• AMD Meta-Analysis: Adverse Events With Bevacizumab and Ranibizumab

    Written By: Jean Shaw
    Selected By: Andrew P. Schachat, MD

    Journal Highlights

    Ophthalmology Retina, September/October 2017

    Download PDF

    Ten years into the anti-VEGF era, Maguire et al. reviewed data on the efficacy and safety of bevacizumab and ranibizumab in patients who received intravitreal injections of the drugs for neovascular age-related macular degeneration (AMD). They found no substantial difference in the risk of serious systemic adverse events (AEs) between the 2 medications.

    For this meta-analysis, the authors identified 6 large-scale multicenter clinical trials and requested individual patient data on serious AEs, assigned drug and dosing regimen, and baseline prognostic factors from the investiga­tors. A 2-stage approach was used to estimate relative risks (RRs) and 95% confidence intervals (CIs) from Cox proportional hazards models adjusting for baseline prognostic factors. The primary outcome measure was the de­velopment of ≥ 1 serious systemic AE; secondary outcome measures included risk of arteriothrombotic events and death.

    Data were received from 5 trials, for a total of 3,052 patients. Of these, 1,513 received bevacizumab, while 1,539 were given ranibizumab. All told, 403 (26.6%) of the bevacizumab group and 366 (23.8%) of the ranibizumab group experienced ≥ 1 serious AE. The adjusted RR (95% CI) for bevacizumab relative to ranibizumab was 1.06 for ≥ 1 serious AE. For secondary outcomes, adjusted RRs were 0.99 for death, 0.89 for arteriothrombotic events, 1.10 for events related to anti-VEGF treatment, and 1.11 for events not related to anti-VEGF treatment.

    At this point in time, the authors wrote, it is unlikely that there will be any additional head-to-head clinical trials of the 2 drugs. Thus, they said, further investigation of differential risk between anti-VEGF drugs will likely be achieved via postmarketing surveillance or in-depth examination of health care databases.

    The original article can be found here.