is less. We then expect to provide anti-PD1 only to the patients who present an immunogenic 1. Immunotherapy has been very successfully cancer. implemented not only in metastatic melano- ma but also in solid tumors such as non-small 5. What are your expectations regarding the cell lung cancer (NSCLC). Do you think that trial´s outcome? breast cancer can be a good candidate for immunotherapy? We expect to provide evidence that using 4-6 weeks of lympho-attractants could render an It’s very likely that immunotherapy (IT) will be marginally immunosensitive tumor into a tumor successfully implemented in BC because these responsive to anti PD-L1. tumor present an inflammatory component and | 41

Interview with Prof. Dr. Fabrice André, Coordinating Investigator of the ULTIMATE trial

ULTIMATE Study Design 42 | Annual Scientific Report 2016 | Recruiting Studies | 43

Recruiting Studies

GBG 89: GeparNuevo 44 GBG 88: GeparX 46 GBG 90: GeparOla 48 GBG 87: PALLAS 50 GBG 82: OLYMPIA 52 GBG 78: PenelopeB 54 GBG 68: GAIN-2 56 GBG 54: MALE 59 GBG 29: Breast Cancer in Pregnancy (BCP) 61 GBG 79: Brain Metastases in Breast Cancer (BMBC) 63 GBG 86: DESIREE 65 GBG 75: INSEMA 67 44 | Annual Scientific Report 2016 | Recruiting Studies

GBG 89: GeparNuevo

A randomised phase II study to investigate the Therefore immunotherapy might be an option addition of PD-L1 antibody MEDI4736 to a tax- to increase the pCR rate in patients with TNBC. ane-anthracycline containing chemotherapy in Two phase I/II studies have recently been re- CONTACT: triple negative breast cancer (GeparNuevo) ported and showed activity of check point inhi Dr. Ursula Wolf bitors in breast cancer. The KEYNOTE-012 study Clinical Project Management NCT02685059 investigated the PD-1 inhibitor pembrolizumab [email protected] in metastatic TNBC, demonstrating a response GeparNuevo is a multicenter, prospective, ran- rate of 20 % and about 50 % having stable dis- domized, double-blinded, placebo controlled ease or a response (Nanda R et al. J Clin Oncol. phase II study that will recruit 174 patients from 2016). In another phase I study of 12 patients approximately 30 sites in Germany within 18 with TNBC treated with the PD-L1 inhibitor MP- months. DL3280A, 33 % had a response. The tolerability was good. Background MEDI4736 is a human monoclonal antibody Triple negative breast cancer (TNBC) is a highly (MAb) of the immunoglobulin G1 kappa (IgG1κ) aggressive disease. Patients who have received that blocks PD-L1 binding to PD-1 and CD80, neoadjuvant therapy with chemotherapy alone preventing PD-L1-mediated inhibition of T-cell and achieved a pathological complete response activation. (pCR), have an excellent survival, however, These data supported an early investigation of women who did not achieve a pCR did signifi- these drugs for treatment of TNBC, when other- cantly worse. wise only chemotherapy is indicated. TNBC and especially the basal like subtype can have a high mutational load with up to 200 mu- Study design and objectives: tations per cancer (Shah S et al. Nature. 2012). GeparNuevo primarily aims to compare the pCR Moreover patients with TNBC and a higher score (ypT0 ypN0) rates of neoadjuvant treatment of of stromal tumor-infiltrating lymphocytes sequential, nab-Paclitaxel followed by Epirubicin (sTILs) have a significantly better disease free and Cyclophosphamide (EC) +/- the PD-L1 anti- survival (DFS) than patients whose TNBC tumor body MEDI4736 in patients with early TNBC. lack sTILs (Adams S et al. J Clin Oncol. 2014). Secondary objectives are to assess the pCR rates

Figure 1: GeparNuevo study design | 45

per arm separately for the stratified subpopula- biology, treatment response and prognosis is tions, rates of ypT0/is ypN0; ypT0 ypN0/+; planned. The association between the germline ypT0/is ypN0/+; ypT(any) ypN0, response rates genotype of the patient with treatment re- of the breast tumor and axillary nodes based on sponse, long term prognosis in both arms and physical examination and imaging tests after with the molecular profile of the tumors will be treatment in both arms, clinical response rate investigated. after taxane in both groups, breast conservation, toxicity and compliance, loco-regional invasive Study report: recurrence free survival (LRRFS), distant-dis- GeparNuevo started recruitment in June 2016. ease-free survival (DDFS), invasive disease-free As of 31st December 2016, a total of 53 patients survival (IDFS), event free survival (EFS) and have been enrolled in the study. Safety interim overall survival (OS) in both arms and according analyses are planned after the first 10, 20 and 30 to stratified subpopulations, to examine quality patients having completed part 2 of the study of life using FACT-Taxane and to examine and and after the first 10, 20, 30 patients having compare pre-specified molecular markers as completed part 3 of the study (six safety interim well as gene expression signatures such as analysis in total). The results of the first safety tumor infiltrating lymphocytes, PD-1, PD-L1, interim analysis will be presented at the Ameri- Ki-67, etc. on core biopsies before chemo can Society of Clinical Oncology (ASCO) June therapy, after the window phase and on surgical 2-6, 2017, Chicago, Illinois, USA. The expected tissue after end of chemotherapy. study duration is 2 years. Furthermore, the GeparNuevo study will also address translational research questions such as Publications: an investigation of the relationship between 1. Loibl S, Schneeweiss A, Burchardi N, et al. A PD-L1, TILs, immunopredict and other immune randomized phase II study to investigate the ad- markers with response. dition of PD-L1 antibody MEDI4673 (Durvalum- ab) to a taxane-anthracycline containing chemo- Pharmacogenetic substudy therapy in triple negative breast cancer A pharmacogenetic substudy on genetic mark- (GeparNuevo). Ann Oncol (2016) 27 (suppl_6): ers from peripheral blood to predict tumor 221 COLLABORATING STUDY GROUPS:

accrual no. 160 31 initiated sites 30 planned no. 140 26 25

120 23 22

100 20 SPONSOR: 80 14 GBG Forschungs GmbH

60 11 10 STUDY CHAIRS: 40 7 Prof. Dr. Sibylle Loibl German Breast Group 20 Prof. Dr. Andreas Schneeweiß 0 0 Nationales Centrum für Jun-16 Aug-16 Oct-16 Dec-16 Feb-17 Apr-17 Jun-17 Aug-17 Oct-17 Dec-17 Tumorerkrankungen Universitäts-Klinikum, Figure 2: GeparNuevo recruitment as of 31st December 2016 Heidelberg We are thanking all participating centers for their commitment and efforts so far. We would kindly Prof. Dr. Michael Untch like to encourage all sites to continue to support the study by recruitment of the patients and by Helios Klinikum Berlin-Buch, providing the remaining biomaterial in a timely manner. Berlin 46 | Annual Scientific Report 2016 | Recruiting Studies

GBG 88: GeparX

Investigating Denosumab as an add-on to neo- Study design and objectives adjuvant chemotherapy in RANK/L-positive or GeparX co-primary aims are to compare the pCR RANK/L-negative primary breast cancer and (ypT0 ypN0) rates of neoadjuvant treatment two different nab-Paclitaxel schedules in a 2x2 with or without denosumab in addition to back- factorial design (GeparX) bone treatment consisting of nP 125mg/m² CONTACT: weekly (Cb) followed by epirubicin (E) and Konstantin Reißmüller NCT02682693 cyclophosphamide (C) or nP 125mg/m² day Clinical Project Management 1,8 q22 (Cb) followed by EC plus anti-HER2 [email protected] GeparX is a multicenter, prospective, 2x2 treatment (i.e. trastuzumab/pertuzumab in case randomized, open-label, phase IIb study that of positive HER2-status) and to compare the pCR will recruit 778 patients from approximately 50 (ypT0, ypN0) rates of nP 125mg/m² weekly (Cb) sites in Germany within 18 months. followed by EC or nP 125mg/m² day 1,8 q22 (Cb) followed by EC plus anti-HER2 treatment. Carbo- Background platin (Cb) will be administered in parallel to nP RANK ligand (RANKL), a key factor for bone re- for patients with triple negative breast cancer. modeling and metastasis, is crucial for the devel- Secondary objectives are to test for interaction opment of mouse mammary glands during preg- of denosumab treatment with RANK expression, nancy. RANKL functions as a major paracrine to assess the pCR rates per arm in subgroups and effector of the mitogenic action of progesterone according to RANK immunohistochemical in mouse and human mammary epithelium via expression (high/low), pCR rates according to its receptor RANK and has a role in ovarian hor- other definitions, response rates of the breast mone-dependent expansion and regenerative tumor and axillary nodes, breast conservation potential of mammary stem cells. Pharmacolog- rate, toxicity and compliance as well as to deter- ic inhibition of RANKL attenuates the develop- mine loco-regional invasive recurrence free ment of mammary carcinoma and inhibits me survival, distant-disease-free survival, invasive tastatic progression in multiple mouse models disease-free survival, event-free survival and (Dougall WC et al. Clin Cancer Res 2012). In a overall survival for all treatment arms and retrospective analysis of 601 patients treated according to stratified subpopulations. Further with anthracyclines-taxane based chemother- secondary objectives are to compare RANK/L apy from the GeparTrio study we showed that expression and Ki67 from baseline to surgery, to an elevated immunohistochemical expression of correlate response measured by best appropri- RANK was present in 14.5 % of patients overall ate imaging method after the first two cycles of (Pfitzner BM et al. Breast Cancer Res Treat 2014). treatment with pCR, to assess mammographic The ABCSG-18 study showed that adjuvant density–changes induced by denosumab and to denosumab (clinically available antibody against assess quality of life with a focus on persisting RANKL) reduces clinical fractures, improves peripheral sensory neuropathy using the FACT- bone health, and can be administered without Taxane questionnaire. added toxicity (Gnant M et al. Lancet 2015). In addition, GeparX offers an opportunity to ad- Moreover denosumab showed a trend in im- dress a range of translational research questions provement of disease-free survival in postmeno- which are summarized below. pausal woman with hormone receptor positive breast cancer (Gnant M et al. SABCS 2015). It Substudies appears therefore reasonable to test denosumab GeparPET substudy in patients with primary breast cancer as an ad- Current breast investigations using ultrasound, junct to neoadjuvant chemotherapy for its abili- mammography and MRI do not provide suffi- ty to increase the pathological complete re- cient sensitivity and specificity to reliably diag- sponse (pCR) rate and improve outcome overall nose response to treatment. Small diagnostic and in relation to the expression of RANK/L. As studies using Positron Emission Tomography- in GeparSepto study nab-paclitaxel (nP) lead to Computed Tomography (PET-CT, Berriolo- an increased pCR rate compared to standard sol- Riedinger et al. 2007; Rousseau et al. 2006; vent-based paclitaxel, nab-paclitaxel has been Schwarz-Dose et al. 2008; Cheng et al. 2012) chosen as backbone chemotherapy. Two differ- suggest that this add-on diagnostic tool has a ent nab-Paclitaxel regimens will be compared. much higher sensitivity and specificity for the presence of tumor lesions and might therefore improve these limitations. The main aim of the | 47

substudy will therefore be to show that presur- corresponding RANK-receptor (Casas et al. gical staging with PET-CT in addition to conven- 2013). Aim of this substudy is to investigate tional presurgical staging methods can decrease whether the application of denosumab in terms the rate of mastectomy in patients treated by of an add-on neoadjuvant treatment eradicate neoadjuvant chemotherapy. DTCs in the bone marrow of breast cancer patients. Pharmacogenetic substudy The direct involvement of genetic markers in the Substudy on urinary miRNA sampling (UMS) metabolism and the pharmacokinetic of a drug Aberrant expression profiles of microRNAs as well as the influence of the inherited genetic (miRNAs) with subsequent functional conse- trait on the molecular profile of the tumor could quences on target gene regulation in physiologi- have an influence on an individual’s prognosis. cal and pathological pathways could already be Aim of this substudy is therefore to analyse set in clear association with breast cancer. Aim potential association between the germline gen- of the substudy is to evaluate a specific microR- otype of the patient and treatment response, NA pattern in urine specimen as an innovative toxicities, long term prognosis, molecular profile tool for subtype-specific diagnosis of breast can- of the tumor and breast cancer risk. cer (HER2-positive vs. TNBC).

DTC Substudy The persistence of disseminated tumor cells Study report (DTC) in the bone marrow after adjuvant chemo- GeparX was approved by the Ethics Committee therapy has been ascertained as an independent in November 2016 and recruitment is expected predictor for poor disease-free survival, can- to start in QI/2017. One interim safety analysis cer-specific and overall survival (Braun et al. will be performed after the first 200 patients 2005; Janni et al. 2011). Furthermore, denosum- have completed the nab-Paclitaxel treatment. ab inhibits osteoclastic differentiation by bind- The expected study duration (after first patient ing to RANK-ligand (RANKL), thereby prevent- in) is approximately 26 months. ing the interaction between RANKL and its

COLLABORATING GeparX: Inves-ga-ng Denosumab as an add-on neoadjuvant treatment for STUDY GROUPS: RANK-posi-ve or RANK-nega-ve primary breast cancer and two diﬀerent nab-Paclitaxel schedules; 2x2 factorial design N=778 pts. Nab-Paclitaxel 125 mg/m² weekly 12x - EC Stra6ﬁca6on - LPBC yes vs no Denosumab 120 mg s.c. q4w 24 weeks - HER2-/HR+ vs Endpoints: TNBC vs HER2+ Nab-Paclitaxel 125 mg/m² weekly 12x - EC Primary: pCR (ypT0 ypN0) - EC q2 vs q3 Secondary: without Denosumab - pCR in RANK high/low - Change of RANK expr R Nab-Paclitaxel 125 mg/m² d 1,8 q22 - EC - Change of Ki67 - other Denosumab 120 mg s.c. q4w 24 weeks

Nab-Paclitaxel 125 mg/m² d 1,8 q22 - EC SPONSOR: without Denosumab GBG Forschungs GmbH 2nd core st 1 core a-er nab-Paclitaxel; STUDY CHAIRS: op6onal PD Dr. Sherko Kümmel Assump-ons for primary endpoint (α=0.2, β=0.2): Treatment backbone: Denosumab: 35 % to 46 % OR 1.58 HER2+: + Trastuzumab + Pertuzumab Kliniken Essen-Mitte nab-Pac: 36 % to 45 %; OR 1.45; TNBC: + Carbopla6n weekly AUC 2 Prof. Dr. Sibylle Loibl German Breast Group Figure 1: GeparX study design 48 | Annual Scientific Report 2016 | Recruiting Studies

GBG 90: GeparOla

A randomized phase II trial to assess the effica- significantly increased the pathological com- cy of paclitaxel and olaparib in comparison to plete response (pCR) rate in patients with TNBC paclitaxel/carboplatin followed by epirubicin/ in two randomized phase II neoadjuvant studies, cyclophosphamide as neoadjuvant chemo the GeparSixto (von Mickwitz et al. Cancer Res CONTACT: therapy in patients with HER2-negative early 2016;76 (4 Suppl): Abstract nr S2-04) and the Dr. Ioannis Gkantiragas breast cancer and Homologous Recombination CALBG 40603 (Silkov et al. J Clin Oncol. 2015) Clinical Project Management Deficiency (HRD patients with deleterious trials. [email protected] BRCA1/2 tumor or germline mutation and/or Olaparib is a potent polyadenosine 5’diphos- HRD score high) phoribose [poly (ADP ribose)] polymerisation (PARP) inhibitor that is being developed as an NCT02789332 oral therapy, both as a monotherapy (including maintenance) and for combination with anti- GeparOla is a multicenter, prospective, rand- cancer agents. The efficacy and safety of olap omized open-label phase II study that will re- arib included in a standard of care regimen like cruit 102 patients from approximately 34 sites paclitaxel weekly followed by epirubicin and in Germany within 24 months. cyclophosphamide is unknown. In the GeparSixto trial it has been demonstrated Background that among patients with homologous recombi- Molecular characterizations have demonstrated nation deficiency (HRD), 67 % had an HRD score a strong association between triple-negative high whereas 30 % had BRCA mutant tumor breast cancer (TNBC) and BRCA1 and lesser with (tBRCA). Moreover among tBRCA mutated/HRD BRCA2 mutations. Tumor cells with a defect in score high patients, 20 % harbored also a BRCA genes have impaired homologous recom- germline BRCA mutation (gBRCA). When using a bination (HR), the only pathway responsible for stricter definition of pCR (ypT0 ypN0), patients the reparation of interstrand crosslinks. Tumors with HRD score high/tBRCA intact or tBRCA with decrease DNA repair capacity are expected mutant experienced the greatest benefit with to show high sensitivity to DNA damaging the addition of carboplatin. When allowing the agents (Ciccia A et al. Mol Cell. 2010; Deans AJ et presence of DCIS in the pCR definition (ypT0/is al. Nat Rev Cancer. 2011). The addition of carbo- ypN0), only patients with HRD score high/ platin to a neoadjuvant chemotherapy regime of tBRCA intact derived benefit by the addition of anthracycline, cyclophosphamide and paclitaxel carboplatin (von Minckwitz SABCS 2016).

Figure 1: GeparOLA study design | 49

Considering the inconsistency of these results pliance with POEC and to compare it with and the absence of a formal interaction between PCbEC. Further objectives are to correlate HRD and carboplatin further studies are needed. co-occurring mutations detected by next gen The GeparOLA study aims to support the deci- eration sequencing in lymphocytes or in tumors sion for a phase III study exploring the addition cells with pCR (exploratory) and potential bio- of olaparib (O) to a paclitaxel (P) followed by markers predicting safety and compliance, like epirubicin (E) and cyclophosphamide (C) sched- SNPs, TILs, PARP, 53BP1, REV7 and other bio- ule (POEC) by providing an estimate on the markers considered for breast cancer. pCR rates in the targeted population but also by providing estimate comparison to paclitaxel (P) Pharmacogenetic substudy and carboplatin (Cb) followed by EC (PCbEC). A pharmacogenetic substudy on genetic mar kers from peripheral blood to predict tumor bi Study design and objectives ology, treatment response and prognosis will be GeparOla primary aims to assess the pCR performed. All patients eligible for the GeparOLA (ypT0/is ypN0) rate of neoadjuvant treatment study having collected a whole blood sample of olaparib and paclitaxel followed by epirubicin will be included. The primary objective is to as- and cyclophosphamide (POEC) in neoadju- sociate the germline genotype of the patient vant patients with early breast cancer and HRD with the treatment response in both randomiza- tumors defined as either tBRCA1/2 mutation or tion arms. Secondary objectives are to associate HRD score high or known gBRCA mutation. the germline genotype of the patient with the Secondary objectives are to assess the pCR rates long term prognosis in both randomization arms of patients in both arms, overall and in stratified and with the molecular profile of the tumors. subgroups, to investigate the pCR rate according other pCR definitions and to compare them be- Study report tween the two arms, to assess the pCR rate in Recruitment started in September 2016. As of HRD high with vs without tBRCA mutation, to 31st December 2016, there are 10 patients en- determine the response rates of the breast tu- rolled in the study. Two interim safety analyses mor and axillary nodes based on physical exami will be performed after 15 and 30 patients hav- nation and imaging tests and to compare them ing completed 2 cycles of treatment. The end of COLLABORATING between the two treatment arms, to determine the study is estimated for the beginning of 2018. STUDY GROUPS: the breast conservation rate, toxicity and com-

100

accrual no. 30 initiated sites 80 planned no.

