Werner Scheithauer Professor of Internal Medicine Cancer Center

Werner Scheithauer Professor of Internal Medicine, Cancer Center at the Vienna University Hospital, Austria

Capecitabine: the new standard chemotherapy in advanced gastric cancer? Werner Scheithauer University Hospital Vienna, Austria

Gastric cancer: a global disease Second most common cause of cancer mortality Worldwide: 934, 000 new cases and 700, 000 deaths/year Gastric cancer incidence 20/100, 000 10 to 20/100, 000 <10/100, 000 www. cancer. gov Kamangar F, et al. J Clin Oncol 2006; 24: 2137– 50

Stage at diagnosis Japan 1 17% 15% Western countries Resectable Locally advanced Metastatic 68% 2 Resectable Locally advanced 10– 15% 25– 30% 30– 35% Metastatic Unstaged 1 http: //www. ncc. go. jp/en/ncch/annrep/2000 2 sanofi-aventis Internal Epidemiology Data

Palliative chemotherapy for advanced gastric cancer (AGC) ü Non-measurable lesions ü Variable study inclusion criteria ü Heterogeneous patient populations ü Inadequate number of patients ü Response rather than survival

Chemotherapy improves survival in metastatic gastric cancer (MGC) Median OS (months) 12 Chemotherapy 10 Best supportive care (BSC) 8 6 4 2 0 FAMTX 1 BSC 1 (n=30) (n=10) FEMTX 2 BSC 2 (n=21) (n=20) OS = overall survival; 5 -FU = 5 -fluorouracil FAMTX = 5 -FU, doxorubicin, methotrexate FEMTX = 5 -FU, epirubicin, methotrexate ELF = etoposide, leucovorin, 5 -FU CLF = cisplatin, leucovorin, 5 -FU 1 Murad ELF 3 BSC 3 (n=31) (n=30) CLF 4 BSC 4 (n=51) (n=52) AM, et al. Cancer 1993; 72: 37– 41; 2 Pyrhonen S, et al. Br J Cancer 1995; 71: 587– 91; 3 Glimelius B, et al. Ann Oncol 1997; 8: 163– 8; 4 Scheithauer W, et al. Second International Conference on Biology, Prevention and Treatment of GI Malignancy, Koln, Germany, 1995; 68 (Abstract)

Survival has improved with conventional regimens BSC 1 4 months 5 -FU monotherapy 2, 3 7 months FAM 2, 4 6– 7 months FAMTX 5– 7 6– 7 months FP 2, 3, 6 and ECF 5, 8, 9 7– 9 months Median OS FAM = 5 -FU, doxorubicin, mitomycin C FP = 5 -FU, cisplatin; ECF = epirubicin, cisplatin, 5 -FU 1 Wagner AO, et al. Cochrane Database Syst Rev 2005; 2: CD 004064; 2 Kim NK, et al. Cancer 1993; 71: 3813– 18 3 Ohtsu A, et al. J Clin Oncol 2003; 21: 54– 9; 4 Wils JA, et al. J Clin Oncol 1991; 9: 827– 31 5 Waters JS, et al. Br J Cancer 1999; 80: 269– 72; 6 Vanhoefer U, et al. J Clin Oncol 2000; 18: 2648– 57 7 Cocconi G, et al. Ann Oncol 2003; 14: 1258– 63; 8 Ross P, et al. J Clin Oncol 2002; 20: 1996– 2004 9 Webb A, et al. J Clin Oncol 1997; 15: 261– 7

Chemotherapy in patients with MGC l Conventional regimens prolong survival and achieve acceptable response rates BUT. . . l Median OS still <12 months l Complete response rate is low l Response duration is short l Inconvenient drug administration l Regimens may be toxic l No standard treatment – all options have similar survival outcomes and safety profiles – FP/ECF most common reference therapy in Europe There is a clear need for new active treatments

New treatment options for AGC aim to improve survival Epirubicin Cisplatin Fluorouracil

New treatment options for AGC aim to improve survival Epirubicin? l Paclitaxel l Irinotecan l Docetaxel Oxaliplatin Capecitabine, S-1

Capecitabine: the new standard chemotherapy in AGC?

