Abstract
Purpose:
Adverse events involving the gastrointestinal (GI) tract have been identified as an area of concern in studies where bevacizumab was administered systemically. The purpose of this study was to synthesize the evidence on the frequency of reported GI events in head-to-head randomised controlled trials (RCTs) of ranibizumab versus bevacizumab administered intraocularly for treatment of age-related choroidal neovascularisation.
Methods:
We collated the results from five recently conducted comparative effectiveness RCTs of ranibizumab versus bevacizumab and undertook a meta-analysis to quantify the incidence and risk of a GI event.
Results:
Data from the CATT, IVAN, Gefal, MANTA and LUCAS RCTs were included. Collectively these trials randomised 1533 patients to ranibizumab and 1496 to bevacizumab. CATT and IVAN followed patients for two years; one year results were available for the Gefal, MANTA and LUCAS trials. On pooling the results from the five trials, 24 (1.6%) patients randomised to ranibizumab and 46 (3.1%) randomised to bevacizumab reported a GI serious adverse event (SAE) during follow-up. The risk of a GI SAE was estimated to be almost double with bevacizumab compared to ranibizumab (relative risk 1.94, 95% CI 1.20 to 3.14), which was statistically significant at the 5% level (p=0.007). Information published in the CATT and IVAN trials on the nature of the SAEs revealed no clear patterns, with vomiting, abdominal pain, intestinal obstruction and perforation all reported.
Conclusions:
The risk of a GI related SAE is significantly higher with bevacizumab compared to ranibizumab, but the overall incidence is low. The nature of the SAE is diverse and the severity is variable.
Keywords: 412 age-related macular degeneration •
462 clinical (human) or epidemiologic studies: outcomes/complications •
754 visual acuity