513P - Impact of tumour site on bevacizumab efficacy in metastatic colorectal cancer (mCRC)

Date 29 September 2014
Event ESMO 2014
Session Poster Display session
Topics Anti-Cancer Agents & Biologic Therapy
Colon Cancer
Rectal Cancer
Presenter Hui-li Wong
Citation Annals of Oncology (2014) 25 (suppl_4): iv167-iv209. 10.1093/annonc/mdu333
Authors H. Wong1, K. Field2, A.J. Lomax3, M. Tacey4, J. Shapiro5, J. McKendrick3, A. Zimet6, D. Yip7, L. Nott8, R. Jennens6, G. Richardson5, J. Tie2, S. Kosmider9, P. Parente10, L. Lim11, P. Cooray11, B. Tran9, J. Desai2, R. Wong11, P. Gibbs2
  • 1Systems Biology And Personalised Medicine Division, Walter and Eliza Hall Institute, 3052 - Parkville/AU
  • 2Department Of Medical Oncology, Royal Melbourne Hospital, Parkville/AU
  • 3Department Of Medical Oncology, Eastern Health, Box Hill/AU
  • 4Melbourne Epicentre, University of Melbourne and The Royal Melbourne Hospital, Melbourne/AU
  • 5Department Of Medical Oncology, Cabrini Health, Malvern/AU
  • 6Department Of Medical Oncology, Epworth Hospital, Richmond/AU
  • 7The Canberra Hospital And The Anu Medical School, Australian National University, Canberra/AU
  • 8Department Of Medical Oncology, Royal Hobart Hospital, Hobart/AU
  • 9Department Of Medical Oncology, Western Hospital, Footscray/AU
  • 10Department Of Medical Oncology, Monash University, Faculty of Medicine, Nursing and Health Sciences, Eastern Health Clinical School, Box Hill/AU
  • 11Department Of Medical Oncology, Eastern Health, 3128 - Box Hill/AU

Abstract

Aim

With an ever-increasing focus on personalised medicine, all factors known to affect treatment response need to be considered when defining optimal therapy for individual patients (pts). While the prognostic impact of primary tumour site on CRC outcomes is established, emerging data suggest potential differences in response to biologic therapies. Here we explore the impact of tumour site on bevacizumab (BEV) efficacy in pts with mCRC.

Methods

Analysis of pts in an Australian prospective multicentre mCRC registry who received first-line palliative chemotherapy from July 2009 to March 2014. Tumour site was defined as: right colon (RCC)–caecum to transverse, left colon (LCC)–splenic flexure to rectosigmoid, and rectum (RC). Kaplan-Meier and Cox models were used for survival analyses.

Results

Of 698 pts, 224 had RCC, 266 LCC and 208 RC primaries. Median age was 69, 66 and 63 years respectively (p < 0.001). RCC was associated with peritoneal (32.1, 19.5 vs 6.7%, p < 0.001) and RC with lung disease (24.6, 26.7 vs 39.4%, p = 0.001). RC pts were most likely to have an intact primary tumour (22.9%, 32.7% vs 48.6%, p < 0.001). 457 (65.5%) pts received BEV: 68.8% of RCC and LCC and 57.7% of RC pts respectively, p = 0.019. BEV-treated pts were younger (median age 64 vs 70 years, p < 0.001) and had better performance status (PS0-1: 91.9 vs 80.1%, p < 0.001). Progression-free survival (PFS) was superior for BEV-treated pts in all groups, though not reaching statistical significance for RC (table). Overall survival was longest in pts with RC, followed by LCC and RCC (median: 28.7, 22.3 and 19 months, p < 0.001).

Median PFS (months) Bevacizumab vs no bevacizumab Hazard ratio* 95% CI P value
RCC 8.7 vs 5.6 0.60 0.42-0.85 0.004
LCC 11.8 vs 8.3 0.69 0.49-0.96 0.028
RC 13.5 vs 9.1 0.74 0.51-1.06 0.101

*Adjusted for age, sex, PS, comorbidity.

Conclusions

Tumour site was prognostic in this cohort, with RC and RCC associated with best and worst outcomes respectively. Patients who received bevacizumab in addition to chemotherapy had superior PFS, with the effect appearing greatest in RCC patients. Although known prognostic factors were adjusted for, other clinicopathologic differences that may contribute to patient outcomes cannot be excluded. Patient accrual is ongoing and molecular analyses are planned.

Disclosure

H. Wong, A.J. Lomax, M. Tacey, J. Shapiro, A. Zimet, D. Yip, R. Jennens, G. Richardson, J. Tie, S. Kosmider

P. Parente, L. Lim, P. Cooray, B. Tran, J. Desai and B. Wong: Roche Products Pty Ltd (Australia) has provided financial assistance for the development, installation and maintenance of this clinical database. K. Field and L. Nott declare: Roche Products Pty Ltd (Australia) has provided financial assistance for the development, installation and maintenance of this clinical database. I have also received honoraria and other renumeration from Roche. J. McKendrick: Roche Products Pty Ltd (Australia) has provided financial assistance for the development, installation and maintenance of this clinical database. I also have an advisory relationship with Roche. P. Gibbs: Roche Products Pty Ltd (Australia) has provided financial assistance for the development, installation and maintenance of this clinical database. I have also received honoraria from Roche.