22 22 22

60 20 18 SPONSOR: GBG Forschungs GmbH 40

10 STUDY CHAIR:

20 Prof. Dr. Peter Fasching Universitätsklinikum Erlangen Prof. Dr. Christian Jackisch 0 0 Sana Klinikum Offenbach Sep-16 Jan-17 May-17 Sep-17 Jan-18 May-18 Prof. Dr. Jens Blohmer Figure 2: GeparOLA recruitment as of 31st December 2016 Charité Berlin We are thanking all participating centers for their commitment and efforts so far. We would kindly Prof. Dr. Sibylle Loibl like to encourage all sites to continue to support the study by recruitment of the patients and by German Breast Group providing biomaterial in a timely manner. 50 | Annual Scientific Report 2016 | Recruiting Studies

GBG 87: PALLAS

PALLAS: PALbociclib CoLlaborative Adjuvant cancer. Secondary objectives are to compare Study: invasive disease-free survival (iDFS) excluding second primary cancers of non-breast origin, A randomized phase III trial of Palbociclib with distant recurrence-free survival (DRFS), locore- CONTACT: standard adjuvant endocrine therapy versus gional recurrences-free survival (LRRFS), overall Dr. Christian Mäurer standard adjuvant endocrine therapy alone for survival (OS) and safety between the two arms. Clinical Project Management hormone receptor positive (HR+) / human epi- The principal translational research objective is [email protected] dermal growth factor receptor 2 (HER2)-nega- to compare baseline tumor tissue to determine tive early breast cancer whether there is prognostic or predictive utility for defined genomic subtypes (luminal A, lumi- NCT02513394 nal B and non-luminal) with respect to iDFS and OS. PALLAS is a multicenter, prospective, interna- Clinical science objectives are to evaluate adher- tional, randomized, open-label, adjuvant phase ence to oral therapy in patients receiving palbo- III study that will recruit 4,600 patients in 25 ciclib and endocrine therapy, to determine the countries within 3 years. association of body mass index (BMI) and race with the efficacy of palbociclib and endocrine Background therapy. Patient reported outcomes objectives Although many patients with HR-positive are to compare patient-reported quality of life, (HR+)/HER2-negative (HER2-) breast cancer fatigue, arthralgia, and endocrine symptoms be- may be cured of their disease with optimal local tween the two arms overall and by subgroups and systemic therapy, a significant number of defined by age at randomization (≤ 50 and > 50) patients with stage II and III disease will experi- and initial endocrine therapy (tamoxifen and AI) ence disease recurrence. Adjuvant endocrine at multiple pre-specified time points. Multiple therapy for breast cancer can be extremely tissue- and blood-based correlative studies are effective, particularly with extension beyond scheduled throughout the course of the PALLAS 5 years, however disease recurrence can occur, trial to evaluate potential markers of response with risk distributed over the decades following and/or resistance in patients receiving endocrine initial diagnosis. Methods to improve the effica- therapy with palbociclib versus endocrine thera- cy of endocrine therapy, and delay the onset of py alone (TRANS-PALLAS). The German Breast resistance, are needed. Group acts as a local study group for Germany HR+ breast cancer biologically may demon- and is responsible for the biobanking organiza- strate features suggestive of sensitivity to tion of PALLAS study. CDK4/6 inhibition with agents such as palbociclib. Given the demonstrated activity and safety Study report of palbociclib in the first-line treatment of me PALLAS started recruitment in October 2015. As tastatic HR+/HER2- breast cancer, supporting of 31st December 2016, 533 patients were rand- FDA approval, there is interest in whether the omized (362 patients in US and 171 patients in benefits of CDK4/6 inhibition may translate into non-US). In Germany, the regulatory authority the adjuvant setting. The purpose of the PALLAS approval was obtained. The study is still under study is to determine whether the addition of review by the Ethics Committee. The first site palbociclib to adjuvant endocrine therapy will activations in Germany are planned for QI/2017. improve outcomes over endocrine therapy alone Two interim efficacy analyses are planned for for HR+/HER2- early breast cancer. Assessment monitoring the futility and superiority, and are of a variety of correlative analysis, including scheduled to occur when 33 % and 67 % of the evaluation of the effect of palbociclib in genom- total number of iDFS events are observed. The ically defined tumor subgroups, is planned. expected study duration is 4.8 years.

Study design and objectives PALLAS primary aims to compare invasive disease-free survival (iDFS) for the combination of at least 5 years endocrine therapy and 2-year palbociclib treatment versus at least 5 years endocrine therapy alone in patients with histologically confirmed HR+/HER2- invasive early breast | 51

Arm A Patient Population Palbociclib (2 yrs) • N = 4600 + • Inclusion Criteria: Endocrine – HR+ and HER2- R – Stage II or III (IIA limited to 1000 patients) A treatment N (5+ yrs) D Diagnosis Surgery O 1:1 Survival/Disease Follow-up M

Neo/Adjuvant I systemic therapy Z Arm B E Endocrine treatment (5+ yrs) COLLABORATING STUDY GROUPS: FFPE Tissue sample PRO & Adherence received at central Monitoring biorepositorysent

Stratifications Factors: Arm A: palbociclib at a dose of 125 mg once daily, day 1-21 • Anatomic stage (IIA vs IIB/III) assessed by pathologic in a 28-day cycle for total duration of 2 years, in staging, or by clinical staging if pre-operative therapy addition to standard adjuvant endocrine therapy was given with the higher stage determining eligibility, Arm B: standard adjuvant endocrine therapy (AI, tamoxifen, • Neo/adjuvant chemotherapy (yes vs no), LHRH agonist) • Age (≤ 50 vs > 50 years), • Geographic region (North America vs Europe vs Other)

Figure 1: PALLAS study design

We are thanking all participating centers for their commitment and efforts so far. We are looking forward to initiate the PALLAS study in Germany.

SPONSOR: AFT (US), ABCSG (Non-US Sites)

INTERNATIONAL STUDY CHAIRS: Dr. Erica L. Mayer Susan F Smith Center for Women’s Cancers (US) Dr. Angela DeMichele Abramson Cancer Center (US) Prof. Dr. Michael Gnant ABCSG (AT)

STUDY CHAIRS (GERMANY): Dr. Sabine Schmatloch Elisabeth Krankenhaus, Kassel 52 | Annual Scientific Report 2016 | Recruiting Studies

GBG 82: OLYMPIA

A randomized, double-blind, parallel group, tients and these patients are treated according placebo-controlled multi-center Phase III study to their hormone receptor and HER2 status. to assess the efficacy and safety of olaparib Polyadenosine 5’diphosphoribase [poly (ADP versus placebo as adjuvant treatment in ribose)] polymerisation (PARP) inhibition is a patients with germline BRCA1/2 mutations and novel approach to targeting tumors with defi CONTACT: high risk HER2 negative primary breast cancer ciencies in DNA repair mechanisms. PARP en Dr. Ioannis Gkantiragas who have completed definitive local treatment zymes are essential for repairing DNA single Clinical Project Management and neoadjuvant or adjuvant chemotherapy strand breaks. Inhibiting PARP leads to the per sistence of single strand breaks, which can then, [email protected] NCT 02032823 during the process of DNA replication, be con verted to the more serious DNA double strand OLYMPIA is a multicenter, double-blind, parallel breaks. Tumors with homologous recombination group, placebo-controlled, randomized phase III deficiencies, such as breast cancers in patients trial that will recruit approximately 1,500 with germline BRCA1/2 mutations, cannot patients from 340 sites in 24 countries within accurately repair the DNA damage, leading to approximately 4 years. cancer cell death. Olaparib is a potent PARP-1, -2 and -3 inhibitor, Background that is being developed as an oral therapy, for Approximately 5 % of breast cancers are asso use as monotherapy (including maintenance) ciated with a BRCA mutation, with approximately and for combination with chemotherapy and 60 % of those being associated with the BRCA1 other anti-cancer agents. Olaparib has been gene (generally presenting with triple negative shown to inhibit selected tumor cell lines in vitro phenotype) and approximately 40 % being and in xenograft and primary explant models as associated with the BRCA2 gene (generally well as in genetic BRCA knockout models, either estrogen/progesterone positive phenotype) Mu as a stand-alone treatment or in combination tations in either gene result in tumors that are with established chemotherapies. Clinical stud deficient in homologous recombination. Currently, ies have shown that olaparib monotherapy in there are no approved treatments specific for patients with germline BRCA mutations offers germline BRCA1/2 mutated breast cancer pa potentially efficacious and less toxic cancer

Post Neoadjuvant gBRCA TNBC patients Non pCR Olaparib 300 mg ER/PgR positive /HER2 negative patients twice daily (bid) Follow- Non pCR AND CPS&EG for 12 months up score ≥3 Randomisation 1:1 Double blind IDFS, Post Adjuvant gBRCA N=1500 distant TNBC patients Placebo IDFS, axillary node-positive (any twice daily (bid) tumour size) or axillary node- for 12 months OS negative tumour > 2cm (pathological size)

ER/PgR positive/HER2 negative patients ≥ 4 pathologically confirmed positive lymph nodes

Figure 1: OLYMPIA study design | 53

treatment compared with currently available current and future germline BRCA mutation chemotherapy regimens (Fong et al. N Engl J Med assays (gene sequencing and large rear range 2009; Tutt et al. Lancet 2010; Gelmon et al. Lancet ment analysis) and the exposure to olaparib (in Oncol 2011; Kaufman et al. J Clin Oncol 2015). plasma) will be determined. The OLYMPIA study investigates for the first time Translational research components include the the efficacy of olaparib compared with placebo exploration of methods for estimating OS in an adjuvant/post-neoadjuvant approach in adjusting for the impact of confounding by patients with germline BRCA1/2 mutations and subsequent therapies, the investigation whether high-risk HER2- disease. resistance mechanisms to olaparib can be identified through analysis of tumor and blood Study design and objectives sample derivatives, and the determination of the OLYMPIA primarily aims to assess the effect of frequency and nature of BRCA mutation/s in adjuvant treatment with olaparib on invasive di tumor samples and comparison with germline sease-free survival (IDFS). In addition, the safety BRCA mutation status. and tolerability of adjuvant treatment with olaparib is a key objective. Secondary objectives Study report are to assess the effect of adjuvant treatment OLYMPIA started recruitment in September with olaparib on overall survival (OS), distant 2014. As of 31st December 2016, a total of 733 disease-free survival (DDFS), the incidence of patients (78 patients in Germany) have been new primary cancers (contralateral invasive enrolled in the study. With an amendment of breast cancer, contralateral non-invasive breast the study protocol, patients with HER2-/HR+ cancer, ovarian cancer, fallopian tube cancer and disease are allowed to take part in the study. An COLLABORATING peritoneal cancer), and patient reported interim analysis for superiority is planned after STUDY GROUPS: outcomes (according to the FACIT- Fatigue and 165 IDFS events have been observed from the EORTC QLQ-C30 questionnaires). Moreover, first 660 patients recruited, which is estimated the efficacy of olaparib will be assessed in for approximately 4.5 years after randomization patients identified as having a deleterious or of the first patient. The end of the study (i.e. last suspected deleterious variant in either of the visit of the last patient randomized) is estimated BRCA genes using variants identified with for 2028.

SPONSOR:

INTERNATIONAL PRINCIPAL INVESTIGATOR: Prof. Dr. Andrew Tutt King's College, London (UK) Figure 2: OLYMPIA recruitment as of 31st December 2016

We are thanking all participating centers for their commitment and efforts so far. We would like to STUDY CHAIR (GERMANY): encourage all sites to continue to support the OLYMPIA study by recruitment of patients and the Prof. Dr. Elmar Stickeler timely provision of biomaterial. Universitätsklinikum Aachen 54 | Annual Scientific Report 2016 | Recruiting Studies

GBG 78: PenelopeB

Phase III study evaluating palbociclib (PD- A 0332991), a Cyclin-Dependent Kinase (CDK) CONTACT: 4/6 Inhibitor in patients with hormone-receptor- positive, HER2-normal primary breast cancer with Dr. Christian Mäurer high relapse risk after neoadjuvant chemotherapy Clinical Project Management [email protected] NCT 01864746

PENELOPEB is a prospective, international, multicenter, randomized, double-blind, placebo‐controlled, post-neoadjuvant phase III study that will recruit 1,100 patients from approximately 300 sites in 13 countries.

Background About one third of hormone‐receptor positive, HER2‐normal breast cancer patients with residual disease after neoadjuvant chemotherapy B have a substantial risk of relapse (von Minckwitz et al. J Clin Oncol 2012). Those patients can be identified using the validated clinical‐pathologic stage–estrogen/grade (CPS‐EG) scoring system (Figure 1A), ranging from 0-6 (Jeruss et al. J Clin Oncol 2008; Mittendorf et al. J Clin Oncol 2011). Palbociclib is an oral, highly selective inhibitor of CDK4/6 kinase activity that prevents cellular DNA synthesis by inhibiting cell cycle progression (Finn et al. Breast Cancer Res 2009). Luminal tumors have shown sensitivity to palbociclib. In a phase II study, palbociclib ex tended progression free survival in combination with letrozole as first-line hormonal treatment for advanced breast cancer (Finn et al. Cancer Res 2012). Palbociclib has also shown single Figure 1: CPS-EG score point assignment (from Jeruss et agent activity in patients with relapsed hormone al. J Clin Oncol 2008) (A) and disease-free survival receptor positive advanced breast cancer according to CPS-EG scores in HER2-/HR+ patients (DeMichele et al. Cancer Res 2011). Phase III studies (Marmé et al. Eur J Cancer 2015) (B). in 1st and 2nd line metastatic breast cancer have been completed and results are awaited for 2015. In addition, iDFS excluding second non-breast In the PenelopeB study we aim to demonstrate cancers, overall, distant disease‐free and local that one year of post-neoadjuvant treatment recurrence-free survival, iDFS per treatment with palbociclib, in addition to standard anti‐ group in patients with luminal-B tumors, com hormonal therapy, provides a superior invasive pliance and safety, patient reported outcomes disease free survival and an acceptable safety (quality of life), health economics, drug-drug profile compared to placebo in women with interaction-potential for the palbociclib – hormone‐receptor positive, HER2‐normal early endocrine combination therapy (in a subset of breast cancer who did not obtain a pathological this patient population) as well as correlations complete response after taxane‐containing between drug exposure and efficacy and safety neoadjuvant chemotherapy and are at high risk of findings will be analyzed. The study includes relapse (CPS‐EG score ≥ 3 or 2 if ypN+) (Figure 1B). post- as well as premenopausal women and allows the use of different endocrine therapies. Study design and objectives Furthermore, the PenelopeB study will also PenelopeB primarily aims to compare invasive address translational research questions, such as disease-free survival (iDFS) between the two the role of biomarkers involved in the CDK4/6 treatment arms. pathway. | 55

Figure 2: PenelopeB study design

With amendment 9, patients with a CPS-EG study and are scheduled to occur when 1/3 of score of 2 are allowed to take part in the study if the total events and 2/3 of the total events re- node positive after neoadjuvant treatment. spectively are observed. Final analysis on the primary endpoint and sec- Study report ondary efficacy endpoints (except for OS) will be PenelopeB started recruitment with the first ran- conducted when 255 iDFS events have been ob- COLLABORATING domized patient in Germany in February 2014. served, which is estimated to occur about 6.5 STUDY GROUPS: The first international patient was randomized years after first patient in. Results are expected in Spain in October 2014. As of 31st December in QIII/2020. 2016, a total of 793 patients from 190 trial sites In the last IDMC meeting (November 9th 2016), worldwide (76 sites in Germany) have been in- the committee did not see any safety concerns cluded. and recommended to continue the trial as Two interim analyses are planned in the PenelopeB planned.

1100 accrual no. 1000 original plan revised protocol C8 900 projection protocol D9

800

700

600

500 SPONSOR: 400 GBG Forschungs GmbH

300 INTERNATIONAL 200 STUDY CHAIR: 100 Prof. Dr. Gunter von Minckwitz

0 German Breast Group Feb-14 Jun-14 Oct-14 Feb-15 Jun-15 Oct-15 Feb-16 Jun-16 Oct-16 Feb-17 Jun-17

Figure 3: PenelopeB recruitment as of 31st December 2016 STUDY CHAIR (GERMANY): Prof. Dr. Toralf Reimer We are thanking all participating centers for their commitment and efforts so far. We would like to Universitätsfrauenklinik encourage all sites to continue to support the PenelopeB study by recruitment of patients and the und Poliklinik am Klinikum timely provision of biomaterial. Südstadt Rostock 56 | Annual Scientific Report 2016 | Recruiting Studies

GBG 68: GAIN-2

Neo-/adjuvant phase III trial to compare in- vent-based taxanes and might thus be preferred tense dose-dense chemotherapy to tailored in an intense dose-dense regimen. dose dense chemotherapy in patients with It is long known from the NSABP-B18 trial and high-risk early breast cancer others that neoadjuvant chemotherapy is as effective as adjuvant chemotherapy in preventing NCT01690702 recurrences (Wolmark et al. J Natl Cancer Inst CONTACT: Monogr 2001). Dr. Cornelia Schneider-Schranz GAIN-2 is a neo-/adjuvant, prospective, multi- The hypothesis studied by GAIN-2 is that in pa- Clinical Project Management center, randomized, open-label phase III trial tients with early node-positive or high-risk [email protected] that will recruit 2,886 patients from 136 sites in node-negative breast cancer, a pre-defined, in- Germany. tense, dose-dense, regimen (EnPC – epirubicin followed by nab-paclitaxel followed by cyclo- Background phosphamide) is more effective compared with Combined chemotherapy regimens always re- a dose-dense regimen, where single doses are quire compromises regarding the doses of each adjusted depending on individual hematological drug and the treatment intervals due to acute and non-hematological toxicities (dtEC-dtD - and cumulative toxicities. The sequential admin- dose-dense, dose-tailored epirubicin and cyclo- istration of monotherapies, however, allows the phosphamide followed by dose-dense, dose-tai- administration of high doses of single substances lored docetaxel). and dose-dense intervals. Such intense, dose- The maximum dose of nab-paclitaxel in this set- dense chemotherapy regimens have shown to ting has been explored in a run-in phase included improve the survival in early breast cancer pain the study design. It has been shown that patients with high risk of recurrence when com- tients can safely be treated with a biweekly dos- pared to conventional dosed chemotherapy age of 330 mg/m2 nab-paclitaxel (Möbus et al. J (Möbus et al. J Clin Oncol 2010; Citron et al. J Clin Oncol 2013) which is now used for the main Clin Oncol 2003). However, both of these dose- phase of the study. dense regimens tested so far used solvent-based An amendment of the study protocol (effective taxanes (paclitaxel and docetaxel) and nowa- as of 1st August 2016) allows treatment of pa- days outdated comparators. tients with the same regimens in the neoadju- Nab-paclitaxel, the nanoparticle albumin-bound vant setting. All neoadjuvant patients with form of paclitaxel, has shown a better toxicity HER2-positive disease receive trastuzumab and profile and higher efficacy compared to sol- optional pertuzumab at doses and duration in

