530PD - Capecitabine (CAPE) +/- Bevacizumab (BEV) in the Adjuvant Treatment of Colorectal Cancer (CRC) - First and Final Toxicity Results from the International Phase III QUASAR2 Trial

Abstract

ABSTRACT Background Bevacizumab is an anti-VEGF antibody approved for use in metastatic colorectal cancer (CRC). However 2 trials have suggested lack of efficacy in combination with dual agent chemotherapy in the adjuvant setting and there has been some indication that the spectrum of toxicity may be different compared to that observed in patients with advanced disease. Methods The international phase III trial, QUASAR2, randomised stage III and high risk stage II CRC patients who had undergone potentially curative resection to receive oral Cape (1250mg/m2 bd, days 1-14 q 21d) alone for 6/12 or Cape (6/12) plus intravenous Bev(7.5mg/kg q 21d) for 12/12. This abstract overviews the final toxicity data (15/12 minimum follow up per patient) from the Q2 trial. Results 1952 patients were consented and randomised and, after excluding patients who did not receive at least one cycle of treatment, 963 patients receiving Cape alone and 959 patients receiving Cape plus Bev were analysed. Baseline characteristics (gender / age / stage of disease) were balanced across the 2 arms. See frequency of selected toxicities, as % patients, in table below. Discussion As expected, the rates of hypertension (all grade or high grade) and proteinuria and wound healing (all grade) were significantly higher among patients receiving Bev. Although the addition of Bev did not increase diarrhoea rates, the frequency and severity of HFS was increased. This is not well recognised. Although cardiac chest pain was not increased by the addition of Bev, both arterial and venous thrombo-embolism (ATE/VTE) rates were increased although the difference only reached significance for VTE in this study. This is in contrast to most results in the advanced disease setting and may be a result of an interaction between post-operative risk of VTE and pro-thrombotic potential of Bev. Of note, an excess of ‘possibly treatment-related’ deaths was found in patients receiving Bev (1.9% vs 0.8%:RR2.3:CI 1.0-5.2) and this just reached significance (p = 0.05). All results will be expanded in detail in the presentation / poster. Cape/ Bev Cape RR/CI/p Hypertension (all grades) 33.4 7.8 4.3/3.4-5.4/p Hypertension (g3/4) 3.8 0.6 6.0/2.6-14.2/p Proteinuria (all grades) 20.5 5.1 4.0/3.0-5.4/p Poor wound healing (all grades) 3.1 1.8 1.8/1.0-3.2/p = 0.05 Diarrhoea (g3/4) 10.8 10.6 1.0/0.8-1.3/p = 0.9 Hand-foot syndrome (g3/4) 26.8 20.9 1.3/1.1-1.5/p = 0.002 Cardiac chest pain SAE 2.7 2.6 1.0/0.6-1.8/p = 0.9 Arterial Thromboembolism SAE 1.1 0.6 1.8/0.7-5.0/p = 0.2 Venous Thromboembolism SAE 4.3 2.3 1.9/1.1-3.1/p = 0.01 Disclosure R.S. Midgley: Auther has received no strings attached educational unrestricted grant for clinical research and has sat on advisory boards for Roche. V. Potter: Auther has received a travel grant and has sat on an advisory board for ROche. E. Segelov: Author has sat on an advisory board for Roche Autralia. D.R. Ferry: Auther has received travel grants, research grants and has sat on advisory boards for Roche. S. Grumett: Auther has received traval grants and has sat on advisory boards for Roche. D.J. Kerr: Auther has received no strings attached educational unrestricted grant for clinical research and has sat on Advisory Boards for ROche. All other authors have declared no conflicts of interest.