GIST and Sarcomas GSF-GETO 21 Juin 2017

Transcription

1 GIST and Sarcomas GSF-GETO 21 Juin 2017

2 GIST GSF-GETO 21 Juin 2017

3 PERSIST-5: Five Year Extended Treatment with Adjuvant Imatinib for Patients with Intermediate/High Risk Primary Gastrointestinal Stromal Tumor (GIST) Chandrajit P. Raut, 1 N. Joseph Espat, 2 Robert G. Maki, 3 Dejka M. Araujo, 4 T. Faith Williams, 5 Ronald P. DeMatteo 6 Abstract 11009

4 Recurrence-Free Survival Censored Relapse Free Survival Probability N = No. of patients still at risk Time to Relapse (Months) Estimated 5-year RFS 90.1% (95% CI, )

5 Patients Who Wrong Recurred Wrong Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Patient 7 Age, years Gender Female Male Male Female Male Female Female Location Small bowel Small bowel Stomach Stomach Small bowel Stomach Small bowel Mutation KIT exon 9a None PDGFRα D842V PDGFRα D842V KIT exon 11 PDGFRα exon 18 On treatment No No No Yes No No No Time of discontinuation, months Reason for discontinuation None AE choice Patient choice Progression AE Completed 5 yrs of treatment sensitive Patient mutation choice Time of relapse, months b Time from discontinuation to relapse, months dose 800mg Wrong Patient patient patient Nonsensitive mutation! Nonsensitive mutation? No a Insertion AY b Patient died of progressive disease (PD) at 21.1 months.

6 Imatinib-sensitive mutations? Mutation N=85*, n (%) Mutation identified 75 (87) KIT Exon 9 5 (6) Exon (68) Should not be in at 400mg/d Exon 13 2 (2) Exon 17 1 (1) PDGFRα Exon 12 1 (1) Exon 18 8 (9) Should not be in *Mutation status at screening for biomarker-consented patients

7 Dr. Raut will edit Analysis of risk groups Inclusion criteria Age 18 years with KIT-positive primary GIST Significant risk of recurrence Primary GIST (any site) 2 cm and a mitotic rate of 5/50 HPF Non-gastric primary GIST 5cm 10-yr risk of recurrence Joensuu et al. Lancet Oncol 2012; 13:265-74

8 Analysis of risk groups Inclusion criteria Age 18 years with KIT-positive primary GIST Significant risk of recurrence Primary GIST (any site) 2 cm and a mitotic rate of 5/50 HPF Non-gastric primary GIST 5cm 10-yr risk of recurrence Joensuu et al. Lancet Oncol 2012; 13:265-74

9 Treatment Duration: Patients Who Did Not Complete 5 Years of Imatinib years N = 46/91 completed the protocol (49%) Months on treatment years Patient choice (n=19) Adverse events (n=15) Protocol deviation (n=5) Lost to fu (n=3) Death (n=2)

10 Pharmacokinetic Analysis Imatinib trough concentration, ng/ml 2500 Male Female * 2000 ** 1500 * *** 87 patients eligible Mean imatinib trough levels were 33% higher in women (at 4 mos and overall) Mean trough levels remained relatively constant during 1 st year of treatment This suggested that imatinib PK remained stable over time 0 Day 29 Month 4 Month 12 Overall *P = not statistically significant; **P ; ***P

11 Clinical activity of BLU-285 in advanced gastrointestinal stromal tumor (GIST) Michael Heinrich 1, Robin Jones 2, Margaret von Mehren 3, Patrick Schoffski 4, Sebastian Bauer 5, Olivier Mir 6, Philippe Cassier 7, Ferry Eskens 8, Hongliang Shi 9, Terri Alvarez-Diez 9, Oleg Schmidt-Kittler 9, Mary Ellen Healy 9, Beni Wolf 9, Suzanne George 10 Abstract no: 11011

12 BLU-285: highly potent and selective targeting of KIT/PDGFRα GIST mutants High kinome selectivity* BLU-285 Binds active conformation *Image reproduced courtesy of CSTI (

13 BLU-285 Phase 1 study Key objectives Part 1: MTD, safety, pharmacokinetics, ctdna analyses, anti-tumor activity Part 2: response rate, duration of response, safety Part 1 Dose escalation completed Advanced GIST MTD 3+3 design with enrichment Dose levels: 30, 60, 90, 135, 200, 300, 400 and 600 mg QD MTD determined to be 400 mg PO QD Part 2 Dose expansion enrolling PDGFRα D842V-mutant GIST (n=50) Unresectable GIST after imatinib and 1 other TKI (n=50)

