Study design
This was a non-randomized, retrospective, observational single center study of treatment-naïve eyes for wet-AMD. Anonymous data were collected from the drug program dedicated electronic database. Records from December 2015 to December 2020 were analyzed.
Data included the assessment of the following: the best corrected visual acuity (BCVA) on a decimal scale made on the basis of a Snellen chart, macular morphology with automatic measurement of the central retina thickness (CRT) from the central subfield of the optical coherence tomography (OCT); percentage share of active area in degenerative lesion, size of degenerative lesion (DA). Type of neovascularization was determined with fluorescein angiography (FA). All OCT scans were performed on certified spectral domain OCT (Spectralis OCT, Heidelberg Engineering, Germany). In addition, possible prior treatment with VEGF inhibitors, disease activity defined as the presence of sub- or intra-retinal fluid in OCT or a new hemorrhage were reported.
Demographic factors extracted from the database were age (years), gender, eye (right/left), date of diagnosis, dates and number of injections with VEGF inhibitors.
Treatment program
The following inclusion criteria were applied in the treatment program: 1) presence of active (primary degeneration, not secondary) occult (type 1), classic (type 2) or mixed (other than pure classic or occult, including retinal angiomatous proliferation – type 3) Subretinal choroidal neovascularization (CNV) occupying more than 50% of the lesion; 2) age \(\ge 45\) years 3) total size of degenerative lesion less than 12 optic disc area (DA); 3) BCVA in the treated eye of 0.1–0.8 (from 2015 to 2017); 0.2–0.8 (from 2017 to present), determined on a decimal scale according to the Snellen chart; 4) absence of dominant geographic atrophy or hemorrhage in the whole lesion (more than 50% of the lesion); 6) informed consent of the patient to undergo intravitreal injections. Patients with scarring or atrophy of the fovea were not eligible for the treatment.
BCVA deterioration to \(\le 0.05\) according to Snellen’s chart lasting longer than 2 months excluded patient from the program. In addition, physician could remove the patient from the program according to the exclusion criteria contained in summary of product characteristics.
Aflibercept or ranibizumab were chosen by the physician at the time of the first injection. Ranibizumab was mainly administrated in cases which were expected to improve rapidly, allowing for a prompt withdrawal from the therapeutic program. All eyes were treated with fixed regimen, according to the “Treatment of neovascular form of age-related macular degeneration” National Health Found program guidelines.6
Three monthly intravitreal injections of aflibercept (2.0 mg/0.05mL) were administrated and then drug was injected every 2 months. After the first 12 months of aflibercept treatment, a pro re nata (PRN) regimen was in force. In patients treated with ranibizumab, after administering the initial dose (0.5 mg/0.05mL) the PRN scheme was applied. Patients received injections monthly until the disease was no active and functional stability was achieved. Then, patients were followed-up for 4–8 weeks. Ranibizumab was given again, when active neovascularization recurred, and functional parameters worsened. Physician could switch drug if the patient met both of the following criteria: 1) BCVA after 7 injections of ranibizumab or aflibercept (since inclusion to the program) was worse than BCVA at baseline; 2) patient still met the inclusion criteria for therapeutic program. Switch from ranibizumab to aflibercept and vice versa was permitted. However, patients which drug was switched during the therapy were excluded from the analysis.
Clinical examinations, OCT and measurement of BCVA were performed at each follow-up visit and recorded in the electronic database. Patients with follow-up period \(cript>\) 12 months were excluded from the analysis. Only, patients with treatment-naïve wet AMD eyes and complete medical records were included in the study.
Changes after 12-month treatment.
The baseline BCVA and CRT were compared with those at 12 months within each treatment group.
Comparisons between the ranibizumab group and aflibercept group.
Eyes were divided into two groups: the ranibizumab (n = 33) and aflibercept groups (n = 44). These were compared in terms of baseline characteristics (BCVA, CRT, percentage share of active area in degenerative lesion, size of degenerative lesion, neovascularization type, age and sex). In addition, the BCVA, CRT and number of injections after 12 months of treatment were compared between groups.
To investigate difference between good visual outcome and poor visual outcome, eyes were divided into two groups on the basis of BCVA at 12 months. These with BCVA \(\ge\) 20/50 were assigned to the “good visual outcome” group. Patients with BCVA \(cript>\) 20/50 were included in the “poor visual outcome” group. Within each treatment group, the visual outcomes were compared in terms of baseline characteristics (mentioned above) and number of injections.
The patients were divided into two groups, according to the number of injections received within 12-month treatment period. Mean number of injections were calculated for ranibizumab and aflibercept groups. Patients, whose number of injections was \(\ge\) than mean in their drug group were compared against these, whose number of injections was \(cript>\) than mean in their drug group. The BCVA and CRT at baseline and at 12 months were compared between these groups.
For the purpose of the study, visual acuity was calculated from the Snellen decimal scale to the number of ETRDS letters.7 The CRT values were reported in µm.
Statistics
Statistical analyses were performed with open-source software JASP, version 0.14.1, https://jasp-stats.org. Data were collected with use of Microsoft Excel software, version 16.46, (Microsoft Corporation, Redmond, WA, USA). The normality of data was examined with Shapiro – Wilk test. Differences between two time points were analyzed using the Wilcoxon signed - rank test. To analyze differences involving parametric values between two groups, either the independent t test or the Mann – Whitney U test were utilized. These that concerned non-parametric values were analyzed using the chi-square test. P values < .05 were considered significant. Data were presented as mean ± standard deviation (SD), where applicable.