565P - Bevacizumab (BEV) + chemotherapy (CT) beyond first progression in patients (pts) with metastatic colorectal cancer (mCRC) previously treated with BE...

Date 01 October 2012
Event ESMO Congress 2012
Session Poster presentation III
Topics Anticancer agents
Colon and Rectal Cancer
Biological therapy
Presenter Jose Maria Vieitez de Prado
Authors J.M. Vieitez de Prado1, C. Borg2, D. Arnold3, R. Greil4, E. Van Cutsem5, R. von Moos6, J. Bennouna7, I. Reyes-Rivera8, B. Bendahmane9, S. Kubicka10
  • 1Medical Oncology, Hospital Central de Asturias (HUCA), Oviedo/ES
  • 2Medical Oncology, University Hospital Besançon, Besançon/FR
  • 3Medical Oncology, Hubertus Wald Tumor Center, University Cancer Center Hamburg (UCCH), Hamburg/DE
  • 4Oncology Centre, III Medizinische Universitätsklinik Salzburg, Salzburg/AT
  • 5Digestive Oncology, University Hospital Gasthuisberg, Leuven/BE
  • 6Medizinische Onkologie Und Hämatologie, Kantonal Hospital Graubünden, Chur/CH
  • 7Medical Oncology, Institut de Cancérologie de l'Ouest, Nantes/FR
  • 8Statistics, Genentech Inc., South San Francisco/US
  • 9Oncology, F. Hoffmann-La Roche, Basel/CH
  • 10Internal Medicine, Cancer Center Reutlingen, Reutlingen/DE



ML18147 evaluated the benefit of continuing BEV + standard CT as second-line (2L) treatment for pts with mCRC progressing after first-line (1L) BEV-containing therapy. Here we report results of pre-specified subgroup and exploratory KRAS mutation analyses.


Pts with unresectable, histologically confirmed mCRC progressing within 3 mo after discontinuing 1L BEV were randomised to 2L fluoropyrimidine + oxaliplatin or irinotecan (crossed over from 1L) ± BEV (2.5 mg/kg/wk equivalent). The primary endpoint was overall survival (OS). Subgroup analyses for OS were performed using the same statistical method as for the primary analysis.


409 pts were randomised to BEV + CT and 411 to CT (1 pt not treated). Median OS was 11.2 mo for BEV + CT vs 9.8 mo for CT (unstratified HR = 0.81; 95% CI 0.69–0.94; p = 0.0062). Subgroup analyses for OS were generally consistent with the overall population (Table). While the treatment effect in female pts appeared to be lower, the treatment-gender interaction test was not statistically significant.

Category Subgroup N HR for OS 95% CI
All All 819 0.81 0.69–0.94
Gender FemaleMale 294525 0.990.73 0.77–1.280.60–0.88
Age <65 y ≥ 65 y 458361 0.790.83 0.65–0.980.66–1.04
ECOG PS 0 ≥ 1 357458 0.740.87 0.59–0.940.71–1.06
First-line PFS ≤9 mo > 9 mo 449369 0.890.73 0.73–1.090.58–0.92
First-line chemotherapy Oxaliplatin-basedIrinotecan-based 343476 0.790.82 0.62–1.000.67–1.00
Time from last BEV dose ≤42 d > 42 d 630189 0.820.76 0.69–0.970.55–1.06
Liver metastases only NoYes 592226 0.810.79 0.67–0.970.59–1.05
No. of organs with metastasis 1 > 1 307511 0.830.77 0.64–1.080.64–0.94

KRAS mutation data were available from an exploratory analysis in 616 pts (75%); median OS for KRAS wild-type (WT) pts was 15.4 mo for BEV + CT vs 11.1 mo for CT (HR = 0.69, 95% CI 0.53–0.90; p = 0.0052); in KRAS mutant (MT) pts median OS was 10.4 vs 10.0 mo, respectively (HR = 0.92; 95% CI 0.71–1.18; p = 0.4969). Median PFS for KRAS WT pts was 6.4 mo for BEV + CT vs 4.5 mo for CT (HR = 0.61; 95% CI 0.49–0.77; p < 0.0001); in KRAS MT pts median PFS was 5.5 vs 4.1 mo, respectively (HR = 0.70; 95% CI 0.56–0.89; p = 0.0027). No treatment interaction by KRAS status was seen for OS (p = 0.1266) or PFS (p = 0.4436).


ML18147 showed that BEV + CT continued beyond progression significantly improves survival vs CT alone. Findings from the subgroup analyses for OS were generally consistent with the overall population.


J.M. Vieitez de Prado: Involved in corporate-sponsored trials for Roche.

D. Arnold: Consultant / Advisory Board: Roche. Honoraria: Roche. Research funding: Roche.

R. Greil: Honoraria: Roche Research support: Roche.

E.J.D. Van Cutsem: Research funding: Roche.

R. von Moos: Consultant/advisory board: Roche. Honoraria: Roche. Research funding: Roche.

J. Bennouna: Consultant / advisory board: Roche. Honoraria: Roche.

I. Reyes-Rivera: Employed by Genentech Inc.

B. Bendahmane: Employed by F. Hoffmann-La Roche.

S. Kubicka: Consultant / advisory board: Roche. Honoraria: Roche.

All other authors have declared no conflicts of interest.