14 Tumor regression across all dose levels in PDGFRα D842-mutant GIST (central radiology review) N=25 60% Recist 100% Choi mpfs NR

15 Dose-dependent tumor reduction across multiple KIT genotypes (central radiographic review) 8% Recist 32% Choi BCR 56% mpfs 9.3m *ctdna results pending Very good toxicity profile, 25% with G3 toxicity of some sort, 8% fatigue

16 The clinical impact of performing routine next generation sequencing in GIST Ciara M. Kelly, Timothy Bowler, Li-Xuan Qin, Sandra P. D Angelo, Mark A. Dickson, Mrinal M. Gounder, Mary Louise Keohan, Taylor Vonya, Cristina R. Antonescu, Sam S. Yoon, Samuel Singer, Ronald P. DeMatteo, Ahmet Zehir, Marc Ladanyi, David B. Solit, Ping Chi, William D. Tap Memorial Sloan Kettering Cancer Center, New York, NY, USA

17 Results 191 samples sequenced, from 177 patients NGS was most commonly performed in setting of metastasis (56%) Primary tumor was the most commonly sequenced tumor type (63%) Percentage Samples Sequenced Clinical Setting When NGS Performed 100% 80% 60% 40% 20% 0% 37% 7% 56% Adjuvant Neoadjuvant Metastatic/LR 37% Tumor Type Sequenced 63% Primary Tumor Metastasis/Loc al Recurrence < 15% of patients with GIST have their tumors genotyped in clinical routine practice

18 Mutational Landscape by Functional Classification Functional Class WNT signalling JAK/STAT signalling Hedgehog signalling DNA-repair Apoptosis Signalling (other) Tyrosine phosphatase mtor signalling MAPK singalling RNA processing Proteosomal degradation PI3K signalling Receptor Tyrosine Kinase Cell cycle Chromatin structure Number of Mutations Metastatic/LR N=33 Primary N=67

19 Paired Sample Analysis KIT Driven GIST - Emergence of Resistance 2012 mutationally quiet tumor at diagnosis? Clinical Course 09/2011 DiagnosisSta ge IV GIST Imatinib 08/2012 Gastrectomy Imatinib 07/2016 Imatinib Refractory - surgery

20 Paired Sample Analysis KIT Driven GIST - Emergence of Atypical Pattern of Resistance 2008 NGS informed therapeutic selection in the majority of GIST cases (79%) NGS adds unique value for clinical care today and research towards better understanding of GIST tomorrow (G. Demetri, discutant) Presented by: Ciara M. Kelly

21 Bone and soft-tissue sarcoma

22 Phase 3 Study of Aldoxorubicin vs Investigators Choice as Treatment for Relapsed/Refractory Soft Tissue Sarcomas Sant Chawla, M.D. Principal Investigator Director, Sarcoma Oncology Center Santa Monica, California

23 Study Design Aldoxorubicin 350 mg/m 2 (260 mg/m 2 doxorubicin equivalents) every 3 weeks vs: Investigators choice of : pazopanib, gemcitabine/docetaxel, dacarbazine, doxorubicin, ifosfamide; administer each per institution standard tx Each site chooses 3 of 5 investigator choice drugs prior to enrollment of patients May use any of the 3 drugs on patients CytRx provides ALL chosen drugs + G-CSF to sites

24 PFS Central Radiology Review Variable Aldoxorubicin (N = 218) Invest s Choice (N = 215) ITT [Time to Progression or Death (Months)] Median p-value Hazard Ratio (95% Confidence Interval) 0.81 (0.64, 1.03) L-Sarcomas [TTP or Death (Months)] Median (95% Confidence Interval) p-value Hazard Ratio (95% Confidence Interval) 0.62 (0.44, 0.88)

25 Aldoxorubicin: Comparing Phase II and Phase III data Phase II trial (ASCO 2014) Phase III trial (ASCO 2017) N=126 N=433 PFS: 5.6 vs 2.7 months PFS: 4.1 vs 2.96 months P=.08 Chawla SP, et al. JAMA Oncol. 2015;17:1-9 Aldoxo vs dox Abstract#11000 Aldoxo vs other regimen (including doxo)

