The anti-VEGF antibody fragment ranibizumab is licensed for the treatment of neovascular age-related macular degeneration (wet AMD) and is approved by NICE for use in this indication, provided certain criteria are met. The purpose of the IVAN non-inferiority study (conducted in the UK) was to compare ranibizumab to bevacizumab (the parent molecule from which ranibizumab is derived), an unlicensed alternative that has been widely used due to the large cost savings.
One-year data from this study was previously reported; however this report includes the definitive two-year data (pre-specified primary efficacy analyses). It also reports a meta-analysis of the results from this study and the CATT study (a similar study conducted in the US). CATT and IVAN are the first and largest head-to-head trials that have compared ranibizumab and bevacizumab for wet AMD.
IVAN included 628 adults aged ≥50 years with active, previously untreated wet AMD, with a best-corrected distance visual acuity (BCVA) of at least 25 letters on a standard vision chart in the study eye. They were randomised to treatment with either ranibizumab (0.5mg) or bevacizumab (1.25mg), in either continuous (monthly) or discontinuous (injections at 0, 1 and 2 months; further 3-dose cycles were then given when required) regimens. The primary outcome was BCVA at 2 years (the non-inferiority margin was 3.5 letters).
Based on the results of a modified intention-to-treat analysis (n=610), bevacizumab was neither inferior nor non-inferior to ranibizumab in terms of BCVA at 2 years (mean difference of -1.37 letters; 95% CI -3.75 to 1.01; p=0.26), as the 95% confidence intervals included zero and the non-inferiority margin. The same applied to the comparison of continuous and discontinuous treatment schedules (mean difference of -1.63 letters; 95% CI -4.01 to 0.75; p=0.18). The mean differences in BCVA between all groups were small and estimated to be within 2.4 letters. The authors note that non-inferiority of bevacizumab to ranibizumab would have been demonstrated if they had used the 5 letter non-inferiority margin set in the CATT study.
BCVA point estimates from a pooled analysis of data from the IVAN and CATT studies (the latter included slightly different PRN regimens) showed that bevacizumab was non-inferior to ranibizumab according to the 3.5 letter non-inferiority margin (weighted mean difference of -1.15 letters; 95% CI -2.82 to 0.51) and that the discontinuous regimen was inferior to the continuous regimen (difference of -2.23 letters; 95% CI -3.93 to -0.53).
The results of the meta-analyses suggest no difference in mortality or arterial thrombotic events between the two drugs. Although bevacizumab does appear to be associated with an increased risk of any systemic severe adverse event (odds ratio favouring ranibizumab of 0.76; 95% CI 0.63-0.93; p=0.008), this finding was inconsistent between the trials (only CATT demonstrated a statistically significant difference).
Analyses failed to identify any differences between drugs in terms of development of new geographic atrophy; however an increased risk was seen for monthly versus discontinuous treatment, which raises the possibility than any visual benefit from monthly benefit may not be maintained in the long-term. The authors say that the finding of greatest concern is that there was a significant difference in mortality favouring continuous treatment – this seems counterintuitive based on conventional dose-response considerations and the authors suggest that the possibility of immunological sensitisation with intermittent dosing needs to be considered.
The authors conclude that the choice of anti-VEGF treatment strategy for wet AMD is less straightforward than previously thought. Although the results of the two comparisons made in this study are inconclusive, non-inferiority of bevacizumab was demonstrated when results were pooled with those from the CATT study. Although the pooled analysis also showed that discontinuous treatment was inferior to continuous treatment, this was not considered to be clinically important. The trade-offs of an observed increased risk of mortality for discontinuous treatment and an increased risk of geographic atrophy with continuous treatment should be discussed with patients (and may favour continuous administration).