N=2,886 N=220***

Pertuzumab i.v. * 3 weeks PINT 2 after cycle 8 Trastuzumab i.v. ** ) Trastuzumab s.c. E E E nP nP nP C C C Abdominal wall Epirubicin nab-Paclitaxel Cyclophosphamide Trastuzumab s.c. 𝐩𝐩𝐩𝐩𝐩𝐩𝐩𝐩𝐩𝐩𝐩𝐩𝐩𝐩𝐩𝐩 150 mg/m2 q2w 330 mg/m2 q2w 2,000 mg/m2 q2w. 600 mg q3w.**** Adjuvant

- / R Pertuzumab i.v. * PINT1 R PINT 2 after cycle 8 𝟎𝟎𝟎𝟎𝟎𝟎 ,

𝟏𝟏 Trastuzumab i.v.** Neo

rest ( E E E E Trastuzumab s.c. D D D D Thigh C C C C week Epirubicin 1 Docetaxel Trastuzumab s.c. Starting dose: 90 mg/m2 q2w Starting dose: 75 mg/m2 q2w 600 mg q3w.**** Dose levels (min./max.) Dose levels (min./max.) 2 2 - Level – 3: 38 mg/m q2w - Level – 1: 60 mg/m q2w. Patient interview before randomization (PINT1) and 2 2 PINT - Level + 2: 120 mg/m q2w - Level + 2: 100 mg/m q2w. after cycle 8 trastuzumab s.c. (PINT2) * Pertuzumab i.v. (if HER2-positive and neoadjuvant): Cyclophosphamide Starting dose 840 mg q3w, thereafter 420mg q3w. 2 Starting dose: 600 mg/m q2w ** Trastuzumab i.v. (if HER2-positive and neoadjuvant or Dose levels (min./max.) adjuvant): Starting dose 8 mg/kg BW q3w, thereafter 2 - Level – 3: 450 mg/m q2w 6 mg/kg BW q3w 2 - Level + 2: 1,200 mg/m q2w *** Thereof, 18 patients per arm will be randomised into the pharmacokinetics sub-substudy. **** Therapy duration trastuzumab i.v. and s.c. totally 1year

Figure 1: Study design of the GAIN-2 main study and the subcutaneous trastuzumab substudy | 57

concordance to current treatment guidelines. pharmacokinetic measurements will be per- GAIN-2 primarily aims to compare invasive dis- formed in 36 patients (18 per group). ease-free survival after neo-/adjuvant chemotherapy with EnPC or dtEC-dtD in patients with Substudy biology of lymph node metastases primary node-positive or high risk node negative The substudy on biology of lymph node metas- breast cancer. In addition, overall, distant dis- tases will examine primary tumors and corre- ease-free, locoregional relapse-free, local responding axillary lymph nodes for biologically lapse-free, regional relapse-free and brain me- relevant factors involved in lymphogenic and tastasis-free survival, compliance and safety, distant tumor cell spread. Written informed con- side-effects of taxanes, pCR rate in patients sent and the availability of primary tumor and treated with neoadjuvant therapy and treat- axillary lymph node tissue are crucial for this ment effects by intrinsic subtypes, number of translational substudy. involved nodes and Ki-67 will be compared between the two treatment arms. Breast conserva- Substudy on SNP (Single Nucleotide tion rate between adjuvant and neoadjuvant Polymorphisms) patients as well as the survival endpoints by pCR This observational substudy aims to associate will be also assessed. the germline genotype of the patient with the In addition, GAIN-2 offers the opportunity to treatment response, long term prognosis and address a range of translational research ques- the molecular profile of the tumors in both ran- tions, which are summarized bellow. domization arms.

Substudies Substudy on ovarian function Substudy subcutaneous trastuzumab Young breast cancer patients (≤ 45years) have a In addition to the main protocol, 220 HER2-pos- risk of developing premature ovarian failure. It itive patients of the GAIN-2 trial will be ran has been shown that the majority of patients domized to receive further trastuzumab subcuta- have regained menstruation, a surrogate for neously (s.c.) instead of intravenously (i.v.) after ovarian function, after 2 years (Gerber et al. J completion of the chemotherapy according to Clin Oncol 2011). However, within this diagnos- current guidelines. The patients will be ran tic ovarian function substudy the combination domized between trastuzumab application into of menstruation with hormones such as FSH, E2, thigh or abdominal wall and the preference of and anti-Mullerian Hormone (AMH) in addition the patients will be determined. In addition, to follicle count measured by ultrasound allows

Top recruiters GAIN-2: accrual no. Gynäkologisch-onkologische Praxis Hannover: N=71 originally planned no. Praxis Dr.Ladda Neumarkt: N=48 planned no. 2500 Universitätsmedizin Mainz: N=46 Klinikum Frankfurt-Höchst: N=42 Onkologische Praxis Pinneberg: N=41

Top recruiters Trastuzumab s.c. substudy: 2000 Klinikum Pforzheim: N=6 Mutterhaus der Borromäerinnen Trier: N=6 Universitätsfrauenklinik Mainz: N=6 Onkozentrum Dresden: N=5 Stadtkrankenhaus Worms: N=5 1500

1000

500 N=2385

0 Sep-12 Feb-13 Jul-13 Dec-13 May-14 Oct-14 Mar-15 Aug-15 Jan-16 Jun-16 Nov-16

Figure 2: GAIN-2 recruitment as of 31st December 2016 58 | Annual Scientific Report 2016 | Recruiting Studies

defining the risk of premature ovarian failure and tients from GAIN-2 have been enrolled into the loss of fertility within modern therapeutic regi- trastuzumab s.c. substudy. No safety data are men. In addition, the rate of amenorrhea can be currently available for the substudy. correlated with survival. Patients of ≤ 45 years and written informed consent can be included. Publications 1. Noeding S, Forstbauer H, Wachsmann G, et Study report al. GAIN2: Adjuvant phase III trial comparing The planned interim safety analysis after 200 an intensified dose-dense adjuvant therapy patients has shown an acceptable toxicity pro- with EnPC compared to a dose-dense, dose- file [1] and the study has been continued with- adapted therapy with dtEC dtDocetaxel in COLLABORATING out changes. The second safety interim analysis patients with primary breast cancer and a STUDY GROUPS: after 900 patients showed no additional or un- high risk of recurrence. Ann Oncol 2014, 25 expected safety signals in the EnPC or dtEC-dtD (suppl_4): iv90. arm. Therefore, no modifications in the conduc- 2. Volker Moebus, Hans-Joachim Lück, Helmut tion of the study were necessary [2]. As of 31st Forstbauer GAIN-2: Adjuvant Phase III Trial to December 2016, a total of 2,385 patients have Compare Intense dose-dense (idd) Treatment been enrolled in the main study (Figure 2), of with EnPC to Tailored dose-dense (dt) them 2,140 in the adjuvant setting and 245 in Therapy with dtEC-dtD for Patients with the neoadjuvant setting. The trastuzumab sub- high-risk Early Breast Cancer: Results of the study has enrolled the first patient in November Second Safety Interim Analyses. Cancer Res 2013. Currently a total of 150 HER2-positive pa- 2016;76(4 Suppl):Abstract P1-13-05 SPONSOR: GBG Forschungs GmbH We are thanking all participating centers for their commitment and efforts so far. We would like to STUDY CHAIR: encourage all sites to continue to support the GAIN-2 study by recruitment of the final patients and Prof. Dr. Volker Möbus by promptly providing the remaining biomaterial. Klinikum Frankfurt Höchst We also encourage all participating sites to support the various substudies. | 59

GBG 54: MALE

A prospective, randomized multi-centre phase available information on adjuvant therapies for II study evaluating the adjuvant, neoadjuvant male breast cancer derives from few retrospective or palliative treatment with tamoxifen +/- studies and retrospective case presentations. GnRH analogue versus aromatase inhibitor + Current clinical management is extrapolated GnRH analogue in male breast cancer patients from principles established for female breast cancer. Since 90 % of male patients have hor NCT 01638247 mone receptor positive breast cancer, anti- hormonal treatment seems indicated, and men CONTACT: MALE is the first phase II study on male breast therefore receive 20 mg tamoxifen as standard Dr. Ursula Wolf cancer in Germany that will prospectively, adjuvant therapy. Clinical Project Management randomize 48 patients from approximately 35 Although women benefit from treatment with [email protected] sites. aromatase inhibitors, there are only a few case reports on men treated with aromatase Background inhibitors. Men under aromatase inhibitors have Breast cancer in men is a rare disease with an shown a suppression of estradiol of only 40- incidence rate of less than 1 % (Siegel et al. CA 50 % with an increase of testosterone of about Cancer J Clin 2013). In Germany, only up to 600 50 % (Mauras et al. J Clin Endocrinol Metab cases are diagnosed each year (Robert Koch 2000). The blockage of the aromatase may Institut, Cancer in Germany 2009/2010). This further activate the hypothalamic-pituitary rarity of male breast cancer leads to a low index feedback loop resulting in an increase of of suspicion of both patients and their doctors substrates for aromatization. This may stimulate which is largely responsible for a delay in the testes, via an increase of FSH (follicle- diagnosis and therefore for a more advanced stimulating hormone), to further produce stage of disease at presentation. This adversely estrogen. Addition of a gonadotropin-releasing affects prognosis with only little improvement hormone (GnRH) analogue, can interrupt the over the years (O'Malley et al. Cancer 2002). negative feedback loop and complete estrogen Treatment strategies are not based on data from suppression can be achieved. prospective, randomized clinical studies and The MALE study will include male breast cancer therefore optimal therapy is unknown. The only patients in the neoadjuvant, adjuvant and

Tamoxifen Arm A

GnRH analogue

N = 48 R Tamoxifen Arm B GnRH analogue

Exemestane Arm C

month 3 month 6

Tamoxifen: 20 mg daily.

Exemestane: 25 mg daily.

Goserelin: 10.8 mg s.c. (see SmPC) a>er randomisa@on and a>er three months. OR

Leuprorelin: 11.25 mg s.c. (see SmPC) a>er randomisa@on and a>er three months.

Figure 1: MALE study design 60 | Annual Scientific Report 2016 | Recruiting Studies

metastatic setting that will be randomized in a Moreover, MALE offers the unique opportunity 3-arm design to receive tamoxifen alone, to address important translational research tamoxifen + GnRH analogue or exemestane + questions, such as the role of cytochrome P450 GnRH analogue for a total duration of 6 months. polymorphisms (involved in tamoxifen meta bolism) and the hormone receptor activity and MALE primarily aims to compare the estradiol aromatase expression of the tumor tissue. suppression between the three treatment arms after three months. Secondary objectives are to Study report compare the estradiol suppression after six The study started recruitment in August 2012. months, treatment compliance, efficacy in The planned time for the total accrual of 48 terms of overall response (for neoadjuvant and patients within two years has been prolonged metastatic patients), changes in hormonal until 2018. As of 31st December 2016, a total of status (testosterone, dihydrotestosterone, SHBG, 44 patients have been recruited in the trial. FSH, LH, osteocalcin and CTX), and safety and side effect parameters between the three treatment arms.

COLLABORATING 40 STUDY GROUPS: accrual no. planned no.

20 SPONSOR: GBG Forschungs GmbH supported by the 10 N=31 Claudia von Schilling Foundation Award 2012 0 Sep-12 Jan-13 May-13 Sep-13 Jan-14 May-14 Sep-14 Jan-15 May-15 Sep-15 Jan-16 May-16 Sep-16 Jan-17 May-17 Sep-17 STUDY CHAIRS: Dr. Mattea Reinisch Figure 2: MALE recruitment as of 31st December 2016 Kliniken Essen Mitte

Prof. Dr. Sibylle Loibl We would like to thank all participating centers for their commitment and efforts so far. We encourage all German Breast Group sites to continue to support the MALE study by recruitment of the final patients and the timely provision of the remaining biomaterial. | 61

GBG 29: Breast Cancer in Pregnancy (BCP)

Prospective and Retrospective Registry Study vestigate breast cancer during pregnancy and to of the German Breast Group (GBG) for increase the evidence for treatment options. Diagnosis and Treatment of Breast Cancer in With an amendment of the original study Pregnancy compared to Young Non-Pregnant protocol, it is now possible to also include a non- Women pregnant control cohort of women diagnosed with breast cancer at or below the age of 40 CONTACT: NCT 00196833 years. Those can be matched to the pregnant breast cancer patients as controls treated in Dr. Ioannis Gkantiragas BCP is a long time retrospective/prospective everyday clinical practice. Clinical Project Management multicenter, international registry that will All patients with histologically confirmed breast [email protected] recruit pregnant breast cancer patients and non- cancer who are pregnant, as well as patients of pregnant young women. 40 years or younger with histologically con firmed breast cancer who are not pregnant and Background have given informed consent for data collection Breast cancer in pregnancy is regarded as a rare and biomaterial collection can be entered into coincidence. However, about 7 % of the women the registry. Retrospective participants can be diagnosed with breast cancer are younger than entered without an informed consent as long as 40 years with a small increase in the incidence in the data are captured anonymously. the last years (Eisemann et al. Geburtsh Frauen heilk 2013; De Santis et al. CA Cancer J Clin Study objectives 2011). The median age of first pregnancy in The BCP study primarily aims to assess the fetal Germany is 30 years (according to the federal outcome 4 weeks after delivery. Secondary statistical office). Since the incidence of breast endpoints will include maternal outcome of cancer under the age of 40 is rising and women pregnancy, tumor stage at presentation and tend to delay pregnancy into later reproductive biological characteristics, breast cancer therapy, years the coincidence of pregnancy and breast type of surgery, mode of delivery (vaginal vs. cancer is increasing. Little is known about the in caesarean), outcome of the new-born 5 years cidence of breast cancer in pregnancy in after diagnosis, and outcome of breast cancer 5 Germany and Western Europe. years after diagnosis. Therefore, in 2003 the German Breast Group In addition the registry allows investigation of launched a registry which was extended translational research questions, using tumor throughout Europe and worldwide (Breast specimen as well as placenta tissue from patients International Group), to systematically in with breast cancer during pregnancy.

1000 Total accrual n = 1452 900 Germany n = 1107 (total) n = 435 (pregnant) n = 672 (non-pregnant)

Foreign countries n = 345 (total) n = 324 (pregnant) n = 21 (non-pregnant) 800 pregnant no. not pregnant no. 700

600

500

400

300

200

100

0 Apr-03 May-04 Jun-05 Jul-06 Aug-07 Sep-08 Oct-09 Nov-10 Dec-11 Jan-13 Feb-14 Mar-15 Apr-16 May-17

Figure 1: BCP recruitment as of 31st December 2015 62 | Annual Scientific Report 2016 | Recruiting Studies

Study report on breast cancer biology at the genomic level. It A first analysis of registry data looking at fetal is believed that breast cancer during pregnancy health for up to 4 weeks after delivery included is biologically not different from breast cancer 447 patients and showed that although infants diagnosed outside pregnancy [5]. Current trans- exposed to chemotherapy in utero had a lower lational research project aims to compare the birth weight and reported more complications mutational pattern as well as the gene expres- compared to their unexposed counterparts, sion profile between pregnant and non-preg- these differences were not clinically significant. nant patients with breast cancer. In our prelimi- Since none of the infants was exposed to chemo nary results we showed that overall the therapy in the first trimester, the differences mutational landscape does not seem to be dif- were most likely related to premature delivery. ferent between pregnant patients enrolled in This led to the conclusion that preterm birth was BCP study and no-pregnant controls obtained strongly associated with adverse events and a from The Cancer Genome Atlas (TCGA) data- full-term delivery seems to be of paramount base [6]. Further analyses using other datasets importance. Moreover, a delay of cancer treat are currently conducted. ment did not significantly affect disease-free survival for mothers with early breast cancer [1]. Publications In a combined analysis of our data with data 1. Loibl S, Han SN, von Minckwitz G, et al. Treat- form the Cancer in Pregnancy registry, Leuven, ment of breast cancer during pregnancy: an Belgium, we were able to demonstrate that observational study. Lancet Oncol 2012; overall survival was similar for patients 13(9):887-96. diagnosed with breast cancer during pregnancy 2. Amant F, von Minckwitz G, Han SN, et al. compared with non-pregnant patients. This Prognosis of women with primary breast can- information is important for counseling patients cer diagnosed during pregnancy: results from and supports the option to start treatment with an international collaborative study. J Clin continuation of pregnancy [2]. Oncol 2013; 31(20):2532-9. We were also able to demonstrate that 3. Loibl S, Han S, Mayer K, et al. Neoadjuvant neoadjuvant chemotherapy for patients with chemotherapy for patients with breast cancer COLLABORATING breast cancer during pregnancy results in the during pregnancy (BCP). J Clin Oncol 2014; STUDY GROUPS: same pCR rate if given during pregnancy or after 32:5s (suppl; abstr 1071). delivery and as in non-pregnant controls. 4. Loibl S, Schmidt A, Gentilini O, et al. Breast Disease free and overall survival was not Cancer Diagnosed During Pregnancy: Adapt- different between the three cohorts [3]. ing Recent Advances in Breast Cancer Care for The general recommendation derived from the Pregnant Patients. JAMA Oncol 2015; registry data so far is that patients with breast 1(8):1145-53. cancer during pregnancy can and should be 5. Loibl S, Han SN, Amant F. Being Pregnant and treated as closely as possible according to pa- Diagnosed with Breast Cancer. Breast Care tients diagnosed outside pregnancy Up-to-date (Basel). 2012; 7(3):204-209. guidelines on breast cancer during pregnancy, 6. Loibl S, Pfarr N, Weber K, et al. Comparison of which also incorporated outcomes from the BCP the mutational landscape of breast cancer SPONSOR: registry, has been published in 2015 [4]. during pregnancy and non-pregnant controls. The project was initially Little is known about the impact of pregnancy Abstract P2-03-09, SABCS 2016 supported by the BANSS- Foundation and German Cancer Consortium (DKTK) Thanks to all participating sites and practices that have entered their patients into the registry and have contributed to this important research so far. We would kindly like to remind all study centers STUDY CHAIR: to provide biomaterial which is urgently needed to answer translational research questions. More Prof. Dr. Sibylle Loibl information and CRF forms are available on the GBG website: German Breast Group http://www.germanbreastgroup.de/de/studien/bcp.php | 63

GBG 79: Brain Metastases in Breast Cancer (BMBC) BMBC BMBC (Brain Metastases in Breast Cancer) is a disease. A multidisciplinary approach with rapid long time retrospective and prospective multi- integration of new treatment strategies is center registry designed to collect tumor char- required for the treatment of patients developing acteristics of the primary and metastatic tumor brain metastases, aiming to prolong survival, as well as treatment data from patients diag- preserve neurologic function and improve CONTACT: nosed with brain metastases of breast cancer quality of life. Udo Pfeil treated in German hospitals. The BMBC registry was initiated to include Clinical Project Management patients with brain metastases and a history of [email protected] Background breast cancer that were diagnosed for brain Brain metastases of breast cancer reduce quality metastases since the year 2000. Registration of of life and prognosis in breast cancer patients. patient data is allowed prospectively after Their incidence has increased during the last obtaining an informed consent. Retrospective years (Frisk et al. Br J Cancer 2012). 10-40% of participants can be entered without an informed patients with metastatic breast cancer will consent if the patient is not able to sign the develop brain metastases during the course of informed consent and as long as the data are disease depending on the biological subtype of anonymously captured. the primary tumor. The prognosis for patients The registry study is performed in collaboration with brain metastases is generally poor. Good with Prof. Dr. Volkmar Müller, Priv. Doz. Dr. performance status and a limited number of Isabell Witzel, and Dr. Elena Laakmann from the brain metastases are factors that can prolong Universitätsklinikum Hamburg-Eppendorf. survival (Ogawa et al. J Neurooncol 2008). Ther apeutic approaches in treating metastases of the Study objectives central nervous system include surgery, radio The BMBC registry aims to collect data to deter- therapy and systemic chemotherapy and the mine the incidence of brain metastases, the combination of these options. number and size of brain metastases, location, Due to the analysis of small and heterogeneous histopathological characteristics of the primary patient cohorts, risk factors for the development tumor and brain metastases, sensitivity of diag- of brain metastases and the impact of early nostic tools (cranial computed tomography (CT) detection of brain metastases have been and magnetic resonance imaging (MRI)), perfor- analyzed insufficiently. Improvement of treat mance status, prognosis, quality of life, and the ment strategies are required as the number of influence of treatment strategies on prognosis brain metastases will increase over the next and neurological function. In addition, the regis- years due to the better control of visceral try allows investigation of translational research