26 Aldoxorubicin Phase III study results Primary endpoint: No statistically significant difference in PFS PFS improved in subsets: L-sarcomas (p=0.007) and North American/ Australian subset Heterogeneity ++++ Pre-randomization Histopathobiological behavior, differences in rate of growth Prior lines of therapy Post-randomization Variability in physician choice by site in control arm(s) Prior use of Doxorubicin in Aldoxorubicin arm: 65% Total NA/AUS EU/LA Dacarbazine 28% 5% Pazopanib 26% 48% GemTax 14% 34% Doxorubicin 31% 3% Ifosfamide 0% 11% 72% 28%

27 Next Generation Sequencing in Sarcomas: On a path towards Precision Abstract 11001: Gounder M et al. Abstract 11002: Italiano A et al. Gounder et al. Number of Patients Study Design Retrospective Retrospective Sequencing genes 310 genes Italiano et al. Clinical Outcomes 2% of study population To be analyze :MULTISARC

28 Can NGS help establish diagnosis? Additional diagnostic insights in 8% of cases Does not replace (but complements) expert pathology consult Evaluation at a high volume center changes sarcoma diagnosis in about 12% of cases (MDACC data)

29 Landscape of Mutations (n = 5749) Germline mutations (9%) ~62,000 mutations ~1200 fusions TP53: 4% BRCA1/2: 6.5% MUTYH : 9% CD36 : 4.5%

30 Results: Italiano et al. Sarcomas with Complex Genomics (14 Histologies) N=331 (56%), 720 mutations in 258 genes (+ fusion genes), ~2 mut/patient Translocation-related Sarcomas (10 Histologies) N=144 (25%), 120 mutations in 88 genes, <1 mut/patient

31 Tumor Mutational Burden for each Sarcoma Sub-types Median TMB = 2.5 (0 328) per megabase

32 Tumor Mutational Burden High tumor mutational burden is widely distributed across sarcomas STS/bone High TMB = 5% of pts

33 Landscape of Actionable Mutations 50% of patients from MSKCC 40% in French study were thought to have actionable mutations 52% of patients who had actionable mutations received matched therapy Biomarkers of Resistance 2% No Actionable Mutations 41% No actionalbe mutations 41% LEVEL 3: Novel Biomarkers and Investigational Drugs 40% LEVEL 1: FDA approved Biomarker/Drug/ Disease 8% LEVEL 2: Known Biomarker + FDA/NCCN approved drugs + New diseases 9% Systematic evaluation of clinical outcomes by histopathology & alterations is acutely needed!!

34 60 year old woman with widely metastatic MPNST TPM4-NTRK3 fusion Enrolled in Phase II trial of larotrectinib ASCO 2017 Oral Developmental Therapeutics Robert C. Doebele et al. Cancer Discovery 2015;5:

35 62 year old man with inoperable chondrosarcoma IDH R132 AG-120 Presented by: William Tap, MD CTOS 2016, Lisbon, Portugal

36 Precision Delivery is critical for success of Precision Medicine Clinical decision support for the oncologist in clinic Getting patient to matched drug Integrated data system for storing/analysis of molecular and clinical data Learning algorithms that can detect insights Knowledge base capable of integrating such insights and delivering them to clinical decision support systems Parikh RB, NEJM 2017; 376: Studies like MULTISARC can reveal the clinical benefit of NGS prospectively

37 MULTISARC Objective To demonstrate the superiority of an innovative treatment strategy guided by high throughput molecular analysis (next generation sequencing exome, RNASeq [NGS], immunological profiling) in patients with advanced soft-tissue sarcomas [STS]. Comparison of 2 treatment strategies: Experimental strategy: standard 1st line anthracycline-based chemotherapy followed by subsequent NGS-guided treatments, Standard strategy: standard 1st line anthracycline-based chemotherapy followed by subsequent treatments not guided by high throughput molecular analysis NGS (exome, RNASeq). 37

