1. Bevacizumab in platinum-sensitive ovarian cancer: OCEANS

2. OCEANS: rationale Bevacizumab has shown single-agent activity in recurrent ovarian cancer (OC) (single-arm studies) 1,2 Carboplatin (C) + gemcitabine (G): phase III AGO/NCIC/EORTC trial in platinum-sensitive OC 3 1 Burger et al. JCO 2007; 2 Cannistra et al. JCO 2007; 3 Pfisterer et al. JCO 2006; EfficacyC (n=178)CG (n=178) Median PFS, months5.88.6 HR for PFS0.72 (p=0.0031) ORR, %3147 p=0.0016 Median OS, months17.318.0 HR for OS0.96 (p=0.7349 HR=hazard ratio, ORR=objective response rate, OS=overall survival, PFS=progression-free survival

3. OCEANS: study schema Stratification variables: Platinum-free interval (6–12 vs >12 months) Cytoreductive surgery for recurrent disease (yes vs no) Placebo q3w until progression Bevacizumab 15mg/kg q3w until progression Platinum-sensitive recurrent epithelial ovarian, primary peritoneal or fallopian tube cancer Measurable disease ECOG 0/1 No prior chemotherapy for recurrent ovarian cancer No prior Bev (n=484) CG + Pl CG for 6 (up to 10) cycles G 1000 mg/m 2, d1 & 8 C AUC 4 G 1000 mg/m 2, d1 & 8 CG + Bev AUC=area under the curve; Bev=bevacizumab; C=carboplatin; ECOG=Eastern Cooperative Group; G=gemcitabine; P=placebo Aghajanian et al. ASCO 2011

4. OCEANS: patient characteristics Characteristic CG + Pl (n=242) CG + Bev (n=242) Median age, years (range) 61 (28−86) 60 (38−87) Age ≥65 years, % 38 35 Race, % White Other 92 8 90 10 ECOG performance status 0, % 7675 Histologic subtype, % Serous Mucinous/clear cell Other 84 3 14 78 5 17 Platinum-free interval, % 6–12 months >12 months 42 58 41 59 Cytoreductive surgery for recurrent disease, %1012 Aghajanian et al. ASCO 2011

5. OCEANS: treatment exposure Treatment delivered CG + Pl (n=233) CG + Bev (n=247) Chemotherapy Median No. of cycles (range)6 (1–10) 6 (1–10) Patients receiving 7–10 cycles, % Carboplatin Gemcitabine 40 46 33 41 Bevacizumab/placebo Median No. of cycles (range)10 (1–36) 12 (1–43) Aghajanian et al. ASCO 2011

6. OCEANS: primary analysis of PFS CG + Pl (n=242) CG + Bev (n=242) Events, n (%) 187 (77.3)151 (62) Median PFS, months (95% CI) 8.4 (8.3–9.7) 12.4 (11.4–12.7) Stratified analysis HR (95% CI) Log-rank p-value 0.484 (0.388–0.605) <0.0001 Proportion progression free 242177451130 2422039233110 CG + Pl CG + Bev No. at risk: 1.0 0.8 0.6 0.4 0.2 0 0612182430 Time (months) Aghajanian et al. ASCO 2011

7. OCEANS: PFS by IRC IRC=independent review committee CG + Pl (n=242) CG + Bev (n=242) Events, n (%) 148 (61)119 (49) Median PFS, months (95% CI) 8.6 (8.3–10.2) 12.3 (10.7–14.6) Stratified analysis HR (95% CI) Log-rank p-value 0.451 (0.351–0.580) <0.0001 1.0 0.8 0.6 0.4 0.2 0 0612182430 Time (months) Proportion progression free 24216831830 242195732270 CG + Pl CG + Bev No. at risk: Aghajanian et al. ASCO 2011

8. Median PFS (months) Baseline risk factor No. of patients CG + Pl (n=242) CG + Bev (n=242)HR (95% CI) CG + Bev better CG + Pl better All patients 4848.412.40.49 (0.40–0.61) Platinum-free interval, months 6–122028.011.90.41 (0.29–0.58) >122829.712.40.55 (0.41–0.73) Cytoreductive surgery for recurrent disease Yes547.516.70.50 (0.24–1.01) No4308.412.30.49 (0.39–0.62) Age, years <653068.512.50.47 (0.36–0.62) ≥651788.412.30.50 (0.34–0.72) Baseline ECOG PS 03678.612.50.47 (0.36–0.60) 11168.310.60.61 (0.39–0.95) OCEANS: PFS subgroup analyses HR 0.20.5125 Aghajanian et al. ASCO 2011

