Long-term visual course after anti-VEGF therapy for exudative AMD in clinical practice evaluation of the German reinjection scheme

Background In phase III trials, the therapeutic efficacy of anti-VEGF therapy with ranibizumab (Lucentis) in patients with choroidal neovascularization due to AMD was demonstrated in a 24-month period with monthly injections. Other studies and models suggested that flexible reinjection regimens can provide similar visual results. The aim of the present study was to evaluate the flexible, predominantly visual acuity-driven ranibizumab retreatment regimen in clinical practice in Germany. Patients and methods Best-corrected visual acuity (VA, logMAR) and central retinal thickness (CRT) were recorded initially and every 4-6 weeks during follow-up (mean follow-up 75.5 weeks) from 152 eyes. All eyes were treated initially 3 times with ranibizumab at 4-weekly intervals, and retreated with another three injections if visual acuity decreased and/or CRT increased (> 100 mu m), and/or if new angiographic leakage and/or new retinal hemorrhages developed. Visual acuity development was analyzed in the whole group. A quartile analysis was also performed, and visual course was correlated with CRT. In all groups, numbers and times of reinjections within the first year were registered and analyzed. Results An increase in mean VA of 0.14 (SD 0.22) logMAR could be observed after 3 months, but during follow-up from months 3 to 12 the mean visual acuity decreased again by 0.14 (SD 0.24) logMAR, and was similar to the initial VA despite several reinjections (mean five injections). Stratification of patients according to the visual effect after 3 months (quartile analysis) demonstrated a differentiation of the visual course. Quartile 1, with the largest increase in VA after 3 months and reduction of the retinal edema, lost this positive effect during follow-up (100% of eyes received further injections). In contrast, quartile 2, with a minor increase, and quartile 3 demonstrated a stabilized response during follow-up (80% reinjections), while quartile 4 demonstrated a further loss in VA despite reinjections initially and during follow-up (60% reinjections). Conclusions The flexible, predominantly visual acuity-driven ranibizumab retreatment regimen employed in clinical practice in Germany generally resulted in a loss of initially gained VA during 12 months of follow-up. Subgroup analysis showed that this negative effect was especially present in patients with relatively bad VA at treatment entry as well as the highest visual gain. Because this result demonstratse that a visual acuity-related retreatment regimen can not preserve the initial positive treatment effects with ranibizumab in exudative AMD, a revision of this schematic retreatment regimen used in Germany and adaptation to more sensitive retreatment parameters is recommended.