38 MULTISARC Schema

39 MULTISARC targets: 16 drugs Expected in Q Arm/ Target Drus(s) Est Prevalence % EGFR/HER2 MET amp/ mutation AL/ROS1 amp/ rearrangement Dual EGFR/HER2 inhibitor 1-2 MET inhibitor 1 ALK/ROS1 inhibitor 1-2 BRCA1 mutation PARP inhibitor 1-2 FGFR1/2/3, FRS2 FGFR inhibitor 1-2 Arm/ Target Drug(s) Est Prevalence % SMO/PTCH1 Smo inhibitor 1 NF1 loss MEK inhibitor 1-2 ckit, CSF1R mutation/ amplification TP53 mutation/loss, CDKN2A, RB1 loss NOTCH1/2/3 KIT inhibitor 4-5 Wee inhibitor Gammasecretase 1-2 mutation/ampl RICTOR mtorc1/mtorc inhibitor NTRK1/2/3 NTRK inhibitor 2-3 amplification, inh High mutational load PDL1 + PARP 4-8 TSC1/2 loss, PTEN loss CDK4,CCND1/2/3 CDK4/6 inhibitor KDR mutation inh VEGFR inhibitor 2 39

40 The sarculator stratified prognosis of patients with highrisk STS of extremities and trunk wall treated with perioperative CT in a randomised controlled trial. Sandro Pasquali, Chiara Colombo, Stefano Bottelli, Paolo Verderio, Javier Martin Broto, Antonio Lopez-Pousa, Stefano Ferrari, Andres Poveda, Antonino De Paoli, Vittorio Quagliuolo, Josefina Cruz, Alessandro Comandone, Giovanni Grignani, Rita De Sanctis, Elena Palassini, Antonio Llombart-Bosch, Angelo Paolo Dei Tos, Paolo Giovanni Casali, Piero Picci, Alessandro Gronchi Multicenter Italy (INT Milano) and Spain

41

42 Results DM Risk Stratification using Sarculator NO difference in OS, LR, DM for 3 vs. 5 cycles OS Use of the Sarculator to prospectively enroll a ptfor perioperative treatments Presented by:

43 Multicenter Phase II Study of Pembrolizumab in Advanced STS and Bone Sarcomas: Final Results of SARC028 and Biomarker Analyses Melissa A. Burgess, MD, Vanessa Bolejack, PhD, Brian A. Van Tine, MD, Scott Schuetze, MD, PhD, PhD, James Hu, MD, Sandra P. D Angelo, MD, Steven Attia, DO, Dennis Priebat, MD, Scott H. Okuno, MD, Richard F. Riedel, MD, Lara E. Davis, MD, Sujana Movva, MD, Damon Reed, MD, Lisa H. Butterfield, PhD, Janos Roszik, PhD, Denise Reinke, NP, MBA, Laurence H. Baker, DO, Robert Maki, MD, PhD, Shreyaskumar Patel, MD, Hussein A. Tawbi, MD, PhD, on behalf of SARC028 Investigators University of Pittsburgh, Cancer Research and Biostatistics, Washington University St Louis, University of Michigan, University of Southern California, Memorial Sloan Kettering Cancer Center, Mayo Clinic Florida, MedStar Research Institute, Mayo Clinic Rochester, Duke University, Oregon Health and Science University, Fox Chase Cancer Center, Moffitt Cancer Center, University of Pittsburgh, UT MD Anderson Cancer Center, SARC, University of Michigan, Mount Sinai Medical Center, UT MD Anderson Cancer Center Abstract Number: 11008

44 RECIST Best Response & Patterns of Response - STS Subtype Best Response CR PR SD PD Total LMS UPS 1 (10%) 3 (30%) LPS 0 2 (20%) Synovial 0 1 (10%) Total 1 (2.5%) 6 (15%) expansion cohorts ongoing in UPS/LPS

45 RECIST Best Response & Patterns of Response - BS Subtype Best Response PR SD PD Total ChondroS 1 (20%) Ewing 0 (0%) Osteosarcoma 1 (5%) Total 2 (5%) Presented by: Melissa A. Burgess, MD

46 Tumor PD-L1 expression/infiltrating T-cytotoxic lymphocytes 3 out of 70 analyzed cases were PD-L1+ PD-L1+ cases were restricted to UPS 2 were evaluable for response- 1 CR, 1PR No post-treatment cases were PD-L1+ Responders had an increased number of infiltrating cytotoxic T cells at baseline

47 Increases in pro-inflammatory cytokines on treatment favored better survival Affymetrix Luminex platform array In the STS cohort, 7 responders & 30 non-responders provided paired blood samples (baseline and week 8) suitable for analysis Week 8-Baseline Differences >median median