2000 1900 1800 accrual no. 1700 1600 1500 1400 1300 1200 1100 1000 900 800 700 600 500 400 300 N=1745 200 100 0 Apr-14 Oct-14 Apr-15 Oct-15 Apr-16 Oct-16 Apr-17 Oct-17

Figure 1: BMBC recruitment as of 31st December 2016 64 | Annual Scientific Report 2016 | Recruiting Studies

questions, using tumor specimen of the primary centers participating in the BMBC registry in and metastatic tumor. Germany [4]. Patients with HR-positive or Planned analyses include treatment patterns in HER2-positive status had a significantly lower Germany, patient outcome, as well as validation number of BM at diagnosis. HER2-positive pa- of prognostic scoring systems in a multicenter tients treated with trastuzumab before the diag- setting and in the context of new targeted nosis of BM demonstrated a lower incidence of therapies. Planned translational research intracranial metastases. Patients with HER2- projects include the impact of glycosylation, positive breast cancer had higher number of cer- resistance mechanisms against HER2-targeted ebellar metastases compared to HER2-negative therapies, the role of the blood brain barrier, patients, whereas patients with triple negative evaluation of markers of radioresistance and breast cancer developed more often leptome- specific genomic alterations associated with ningeal disease. These findings suggest a differ- brain tropism of breast cancer cells. ent tumor cell homing to different brain regions depending on molecular subtype and treatment. Study report Registration of patients is ongoing. The study was opened for documentation in April 2014 with more than 50 participating Publications centers. So far, 1,745 patients have been re 1. Witzel I, Laackman E, Fehm T, et al. Brain COLLABORATING gistered and 233 tissue samples have been Metastases in Breast Cancer Network Ger- STUDY GROUPS: received. many (BMBC, GBG 79): Multicentric, retro- Recent analysis of treatment patterns and clini- and prospective collection of patient data cal outcome in 1,105 breast cancer patients with and biomaterial from breast cancer patients brain metastases (BM) has confirmed our previ- with brain metastases. Cancer Res 2015; 75 ous findings [1, 2] and has shown that HER2-posi (9 Suppl): Abstract OT2-5-01 tive patients had the longest median survival 2. Witzel I, Loibl S, Laakmann E, et al. Brain with 12.1 months (95 % CI, 10.2–13.7) compared Metastases in Breast Cancer Network Ger- to 5.5 months (95 % CI, 4.1–7.1) for luminal pri- many (BMBC, GBG 79): First analysis of 1004 mary tumors and 4.1 months (95 % CI, 3.1–4.8) patients from the multicenter registry. for triple-negative patients (p < 0.001) [3]. Me- Cancer Res 2016;76(4 Suppl): Abstract P6-17- dian overall survival (OS) of patients with BM 08 diagnosed between 2000 and 2006 was 7.2 3. Mueller V, Loibl S, Laakmann E et al. Doris Au- months (95 % CI, 5.3-9.6) compared to 6.7 gustin,Brain Metastases in Breast Cancer months (95 % CI, 5.5-8.1) of patients diagnosed Network Germany (BMBC, GBG 79): Treat- between 2007 and 2012 (p = 0.848). However, ment patterns and clinical outcome of more we could not observe improvement of survival than 1000 patients with brain metastases over the study period and the time intervals cho- from breast cancer. J Clin Oncol 34, 2016 sen in any tumor subtype. Hence, larger patients (suppl; Abstract 2070) cohorts are needed to detect the survival differ- 4. Laakmann E, Witzel I, Scriba V et al. Radiolog- SPONSOR: ence. ical Patterns of Brain Metastases in Breast GBG Forschungs GmbH In addition, a subproject has retrospectively ana Cancer Patients: A Subproject of the German lysed clinical data, CT and MRI scans obtained Brain Metastases in Breast Cancer (BMBC) from 300 breast cancer patients with BM from 4 Registry. Int J Mol Sci. 2016 23;17(10). STUDY CHAIR: PD Dr. Isabell Witzel

and Prof. Dr. Volkmar Müller We encourage all study centers and practices to enter eligible patients into the registry. Universitätsklinikum We thank all participating sites that have entered their patients into the registry and have con- Hamburg-Eppendorf tributed to this important research so far. Klinik und Poliklinik für We would like to kindly remind all sites to provide biomaterial which is urgently needed to answer Gynäkologie translational research questions. | 65

GBG 86: DESIREE

A multicenter, randomized, double-blind, phase to findings from BOLERO-2, where the number II study to evaluate the tolerability of an induc of patients still taking full-dose (10 mg) of tion dose escalation of everolimus in patients everolimus at 4, 8, and 12 weeks is 77.8 %, with metastatic breast cancer 75.6 %, and 75.6 %, respectively. These findings are in concordance with non-interventional NCT 02387099 studies. In the non-responder part (setting III) of the DESIREE is a multicenter, double-blind, randomized neoadjuvant GeparQuinto study, everolimus CONTACT: phase II trial that will recruit 156 patients from 60 was given as salvage treatment in combination Dr. Ioannis Gkantiragas sites in Germany within approximately 24 months. with paclitaxel for patients without response to Clinical Project Management 4 cycles epirubicin/cyclophosphamide +/- be [email protected] Background vacizumab. A dose-escalation schema was The BOLERO-2 study demonstrated an successfully used to improve tolerability of enormous benefit for patients who received everolimus together with the cytotoxic agents everolimus in addition to exemestane and who (von Minckwitz Ann Oncol 2011; von Minckwitz progressed during/after a non steroidal aromatase Ann Oncol 2014). inhibitor (NSAI) (Baselga N Engl J Med 2012), The palliative DESIREE study compares the which led to approval of everolimus in this cumulative rate of mucositis/stomatitis grade indication. However, experience from routine 2-4 (WHO’s oral toxicity scale (OTS) at 12 weeks use has shown a high rate of intolerability of this after start of treatment using a conventional innovative treatment approach especially during and a dose-escalating schema of everolimus in the first 12 weeks of treatment. Most common combination with exemestane in patients with side effect is mucositis/stomatitis which is con metastatic breast cancer and progression or sidered the leading cause for treatment dis relapse after non-steroidal aromatase-inhibitor continuation not related to tumor progression. treatment. This outside clinical trial experience is contrary

Everolimus 10 mg

N=156 week 1-3: 4x2.5 mg/day (blinded) week 4-24: 10mg/day (open) R

Everolimus dose-escalated

week 1: 1x2.5 mg verum + 3 x placebo/day + exemestane week 2: 2x2.5 mg verum + 2 x placebo/day week 3: 3x2.5 mg verum + 1 x placebo/day week 4-24: 10 mg/day (open)

Eligibility according to current label

Figure 1: DESIREE study design 66 | Annual Scientific Report 2016 | Recruiting Studies

Study design and objectives to grade ≥ 2 mucositis/stomatitis, cumulative DESIREE primarily aims to assess the cumulative dose at 4 weeks, relative dose intensity for rate of mucositis/stomatitis grade 2-4 (OTS) at everolimus and quality of life using the FACT-B 12 weeks after start of treatment using a questionnaire and the QSDQ. Potential biomarkers conventional and a dose-escalating schema of predicting safety and compliance will be de everolimus in combination with exemestane. termined after completion of study treatment. Secondary objectives are: the cumulative rate of mucositis/stomatitis grade 2-4 (OTS), cumulative Study report rate of mucositis/stomatitis grade 1 and any grade DESIREE started recruitment in June 2015. As of (OTS) at 12 and 24 weeks after start of treatment, 31st December 2016, a total of 56 patients have rate of patients on 10mg daily at 12 weeks and 24 been included. The end of the study (i.e. last visit weeks, clinical benefit rate at 24, safety with of the last patient randomized) is initially regard to other organ signs and symptoms, time estimated for October 2017.

40 40 40 40 40 40 40 38 38 38 38 37 140 35 35 35 35 35 33

120 29 30 accrual no. initiated sites 100 25 23 23 planned no. 21 revised projection 80 20

60 15 COLLABORATING

STUDY GROUPS: 40 10

20 5

0 0 Jun-15 Oct-15 Feb-16 Jun-16 Oct-16 Feb-17 Jun-17

Figure 2: DESIREE recruitment as of 31st December 2016 SPONSOR: GBG Forschungs GmbH

STUDY CHAIR: We are thanking all participating centers for their commitment and efforts so far. We would like to Prof. Dr. Sibylle Loibl encourage all sites to continue to support the DESIREE study by recruitment of patients and provi German Breast Group sion of biomaterial in a timely manner. | 67

GBG 75: INSEMA INSEMA

Comparison of axillary sentinel lymph node tion and in an 1:1 allocation for the second rand- biopsy versus no axillary surgery in patients with omization. Duration of recruitment is 4 years. early-stage invasive breast cancer and breast- The aim of the trial is to compare the invasive conserving surgery: a randomized prospective disease-free survival after breast-conserving surgical trial surgery between patients who received no axil- CONTACT: lary surgery vs. patients who received sentinel Dr. Cornelia Schneider-Schranz NCT02466737 lymph node biopsy (SLNB) and between node Clinical Project Management positive patients who received SLNB alone vs. [email protected] INSEMA is a prospective, multicenter, ran patients with completion of axillary lymph node domized trial that will recruit 5,940 patients dissection (ALND). Secondary objectives of the from approximately 150 sites in Germany and study are to compare the invasive disease-free approximately 20 sites in Austria. survival after breast-conserving surgery between patients with no axillary surgery vs. node Background negative patients, between patients with no ax- Although there is no doubt that the presence of illary surgery vs. node positive patients who re- lymph node metastases worsens prognosis of a ceived SLNB alone, and between patients with patient, there is a lack of unambiguous evidence no axillary surgery vs. node positive patients to support lymph node dissection. Axillary with completion of ALND. Furthermore, the surgery for breast cancer is now considered as a study allows comparison of overall survival, lo- staging procedure that does not seem to coregional disease-free survival (no tumor in the influence breast cancer mortality, since the risk ipsilateral breast or ipsilateral supraclavicular, of developing metastases depends mainly on subclavicular, internal mammary or axillary the biological behavior of the primary tumor nodes), ipsilateral axillary recurrence rate, dis- (seed-and-soil model). tant disease-free survival, and quality-of-life be- Women with breast cancer have benefitted tween arms as well as the event-free survival in greatly from a series of carefully conducted subgroups according to age (< 65 vs. ≥ 65 years), randomized controlled trials focusing on axillary grading G1/2 vs. G3), tumor size (≤ 2 vs. > 2 cm), surgery. Each successive trial showed that less and study site (German vs. Austrian sites in ran- surgery was better, as outcomes were the same domization 2). and less surgical intervention resulted in fewer INSEMA also has an attached translational pro- surgical complications (Fisher et al. N Engl J Med gram including biobanking of tumor tissue and 2002; Rudenstam et al. J Clin Oncol 2006; serum samples. One translational objective is to Martelli et al. Ann Surg 2005; Veronesi et al. Ann determine the value of Memorial Sloan-Ketter- Oncol 2005; Giuliano et al. Ann Surg 2010; ing Cancer Center nomograms in predicting in- Giuliano et al. JAMA 2011). volved sentinel nodes and positive non-sentinel A high rate of loco-regional control could be nodes after positive SLNB. achieved with multimodality therapy, even An amendment of the study protocol (version without axillary lymph node dissection (ALND). 15.09.2016) includes the following changes: a) Despite increasing evidence disfavoring ALND, it within the inclusion criteria: patients with age at remains part of widely recognized guidelines for diagnosis ≥ 18 years can be enrolled in the study; breast cancer treatment. The modern approach histologically confirmation of the unilateral pri- in breast cancer care, which includes improved mary invasive carcinoma of the breast can be imaging, more detailed pathological evaluation, also done by open biopsy; multifocal or multi- improved planning of surgical and radiation centric tumors are allowed if breast-conserving therapy, and more effective systemic treatment, surgery is planned; patients with SLNB and pN+ emphasizes the need for ongoing re-evaluation (sn) (1-3 macrometastases, stage pN1a) will un- of “standard” local therapy. The postsurgical dergo a second randomization to either SNLB therapy should be considered on the basis of alone or completion ALND; patients with ≥ 4 biologic tumor characteristics rather than nodal metastatic SLN should undergo completion involvement. ALND; b) within the exclusion criteria: patients with history of malignancy within last 5 years as Study design and objectives well as pregnant or lactating patients are ex- INSEMA is a prospective randomized surgical tri- cluded from the study; c) adaptation of the al. Patients are randomized into two treatment postoperative radiotherapy: patients with arms in an 1:4 allocation for the first randomiza- ≥pN2a (≥ 4 involved axillary lymph node metas- 68 | Annual Scientific Report 2016 | Recruiting Studies

tases) should receive regional nodal irradiation from the German study sites can also be en- and d) changes in the randomization 2: patients rolled directly in the randomization 2.

N=5,940 patients, Stratification: cN0/iN0, ≥18 years, iT1 or iT2 (≤5cm) • Age (<65 vs ≥65 years) planned BCS and postoperative irradiation • Tumor grading (G1/2 vs G3) • Tumor size (≤2 vs >2cm)

R 1:4

no SLNB SLNB (n≈1,188) (n≈4,752)

SLN negative SLN positive SLN positive or pN1mi (1-3 macrometastases) (≥4 macrometastases, (≈70%) (≈25%) ALND) (≈5%)

cN0 clinically nodal negative R 1:1 iN0 imaging nodal negative (if cN0/iN+: fine needle SLNB alone Completion aspiration or core biopsy (n≈984) ALND (n≈984) should be done) iT1/2 imaging tumor size BCS breast-conserving surgery SLNB sentinel lymph node biopsy plus 800 patients with enrolment directly into 2nd ALND axillary lymph node dissection randomisation (German or Austrian study sites)

Figure 1: INSEMA study design after amendment 4

Study report INSEMA recruitment started in September 2015 and is planned for 48 months. As of 31st Decem- ber 2016, a total of 1,768 patients from 137 centres have been enrolled in the first randomization and 144 patients in the second randomization (Figure 2). After a 5 year follow-up, final analysis is planned for 2024. | 69

150 accrual no. rando 1 137 137 accrual no. rando 2 136 140

5500 132 activated sites 130 COLLABORATING planned no. rando 1 124 130 5000 planned no. rando 2 118 STUDY GROUPS:

114 114 120 4500 106 110 101

4000 97 100 93 89 3500 90 80 3000

64 70 2500

54 60

2000 45 45 50

1500 40 30 1000 20 500 10

0 0 Sep-15 Mar-16 Sep-16 Mar-17 Sep-17 Mar-18 Sep-18 Mar-19

Figure 2: INSEMA recruitment as of 31st December 2016

Publication 1. Reimer T, von Minckwitz G, Loibl S et al., Comparison of axillary sentinel lymph node biopsy versus no axillary surgery in patients with early-stage invasive breast cancer and breast-conserving surgery: a randomized pro- SPONSOR: spective surgical trial. The Intergroup-Senti- University of Rostock nel-Mamma (INSEMA)-Trial; Poster OT2-04- 02, SABCS 2016 STUDY CHAIR: Prof. Dr. Toralf Reimer We are thanking all participating centers for their commitment and efforts so far. We would kindly Universitätsfrauenklinik like to encourage all sites to continue to support the INSEMA study by recruitment of patients and und Poliklinik am Klinikum the timely provision of biomaterial. Südstadt Rostock 70 | Annual Scientific Report 2016 | Studies in Follow-up | 71

Studies in Follow-up

GBG 34: IBIS-II 72 GBG 53: PANTHER 74 GBG 59: SOLE 76 GBG 67: APHINITY 78 GBG 69: GeparSepto 80 GBG 77: Katherine 83 GBG 81: Brightness 85 GBG 84: GeparOcto 87 72 | Annual Scientific Report 2016 | Studies in Follow-up

GBG 34: IBIS-II

International Breast Cancer Invention Study-II New Engl J Med 1989) and its very low toxicity makes it an attractive agent to consider in a NCT 00072462 (DCIS) preventive setting. NCT 00324714 (prevention) A number of trials have previously been conducted to examine the role of tamoxifen in IBIS-II is an international multicenter, ran prevention and a reduction in incidence of breast CONTACT: domized phase III trial investigating chemo cancer could be shown (King et al. JAMA 2001, Dr. Ioannis Gkantiragas preventive strategies in women with ductal Cuzick et al. Lancet 2002). However, due to the Clinical Project Management carcinoma in situ (DCIS) or at increased risk of side effect profile which includes a small increase [email protected] breast cancer (prevention). A total of 6,844 in endometrial cancer, gynecological problems patients were enrolled, 2,980 in the DCIS part of and thromboembolic disease, other strategies the trial and 3,864 patients in the prevention need to be considered. part of the trial. The new third generation aromatase inhibitors, e.g. anastrozole have shown good efficacy in Background advanced breast cancer and have a very low The success of tamoxifen in preventing re toxicity profile (Buzdar Ann Oncol 2000; Jonat currences and reducing mortality is established European Journal of Cancer 1996; Nabholtz in breast cancer (EBCTCG. Lancet 1998). Its J Clin Oncol 2000). They offer an alternative ability to prevent or delay a substantial number approach to local control, prevention of of new contralateral tumors (Cuzick et al. Lancet recurrence and the prevention of primary 1985; Fornander et al. Lancet 1989; Fisher et al. breast cancers, which may be superior and/or

Anastrozole 1mg daily Women at high risk of R breast cancer Placebo

Total treatment duration 5 years, followed by 5 years follow-up

Figure 1: IBIS-II Prevention study design

Anastrozole 1mg daily + (Tamoxifen) Placebo Women with DCIS R Tamoxifen 20mg daily + (Anastrozole) Placebo