Can XP improve PFS versus FP? International phase III study in MGC (ML 17032) AGC and/or MGC (n=316) R A N D O M I S A T I O N n=156 XP Capecitabine 1, 000 mg/m 2 b. i. d. day 1 14 cisplatin 80 mg/m 2 day 1 every 3 weeks FP 5 -FU c. i. 800 mg/m 2 day 1 5 cisplatin 80 mg/m 2 day 1 every 3 weeks n=160 l Primary endpoint: non-inferiority in PFS = progression-free survival; b. i. d. = twice daily c. i. = continuous infusion Kang Y-K, et al. Ann Oncol 2006; 17(Suppl. 6): vi 19 (Abstract O-003)

Why should XP be more effective than FP? l FP is a reference regimen for AGC l Capecitabine monotherapy is effective and well tolerated in AGC 1– 4 – ORR: 19– 34%; SD: 30– 56% – median TTP: 2. 8– 4. 8 months – median OS: 8. 1– 10. 0 months l XP was effective and well tolerated in phase II study 5 – ORR: 55%; SD: 16. 7% – median TTP: 5. 8 months – median OS: 10. 1 months ORR = overall response rate SD = stable disease TTP = time to progression 1 Hong YS, et al. Ann Oncol 2004; 15: 1344– 7 2 Leon-Rodriguez E, et al. Ann Oncol 2002; 13(Suppl. 5): 191 (Abstract 708) 3 Sakamoto J, et al. Anticancer Drugs 2006; 17: 231– 6 4 Koizumi W, et al. Oncology 2003; 64: 232– 6 5 Kim TW, et al. Ann Oncol 2002; 13: 1893– 8

Primary endpoint met: XP improves PFS versus FP (HR=0. 81) Estimated probability 1. 0 XP (n=139) FP (n=137) 0. 8 0. 6 HR=0. 81 (95% CI: 0. 63– 1. 04) Compared to HR upper limit 1. 25, p=0. 0008 0. 4 0. 2 0 5. 0 0 2 4 5. 6 6 8 10 Per protocol HR = hazard ratio; CI = confidence interval 12 14 Months 16 18 20 22 24 26 Kang Y-K, et al. Ann Oncol 2006; 17(Suppl. 6): vi 19 (Abstract O-003)

PFS: robust results in subgroups Favours XP Favours FP PP population Prior chemotherapy 276 28 No prior chemotherapy 248 Male 185 Female 91 65 years 238 >65 years 38 KPS <80% 30 KPS 80% 246 1 metastatic site 2 metastatic sites 0. 2 0. 6 1. 0 1. 4 1. 8 HR (95% CI) KPS = Karnofsky performance status 2. 2 98 178 2. 6 Kang Y-K, et al. Ann Oncol 2006; 17(Suppl. 6): vi 19 (Abstract O-003)

XP versus FP also shows non-inferiority in OS (HR=0. 85) Estimated probability 1. 0 XP (n=139) FP (n=137) 0. 8 0. 6 HR=0. 85 (95% CI: 0. 64– 1. 13) Compared to HR upper limit 1. 25, p=0. 0076 0. 4 0. 2 0 9. 3 0 Per protocol 2 4 6 8 10. 5 10 12 14 16 18 20 22 24 26 28 30 32 Months Kang Y-K, et al. Ann Oncol 2006; 17(Suppl. 6): vi 19 (Abstract O-003)

Superior response rate with XP versus FP RECIST confirmed response (%) XP (n=160) FP (n=156) ORR 95% CI 41 (33– 49) 29 (22– 37) 0. 033 CR 2 3 0. 720 PR 39 26 0. 022 SD 37 42 0. 421 PD 10 18 0. 051 p value Intent-to-treat (ITT), investigators’ assessment RECIST = Response Evaluation Criteria in Solid Tumours CR = complete response PR = partial response Kang Y-K, et al. Ann Oncol PD = progressive disease 2006; 17(Suppl. 6): vi 19 (Abstract O-003)

Similar incidence of grade 3/4 adverse events: XP versus FP Grade 3/4 adverse events (AEs) 60 Patients (%) 50 XP (n=156) FP (n=155) 40 30 20 10 0 tro u Ne ia en p e ril ia b Fe pen tro u ne HFS = hand-foot syndrome / ea g s au itin N m vo a is it at S m to rr a Di e ho S HF ia en p o c eu L Kang Y-K, et al. Ann Oncol 2006; 17(Suppl. 6): vi 19 (Abstract O-003)

XP: a potential new standard chemotherapy in AGC l As effective as FP l Improved tolerability l Avoids inconvenience and complications associated with infused 5 -FU