48 Alliance A091401: A multi-center phase II study of nivolumab +/- ipilimumab for patients with metastatic sarcoma Sandra P. D Angelo 1, Michelle R. Mahoney 2, Brian A. Van Tine 3, James Atkins 4, Mohammed M. Milhem 5, William D. Tap 1, Cristina R. Antonescu 1, Elise Horvath 6, Gary K. Schwartz 7, Howard Streicher 8 1. Memorial Sloan Kettering Cancer Center 2. Alliance Statistics and Data Center, Mayo Clinic 3. Washington University School of Medicine 4. Southeast Clinical Oncology Research Consortium NCORP, Winston-Salem, NC 5. University of Iowa/Holden Comprehensive Cancer Center 6. Astellas 7. Herbert Irving Comprehensive Cancer Center 8. National Cancer Institute, Cancer Therapy Evaluation Program, Investigational Drug Branch, Bethesda, MD

49 Study Design Eligible patients with advanced sarcoma R 1:1 Nivo 3 mg/kg + Ipi 1 mg/kg Nivo 3 mg/kg Q3Wx4 Cross over Nivo 3 mg/kg Q2W Q2W Treatment until: PD* Toxicity Up to 2 years * Treatment beyond PD allowed in 1 st 12 wks; 4 wk confirmation required to continue.

50 ORR: 3% Monotherapy, 16% Combination Nivo 3 Nivo 3 + Ipi 1 ORR 3% PR ASPS LMS Sarcoma, NOS ORR 16% CR Myxofibrosarcoma (1) Uterine LMS (1) PR UPS (3) LMS (1) Angiosarcoma (1) Expansions in LPS, UPS approved

51 PFS: 2.1m Monotherapy, 4.4m Combination % Alive & Progression Free Total (Events) 38 (34) 20 Nivo 3 10 Nivo 3 + Ipi 1 Patients at Risk Time (Months) Median (95% CI) 2.1 ( ) + Censor Total (Events) 38 (30) Patients-at-Risk Time (Months) Median (95% CI) 4.4 ( ) + Censor

52 Safety Overview (Treated Patients) Nivolumab n=42 (%) Nivolumab +Ipilimumab n=42 (%) Any grade Grade 3-4 Any grade Grade 3-4 Adverse Events (AEs) 42 (100) 17 (40) 42 (100) 20 (48) Treatment Related AEs 28 (67) 3 (7) 29 (69) 6 (14) Serious Adverse Events (SAEs) 19 (45) 13 (31) 20 (42) 12 (29) Treatment Related SAEs 3 (7) 1 (2) 6 (14) 4 (10) * There were 11 deaths (5 Single Agent, 6 Dual Agent) unrelated to study treatment. Presented by: Sandra P. D Angelo

53 OS: 10.7m Monotherapy, 14.3m Combination Nivo 3 Nivo 3 + Ipi 1 40% alive at 12m Total (Events) Median (95% CI) 38 (26) 10.7 ( ) + Censor Patients at Risk Time (Months) % alive at 12m Total (Events) Median (95% CI) 38 (20) 14.3 (9.6-NE) +Censor 0 Patients at Risk Time (Months)

54 Immune response, safety, and survival impact from CMB305 in NY-ESO-1+ recurrent soft tissue sarcomas (C131 Study) Neeta Somaiah, Sant P. Chawla, Matthew Stephen Block, John Morris, Khanh Do, Joseph W. Kim, Mihaela Druta, Kamalesh Kumar Sankhala, Patrick Hwu, Sacha Gnjatic, Hailing Lu, Richard T. Kenney, Chet Bohac, Seth Pollack

55 CMB305: Prime-Boost Immunotherapy Targeting NY-ESO-1 No ex vivo manipulation or HLA matching Sequential administration of LV305 and G305 (3 months) followed by monthly G305 LV305 Priming: Dendritic cell (DC) targeting NY-ESO-1 lentiviral vector encoding full length NY-ESO-1 Integration deficient, replication incompetent Induces and expands NY-ESO-1 specific CD8 and CD4 T Cells G305 boosting: Potent TLR-4 agonist co-formulated with NY-ESO-1 full length protein Enhances LV305 immunogenicity and triggers anti- NY-ESO-1 antibodies 55