Total treatment duration 5 years, followed by 5 years follow-up

Figure 2: IBIS-II DCIS study design | 73

complimentary to the use of SERMs (Selective For the DCIS part of the study, 2,980 post Estrogen Receptor Modulators). menopausal women from 236 centres in 14 coun The IBIS-II trial investigates the use of an tries have been enrolled between 2003 and astrozole in two different settings: in disease 2012 and randomly assigned to receive anas- prevention and in patients with ductal carcinoma trozole (1,449 analysed) or tamoxifen (1,489 in situ who have a high recurrence risk. analysed). After a median follow-up of 7.2 years and 144 breast cancer recurrences, there was no Study design and objectives statistically significant difference in overall The prevention part of IBIS-II aimed to determine recurrence (67 recurrences for anastrozole vs 77 if anastrozole was effective in preventing breast for tamoxifen; hazard ration 0.89 [95 % CI 0.64- cancer in postmenopausal women at increased 1.23]). The non-inferiority of anastrozole was risk of the disease. The DCIS part of the study, in- established (upper 95 % CI < 1.25), but its vestigated if anastrozole was at least as effective superiority to tamoxifen was not (p=0.49). There as tamoxifen in local control and prevention were a total of 69 deaths (33 for anastrozole of contralateral disease in women with locally vs 36 for tamoxifen (hazard ration 0.93 [95 % CI excised hormone-receptor-positive ductal car 0.58-1.50], p=0.78) with no specific cause being cinoma in situ. more common in one group than the other. The Moreover, effectiveness according to hormonal number of women reporting any adverse event receptor status, breast recurrence, new con was similar between anastrozole (1,323 women, tralateral tumor, mortality, side effect profile, 91 %) and tamoxifen (1,379 women, 93 %); the tolerability and acceptability were investigated. side-effect profiles of the two drugs differed, Particular attention has been paid on cardio- with more fractures, musculoskeletal events, vascular disease, fracture rates, and non-breast hypercholesterolaemia, and strokes with anas cancer deaths. Quality of life, cognitive function, trozole and more muscle spasm, gynaecological bone density and other disease markers were cancers and symptoms, vasomotor symptoms, also studied. and deep vein thromboses with tamoxifen. In conclusion, no clear efficacy differences were Study report seen between the two treatments. Anastrozole Between 2003 and 2012, 1,920 postmenopausal offers another treatment option for post women at increased risk of breast cancer have menopausal women with hormone-receptor- been recruited from 18 countries for the positive DCIS, which may be more appropriate prevention part of IBIS-II. After a median follow- for some women with contraindications for COLLABORATING up of 5 years, 40 women in the anastrozole tamoxifen. Longer follow-up will be necessary to STUDY GROUPS: group (2 %) and 85 in the placebo group (4 %) fully evaluate treatment differences [2]. had developed breast cancer (hazard ration 0.47 [95 % CI 0.32-0.68], p<0.001). The predicted Publications cumulative incidence of all breast cancers after 7 1. Cuzick J, Sestak I, Forbes JF, et al. Anastrozole years was 5.6 % in the placebo group and 2.8 % for prevention of breast cancer in high-risk in the anastrozole group. About 18 deaths were postmenopausal women (IBIS-II): an inter- reported in the anastrozole group and 17 in the national, double-blind, randomized placebo- placebo group, and no specific causes were more controlled trial. Lancet. 2014; 383(9922): common in one group than the other (p=0.836). 1041-8.8. Anastrozole effectively reduced incidence of 2. Forbes JF, Sestak I, Howell A, et al. Anastrozole breast cancer in high-risk postmenopausal versus tamoxifen for the prevention of loco SPONSOR: women. This finding, along with the fact that regional and contralateral breast cancer in Cancer Research UK most of the side-effects associated with postmenopausal women with locally excised London, UK estrogen deprivation were not attributable to ductal carcinoma in situ (IBIS-II DCIS): a the treatment, provides support for the use of double-blind, randomised controlled trial. STUDY CHAIRS: anastrozole in postmenopausal women at high Lancet. 2016; 387(10021):866-73. Prof. Dr. Wolfgang Eiermann risk of breast cancer [1]. Frauenklinik, München (DCIS) Prof. Dr. Gunter von Minckwitz Sites are highly encouraged to invite patients to join the patient self-reported outcome registry so German Breast Group that long term relapse and survival data can be obtained in a timely manner. (Prevention) 74 | Annual Scientific Report 2016 | Studies in Follow-up

GBG 53: PANTHER

A randomized phase III study comparing Study design and objectives biweekly and tailored epirubicin plus cyclo After surgery, patients were randomized to phosphamide followed by biweekly tailored receive either dose-dense, dose-adapted epi docetaxel versus three weekly epirubicin plus rubicin/cyclophosphamide for 14 days followed cyclophosphamide, 5-fluorouracil followed by by 14 days of dose-dense, dose-adapted doce docetaxel in lymph node positive or high risk taxel or 5-fluorouracil, epirubicin and cyclo CONTACT: lymph node negative breast cancer patients phosphamide for 21 days followed by 21days of Konstantin Reißmüller docetaxel (Figure 1). Clinical Project Management NCT 00798070 PANTHER primarily aimed to compare breast [email protected] cancer relapse-free survival (BCRFS) between PANTHER is an adjuvant, open-label, pro treatment arms. A total of 225 events (overall spective, randomized phase III trial that has in the two arms) were therefore needed to recruited 2,017 patients, including 772 from detect a five-year breast cancer relapse-free Germany. survival difference of 0.710 to 0.790. Secondary objectives of the study are to compare distant Background disease free survival, event-free survival overall In the adjuvant setting, a number of trials have survival, health related quality of life and toxicity demonstrated that the addition of a taxane to and outcome in relation to tumor biological chemotherapy regimen leads to a survival gain factors and polymorphism patterns between the (Henderson et al. J Clin Oncol 2003; Martin et al. two treatment arms. Moreover, dose intensity N Engl J Med 2005; Roché et al. J Clin Oncol. and side effects were analyzed. 2006), however, relapse risk remains significant especially in lymph-node positive disease. Study report Therefore, various dose and schedule strategies The PANTHER trial recruited a total of 2017 pa- have been investigated. Dose-dense and se tients between February 2007 and September quential designs have been shown to improve 2011 from 86 sites in Sweden, Germany and clinical outcomes (Citron J Clin Oncol 2003; Austria. Estimated study completion date is Jan- Möbus et al. J Clin Oncol 2010). Long term uary 2022. The final analysis of the primary end- results confirmed that the sequential use of do point included 2000 patients (1001 in the tai- cetaxel compared with concurrent doxorubicin- lored dose-dense group and 999 in the control docetaxel resulted in a better disease free group). After a median follow-up time of 5.3 survival and significantly better overall survival years (interquartile range (IQR), 4.5-6.1 years), (Oakman et al. Ann Oncol 2013). The dose- 269 BCRFS events were reported, 118 in the tai- dense and tailored epirubicin/cyclophosphamide lored dose-dense group and 151 in the control followed by docetaxel regimen was found to group (HR 0.79 [95 % CI, 0.61-1.01]; log-rank produce manageable feasibility, with an ac p = 0.06; 5-year BCRFS, 88.7 % vs 85.0 %). Thus, ceptable incidence of grade 4 infection/febrile the use of tailored dose-dense chemotherapy neutropenia (Margolin et al. Acta Oncol 2011). compared with standard adjuvant chemo All patients received granulocyte colony therapy did not result in a statistically signifi- stimulating factor (G-CSF) support and prophy cant improvement in BCRFS. The EFS was signifi lactic ciprofloxacin. G-CSF has the capacity to cantly better in tailored dose-dense group than reduce the depth and duration of granu in the control group (HR 0.79 [95 % CI, 0.63- locytopenia and to reduce the risk of granu 0.99]; p = 0.04; 5-year EFS, 86.7 % vs 82.1 %). locytopenic fever and infection and its use is The groups did not differ in OS (HR 0.77 [95 % mandatory in dose-dense regimen. CI, 0.57-1.05]; p = 0.09; 5-year OS, 92.1 % vs PANTHER aimed to investigate efficacy of a dose- 90.2 %) or DDFS (HR, 0.83 [95 % CI, 0.64-1.08]; dense and dose-adapted sequence of epirubicin/ p = 0.17; 5-year DDFS, 89.4 % vs 86.7 %). High- cyclophosphamide and docetaxel in the adjuvant grade non-hematological toxicities were more setting in a cohort of lymph node positive or frequent in the tailored dose-dense group high-risk lymph node negative patients (52.6 %) as compared to control group (36.6 %) compared to a standard anthracycline/taxane [1]. containing regimen. | 75

dt EC q14 dt T q14

+1 week Arm A

Individual adaptation with starting doses of E 90mg/m2, q2w, C 600mg/m2, q2w E: epirubicin 2 T 75mg/m , q2w* C: cyclophosphamide R T: docetaxel F: 5-fluorouracil SURGERY SURGERY FEC q21 T q21 COLLABORATING STUDY GROUPS: Arm B

F 500 mg/m² E 100 mg/m² T 100mg/m² C 500 mg/m²

Prophylaxis: Ciprofloxacin days 5-12 after each EC; G-CSF during ECàT chemotherapy * if toxicity ≤grade 2 docetaxel can be escalated to 100mg/m2

Figure 1: PANTHER study design

Publication 1. Foukakis T, von Minckwitz G, Bengtsson NO, et al. Effect of Tailored Dose-Dense Chemo- therapy vs Standard 3-Weekly Adjuvant Chemotherapy on Recurrence-Free Survival Among Women With High-Risk Early Breast Cancer: A Randomized Clinical Trial. JAMA SPONSOR: 2016; 316(18):1888-1896. Karolinska University Hospital Department of Oncology We would like to thank all participating sites for their efforts so far. Sites are highly encouraged to Stockholm, Sweden transfer their patients to the self-reported outcome registry so that long term relapse and survival data can be obtained in a timely manner. INTERNATIONAL STUDY CHAIRS: Prof. Dr. Jonas Bergh Scandinavian Breast Group Prof. Dr. Michael Gnant Austrian Breast and Colorectal Cancer Study Group

STUDY CHAIRS GERMANY: Prof. Dr. Volker Möbus AGO-B Prof. Dr. Gunter von Minckwitz German Breast Group 76 | Annual Scientific Report 2016 | Studies in Follow-up

GBG 59: SOLE

A phase III trial evaluating the role of continuous Questions remain about the optimal duration letrozole versus intermittent letrozole follow and best schedule of AIs in the extended ing 4 to 6 years of prior adjuvant endocrine adjuvant setting. The SOLE trial tests the hy- therapy for postmenopausal women with pothesis that introducing 3-months treatment- hormone-receptor positive, node positive early free intervals during the course of five years of CONTACT: stage breast cancer extended adjuvant letrozole will improve Dr. Ioannis Gkantiragas disease-free survival. This hypothesis is based on Clinical Project Management NCT 00553410 the theoretical principle that letrozole [email protected] withdrawal for 3 months will permit some SOLE is a multicenter, non-blinded, randomized, estrogenic stimulation which makes residual international phase III trial that has recruited a resistant disease susceptible to letrozole re- total of 4,884 patients internationally, including introduction (Lewis et al. J Steroid Biochem Mol 291 patients from Germany. Biol 2005). Moreover clinical observations also indicate an antitumor activity of estradiol, Background supporting the role for intermittent treatment In 2006, the standard duration of adjuvant with AIs (Lonning et al. Breast Cancer Res Treat endocrine therapy for breast cancer with either 2001). The introduction of a regimen that selective estrogen-receptor modulators (SERMs) decreases the yearly amount of letrozole may or aromatase inhibitors (AIs) was five years. Pa improve the cost-effectiveness of extended tients who received extended adjuvant letrozole letrozole administration. for five years following approximately five years of tamoxifen demonstrated a reduction in the Study design and objectives risk of locoregional and distant breast cancer The aim of SOLE is to compare continuous recurrence, compared with five years of ta letrozole for five years with intermittent letro moxifen alone (Goss et al. N Engl J Med 2003; zole over a five year period for postmenopausal Goss et al. J Natl Cancer Inst 2005). Similarly, a women who are disease-free following 4-6 years benefit has been demonstrated for switching of prior adjuvant endocrine therapy with from tamoxifen to an AI after 2 to 3 years of ta SERM(s) and/or AI(s) for endocrine-responsive, moxifen in order to complete five years of endo node-positive, operable breast cancer. The crine therapy (Coombes et al. N Engl J Med 2004; primary endpoint is disease free survival. Jakesz et al. Lancet 2005; Boccardo et al. J Clin Secondary endpoints are overall survival, distant Oncol 2005), as well as initiating therapy with AI disease-free survival, breast cancer free interval, following surgery and administering the AI for sites of first failure, second (non-breast) ma five years (ATAC Trialists´group. Lancet 2005; lignancies, deaths without prior cancer events, BIG 1-98 Collaborative group. N Engl J Med 2005). and adverse events.

Continuous letrozole for 5 years

Stratification: A • institution • prior adjuvant endocrine therapy (AI, SERM): AI alone, SERM alone, Intermittent letrozole over 5years both SERM and AI B 9 mths 9 mths 9 mths 9 mths 12 mths Randomisation

0 6 12 18 24 30 36 42 48 54 60

A: continuous letrozole, 2.5mg daily for 5 years B: intermittent letrozole, 2.5mg daily for the first 9 months of years 1 through 4, followed by 12 months in year 5

Figure 1: SOLE study design | 77

Substudy QoL domains were observed. An extension of COLLABORATING Patient reported symptoms and Quality of treatment implies a continuation of symptoms, STUDY GROUPS: Life (QoL) substudy which may be a burden to the patient. Patient- Information about concomitant and late reported symptoms will therefore be evaluated symptoms of adjuvant endocrine therapies for in a sub-sample of patients participating in the breast cancer is crucial to prepare women for SOLE trial by comparing the occurrence and physical and psychological consequences of severity of symptoms between the two different treatment and for making informed decisions. administration schedules, and their relative Several randomized trials in the adjuvant setting impact on global QoL indicators. reported a worsening of symptoms related to endocrine therapy (Fallowfield et al. J Clin Oncol Study report 2004; Cella et al. Breast Cancer Res Treat 2006; Recruitment of patients has started (first patient Fallowfield et al. J Clin Oncol 2006; Whelan et in) in July 2009. A total of 4,884 patients have al. J Clin Oncol, 2005). Although these studies been recruited internationally (including 291 revealed no major effect on overall QoL by the patients from Germany) within approximately different endocrine agents, differences in some 3 years. Follow-up is still ongoing.

SPONSOR: We are thanking all participating centers for their commitment and efforts so far. We would like International Breast Cancer to encourage all sites to continue to support the SOLE study by promptly providing data for the Study Group (IBCSG) patients in follow-up.

INTERNATIONAL STUDY CHAIR: Dr. Marco Colleoni European Institute of Oncology, Italy

COORDINATING INVESTIGATOR: Prof. Dr. Christian Jackisch SANA Clinic Offenbach, Germany 78 | Annual Scientific Report 2016 | Studies in Follow-up

GBG 67: APHINITY

A randomized multicenter, double-blind, HER2-positive breast cancer (Wong et al. Int J placebo-controlled comparison of chemo Breast Cancer 2012). Pertuzumab is a humanized therapy plus trastuzumab plus placebo versus monoclonal antibody that is designed to inhibit chemotherapy plus trastuzumab plus per HER2-dimerization and induce antibody- tuzumab as adjuvant therapy in patients with dependent cell-mediated cytotoxicity with a operable HER2-positive primary breast cancer complementary mechanism of action to trastu CONTACT: zumab. In HER2-positive advanced breast cancer Dr. Ioannis Gkantiragas NCT 01358877 the combination trastuzumab and pertuzumab Clinical Project Management was shown to be active in patients who previ ously [email protected] APHINITY is an adjuvant, prospective, two-arm, progressed on trastuzumab (Baselga et al. J Clin randomized, multicenter, international, double- Oncol 2010; Portera et al. Clin Cancer Res 2008). blind, placebo-controlled phase III trial that has In the neoadjuvant setting, trastuzumab and recruited 4810 patients, including 459 from pertuzumab in combination with chemotherapy Germany. nearly doubled the pathological complete response rate compared to either trastuzumab Background or pertuzumab administered in combination Approximately 20% of breast cancer patients with chemotherapy (45.8% vs 29% vs 24%, have HER2-positive tumors which are associated respectively) (Gianni et al. Lancet Oncol 2012). with a poorer prognosis if untreated (Slamon et Therefore, the potential for a more com al. Science 1987). Numerous studies have shown prehensive HER2 blockade with two anti-HER2 that adjuvant use of the anti-HER2 humanized monoclonal antibodies warrants further in monoclonal antibody trastuzumab improves vestigation in the adjuvant setting. disease-free and overall survival (Joensuu et al. J APHINITY aims to investigate safety and efficacy Clin Oncol 2009; Slamon et al. N Engl J Med of a combination therapy with two anti-HER2 2011; Gianni et al. Lancet Oncol 2011; Perez et al. agents (trastuzumab and pertuzumab) in ad J Clin Oncol 2011). However, not all patients dition to chemotherapy in the adjuvant setting, treated with this agent benefit from this therapy compared to chemotherapy and trastuzumab and resistance is a challenge in the treatment of alone.

R F A O N=4800 Chemotherapy* plus trastuzumab L N and pertuzumab D L S O U O W R Central M U G conﬁrma3on I of HER2 P E Z status 10 R A Y T Y I Chemotherapy* plus trastuzumab E O and placebo A R N S

Start treatment An3 HER2 therapy for a total of 1 year (52 weeks) within 8 weeks of surgery Start treatment *Inves5gator’s choice of 6-8 cycles of standard within chemotherapy (non-anthracycline or anthracycline based) 1 week Radiotherapy and/or endocrine therapy may be started of randomiza5on at the end of adjuvant chemotherapy 1 1

Figure 1: APHINITY study design | 79

Study design and objectives recurrence-free interval, cardiac and overall COLLABORATING After surgery, patients were randomized to safety as well as health related quality of life in STUDY GROUPS: receive either pertuzumab (840mg loading dose the two treatment arms. in cycle 1, followed by 420mg) or placebo in travenously every 3 weeks for one year, in addi Study report tion to 6-8 cycles of chemotherapy and 1 year of APHINITY randomized a total of 4,810 patients trastuzumab (8mg/kg loading dose in cycle 1, between November 2011 and August 2013 and followed by 6mg/kg) intravenously every 3 the last patient completed treatment in August weeks. Total duration of anti-HER2 treatment 2014. The study is now in follow-up with pa- was 52 weeks. Patients will be followed up for 10 tients being followed at approximately years. 3-monthly intervals for 2 years, then every APHINITY primarily aimed to compare invasive 6 months during years 3 to 5 and annually there- disease-free survival, excluding second primary after. The study will continue until 10 years after SPONSOR: non-breast cancers between treatment arms. the randomization of the last patient (around F. Hoffmann-La Roche Ltd. Secondary objectives of the study were invasive September 2023). Primary analysis for the inva- and Genentech, Inc. disease-free survival including second primary sive disease-free survival is planned for the non-breast cancers, disease-free survival, over- beginning of 2017. Due to the long follow-up pe- STUDY CHAIRS (International): all survival, recurrence-free interval, distant riod, results will be a long time coming. Prof. Dr. Gunter von Minckwitz German Breast Group Jose Baselga, MD, PhD We would like to thank all participating sites for their efforts so far. Sites are highly encouraged to Massachusetts General Hospital provide follow-up documentation of their patients in order to avoid delays in the study analysis. We Boston, USA would like to remind all participating sites to provide a tumor sample at disease recurrence in order to increase the robustness of data. José Bines, MD National Cancer Institute Rio de Janeiro, Brazil 80 | Annual Scientific Report 2016 | Studies in Follow-up

