A randomized phase II study of paclitaxel/carboplatin/bevacizumab, - PowerPoint PPT Presentation

A randomized phase II study of paclitaxel/carboplatin/bevacizumab, paclitaxel/carboplatin/temsirolimus and ixabepilone/carboplatin/bevacizumab as initial therapy for measurable stage III or IVA; stage IV or recurrent endometrial cancer, GOG-86P C Aghajanian, VL Filiaci, DS Dizon, J Carlson, MA Powell, AA Secord, KS Tewari, D Bender, DM O’Malley, A Stuckey, J Rotmensch, DA Levine, HA Lankes, KN Moore

Background • Paclitaxel/carboplatin (PC) is a standard initial therapy for advanced endometrial cancer (EC) • GOG209, a randomized noninferiority trial of initial therapy (n=1381), showed that PC was not inferior to TAP in terms of PFS and OS • Bevacizumab, temsirolimus and ixabepilone have shown single agent activity in recurrent EC TAP, Paclitaxel + Doxorubicin + Cisplatin Gynecol Oncol 2012,125(3) Suppl 1:S2-188, page 771; J Clin Oncol 2011;29:2259-65; Gyn Oncol 2013;129:22-7; J Clin Oncol 2011;29:3278-85; Gynecol Oncol 2014;132:585-92; J Clin Oncol 2009;27:3104-8

Open: 9/14/09 GOG86P: Schema Closed: 1/9/12 One Cycle = 3 weeks 6 cycles Maintenance Paclitaxel 175 mg/m 2 IV 3 hours Carboplatin AUC 6 IV Bevacizumab 15 mg/kg IV Bevacizumab 15 mg/kg IV* Endometrial Cancer Paclitaxel 175 mg/m 2 IV 3 hours No prior chemotherapy Temsirolimus 25 mg IV Carboplatin AUC 5 IV -Stage III or IVA (measurable disease) days 1, 8 and 15 Temsirolimus 25 mg IV days 1 & 8* -Stage IVB (measurable disease or not) -Recurrent (measurable disease or not) Ixabepilone 30 mg/m 2 IV 1 hour Stratification: Carboplatin AUC 6 IV Bevacizumab 15 mg/kg IV • performance status (< 1 vs 2) • recurrent disease (yes/no) Bevacizumab 15 mg/kg IV* • measurable disease (yes/no) • prior EBRT (yes/no) Treatment until disease progression or adverse events prohibit further therapy *Starting cycle 2 when within 12 weeks of surgery EBRT, External Beam Radiation Therapy

GOG86P: Statistical Design • Primary endpoint: PFS • Secondary endpoints: ORR, OS, safety – Matched group from GOG209 PC Arm used as historical control – 3 experimental Arms of GOG86P are compared individually to historical control – RECIST 1.1 – Common Terminology Criteria for Adverse Events (CTCAE) v3.0 • Planned sample size: 330 • PFS: 35% decrease (HR=0.65) considered significant – 12 month PFS from 39% to 54% – 58 PFS events in an experimental arm, 85% power, 3.9% type I error – 6, 18 week time intervals evaluated to lessen bias (as comparisons made to matched group from GOG209)

GOG86P: Patient Characteristics PC + Bevacizumab PC + Temsirolimus IC + Bevacizumab Enrolled (n=349) 116 115 118 Eligible and Treated (n=329) 108 111 110 Median Age, years (range) 62 (36-87) 63 (38-82) 65 (37-89) Performance Status 0-1 106 (91%) 109 (95%) 113 (96%) Stage III (measurable disease) 12 (10%) 13 (11%) 10 (9%) IV 58 (50%) 58 (51%) 62 (52%) Recurrent 46 (40%) 44 (38%) 46 (39%) Histology (Central pathology review) Endometrioid, grade 1 17 (15%) 13 (11%) 15 (13%) Endometrioid, grade 2 36 (31%) 24 (21%) 27 (23%) Endometrioid, grade 3 30 (26%) 30 (26%) 22 (19%) Serous 16 (14%) 26 (23%) 31 (26%) Clear Cell 6 (5%) 4 (3%) 6 (5%) Other 11 (9%) 18 (16%) 17 (14%) PC, paclitaxel + carboplatin; IC, ixabepilone + carboplatin

GOG86P: Patient Characteristics PC + Bevacizumab PC + Temsirolimus IC + Bevacizumab Eligible and Treated (n=329) 108 111 110 Prior EBRT No 99 (85%) 95 (83%) 98 (83%) Yes 17 (15%) 20 (17%) 20 (17%) Measurable Disease No 27 (23%) 30 (26%) 33 (28%) Yes 89 (77%) 85 (74%) 85 (72%)