Multicentre phase III REAL-2 trial: can capecitabine improve OS versus 5 -FU? Locally advanced or metastatic oesophagogastric cancer (n=1, 002) RANDOMISATION Epirubicin Cisplatin Fluorouracil Epirubicin Oxaliplatin Fluorouracil Epirubicin Cisplatin Oxaliplatin Capecitabine l Epirubicin 50 mg/m 2 day 1 l Cisplatin 60 mg/m 2 vs oxaliplatin 130 mg/m 2 day 1 l 5 -FU 200 mg/m 2 c. i. daily vs capecitabine 500– 625 mg/m 2 b. i. d. p. o. daily l For 24 weeks: eight cycles every 3 weeks p. o. = orally Cunningham D, et al. N Engl J Med 2008; 358: 36– 46

Why should EOC be more effective than ECF? l Capecitabine is effective and well tolerated in AGC l XP has become a new reference regimen for AGC l Oxaliplatin has shown promising activity in phase II trials Regimen n ORR (%) TTP OS (months) Reference FOLFOX 49 45 6. 2 8. 6 FU(inf)/Ox 55 56 5. 2 10. 0 Chao et al. 2004 FUFOX 48 54 6. 5 11. 4 Dyster et al. 2005 FOLFOX 4 61 38 7. 1 11. 2 De. Vita et al. 2005 FLO 41 43 5. 6 9. 6 Louvet et al. 2002 Al-Batran et al. 2005 EOC = epirubicin, oxaliplatin, capecitabine FOLFOX = 5 -FU, leucovorin, oxaliplatin; FU(inf)/Ox = 5 -FU infusion, oxaliplatin FUFOX = 5 -FU, oxaliplatin; FLO = 5 -FU, leucovorin, oxaliplatin

REAL-2: can capecitabine replace 5 -FU, can oxaliplatin replace cisplatin? l In combination treatments of oesophagogastric cancer – can capecitabine replace 5 -FU? – can oxaliplatin replace cisplatin? l Primary endpoint of non-inferiority in OS for capecitabine versus 5 -FU, oxaliplatin versus cisplatin (PP population) – based on ECF 1 -year survival of 35%, 1, 000 patients (250 per arm) needed to show non-inferiority (80% power, one-sided a=0. 05) – upper limit of HR for experimental arms compared with standard arms should be <1. 23 – secondary endpoint: superiority in OS among the four regimens (intent-to-treat) Cunningham D, et al. N Engl J Med 2008; 358: 36– 46

REAL-2: primary endpoint met – non-inferior OS with capecitabine versus 5 -FU Estimated probability 1. 0 Capecitabine combinations (n=480) 5 -FU combinations (n=484) 0. 8 0. 6 HR=0. 86 (95% CI: 0. 80– 0. 99) 0. 4 0. 2 0 9. 6 0 Per protocol 10. 9 12 24 36 Months 48 60 72 Cunningham D, et al. N Engl J Med 2008; 358: 36– 46

Similar OS with oxaliplatin versus cisplatin Estimated probability 1. 0 Oxaliplatin combinations (n=474) Cisplatin combinations (n=490) 0. 8 0. 6 HR=0. 92 (95% CI: 0. 8– 1. 1) 0. 4 0. 2 0 10. 4 0 Per protocol 12 24 36 Months 48 60 72 Cunningham D, et al. N Engl J Med 2008; 358: 36– 46

Superior OS with EOC versus ECF Median OS (months) Estimated probability 1. 0 ECF (n=249) EOC (n=239) 0. 8 9. 9 11. 2 0. 6 HR=0. 80 (95% CI: 0. 66– 0. 97) Log-rank p=0. 02 0. 4 0. 2 0 0 Intent-to-treat 12 24 36 Months 48 60 72 Cunningham D, et al. N Engl J Med 2008; 358: 36– 46

Capecitabine-based regimens are well tolerated 60 Grade 3 / 4 AEs ECF (n=236) Patients (%) 50 ECC (n=241) * 40 EOF (n=235) EOC (n=239) * * 30 20 * * 10 0 tro a ni e p e ril ia b Fe pen ro ut eu N *p<0. 05 compared with ECF ne / is ea g tit s a au itin m N m o St o v rr a Di a oe h * * ic ol s b t em ven bo e S HF m ro h T Cunningham D, et al. N Engl J Med 2008; 358: 36– 46 Starling N, et al. Proc ASCO GI 2007 (Abstract 74)

REAL-2 versus ML 17032: consistent efficacy with capecitabine regimens ORR (%) 46 41 48 42 XP 2 FP 2 OS* (months) 9. 9 11. 2 9. 3 41 29 ECC 1 ECF 1 EOC 1 EOF 1 PFS* (months) 6. 7 6. 2 7. 0 6. 5 5. 6 5. 0 10. 4 8. 9 l Response evaluated every 3 months in REAL-2, 1 every 1. 5 months in ML 170322 l In REAL-2, 1 maximum treatment duration: 6 months 1 Cunningham * * Intent-to-treat 2 Kang D, et al. N Engl J Med 2008; 358: 36– 46 Y-K, et al. Ann Oncol 2006; 17(Suppl. 6): vi 19 (Abstract O-003)