GBG 69: GeparSepto

A randomized phase III trial comparing phase II trial, weekly (3 weeks on, 1 week off) nanoparticle-based paclitaxel with solvent- nab-paclitaxel at doses of 150 mg/m² was more based paclitaxel as part of neoadjuvant effective in terms of response rate, even chemotherapy for patients with early breast including a faster response, and overall survival cancer than 100 mg/m² nab-paclitaxel or 3-weekly docetaxel 100 mg/m² in patients with metastatic CONTACT: NCT 01583426 breast cancer. Tolerability of 150 mg/m² weekly Konstantin Reißmüller nab-paclitaxel appeared acceptable (Gradishar J Clinical Project Management GeparSepto is a neoadjuvant, prospective, multi- Clin Oncol 2009). In the neoadjuvant setting, a [email protected] center, open-label, randomized phase III trial that phase II study including 66 patients that received has recruited 1,229 patients from 69 centers in nab-paclitaxel (100 mg/m² for 12weeks) followed Germany. by FEC (5-fluorouracil/epirubicin/cyclophos phamide) (+Trastuzumab for HER2-positive Background patients) showed that nab-paclitaxel was well Solvent-based taxanes (paclitaxel, docetaxel) tolerated and achieved a pCR rate of 29 % cause severe toxicities not only by the active (Robidoux et al. Clin Breast Cancer 2010). agents themselves but also by solvents like The Neo-tAnGo study showed a significantly cremophor. Nab-paclitaxel (Abraxane®) is a sol higher pathological complete response rate with vent-free formulation of paclitaxel encapsulated the reverse sequence of taxane followed by an in albumin. It does not require premedication anthracycline containing regimen (Earl et al. with corticosteroids or antihistamines to prevent Lancet Oncol 2014). the risk of solvent-mediated hypersensitivity Given this evidence, GeparSepto aimed to reactions. This new formulation improves the evaluate the efficacy and safety of nab-paclitaxel safety profile, allows higher dosing with shorter compared with solvent-based paclitaxel as part infusion duration, and produces higher tumor of neoadjuvant treatment of operable and drug concentration (Ibrahim et al. J Clin Oncol locally advanced breast cancer. 2005; Desai et al. Clin Cancer Res 2006). In the pivotal phase III trial, nab-paclitaxel at a Study design and objectives dose of 260 mg/m² was more effective than The GeparSepto study primarily aimed to com- solvent-based paclitaxel at a dose of 175 mg/m² pare pCR rates (ypT0 ypN0) between nab-Pac or (both given every 3 weeks), with a favorable solvent-based Pac. safety profile (Gradishar J Clin Oncol 2005). In a Secondary objectives were to assess pCR rates in

N = 60 N = 1200 (HER2 positive) Arm A If HR positive: tamoxifen, aromatase inhibitors R* 12 weeks 12 weeks R* Surgery Core biopsy* before study entry) Arm B (before study entry) Core biopsy* (after anti-HER2 treatment / If HER2 positive: trastuzumab acc. to AGO Guidelines

6 weeks Core biopsy Core biopsy *Centrally confirmed: optional optional - Subtypes HER2/HR - Ki67 - SPARC If HER2 positive:

Trastuzumab 8 mg/kg (loading dose) Paclitaxel nab-Paclitaxel Epirubicin 90 mg/m2 2 2 followed by 6 mg/kg Core biopsy 80 mg/m 125 mg/m 2 weekly weekly Cyclophosphamide 600 mg/m Pertuzumab (absolute dose per application) 840 mg (loading dose) * Randomizations carried out simultaneously followed by 420 mg

Figure 1: GeparSepto study design, including window substudy | 81

different pre-specified subpopulations and allows defining the risk of premature ovarian according to different definitions, to determine failure and loss of fertility within modern thera- the response rates of breast tumor and axillary peutic regimen. nodes, to assess compliance and toxicity with a special interest in neuropathy, and to determine Study report loco-regional invasive recurrence-free survival, The GeparSepto study aimed to improve the distant disease-free survival, invasive dis- efficacy as well as the safety profile by using ease-free survival, and overall survival in both solvent-free formulation of Pac encapsulated in arms. albumin (nab-Pac). With amendment 3 of the study protocol (10th After a safety analysis found higher rates of dose June 2015) the study has been re-opened in reductions and treatment discontinuations with order to collect information on quality of life nab-Pac compared to Pac, nab-Pac dose was re- with a special focus on persisting peripheral neu- duced from 150 to 125 mg/m² weekly (Jackisch ropathy and cardiac toxicity, as well as on smok- et al. J Clin Oncol 2013). ing habits and alcohol consumption before and The recruitment phase took 18 months and 1229 after treatment in order to compare these with patients were enrolled from 69 German centers. safety and efficacy of study treatment and to Final analysis on the primary endpoint pCR in- assess genetic changes in the tumor; BRCA sta- cluded 1,206 patients (606 treated with nab-Pac tus and other mutations. and 600 treated with Pac) and baseline characteristics were well balanced. pCR rate (ypT0 Integrated window ypN0) was 38.4 % in the nab-Pac compared to A window-of-opportunity study was integrated 29.0 % in the Pac arm (OR 1.53 [95 % CI 1.20- to investigate specific biomarkers in patients 1.95], p<0.001). The results were very homoge- with HER2-positive breast cancer treated with nous across subgroups, with TNBC patients de- different anti-HER2 therapies (trastuzumab, riving the highest benefit (OR 2.61 [95 % CI pertuzumab or combination) alone and compare 1.57-4.33], p<0.001) and an increase in pCR responders to non-responders. Clinical response from 26% to 48 %. In terms of safety 281 (23 %) after 6 weeks and biomarker profile were com- patients reported at least one SAE (126 in Pac pared between the groups and correlated to the and 155 in nab-Pac arm) and 4 patients died on response after the addition of 4 cycles tax- study [1]. Importantly nab-Pac 125 mg/m² was ane-based chemotherapy and at the end of the associated with a significantly better safety pro- main study. file and compliance without compromising the efficacy compared to nab-Pac 150 mg/m² [2]. Substudies Moreover, patients with PIK3CA mutant HER2- Phamacogenetic substudy positive breast cancer have a significantly lower The aim of this substudy is to perform genetic pCR rate compared to patients with wild-type associations, i.e. germline genotype of the pa- tumors irrespective of the hormone receptor tient with treatment response, long term prog- status. PIK3CA mutation in HER2-positive breast nosis or the molecular profile of the tumors. cancer predicts resistance to anti-HER2 therapy. In addition, PIK3CA mutations were found to Surgical substudy predict response to nab-Pac. [3]. This surgical substudy offers the opportunity to We were also able to demonstrate that the spare patients surgery, which always carries the HER2-positive cohort, receiving dual HER2 risk of complications. If it can be shown at an in- blockade with pertuzumab and trastuzumab, terim analysis that the positive predictive value had higher rates of pCR (ypT0 ypN0) than the for a pCR of (at least 3) negative core biopsies HER2-negative tumors (57.8 % vs. 22.0 %), with before surgery in patients with complete clinical the highest rate in the HER2-positive/ hormone response is >90 %, these patients might opt for receptor negative cohort. Furthermore, HER2- having no further breast surgery. positive patients experienced more high-grade toxicity (85.4 % vs. 78.0 %; p=0.003) than Ovarian Function substudy HER2-negative patients. A sensitivity analysis in The ovarian function substudy which will ana- HER2-positive cohort showed a pCR rate of lyze the combination with hormones such as 60.1 % with nab-Pac and 52.7 % with Pac FSH, E2, and anti-Mullerian Hormone in addi- (p=0.212). Although the HER2-positive patients tion to follicle count measured by ultrasound experienced more noteworthy toxicity, this 82 | Annual Scientific Report 2016 | Studies in Follow-up

treatment regime has acceptable toxicity and 2. Minckwitz G, Untch M, Jackisch C, et al. achieved higher rates of pCR in the HER2-posi- nab-Paclitaxel at a dose of 125 mg/m² weekly tive cohort [4]. is more efficacious but less toxic than at 150 In a follow-up analysis of peripheral sensory mg/m². Results from the neoadjuvant rand- neuropathy (PSN), nab-Pac 125 mg/m² was as- omized GeparSepto study (GBG 69). Cancer sociated with a lower occurrence of grade 3-4 Res 2016; 76 (4 Suppl) Abstract P1-14-11 PSN compared to nab-Pac 150 mg/m² (14.5 % vs. 3. Loibl S, Budczies J, Weichert W, et al. PIK3CA 8.1 %, p=0.019) but higher PSN than Pac 80 mg/m² mutations predict resistance to trastuzumab/ (2.7 %). The grade 2-4 PSN occurred with pertuzumab and nab-paclitaxel in primary nab-Pac 125 mg/m² was associated with a more HER2-positive breast cancer – massive paral- rapid resolution compared to nab-Pac 150 mg/m² lel sequencing analysis of 293 pretherapeutic (6.4 vs. 12.7 weeks, p=0.001). The follow-up core biopsies of the GeparSepto study. Can- of GeparSepto is ongoing and long term data cer Res 2016; 76 (4 Suppl) Abstract P3-07-03. analysis of PSN is planned. [5]. 4. Loibl S, Jackisch C, Schneeweiss A, et al. Dual HER2-blockade with pertuzumab and trastu- Publications zumab in HER2-positive early breast cancer: a 1. Untch M, Jackisch C, Schneeweiß A, et al. A subanalysis of data from the randomized randomized phase III trial comparing neoad- phase III GeparSepto trial. Ann Oncol. 2016; juvant chemotherapy with weekly nanoparti- [Epub ahead of print] cle-based paclitaxel with solvent-based pacl- 5. Furlanetto J, von Minckwitz G, Jackisch C, et itaxel followed by anthracyline cyclo- al. Peripheral Sensory Neuropathy Occur- phosphamide for patients with early breast rence and Resolution: Results from the neo- cancer (GeparSepto). Lancet Oncol. adjuvant randomized GeparSepto study 2016; 17(3):345-56 (GBG 69). Abstract P5-16-03, SABCS 2016

COLLABORATING STUDY GROUPS:

N % Death + no 1115 92 ∆ yes 91 8

SPONSOR: GBG Forschungs GmbH

CO-ORDINATING INVESTIGATOR: Prof. Dr. Michael Untch Date randomisation Berlin Figure 2: GeparSepto follow-up as of 31st December 2016 STUDY CO-CHAIRS: Prof. Dr. Christian Jackisch Offenbach Sites are highly encouraged to enter study participants in the General Follow-up or invite them to Prof. Dr. Gunter von Minckwitz join the patient self-reported outcome registry so that long term relapse and survival data can be Neu-Isenburg obtained. | 83

GBG 77: Katherine

A Study of Trastuzumab Emtansine Versus linker (Lewis Phillips et al. Cancer Res 2008). Trastuzumab as Adjuvant Therapy in Patients T-DM1 prolonged progression free survival with With HER2-Positive Breast Cancer Who Have a more favorable toxicity profile compared with Residual Tumor in the Breast or Axillary Lymph trastuzumab and docetaxel in a phase II study in Nodes Following Preoperative Therapy patients with previously untreated HER2- positive metastatic breast cancer (Hurvitz et al. J CONTACT: NCT 01772472 Clin Oncol 2013). Moreover, the phase III EMILIA Dr. Christian Mäurer trial has shown that T-DM1 prolonged pro Clinical Project Management Katherine is a randomized, multicenter, open- gression-free and overall survival with less tox- [email protected] label phase III study that has recruited 1,487 icity in patients with previously treated HER2- patients from 328 international sites (57 in positive metastatic breast cancer compared Germany) within 3 years. with capecitabine and lapatinib (Verma et al. N Engl J Med 2012). Background Given the results with T-DM1 in the metastatic HER2-positive patients who do not achieve setting, the Katherine trial aims to explore the pathologic complete response (pCR) following efficacy and safety of single-agent T-DM1 com neoadjuvant therapy have a worse prognosis pared with trastuzumab in patients with HER2- compared to patients who achieve a pCR (Untch positive primary breast cancer. et al. J Clin Oncol 2011; von Minckwitz et al. J Clin Oncol 2012). HER2-targeted neoadjuvant Study design and objectives therapy has shown to improve pCR rates in Katherine aimed to investigate whether adjuvant patients with HER2-positive (Gianni et al. Lancet T-DM1 was more effective than trastuzumab in 2010; von Minckwitz et al. Breast Cancer Res patients with HER2-positive primary breast Treat 2011) and a double HER2 blockade can cancer who have received neoadjuvant chemo lead to an even further increase (Gianni et al. therapy including trastuzumab and have residual Lancet Oncol 2012). invasive disease after surgery. Patients with residual disease after neoadjuvant The study primarily aimed to compare invasive chemotherapy and anti-HER2 treatment are at disease-free survival between the two treatment an increased risk of recurrence and mortality arms. Secondary endpoints included disease-free and it is unknown whether the application of survival, overall survival, distant recurrence-free additional agents following surgery may provide interval, quality of life, and pharmacokinetics. further benefit. Safety objectives included all adverse events, The antibody–drug conjugate trastuzumab abnormal laboratory values, cardiac events, and emtansine (T-DM1) combines the antitumor LVEF (left ventricular ejection fraction). Patient activities of trastuzumab with intracellular reported outcomes using different quality of life delivery of the cytotoxic agent DM1 via a stable questionnaires were also assessed.

N=1484 Trastuzumab emtansine *

Main inclusion criteria: 3.6 mg/kg given i.v. q3w • centrally confirmed HER2-positive for 14 cycles • Clinical stage T1-4 N0-3 M0 (T1a/ bN0 not eligible) • residual invasive disease after R preoperative chemotherapy: • ≥6 cycles with total duration ≥16 Trastuzumab * weeks • ≥9 weeks of trastuzumab and 6 mg/kg given i.v. q3w taxane-based chemotherapy for 14 cycles

* plus concomitant radiotherapy and/or hormonal therapy if clinically indicated

Figure 1: Katherine study design 84 | Annual Scientific Report 2016 | Studies in Follow-up

COLLABORATING Moreover, Katherine assessed the pharma Study report STUDY GROUPS: cokinetics of T-DM1 in T-DM1–treated patients Katherine randomized a total of 1,487 patients and trastuzumab in trastuzumab-treated pa (292 patients in Germany) between April 2013 tients, which permited an intra-study com and December 2015. The study is now in follow- parison of trastuzumab exposure in the 2 up. treatment arms. Exposure–effect (efficacy and The biomaterial collection was a crucial part of safety) relationships were also explored in this the study and has been led by the German Breast patient population. Group. The first interim analysis is planned after A range of translational research questions were 257 events have occurred which is estimated for also addressed within the Katherine study, such 2017. The final analysis for disease-free survival as correlations between biomarker status and will be performed after 384 events have efficacy and/or safety or the incidence of occurred. The final analysis for overall survival SPONSOR: antitherapeutic antibodies and their effect on will be performed at the end of 10 years of F. Hofmann-La Roche Ltd pharmacokinetics, safety, and efficacy. follow-up.

INTERNATIONAL We are thanking all participating centers for their commitment and efforts so far. We would like to STUDY CHAIR: encourage all sites to continue to support the Katherine study by promptly providing the required Prof. Dr. Charles Geyer biomaterial. NSABP

STUDY CHAIR GERMANY: Prof. Dr. Gunter von Minckwitz German Breast Group | 85

GBG 81: BRIGHTNESS

A Randomized, Placebo-Controlled, Double- cases. In patients with a positive family history, Blind, Phase 3 Study Evaluating Safety and up to 20 % harbor a germline mutation of the Efficacy of the Addition of Veliparib Plus BRCA 1/2 gene. Chemotherapy is still the most Carboplatin Versus the Addition of Carboplatin important treatment for TNBC irrespective of a to Standard Neoadjuvant Chemotherapy germline BRCA mutation. Recently, it could be Versus Standard Neoadjuvant Chemotherapy shown in the metastatic setting that carboplatin CONTACT: in Subjects with Early Stage Triple Negative monotherapy is as effective as docetaxel in Breast Cancer (TNBC) patients with TNBC, but it seems that in Carmen Schmidt-Rau particular women harboring a germline BRCA Clinical Project Management NCT 02032277 mutation benefit from the platinum agent as [email protected] first-line therapy (Tutt et al Cancer Res 2015). BRIGHTNESS is a multicenter, double-blind, Platinum-containing chemotherapy combina placebo-controlled, randomized phase III trial tions in patients with TNBC and germline BRCA that has globally recruited 624 patients (55 mutation have been shown to be very effective patients in Germany) from 250 sites (34 in in achieving high pathological complete re Germany) in 18 countries within approximately sponse (pCR) rates (von Minckwitz et al. Lancet 22 months. Oncol 2014; Sikov et al J Clin Oncol 2015). The success of the PARP inhibitors in platinum Background sensitive ovarian cancer (Liu et al. Lancet Oncol. Patients with triple-negative breast cancer 2014; Oza et al. Lancet Oncol. 2015) and the (TNBC) represent about 15 % of all breast cancer data in metastatic breast cancer using a PARP

Veliparib

N=624 Placebo Post Surgery R* Follow-up 0 2:1:1 Surgery

Placebo Placebo

12-16 weeks 8-12 weeks 10 years

Veliparib/Placebo Paclitaxel Carboplatin/Placebo Doxorubicin Cylcophosphamide 50 mg BID/daily 80 mg/m2 q1w AUC 6 q3w 60 mg/m2 q2w or q3w 600 mg/m2 q2w or q3w *Stratification by • BRCA status (BRCA1 and/or BRCA2 mutation vs no BRCA mutation vs unknown) • lymph node stage (N0 vs N1-2) • planned schedule of doxorubicin/cyclophosphamide (q2w vs q3w)

Figure 1: BRIGHTNESS study design 86 | Annual Scientific Report 2016 | Studies in Follow-up

inhibitor (Kaufman et al. J Clin Oncol 2015) servation after therapy. Further objectives were support the idea to test these agents in primary to assess event free survival (EFS), overall breast cancer. survival (OS), clinical response rate (CRR) at 12 The neoadjuvant BRIGHTNESS study compares weeks, the incidence of pCR plus minimal paclitaxel plus carboplatin plus the PARP residual disease (defined as residual cancer inhibitor veliparib with paclitaxel plus car burden [RCB] class I), ECOG performance status, boplatin and with paclitaxel alone, each followed and breast cancer related quality of life. by standard neoadjuvant chemotherapy with doxorubicin/cyclophosphamide. TNBC patients Study report are included irrespective of their germline BRCA BRIGHTNESS has randomized globally a total of status, but this information is used to stratify 624 patients and 55 patients in Germany patients prior to randomization. between February 2015 and March 2016. To ensure patient safety, an Independent Data Study design and objectives Monitoring Committee (IDMC) has reviewed BRIGHTNESS primarily aimed to assess the in unblinded safety data. The final interim analysis cidence of pCR in breast and ipsilateral axillary has been finished in QIV/2016 and the first tissue of veliparib + carboplatin + paclitaxel results will be presented at the American Society compared with the two placebo-containing of Clinical Oncology (ASCO), June 2-6, 2017, arms. The secondary objective of the study was Chicago, Illinois, USA. Post-surgery follow-up to assess the rate of eligibility for breast con period of 10 years is now ongoing.

COLLABORATING STUDY GROUPS:

120 accrual no. 110 planned no.