GOG86P: Treatment Exposure Treatment Delivered PC + Bevacizumab PC + Temsirolimus IC + Bevacizumab (n=116) (n=115) (n=118) Chemotherapy Median number of cycles, range 6 (0-6) 6 (0-6) 6 (0-6) Paclitaxel/Ixabepilone Median number of cycles, range 6 (0-6) 6 (0-6) 6 (0-6) Carboplatin Bevacizumab/Temsirolimus Median number of cycles, range 12 (0-78) 8 (0-62) 9 (0-53)

GOG86P: Overview of AEs PC + Bevacizumab PC + Temsirolimus IC + Bevacizumab Patients, % n=112 n=113 n=114 Any AE, Any grade 100 100 100 Any AE, Grade > 3 93.7 98.2 95.6 Any AE, Grade 5 3.6 5.3 5.3 AE Leading to Study Drug 26.8 (bev) 23 (tem) 24.6 (bev) Discontinuation 15.8 (ixa) SAE 42.8 50.4 46.5

GOG86P: AEs of Special Interest Fisher’s Exact PC + PC + IC + Bevacizumab Bevacizumab Temsirolimus n=114 Test Patients, % n=112 n=113 (2 sided)* ATE, grade > 3 0.9 0 0.9 p=0.554 VTE, grade > 3 8 9.7 7.9 p=0.681 Non-CNS Bleeding, grade > 3 2.7 0.9 4.4 p=0.281 GI Fistula, Leak, Perforation, Any grade 2.7 1.8 4.4 p=0.505 HTN, grade > 3 16.1 2.7 16.7 p<0.001 Proteinuria, grade > 3 5.4 0 4.4 p=0.018 Pneumonitis, any grade 0 6.2 0.9 p=0.002 Mucositis, oral, grade > 2 4.5 15.9 2.6 p<0.001 Rash, grade > 2 2.7 16.8 3.5 p<0.001 Hypertriglyceridemia, grade > 3 0 4.4 0 p=0.004 *Test comparing PC or IC + bevacizumab vs PC + temsirolimus

GOG86P: PFS Comparison to Historical Reference Hazard 92.2% Hazard Ratio Ratio Confidence Limits Arm 1 0.805 0.633 1.023 Hazard 92.2% Hazard Ratio Ratio Confidence Limits Arm 2 1.222 0.961 1.554 Hazard 92.2% Hazard Ratio Ratio Confidence Limits Arm 3 0.871 0.685 1.107

GOG86P: Objective Response Rate Measurable Disease Patients PC + Bevacizumab PC + Temsirolimus IC + Bevacizumab GOG209 PC (n=116) (n=115) (n=118) Historical Reference (n= 462) Complete Response (CR) 22 (24.7%) 14 (16.5%) 9 (10.6%) 40 (10.8%) Partial Response (PR) 31 (34.8%) 33 (38.8%) 36 (42.4%) 149 (40.4%) Objective Response 53 (59.5%) 47 (55.3%) 45 (52.9%) 189 (51.2%) Rate (CR + PR)

GOG86P: OS Arm Median Point Estimate 1 34.0 (p<0.039) 2 25.0 3 25.2 Reference 22.7

Translational science 25 most commonly mutated genes from TCGA endometrial cancer project sequenced to a minimum mean depth of 600X Top mutated genes by histology (primary samples only, n=219) Profile of serous cases (n=45) Endometrioid, G1 Endometrioid, G2 Endometrioid, G3 Serous Gene (n=37) (n=69) (n=59) (n=45) PTEN 73% 71% 58% 9% PIK3CA 38% 54% 53% 29% TP53 14% 20% 44% 87% ARID1A 38% 48% 39% 2% CTNNB1 57% 32% 31% 2% PIK3R1 27% 26% 19% 7% KRAS 19% 20% 24% 2% CHD4 27% 12% 20% 11% PPP2R1A 11% 4% 5% 29% FBXW7 0% 6% 10% 16% MTOR 5% 9% 14% 4% POLE 11% 6% 8% 2% Douglas A. Levine, Fanny Dao

Conclusions • PFS is not significantly increased in any Arm • OS is significantly increased in the PC + Bevacizumab Arm when compared with historical controls • No new safety signals • Integration of the translational research findings and clinical endpoints is ongoing

Thank you We gratefully acknowledge: All the patients and their caregivers who participated in this trial, All the Investigators and staff who contributed their time and effort to this study, and NRG Oncology