9. Median PFS (months) Baseline risk factor No. of patients CG + Pl (n=242) CG + Bev (n=242)HR (95% CI) CG + Bev better CG + Pl better All patients Age, years ECOG PS Race Primary site <65 ≥65 0 1 White Non-white Fallopian tube carcinoma Ovarian carcinoma Primary peritoneal carcinoma 484 306 178 367 116 440 34 29 407 48 8.4 8.5 8.4 8.6 8.3 8.4 8.3 8.5 8.3 12.4 12.5 12.3 12.5 10.6 12.5 10.6 14.6 12.5 9.4 0.49 (0.40–0.61) 0.47 (0.36–0.62) 0.50 (0.34–0.72) 0.47 (0.36–0.60) 0.61 (0.39–0.95) 0.45 (0.36–0.57) 0.68 (0.32–1.46) 0.66 (0.27–1.62) 0.46 (0.36–0.58) 0.55 (0.28–1.09) 0.20.5125 OCEANS: PFS subgroup analyses HR Aghajanian et al. ESMO 2012

10. Median PFS (months) Baseline risk factor No. of patients CG + Pl (n=242) CG + Bev (n=242)HR (95% CI) CG + Bev better CG + Pl better Recurrence since last platinum therapy (mo) SLD of target lesions (median=59mm) CA-125 (U/mL) Prior biologic therapy Prior hormonal therapy 6–<12 12–24 >24 ≤Median >Median ≤35 >35 Yes No Yes No 171 209 104 244 240 120 338 10 474 22 462 7.4 8.6 11.6 8.5 8.4 10.2 8.3 10.2 8.4 8.3 8.4 12.5 12.3 16.6 12.6 11.4 12.5 12.3 22.8 12.3 18.6 12.4 0.36 (0.25–0.53) 0.52 (0.37–0.72) 0.62 (0.38–1.01) 0.49 (0.36–0.66) 0.48 (0.35–0.66) 0.51 (0.32–0.80) 0.47 (0.37–0.61) 0.00 (0.00–NE) 0.51 (0.41–0.63) 0.68 (0.25–1.83) 0.48 (0.38–0.60) 0.20.5125 HR OCEANS: PFS subgroup analyses NE, not estimable; SLD, sum of longest diameters Aghajanian et al. ESMO 2011

11. 100 80 60 40 20 0 OCEANS: objective response Aghajanian et al. ASCO 2011 Duration of response CG + Pl (n=139) CG + Bev (n=190) Median, months 7.410.4 HR (95% CI)0.534 (0.408–0.698) p<0.0001 a % 78.5 57.4 PR = 61 PR = 48 CR = 17 CR = 9 Difference: 21.1% p<0.0001 a Compared for descriptive purposes only CG + Pl (n=242) CG + Bev (n=242)

12. OCEANS: AEs leading to study drug discontinuation CG + Pl % (n=233) CG + Bev % (n=247) Timing of study drug discontinuation During concurrent phase3.4 (8)14.6 (36) During single-agent phase1.3 (3)5.3 (13) Adverse events Hypertension03.6 Proteinuria02.4 GI disorders (not GIP, fistula, abscess)0.92.4 Thrombocytopenia0.91.6 RPLS01.2 Epistaxis01.2 Skin disorders01.2 GI=gastrointestinal; GIP=gastrointestinal perforation; RPLS=reversible posterior leukoencephalopathy syndrome Nycum et al. ESGO 2011

13. OCEANS: AEs of special interest ATE=arterial thromboembolic event; CHF=congestive heart failure; GI=gastrointestinal; RPLS=reversible posterior leukoencephalopathy syndrome; VTE=venous thromboembolic event a Two GI perforations occurred 69 days after last BV dose Patients, % CG + Pl (n=233) CG + Bev (n=247) ATE, all grades 13 VTE, grade ≥3 34 CNS bleeding, all grades <11 Non-CNS bleeding, grades ≥3 16 CHF, grades ≥3 11 Neutropenia, grade ≥3 5658 Febrile neutropenia, grade ≥3 22 Hypertension, grade ≥3 <117 Fistula/abscess, all grades <12 GI perforation, all grades 00a0a Proteinuria, grade ≥3 19 RPLS, all grade 01 Wound-healing complication, grades ≥301 Aghajanian et al. ASCO 2011

14. OCEANS: conclusions Bevacizumab + carboplatin + gemcitabine followed by bevacizumab until progression provides a clinically meaningful benefit over chemotherapy alone in recurrent OC –improved PFS: HR 0.484 (p<0.0001); median 8.4 → 12.4 months –improved ORR and duration of response –OS data not yet mature Safety data consistent with bevacizumab profile –no GI perforations and no new safety signals This regimen should be considered a new option for recurrent platinum- sensitive OC