100

SPONSOR: 70

40 INTERNATIONAL STUDY CHAIRS: 30 Prof. Dr. Sibylle Loibl 20 German Breast Group 10 Prof. Dr. Charles Geyer 0 Jan-15 Mar-15 May-15 Jul-15 Sep-15 Nov-15 Jan-16 Mar-16 May-16 University Richmond, USA

Figure 2: BRIGHTNESS final recruitment in Germany CO-ORDINATING INVESTIGATOR: Prof. Dr. Jens Huober We are thanking all participating centers for their commitment and efforts so far. We would like to Universitätsfrauenklinik encourage all sites to continue to support the BRIGHTNESS study by entering study participants in Ulm, Germany the General Follow-up or invite them to join the patient self-reported registry. | 87

GBG 84: GeparOcto

A randomized phase III trial comparing two al. J Clin Oncol 2010; Untch et al. Lancet Oncol dose-dense, dose-intensified approaches 2012; Gianni et al. Lancet 2010). Pertuzumab in (ETC and PM(Cb)) for neoadjuvant treatment combination with trastuzumab has shown of patients with high-risk early breast cancer impressive activity in combination with (GeparOcto) docetaxel and/or carboplatin as neoadjuvant treatment in the NeoSphere study (Gianni et al. CONTACT: NCT 02125344 Lancet Oncol 2012) and in the Tryphaena study Konstantin Reißmüller (Schneeweiss et al. Ann Oncol 2013). Clinical Project Management GeparOcto is a multicenter, prospective, ran In addition the supportive treatment of [email protected] domized open-label phase III study that has chemotherapy-induced iron deficiency anemia recruited 961 patients from 57 sites in Germany was investigated. Iron substitution is currently within 18 months. Moreover, a total of 123 pa- mostly given as an oral supplement in the daily tients were randomized for a substudy on clinical practice. However, parenteral iron supportive anemia treatment. substitution is assumed to be more efficient in adjusting iron hemostasis and hemoglobin, as Background oral preparations are less efficiently absorbed Two regimen are currently considered to be and more frequently cause gastro-intestinal among the treatments with the highest efficacy adverse events, leading to non-compliance. The in patients with high-risk early stage breast diagnosis and treatment of iron deficiency is at cancer: sequential treatment of high dose present not integrated in the routine medical epirubicin, taxane, and cyclophosphamide (ETC) care of chronic disease, although iron deficiency concomitantly with or without a dual HER2- is a frequent comorbidity in cancer patients and blockade mainly based on the AGO ETC adjuvant the understanding of iron physiology and study (Moebus et al. J Clin Oncol 2010), and pathology have recently gained major insights. weekly treatment with paclitaxel/non-pegylated The neoadjuvant GeparOcto study - compared a liposomal doxorubicin with dual HER2-blockade sequential, dose-dense, dose-intensified ETC or carboplatin (PM(Cb)) based on the GeparSixto (epirubicin, paclitaxel, cyclophosphamide) study (von Minckwitz et al. Lancet Oncol 2014). treatment vs. weekly PM (Cb) (paclitaxel, The aim of the GeparOcto study is to compare liposomal doxorubicin (carboplatin)) treatment those two regimens. Moreover, patients with in patients with high-risk operable or locally HER2 positive breast cancer have received anti- advanced breast cancer with the addition of HER2 treatment with trastuzumab and trastuzumab and pertuzumab in HER2 positive pertuzumab. In clinical trials, preoperative patients. Moreover, the use of parenteral ferric trastuzumab leads to increased pathological carboxymaltose versus physician’s choice for the complete response (pCR) rates in the range of treatment of chemotherapy-induced anemia in 39-62 % (Untch et al. J Clin Oncol 2011; Untch et patients with iron deficiency was compared.

PM(Cb) Paclitaxel 80 mg/m2 q1w or 225 mg/m2 q2w

Epirubicin 150 mg/m2 q2w N=950 TNBC NPLD 20 mg/m2 q1w

or R Cyclophosphamide 2g/m2 q2w HER2 pos

Surgery Trastuzumab (8), 6 mg/kg q3w or HER2 pos: high risk Pertuzumab (840), 420 mg q3w

HR+/HER2- TNBC: Carboplatin AUC 1.5 q1w

iddEnPC ferric carboxymaltose

physician’s choice of anemia treatment anemia grade ≥2 R

Figure 1: GeparOcto study design 88 | Annual Scientific Report 2016 | Studies in Follow-up

Study design and objectives receiving supportive treatment for iron defi GeparOcto primarily aimed to compare the pCR ciency with parenteral ferric carboxymaltose (ypT0/is ypN0) rates between the two treatment versus physician’s choice (no supportive treat arms. The secondary objective of the study was ment, oral iron substitution, erythropoiesis- to assess the pCR rates per arm separately for stimulating agent, or both). the stratified subpopulations. Further objectives were to determine pCR according to other Substudies definitions, the response rates of the breast GeparPET Substudy tumor and axillary nodes based on physical Aim of the study is to demonstrate that PET-CT examination and imaging tests, the breast before surgery in addition to conventional conservation rate, toxicity and compliance, presurgical staging methods can decrease loco-regional invasive recurrence free survival the mastectomy rate in patients receiving (LRRFS), distant-disease-free survival (DDFS), neoadjuvant chemotherapy for breast cancer. invasive disease-free survival (IDFS), and overall survival (OS) in both arms and according to Pharmacogenetic Substudy stratified subpopulations, regional recurrence Aim of this study is to analyze potential free survival (RRFS) in patients with initial node- associations between the germline genotype of positive axilla converted to negative at surgery the patient, treatment response, toxicities, long (ypN0) and treated with sentinel node biopsy term prognosis and molecular profile of the alone, pCR rate and local recurrence free survival tumors. (LRFS) in patients with a clinical complete response and a negative core biopsy before Ovarian function substudy surgery and to correlate response (complete vs. It has been previously shown that women with partial vs. no change) measured by the best amenorrhea, a surrogate for ovarian function, appropriate imaging method after 6 weeks of had an improved survival. Aim of this study is to treatment with pCR. evaluate the rate of premature ovarian failure 2 For those patients randomized for the supportive years after the end of chemotherapy in women anemia treatment the primary objective was to ≤ 45 years. compare the frequency of patients reaching hemoglobin (Hb) levels ≥ 11 g/dl 6 weeks after Study report treatment start of a first episode of anemia GeparOcto randomized a total of 961 patients grade ≥ 2 (Hb < 10 g/dl) between patients between December 2014 and June 2016 and of

Top recruiters G8: accrual no. 900 University Hospital Heidelberg (N=71) Medical practice at Bethanien KH FFM (N=50) anemia question accrual no. Henriettenstiftung Hannover (N=46) planned no. 800 Medical practice Kittel/Klare, Berlin (N=42) UFK Dresden (N=41) anemia question planned no. 700

600

500

400

300

200

100

0 Dec-14 Feb-15 Apr-15 Jun-15 Aug-15 Oct-15 Dec-15 Feb-16 Apr-16 Jun-16

1 Figure 2: GeparOcto final recruitment | 89

those 123 patients were enrolled in the anemia COLLABORATING treatment substudy. Final data cleaning will be STUDY GROUPS: finished in QII/2017. The first results will be presented at the American Society of Clinical Oncology (ASCO), June 2-6, 2017, Chicago, Illinois, USA.

We are thanking all participating centers for their commitment and efforts so far. We would like to encourage all sites to continue to support the GeparOcto substudies by providing biomaterials in a timely manner.

SPONSOR: GBG Forschungs GmbH

STUDY CHAIR: Prof. Dr. Andreas Schneeweiss

STUDY CO-CHAIRS: Prof. Dr. Gunter von Minckwitz (TNBC) Prof. Dr. Volker Möbus (HER2-pos) 90 | Annual Scientific Report 2016 | Completed Studies | 91

Completed Studies

GBG 74: GENEVIEVE 92

GBG 80: SenSzi 94 92 | Annual Scientific Report 2016 | Completed Studies

GBG 74: GENEVIEVE

Randomized, open-label, phase II study bules against cold-induced depolymerization as comparing the efficacy and the safety of ca efficiently as docetaxel and was selected for bazitaxel versus weekly paclitaxel given as neo- development based on a better antiproliferative adjuvant treatment in patients with operable activity on resistant cell lines than docetaxel. A Triple Negative or luminal B/HER2 normal phase 2 study in patients with taxane- and/or Breast Cancer anthracycline-resistant metastatic breast cancer CONTACT: demonstrated that cabazitaxel was active and Dr. Gemma Bruno NCT 01779479 well tolerated (Pivot et al. Ann Oncol 2008). Clinical Project Management These clinical data support to assess cabazitaxel [email protected] GENEVIEVE is a neoadjuvant, prospective, multi versus an established taxane such as paclitaxel center, open-label, randomized phase II trial that in breast cancer using the dose registered for has recruited 333 patients from 44 sites in metastatic castration-resistant prostate cancer. Germany. In the GENEVIEVE study, cabazitaxel has been compared against weekly paclitaxel which is the Background currently most widely used treatment for breast Paclitaxel is amongst the most active agents cancer patients. A head-to-head comparison in in the treatment of metastatic breast cancer, the neoadjuvant setting was sought to allow a with response rates ranging from 30 to 60% rapid and precise comparison of efficacy and when used as a single agent. Moreover, in the tolerability of cabazitaxel versus paclitaxel to neoadjuvant setting weekly paclitaxel given decide whether further development of this before an anthracycline-based regimen in taxoid in breast cancer is reasonable. patients with resectable tumors induced a GENEVIEVE primarily aimed to compare the significantly higher pCR rate (Green et al. J Clin pathological complete response (pCR, ypT0/is Oncol 2005). ypN0/+) in patients with operable HER2- Cabazitaxel is a new taxoid which promotes the negative (triple negative or luminal B/HER2-) tubulin assembly in vitro and stabilizes microtu primary breast cancer treated with either

Immediately before surgery: • Presurgical core biopsy (op=onal) Cabazitaxel 25mg/m², q3w yes cCR in R breast and nodes If no pCR and no EC before: no e.g. 4xEC Surgery neg Paclitaxel 80mg/m², q1w Core Biopsy e.g. pos 4xEC

before Rando: a7er 2nd cycle: EOT all paAents: At surgery if pCR/cCR • FFPE =ssue block • FFPE =ssue block (op=onal) • 10 ml blood (serum) (if pCR: To be taken at EOT or surgery): • RNAlater conserved =ssue • 10 ml blood (serum) • 10 ml blood (plasma) • FFPE =ssue block (op=onal) • 10 ml blood (plasma) • RNAlater conserved =ssue (op=onal) • Fresh frozen =ssue (op=onal) and if no pCR: • Fresh frozen =ssue (op=onal) • 10 ml blood (serum) • FFPE =ssue block • 10 ml blood (plasma) • RNAlater conserved =ssue • 10ml blood (SNP analysis) (op=onal) (op=onal) • Fresh frozen =ssue (op=onal)

Figure 1: GENEVIEVE Study Design (Amendment 2) | 93

cabazitaxel or weekly paclitaxel. In addition, or luminal B/HER2−, primary breast cancer. pCR according to other definition and in High-grade toxicity (hematological and non- stratified subgroups, objective response rate, hematological) was significantly more common pCR and local recurrence free survival in patients in the cabazitaxel arm as compared to the with a clinical complete response (cCR) and paclitaxel arm, while the drug exposure and negative core biopsy before surgery, breast patient compliance did not differ between the conservation rate, toxicity, compliance, invasive two arms [1]. Concerning the pharmacogenetic locoregional recurrence-free, distant-disease- and the ovarian function substudies, data free, invasive disease-free and overall survival collection and evaluation are still ongoing and were compared between the two treatment therefore the results are not available yet. Data arms. analysis for the surgical substudy is also still Patients without clinical complete response outstanding. could undergo a core biopsy to demonstrate that a pCR has not been obtained. Patients Publications without response could then continue with an 1. Paepke S, Huober J, Kümmel S, et al. anthracycline based chemotherapy (Figure 1). Randomized, open-label, phase II study GENEVIEVE also offered the opportunity to comparing the efficacy and the safety of conduct translational research in order to cabazitaxel versus weekly paclitaxel given as explore the existence of biomarkers and profiles neoadjuvant treatment in patients with potentially predicting response to treatment. operable triple negative or luminal B/HER2 normal breast cancer (GENEVIEVE). Eur J Study report Cancer 2016; 57 (suppl.2):S99 (Poster 350) The study has recruited a total of 333 patients between March 2013 and June 2015 from 44 German sites. The first study results showed no short-term effect of cabazitaxel in triple negative COLLABORATING STUDY GROUPS: We are thanking all participating centers for their commitment and efforts.

SPONSOR: GBG Forschungs GmbH

STUDY CHAIR: Prof. Dr. Gunter von Minckwitz German Breast Group

STUDY CO-CHAIRS: Dr. Stefan Paepke TU München PD Dr. Sherko Kümmel Kliniken Essen-Mitte 94 | Annual Scientific Report 2016 | Completed Studies

GBG 80: SenSzi SenSzi

Randomized, controlled, multi-center, operative third of the patients no sentinel lymph nodes study to evaluate the role of a pre-operative could be identified (Rousseau et al. Ann Surg lymphoscintigraphy prior to axillary sentinel Oncol 2005; Goyal et al. Eur J Cancer 2005). In CONTACT: lymphadenectomy in primary breast cancer this case a gamma probe could discover sentinel Dr. Christian Mäurer lymph nodes in 78-93% of patients, in particular Clinical Project Management SenSzi is an operative controlled, multicenter, if used by an experienced team (McMasters et al. [email protected] randomized trial that has recruited 1,198 Ann Surg 2000). Several studies have shown patients from 15 sites throughout Germany. that SNB can be used as a reliable method even without prior lymphoscintigraphy, (Wang et al. Background Ann Surg Oncol 2007) leading to additional Sentinel lymph node biopsy (SNB) is a well- benefits in terms of improved effectiveness established staging tool within the surgical in pre-surgical organization and cost-reduction treatment of breast carcinoma (Veronesi et al. N for the health care system. In Germany, a pre- Engl J Med 2003; Krag et al. Lancet Oncol 2010). operative lymphoscintigraphy is not definitely Lymphoscintigraphy is not a standard procedure requested according to the current S3 guidelines, yet. The detection rate of lymph nodes with this the procedure is however a well-established code approach ranges from 72 to 99.3% according to of practice (Kreienberg et al. S3 guidelines 2012). different studies (Sun et al. J Clin Oncol 2010; The SenSzi study aimed to evaluate the necessity Kim et al. Cancer 2006), however in up to one of pre-operative lymphoscintigraphy using a

prior to sentinel lymph node biopsy : Injection of radionuclide according to current guidelines, lymphoscintigraphy Main inclusion criteria: • Histologically (core biopsy) confirmed breast carcinoma or DCIS (>5cm or >2.5cm and G3) R • cT1-3, cN0, cM0 • Karnofsky > 70% and ECOG < 1 N=1102

Arm A Arm B

Surgeon: Surgeon: with knowledge on without knowledge on lymphoscintigraphy lymphoscintigraphy outcome outcome

sentinel lymph node biopsy

Figure 1: SenSzi study design | 95

prospective approach by blinding the surgeon level I/II) and, the rate of complete axillary regarding the outcome of the lymphoscintigraphy lymphadenectomy in pN+ patients without in order to add a recommendation to future prior neo-/adjuvant systemic therapy (endocrine/ guidelines. SenSzi primarily aimed to compare chemotherapy). Furthermore, identification the histological identification rate, i.e. the rates between imaging using lymphoscintigra average number of histologically detectable phy vs the intraoperative gamma probe and sentinel lymph nodes per patient with vs. histological detection rate are compared, without the surgeon’s pre-operative knowledge between with vs without the surgeon’s pre- on lymphoscintigraphy outcome. operative knowledge on lymphoscintigraphy Moreover, SenSzi compared the rate of node- outcome. positive (pN+) disease detected with SNB with vs. without the surgeon’s pre-operative knowl Study report edge on lymphoscintigraphy outcome in the The study has randomized a total of 1,198 overall cohort and in predefined subgroups patients between July 2014 and October 2015 (periareolar vs. peritumoral injection of radio from 22 sites (21 German and 1 Swiss). The first nuclide, 1 vs 2-day protocol, with or without study results will be presented at the American additional blue dye, with or without intraoper Society of Clinical Oncology (ASCO) June 2-6, ative frozen section analysis, with or without 2017, Chicago, Illinois, USA. complete axillary lymphadenectomy according

We are thanking all participating centers for their commitment and efforts.

COLLABORATING STUDY GROUPS:

SPONSOR: Research is supported by the individual participating centers

STUDY CHAIR: PD Dr. Sherko Kümmel Kliniken Essen-Mitte 96 | Annual Scientific Report 2016 | Follow-up Activities | 97

Follow-up Activities

Patient self-reported outcome (PSRO) 98 General Follow-up Database and eCRF 98 Current trials in follow-up 99 98 | Annual Scientific Report 2016 | Follow-up Activities

Follow-up Activities

Long-term follow-up of early breast cancer financially and organizationally independent trials is considered highly important as treat from the GBG. KKS is handling names and ment effects might increase, maintain or de addresses of the patients with a database strictly crease over time and have to be put into not accessible by GBG. relation with late or chronic toxicities. How ever, in Germany without access to national Triggered by GBG, the trustee sends a cancer registries, collection of long-term questionnaire asking for current health status, follow-up is very often an unaccomplishable including date and site of relapse, second task due to the logistic and financial burden malignancies, and date of death. The for study sites and sponsors. questionnaires are allowed to be filled in by a third person in case of death. Forms are to be Patient self-reported outcome sent to GBG using only the unique study (PSRO) identifier as pseudonym. For address changes or withdrawal of consent another form can be To improve follow-up and reduce the workload returned to the trustee. for the trial sites, we developed a concept to use patient self-reported outcome (PSRO) for long- GBG links updated data with the original study term follow-up in the GBG trials on early breast database and informs the site about their cancer. patients.