56 C131 Study: Disease Control in STS Patients SS: n=14, MRCLPS, n=9, others: n=2 STS patients with disease progression at entry: Experienced durable tumor growth arrest STS Patients n=25 ORR, pt (%) 0 SD, pt (%) 16 (64) DCR, pt (%) 16 (64) Median PFS 4.7 m Patients with induced anti-ny-eso-1 immune response 6 mos PFS 36.4% Most common EAEs: fatigue, injection site pain, influenza like illness, myalgia, injection site reaction

57 C131 Study: Overall Survival (OS) in STS pts Standard Therapy for 2nd Line STS Study OS Pazopanib months Eribulin months Trabectedin months 1 van def Graaf, 2012, 2 Schoffski, STS patients n=25 Median OS NA 12 mos OS Rate, % mos OS Rate, % 76.2 P, 2016, 3 Demetri, G. et al Standard Therapy for 2nd Line Synovial Sarcoma Study OS METASARC months Sarcoma WIP Registry 1 Savina, et al., 2017 (synovial sarcoma)

58 NY-ESO-1 Immunity and STS Patient Survival Induction of anti-ny-eso-1 immunity have a trend to a better survival Induction of anti-ny-eso-1 immune response in STS patients with baseline NY- ESO-1 + is associated with a better survival All patients with STS treated with LV305 or CMB305 ID/IM and who had biomarker samples were analyzed (n=38); 30/38 pts (79%) of patients had an induced anti-ny-eso-1 immune response on LV305 or CMB305 therapy 58

59 Conclusions Immunotherapy works in sarcoma! in immune subsets of molecular subsets of histological subgroups

60 Tailored Approaches to Rare Sarcomas Regorafenib in Ewing sarcoma (Dr. Attia) Pazopanib in Solitary Fibrous Tumor (Dr. Martin-Broto) Cediranib in Alveolar Soft Part Sarcoma (Dr. Judson) Evidence for Activity of VEGFR Inhibitors in ASPS, SFT and ESFT? Rare sub-type sarcoma trials are feasible

61 Ewing sarcoma outcomes comparison Agent / Study # PTS RR 8-WK PFS Median PFS Median OS Regorafenib/SARC % 60% 3.6 mo Not reported Imatinib/SARC % 8% Not reported Not reported Dasatinib/SARC % 24% 1.7 mo* 7.5 mo* R1507/SARC011 (IGF1-R) % 42% (6 wk)* 1.3 mo 7.6 mo Robatumumab (IGF1-R) 116 7% 34% NR 6.9 mo Pembrolizumab/SARC % 15% 1.75 mo mo Olaparib (PARP) 12 0% 33% 1.4 mo Not reported *Data provided by SARC

62 SFT outcomes comparison Drug # pts Choi RECIST ORR mpfs 6-mPFS ORR Pazopanib GEIS % 5.5 mo 40% 3% Pazopanib RMH % NA NA 9% Sunitinib 31 48% 7 mo NA 6% TMZ/BEV 14 79% 9.7 mo 79% 14% Dasatinib 25 20% 2 mo 30% 0% Adapted from S.M. Schuetze et al. Cancer 2017 M. Maruzzo et al. Clin Sarcoma Res 2015 Preclinical xenograft mouse model suggests regorafenib is more active in SFT (Stacchiotti. EJC 2014; 50: )

63 ASPS outcomes comparison Drug # PTS RECIST ORR Median PFS 24 wks Cediranib (CASPS) 28 21% 10.8 mo 61% Placebo 14 0% 3.7 mo 40% Cediranib 43 35% Not reported 84% Sunitinib 9 56% 17 mo 88% Sunitinib 15 40% 19 mo 86% Dasatinib 12 0% 11 mo 62% Adapted from S. Schuetze et al. Cancer 2017 P. Jagodzinska-Mucha et al. Tumori 2017 S. Kummar et al. JCO :

64 Overall Survival for Olaratumab + Doxorubicin versus Doxorubicin Alone (ITT Population) INVESTIGATOR assessment Olara + Dox Dox Patients/Events 66/39 67/52 Median, months (95% CI) 26.5 (20.9, 31.7) 14.7 (9.2, 17.1) HR (95% CI) 0.46 (0.30, 0.71) Stratified p-value.0003 Olara + Dox Dox Number at Risk Olara + Dox Dox Tap WD et al. Lancet 2016;388:488-97

65 An analysis of efficacy and safety of olaratumab+ doxorubicin or doxorubicin alone in patients with LMS Brian A Van Tine1, Women Cancer, March 2017

66 Merci!! GSF-GETO 21 Juin 2017