665 Background: Clinical trials support the addition of bevacizumab (Bev) to first line chemotherapy for metastatic colorectal cancer (mCRC). However, given that there are alternate biological agents that can be used with chemotherapy; and rare but potentially serious toxicities with Bev, there has been variable use of the drug in routine care. The purpose of this study was to prospectively explore 1st line use of Bev in routine clinical care. Methods: A multi-site data collection was initiated in July 2009 when Bev became publicly available in Australia. A particular focus was capturing information regarding clinician decision-making that is not routinely recorded or is often poorly documented. Comprehensive comorbidity data was collected, along with specific factors, both patient (pt) and disease-related, that may impact on treatment decisions. Major adverse events that may be related to Bev were also recorded. Results: For the 2-year period from July 2009-June 2011, a database search identified 278 pts diagnosed with mCRC at five participating hospitals. 240 pts (86%) had received 1st line chemotherapy. 102 (43%) also received Bev, the majority in combination with oxaliplatin based chemotherapy (n = 81, 79%). Bev use increased significantly from 40.4% (34/84) of all pts receiving chemotherapy in July-December 2009, to 65.8% (25/38) in January-June 2011 (p=0.0114). Overall the most frequent reasons for clinicians advising against the use of Bev were a bleeding or asymptomatic primary in situ (n=28, 23%), known hypertension (n=14, 11%), clinical trial participation (n=14, 11%), poor performance status (n=9, 7%), and perceived risk of arterial ischemic event (n=8, 6.5%). There have been four episodes of gastro-intestinal perforation, 3 in pts receiving Bev (2.9% of Bev treated patients). Conclusions: Bev use in first line mCRC has significantly increased over time, presumably as further positive safety data emerges regarding safety in older and frailer pts, and clinician prescribing experience increases. The major grounds for not using Bev in the 1st line mCRC setting were specific clinical contexts where the risk of adverse events may be increased.