Study participants are invited by the site Currently over 7,600 participants from 20 trials investigator to join the PSRO registry. They and 242 sites are included in this registry. consent that their name, address, and the unique study identifier are being collected and to General Follow-up Database regularly receive health status questionnaires. and eCRF

German privacy law (BDSG) and good clinical Follow-up documentation over different studies practice (GCP) regulations forbid the storage of and long timespans is a burden for the sites due patient-identifying data by the sponsor. We to different systems, case report forms (CRFs), therefore developed the model with a strict schedules and procedures. To mitigate this we separation of patient-identifying data and developed a general follow-up database to pseudonymised medical data. The data trustee, document follow-up for all trials with the same based at the KKS (Koordinierungszentrums für electronic CRF (eCRF). This eCRF is simplified as klinische Studien), University of Marburg, is much as possible to collect only the basic

9000

8000

7000

6000

5000

4000

number of par,cipants 3000

2000

1000

0 Jul. 09 Jul. 10 Jul. 11 Jul. 12 Jul. 13 Jul. 14 Jul. 15 Jul. 16 Jan. 11 Jan. 15 Jan. 10 Jan. 12 Jan. 13 Jan. 14 Jan. 16 Date Figure 1: PSRO participants | 99

information necessary for analysis of the end accurate. Data collected by PSRO registry were points of our neoadjuvant and adjuvant trials. compared with those reported by the in trials. All these items can be collected during vestigator. routine aftercare without trial specific PSRO data of 5,417 patients from 18 trials have examinations. included relapses/second malignances (5 %), deaths (2.5 %), and no answer/withdrawal Current trials in follow-up (14 %). These data have been sent for a validation to the respective sites. Median FU was 1.5 (range GBG 27: REACT 0.1-6) years. We have received answers from 67 The recruitment of this study has finished in (32 %) sites with feedback on 1,161 (21 %) November 2012 with a total of 2,639 patients, patients. Sites have confirmed the PSRO records including 814 from Germany. Patients are in 675 (58 %) cases. In addition, sites were able followed three-monthly in the first year, to provide information for 157/228 (70 %) 6-monthly in year 2 and annually thereafter patients not responding via PSRO. Overall PSRO until year 10. Follow-up is still ongoing. After a registry seems to provide accurate information median follow-up (FU) of 5 years, an amendment on disease-free and overall survival of breast of the study protocol allowed long term follow- cancer patients at much lower costs and efforts. up data from patients to be collected via PSRO However, a combination of PSRO with the site’s registry. Currently 77 % of the patients already efforts to collect data from those patients not consented to the PSRO registry. responding or reporting incomplete data seams to be an effective approach to obtain high- PSRO Validation quality FU data for further analyses and In 2016 we have validated the quality of our reporting. PSRO registry for ensure it is complete and

Trial N (patients) FU completeness GBG 18: GeparDuo 907 42,9 % GBG 24: GeparTrio 2,357 48,2 % GBG 32: ICE 1,358 55,6 % GBG 33: GAIN 2,995 74,9 % GBG 36: Natan 693 57,7 % GBG 40: GeparQuattro 1,495 57,9 % GBG 44: GeparQuinto 2,572 68,0 % GBG 49: TECHNO 733 63,7 % GBG 50: Prepare 217 62,8 % GBG 52: ICE-2 391 68,7 % GBG 66: GeparSixto 588 75,9 % GBG 68: GAIN-2 2,169 57,2 % GBG 69: GeparSepto 1,203 71,1 % GBG 70: DAFNE 65 68,8 % GBG 74: Genevieve 333 52,9 % GBG 75: Insema 1,417 27,4 % GBG 84: GeparOcto 948 22,4 %

Table 1: Status of our trials in follow-up (completeness according to Clark, Lancet 2002;359:1309) 100 | Annual Scientific Report 2016 | Translational Research | 101

Translational Research

Translational Research Activities 102 GBG Biobanking 102 Central Pathology and biomarker analysis 103 Collaborative Translational Research Projects and Substudies 103 102 | Annual Scientific Report 2016 | Translational Research

Translational Research

Translational Research Activities hort. The wide area of biospecimens including blood, plasma and purified DNA maintained in Translational research is the link between clini- biobanks can be described as libraries of the hu- cal and experimental science. It implements sci- man organism. The Declaration of Helsinki entific knowledge from basic research into clini- states that all biobanks must take donated ma- cal practice but also transfers clinical issues back terials via a process of informed consent. to the laboratory. Translational research pro- GBG’s main Biobank facilities are kept in Berlin jects are aimed at increasing scientific knowl- (Charité-tissue bank) and Sulzbach/Saar (BioKryo edge for the understanding of the origin and de- GmbH - liquids and frozen tissue). Standardized velopment of a disease. The identification and storage and administration of samples are guar- validation of biological markers is accomplished anteed by both institutions. Additionally, our to improve the treatment decision for our breast cooperating laboratories in Hamburg and Erlangen cancer patients. We therefore contribute to the store and process samples for further analyses. development of innovative diagnostics and new, purposive cancer therapies for a more personal- The collected biomaterials include: ized medicine. • FFPE (formalin-fixed paraffin-embedded) tissue • Fresh frozen tissue GBG Biobanking • RNAlater® conserved tissue • whole blood (DNA extraction, Single An important pre-requisite for performing trans- Nucleotide Polymorphism (SNP) analyses, lational research projects is the availability of RNA analysis) high-quality biomaterial collections. Together • serum with the clinical data, the tissue and blood-de- • plasma rived samples are the key aspects for testing sci- • circulating tumor cell samples entific hypotheses in a well-defined patient co- • urine

GeparOcto - Collec5on of biomaterial 2016

1000 961 931 908 912 900

783 789 800

700 673 674

Collec5on rate 60% 600

500 Ovarian Func5on Substudy 387 400 364 Number of collected samples 300

200 149 117 121 87 100 68 77 14 17 7 8 1 0 0

Figure 1: Overview of collected biomaterial within the GeparOcto main study as well as within ovarian function substudy as of December 2016. | 103

During a trial, collection of certain biomaterials allelic imbalance) and LST score (long segment is mandatory and defined in the individual study transitions). A high HRD score is defined as ≥ 42 protocols. Even if the respective clinical trials are and a BRCA mutation is defined as a deleterious already closed for recruitment, the coordinating mutation of BRCA1 or BRCA2 in the tumor. Thus centers can still collect biomaterials, mostly HRD is defined as either HRD score ≥ 42 or a FFPE tissues. BRCA mutation. In order to obtain meaningful results from statistical analyses, the investigated samples Collaborative Translational Research should represent the majority of patients includ- Projects and Substudies ed in the trial. Therefore, the sample collection rate should be at least 60 %. This goal is usually Based on our standardized pathological evalua- reached and also exceeded for most samples tion of tumor specimens, the biomaterials com- collected in GBG trials (e. g. FFPE tissue, SNP, bined with the clinical data from our studies be- serum samples). As an example an overview of come more and more interesting for further all collected biomaterials within the GeparOcto translational research approaches. Evaluated study is shown in Figure 1. biomarkers have been shown to further enhance diagnostic sensitivity, prediction rate and thera- Central Pathology and biomarker peutic response. The investigation of the bioma- analysis terials within the framework of our scientific projects is carried out in cooperation with our The Institute of Pathology at Charité University national and international partners. Hospital, Berlin performs an independent central pathological assessment of standard breast Ovarian function substudy cancer biomarkers such as estrogen receptor The substudy is supported by the Walter Schulz (ER), progesterone receptor (PR), HER2, Ki-67 as Stiftung. well as tumor infiltrating lymphocytes (TILs) im- Women below the age of 45 years with breast plemented as inclusion criteria or stratification cancer have a risk of developing premature ovar- factors in the new GBG clinical trials. This inde- ian failure. To define the risk of premature ovari- pendent central histologic assessment allows a an failure and loss of fertility with modern ther- standardized definition of the respective patient apeutic regimen, the hormone levels of estradiol, cohorts as well as comparison of treatment out- Follicle-Stimulating Hormone (FSH) and An- come within and across GBG clinical trials. Such ti-Müllerian Hormone (AMH) in addition to an- standardization can also promote the submis- tral follicle counts measured by ultrasound are sion of specimens for translational research. All assessed in GeparSixto, GeparSepto, GENE- these central testing methodologies are per- VIEVE, GeparOcto and Gain-2 trials. An over- formed according to the latest standards and view of all collected biomaterials within the guidelines. ovarian function substudy is shown in Figure 2.

In addition to assessment of standard TransLuminal B biomarkers, the Homologous Recombination The project is being supported in funding from Deficiency (HRD) status of tumor samples is German Cancer Aid that was awarded as part of recently evaluated in some trials (GeparSixto, its “Translational Oncology” priority programme Olympia, GeparOla). In general, HRD tumors and started in 2015. This project is being led and have lost the ability to repair double-stranded coordinated by Prof. Dr. Carsten Denkert, Insti- DNA breaks, resulting in increased susceptibility tute of Pathology at Charité University Hospital, to DNA- damaging drugs such as platinum Berlin. The partners involved in the project are agents. Genomic instability and a high frequency academic institutions from Heidelberg, Regens- of germline BRCA1 and BRCA2 mutations are burg and the GBG, Neu-Isenburg. The aim of this commonly associated with triple-negative project is to investigate the molecular mecha- breast cancer (TNBC). In the GeparOla trial nism involved in drug resistance and metastasis BRCA mutation status as well as HRD score of of high risk luminal B breast tumors and come up the tumor samples are tested centrally using with clinically relevant risk stratification. Blood Myriad’s HRD test. The HRD score is defined as circulating tumor cells (CTCs), disseminated tu- an algorithmic assessment of the LOH score mor cells (DTCs) in bone marrow and residual (loss of heterozygosity), TAI score (telomeric tumor cells (RTCs) in tumor tissue after neoad- 104 | Annual Scientific Report 2016 | Translational Research

juvant therapy will be analyzed to identify key (n=906), HER2-positive (n=1,379) and luminal/ molecular alterations in individual cancer cells. HER2-negative (n=1,366) tumors from 6 GBG This will allow physicians to treat breast cancer neoadjuvant clinical trials. The first results in a more targeted way, saving valuable time. showed that TILs are linked to increased pCR in all subtypes and most clinical subgroups TILs meta-analysis (Denkert et al. Abstract S1-09, SABCS 2016). Increasing evidence indicates the immune mark- High TILs improved survival in HER2-positive ers as predictive and prognostic factors in differ- and TNBC, whereas low TILs improved overall ent types of breast cancer, especially in the set- survival (OS) in luminal tumors. These findings tings of TNBC and HER2-positive breast cancer. suggest a different prognostic potential of TILs in Tumor-infiltrating lymphocytes (TILs) can influ- TNBC and luminal tumors, especially in non- ence tumor prognosis and the chemotherapeu- pCR groups. In the TNBC and HER2-positive tu- tic response. Current options on the features of mors TILs can be used as a positive predictive TILs and their prognostic potential in breast can- and prognostic factor while in luminal breast cer are inconsistent. Therefore the aim of this cancer TILs may be linked to reduced endocrine substudy (meta analysis) is to evaluate the pre- response. However, further validation studies dictive and prognostic potential of TILs in differ- are needed. ent subtypes of breast cancer treated with neoadjuvant therapy. An immunohistochemistry of TH4RESPONS (TRANSCAN-2, ERA-NET) TILs was performed on pretherapeutic core biop- TRANSCAN-2 is a 5 year project (2015-2019) sies derived from 3,771 patients with TNBC funded by the European Commision under the

Figure 2: Overview of collected biomaterial within the ovarian function substudy in GeparSixto, GeparSepto, GeparOcto, GENEVIEVE and GAIN-2 as of December 2016. | 105

EU framework programme Horizon2020. It is had significantly lower pCR rates compared with the continuation of the previous ERA-NET on wild-type tumours (16.2 % vs. 29.6 %; P < 0.001), translational cancer research TRANSCAN, fund- (Loibl et al. Ann Oncol. 2016). Within the hor- ed under the FP7 (2011-2014). TH4RESPONS is mone-receptor positive subgroup, the PIK3CA a subproject under the TRANSCAN-2 pro- mutant group had a pCR rate of only 7.6 % gramme with a focus on tumor heterogeneity compared with 24.2 % in the wild-type group (TH) in triple negative breast cancer (TNBC) and (P < 0.001). No definite conclusions can be high-grade serous ovarian carcinoma (HGSOC) drawn regarding survival. for prediction of therapy response. Project partners are Spain (Vall d' Hebron Institute of Oncol- DKTK ogy, Experimental Therapeutics Group, Barcelo- The DKTK (Deutsches Konsortium für Transla- na), Italy (European Institute of Oncology, tionale Krebsforschung) is a consortium build up Department of Pathology, Milan), France (Insti- of academic institutions with outstanding ex- tut Gustave Roussy Gynecology Unit, Transla- pertise in the field of cancer research at eight tional Research Laboratory INSERM U981, Ville- locations throughout Germany. Within the juif) and Germany (Institute of Pathology, DKTK, cancer researchers and medical practi- Charité University Hospital Berlin and GBG, tioners collaborate closely to speed up the Neu-Isenburg). The project coordinator is Man- transfer of basic scientific research to clinical fred Dietel (Institute of Pathology, Charité Uni- practice of new diagnostic and treatment ap- versity Hospital Berlin, Germany). proaches. The cooperation network of GBG’s Aims of this project are (1) to define molecular DKTK partners include Berlin, Heidelberg, Frei- features of TH as drivers of chemotherapy resist- burg, Düsseldorf and Munich. In the framework ance in TNBC and HGSOC, (2) to profile TH and of DKTK, the RNA, DNA and metabolic profile of tumor evolution in tissue samples and liquid bi- FFPE tumor samples of GeparTrio, GeparSixto, opsies of ovarian cancer, (3) to assess the clinical GeparSepto, Gain and BCP trials will be investi- utility of differences in molecular alterations in gated. The aim of this project is to identify new breast cancer between primary tumor and me- DNA-based mutations and to define RNA-based tastases, (4) to identify resistance mechanisms intrinsic subtypes. Ultimately, standardized against targeted therapy in patient-derived xen- FFPE based diagnostic tools will allow new uni- ograft (PDX) models, and (5) to integrate the fied classification of breast cancer. results for further validation including prospective validation studies as well as development of HTG EdgeSeq system concepts for future validation trials. The HTG EdgeSeq has been recently developed system to automate library preparation plat- PIK3CA meta-analysis form for targeted RNA sequencing enhancing The aim of this substudy was to evaluate PIK3CA productivity and ease of use. It is a nucleic acids mutational pattern and its association with extraction-free chemistry system that enables pathological complete response (pCR), dis- quantitation from very small, clinically relevant ease-free survival (DFS) and overall survival FFPE tissue. Furthermore, the HTG EdgeSeq sys- (OS) in a cohort of 967 patients with HER2-pos- tem creates opportunities for new treatment itive breast cancer enrolled in 5 GBG neoadju- approaches in cancers where initial diagnostic FURTHER INFORMATION: vant clinical trials. Patients have received either biopsies do not yield sufficient tissue for current Dr. Bärbel Felder trastuzumab (T), lapatinib (L) or the combina- tumor profiling methods. Translational Research tion T/L in addition to a taxane-based chemotherapy. PIK3CA was genotyped in tumour biop- Proposals may also be submitted by groups that Phone: +49 6102 7480-217 sies taken before therapy. The results showed are currently not represented in any board. Fax: +49 6102 7480-440 that after a median follow-up of 47 months www.gbg.de/de/forschung/translationale- [email protected] overall PIK3CA mutant/HER2-positive tumours forschung.php www.gbg.de/trafo 106 | Annual Scientific Report 2016 | Studienfinder

GBG Studienfinder

Therapieoptimierungsmaßnahmen bei Patientinnen mit primärem Mammakarzinom*:

Brustkrebs in der Schwangerschaft oder < 40 Jahre außerhalb der Schwangerschaft:

BCP (GBG 29) (retro- und prospektive Registerstudie)

Operative Studien:

INSEMA: axilläre SLNB vs. keine axilläre Operation (1. Rando), falls SLNB positiv: SLNB alleine vs. ALNB (2. Rando)

Präoperative (neoadjuvante) Therapieentscheidungen (M0):

Hochrisiko-Brustkrebs-Patientinnen: HER2+ oder TNBC GAIN-2: EnPC vs tailored dose-dense EC-Doc oder Luminal B mit ≥ N1 oder Luminal A mit ≥ 4 LK (bei HER2+ zudem Trastuzumab+Pertuzumab)

GeparX: 2x2 nab-Paclitaxel (weekly vs 1,8q22) gefolgt Mamma-Ca mit cT2 - cT4a-d oder cT1c wenn TNBC, von EC q2/q3 +/- Denosumab HER2+, N+ oder KI67 > 20 %, M0 (TNBC: Carboplatin; HER2+:Trastuzumab+Pertuzumab) GeparNuevo: nab-Paclitaxel weekly gefolgt von TNBC mit cT1b - cT4a-d, N0-3, M0 EC q2 +/- Durvalumab (anti-PD-L1) HER2neg Mamma-Ca mit mit homologen rekombinanten GeparOla: 12x Paclitaxel weekly + Olaparib gefolgt von EC q2/q3 Mangel (HDR): BRCA1/2 Mutation und/oder HDR-Score pos. vs 12x Paclitaxel + Carboplatin weekly gefolgt von EC q2/q3 HR+/HER2neg Mamma-Ca mit cT2–cT4, N0-3, M0, ULTIMATE: Exemestan/Durvalumab (anti-PD-L1) bei Pat. mit postmenopausal CD8+ T-Zellinfiltration nach Immun-Stimulation

Postoperative Therapieentscheidungen (M0):

gBRCA 1-2+; HER2-, nach adj oder neoadj CHT, N-/N+ OLYMPIA: Olaparib vs. Placebo Hochrisiko-Brustkrebs-Patientinnen: HER2+ oder TNBC GAIN-2: EnPC vs tailored dose-dense EC-Doc (bei HER2+ zudem oder Luminal B mit ≥ N1 oder Luminal A mit ≥ 4 LK Trastuzumab) High risk, non-pCR nach PST, HR pos, HER2 neg PenelopeB: endokrine Therapie ± Palbociclib

HR pos und HER2 neg PALLAS: adjuvante ET (5 Jahre) +/- Palbociclib (2 Jahre)

Mamma-Ca beim Mann MALE: Tamoxifen vs. Tamoxifen+GnRH vs Exemestan+GnRH

Therapieoptimierungsmaßnahmen beim metastasierten Mammakarzinom*:

Vortherapie AURORA: Gewebesammlung Primärtumor und Metastase incl. Blutproben beim Mann MALE: Tamoxifen vs. Tamoxifen+GnRH vs Exemestan+GnRH

PATINA: HER2+/HR+ Kollektiv: Anti-HER2 und endokrine Erhaltungstherapie +/- Palbociclib nach Induktionstherapie mit Chemo- bzw. HER2- Therapie 1st line PADMA: HR+/HER2- Kollektiv: endokrine Therapie + Palbociclib vs Monochemotherapy +/- endokriner Erhaltungstherapie (physician`s choice) DESIREE: HR+/HER2- Kollektiv: dose-escalation Everolimus über 4 Wochen vs Standard > 1st line: Everolimus plus Exemestan Cerebrale Filialisierung Prospektive Registerstudie incl Biomaterialsammlung bei mBC mit Hirnmetastasen

*Weitere Studien sind derzeit in Planung und werden auf unserer Homepage zu finden sein: www.GBG.de | 107

Brustkrebs in der Schwangerschaft oder < 40 Jahre außerhalb der Schwangerschaft:

BCP (GBG 29) (retro- und prospektive Registerstudie)

Operative Studien:

INSEMA: axilläre SLNB vs. keine axilläre Operation (1. Rando), falls SLNB positiv: SLNB alleine vs. ALNB (2. Rando)

Präoperative (neoadjuvante) Therapieentscheidungen (M0):

Hochrisiko-Brustkrebs-Patientinnen: HER2+ oder TNBC GAIN-2: EnPC vs tailored dose-dense EC-Doc oder Luminal B mit ≥ N1 oder Luminal A mit ≥ 4 LK (bei HER2+ zudem Trastuzumab+Pertuzumab)

Postoperative Therapieentscheidungen (M0): gBRCA 1-2+; HER2-, nach adj oder neoadj CHT, N-/N+ OLYMPIA: Olaparib vs. Placebo Hochrisiko-Brustkrebs-Patientinnen: HER2+ oder TNBC GAIN-2: EnPC vs tailored dose-dense EC-Doc (bei HER2+ zudem oder Luminal B mit ≥ N1 oder Luminal A mit ≥ 4 LK Trastuzumab) High risk, non-pCR nach PST, HR pos, HER2 neg PenelopeB: endokrine Therapie ± Palbociclib

HR pos und HER2 neg PALLAS: adjuvante ET (5 Jahre) +/- Palbociclib (2 Jahre)

Mamma-Ca beim Mann MALE: Tamoxifen vs. Tamoxifen+GnRH vs Exemestan+GnRH 108 | Annual Scientific Report 2016 | Notes

Notes | 109 110 | Annual Scientific Report 2016 | Notes

Notes | 111 112 | Annual Scientific Report 2016 | Notes

Notes | 113 114 | Annual Scientific Report 2016 | Notes

Notes

Impressum: GBG Forschungs GmbH Martin-Behaim-Strasse 12 63263 Neu-Isenburg GERMANY www.GBG.de