Ranibizumab treatment in patients with neovascular age-related macular degeneration and very low vision

Abstract

Editor, B ased on large clinical trials, treatment with antivascular endothelial growth factor (anti-VEGF) substances has become the standard treatment of choroidal neovascularisation (CNV) because of age-related macular degeneration (ARMD) (Gragoudas et al. 2004; Rosenfeld et al. 2006). However, information is limited regarding the effect of anti-VEGF treatment in patients with very low visual acuity (VA). Therefore, we evaluated the efficacy of ranibizumab in patients with ARMD and low initial VA. All patient cases treated with ranibizumab because of CNV because of ARMD with VA of 0.1 or less and with a follow-up period of at least 6 months were retrospectively reviewed. Test for VA was carried out as part of routine clinical setting using the Snellen Chart and on the ETDRS chart. All VA measurements were transformed to logMAR values for statistical analysis. Patients initially received three injections of 0.5 mg ranibizumab intravitrously approximately 4–6 weeks apart. At clinical follow-up, it is decided whether to continue treatment – based on signs of activity – or to reevaluate the patient again every other 4–6 weeks. The off-label use was discussed with patients, and informed consent was obtained. A total of 135 patients (135 eyes) (84 women and 51 men were included; median age 79 range 50–98). Two of the patients had previously been treated with photodynamic therapy. Test for normal distribution was carried out using Kolmogorow–Smirnoff test. Because the data were not normally distributed, nonparametric tests were performed and data are presented as median with range. For related samples, the Wilcoxon test was carried out and correlations were tested using Spearman Rank correlations. A significance level of p < 0.05 was employed. Median months of follow-up was 14 (range 6–27), and median number of treatments was 5.6 (range 2–14). We found a significant increase in median VA after treatment logMAR median 0.98 (range 0–2) (Snellen 0.14 ⁄ETDRS 40 letters) compared to VA before treatment logMAR median 1.1. (range 1–2) (Wilcoxon: p < 0.001) (Snellen 0.08 ⁄ETDRS 30 letters) (Table 1). There was a significant correlation between initial VA and final VA (p < 0.05, Spearman rho = 0.4). Fifty-four per cent of the patients improved in VA, 23% maintained their vision, and 23% lost vision. We found a significant decrease in central retinal thickness as measured by OCT initial median thickness 353 lm compared to central retinal thickness after treatment 264 lm (p < 0.001, Wilcoxon). No correlations regarding changes in VA and CNV subtype, total lesion size and ⁄or presence of PEDwere observed. Thirty-three patients had a VA of less than 20 ⁄ 230, lower than VA allowed in the clinical trials. In this group, we also found a significant increase in VA (Table 1). LogMAR VA before treatment was median 1.3 (range 1.1–2) (Snellen 0.05 ⁄ETDRS 20 letters) VA; and after treatment, it was logMAR median 1 (range 0.3–2) (Snellen 0.1 ⁄ETDRS 35 letters) (p < 0.001, Wilcoxon). In this group, 76% improved vision, 9% maintained vision and 15% lost vision. Our data suggests that ranibizumab may improve VA in patients with low VA. Two previous studies suggested that bevacizumab (Avastin ; Genentech, San Francisco, CA, USA) could be beneficial for patients with longstanding ARMD and ⁄or vision below 20 ⁄ 150 (Ehrlich et al. 2008), and in patients with VA of less than 20 ⁄ 150 (Spaide et al. 2006). A recent article studied the effect of bevacizumab in 44 patients with VA of 0.1 or less, showing that 57% of patients improved (Galbinur et al. 2009), comparable to the 54% we find. No study has reported on the effect of ranibizumab in CNV and low VA. Even though information regarding duration of symptoms and degree of subretinal fibrosis was unavailable, ranibizumab seems to improve VA in some patients with very low vision because of ARMD and CNV.