Capecitabine plus platinum-based chemotherapy approved in Europe BSC 1 FAMTX 2 FP 3 IF 4 ECF 5 EOF 5 DCF 3 XP 6 ECx 5 EOx 5 0 IF = irinotecan + 5 -FU 1 Murad 2 4 6 Months 8 10 12 AM, et al. Cancer 1993; 72: 37– 41 2 Vanhoefer U, et al. J Clin Oncol 2000; 18: 2648– 57; 3 Van Cutsem E, et al. J Clin Oncol 2006; 24: 4991– 7 4 Dank M, et al. J Clin Oncol 2005; 23(Suppl. 16 S): 403 s (Abstract 4003); 5 Cunningham D, et al. N Engl J Med 2008; 358: 36– 46; 6 Kang Y-K, et al. Ann Oncol 2006; 17(Suppl. 6): vi 19 (Abstract O-003)

Putting the evidence into practice: case study l 45 -year-old male; ECOG l l PS: 0 Chief complaint: epigastric pain No dysphagia or weight loss Gastroscopy: large ulceroinfiltrative lesion encircling the antral lumen Biopsy: tubular adenocarcinoma, M/D ECOG = Eastern Cooperative Oncology Group; PS = performance status

Abdomen and pelvis computed tomography (CT)

Patient received EOC chemotherapy Before treatment 22 August 2004 After cycle 4: ++PR 10 November 2004 After cycle 8: CR? After cycle 2: PR 01 October 2004 07 February 2005

CEA (ng/m. L) Carcinoembryonic antigen (CEA) levels 18 16 14 12 10 8 6 4 2 0 Upper limit of normal 04 t Au us g m 04 r be S e pt e ct O be o 04 r N m ve o be 04 r m ce De be y n Ja r ua 05 r eb F r ua y 05

Patient received EOC chemotherapy Before treatment After cycle 7 Biopsy: no cancer

Patient was in remission for 9 months without chemotherapy l Chemotherapy stopped in February 2005 l Patient followed-up every 3 months – abdomen and pelvis CT – gastrofiberscopy (GFS) – CEA

CT scans 22 September 2006 revealed disease progression

Patient received second-line chemotherapy l Re-induction of EOC – four cycles: SD (for 4 months) l Irinotecan 150 mg/m 2 day 1, 14 plus mitomycin C 8 mg/m 2 day 1 every 4 weeks 1, 2 – four cycles: SD (for 2 months) l Docetaxel 75 mg/m 2 every 3 weeks 3 – three cycles: PD 1 Giuliani F, et al. Am J Clin Oncol 2005; 28: 581– 5 A, et al. J Chemother 2003; 15: 275– 81 3 Lee JL, et al. Cancer Chemother Pharmacol 2008; 61: 631– 7 2 Bamias

4 December 2007: patient with PD, alive no longer receiving treatment CEA: 20. 9 ng/m. L

What other capecitabine-based options are available?

Integrating docetaxel in the treatment of AGC: high efficacy with capecitabine Efficacy DCX 1 (n=40) DCF 2 (n=227) ORR (%) 95% CI 68 52– 83 37 30– 43 TTP/PFS (months) 7. 8 5. 6 OS (months) 16. 9 9. 2 1 Kang Y-K, et al. J Clin Oncol 2004; 22(Suppl. 14 S): 329 s (Abstract 4066) 2 Van Cutsem E, et al. J Clin Oncol 2006; 24: 4991– 7

Capecitabine: the backbone of future standards in gastric cancer Trial Setting n Advanced disease XP or FP ± trastuzumab (TOGA) Capecitabine + cisplatin ± bevacizumab ECX FOLFIRI vs FOLFIRI ECX ± panitumumab (REAL-3) HER 2+ AGC 374 MGC 760 AGC/MGC 416 AOGC 660 Operable disease Perioperative ECX ± bevacizumab (STO-3) Adjuvant 1, 100 XELOX vs observation (CLASSIC) Adjuvant 1, 024 HER 2 = human epidermal growth factor receptor 2; AOGC = advanced oesophagogastric cancer XELOX = capecitabine + oxaliplatin

Conclusions l Capecitabine is effective and well tolerated – can replace 5 -FU in current treatment regimens for AGC l Further data will support the use of capecitabine in this indication l Capecitabine is an effective, safe and convenient oral therapy for gastric cancer