Author: Pietro Monaco
Co-authors: Michele Del Borrello, Matteo Airaldi, Luigi Tozzi, Domenico De Robertis, Luigina Tollot, Alessandro Invernizzi
Abstract
Purpose:To compare treatment efficacy of anti-Vascular Endothelial Growth Factor (VEGF) medications following pro re nata (PRN, “as needed”, monthly injections only in case of active disease) or treat and extend (T&E, progressive extension of treatment intervals up to 12 weeks depending on the clinical findings) treatment protocols in real-practice conditions over a period of 24 months.
Setting/Venue:
Eighty-nine treatment-naïve eyes from 78 exudative age-related macular degeneration patients enrolled in the Fight Retinal Blindness! registry were identified and retrospectively reviewed between January 2017 and December 2019 (47 eyes) (PRN group) and between January 2020 and January 2022 (42 eyes) (T&E group).
Methods:
Patients with different choroidal neovascularization (CNV) lesions (I, II, III, PVC) receiving anti-VEGF treatment with either PRN or T&E were included; subjects having a follow-up of at least 24 months were defined as completers.
The primary outcome was visual acuity (VA) change at 2 years, other outcomes included baseline characteristics, number of intravitreal anti-VEGF injections, proportion of active visits and incidence of macular atrophy and subretinal fibrosis.
Calculation of visual outcomes over 24 months used the last-observation-carried-forward (LOCF) for non-completers.
Descriptive data were summarized using the mean (SD), median (IQR), and number (%) where appropriate. Two-tailed t-tests, Mann-Whitney U tests, Fisher's exact tests and Chi-square tests were used to compare baseline characteristics and crude outcomes between PRN and T&E patients.
Generalized linear, logistic, Poisson and Cox proportional mixed-effects models were used to compare VA change at 24 months, number of injections, proportion of active visits and incidence of macular atrophy and subretinal fibrosis. All regression models were adjusted for activity pattern (mostly inactive, mostly active [subretinal fluid only] and mostly active), baseline age and baseline VA (fixed effects), and nesting of outcomes for bilateral cases (random effects).
Results:
Among 89 eyes investigated, 32 (68%) PRN and 12 (29%) T&E have completed 2 years of follow-up.
Baseline VA was not significantly different between groups (median [IQR], 65 [50, 73.5] vs. 59 [43.5, 70] LogMAR letters, p = .24); other baseline variables were comparable.
Final VA using LOCF for non-completers was similar (median [IQR], 65 [55, 75] vs. 62 [35, 71.5] LogMAR letters, p = .35), as well as adjusted VA change from baseline (mean [95% CI], 0.6 [-4.4, 5.7] vs. -2.2 [-7.4, 3] LogMAR letters, p = .44).
Patients in the T&E group received significantly more injections (median [IQR], 7.5 [5, 9.2] vs. 10.5 [8.8, 12], p < .001) and were less frequently graded as active (66% vs 61% of visits, p < .001).
T&E regimen was significantly associated with the development of subretinal fibrosis compared to PRN (31% vs 63% at 2 years, HR = 4, p < .001) but not of macular atrophy (12% vs. 52% at 2 years, HR = 1.8, p = .28).
Less active lesions (mostly SRF only and mostly inactive) were less likely to develop macular atrophy and more likely to develop subretinal fibrosis than mostly active ones, but these results did not reach statistical significance.
Conclusions:
In our clinical practice the T&E “pro-active” protocol warranted a reduction of the ophthalmic evaluations number improving the patient’s compliance. The mean change in VA between the two groups was similar with a dropping in VA after 18 months in the T&E group. The VA trend was more stable with less fluctuations until 18 months in the T&E group. Lastly, following the hypothesis that subretinal fluid could protect against macular atrophy and extrafoveal subretinal fibrosis, the greatest rate of subretinal fibrosis in T&E arm could be related to the less tolerance of subretinal fluid accumulation with this kind of regimen.
