Intravitreal Faricimab for Retinal Vein Occlusion: Real-World Outcomes from the Multicenter FARTURK-RVO Study in Türkiye Authors: Fevzi Akkan, Ebru Görgün, Burak Erden, Yavuz Kamil Bardak, Nigar Şerif, Mahmut Öztürk, Erkan Çelik, Selim Bölükbaşı, Şefik Can İpek, Tahsin Uzundede, Gamze Karataş, Murat Karapapak, Ece Özal, Osman Murat Uyar, Doğukan Cömerter, Mehmet Numan Alp, Serhat Ermiş, Anıl Korkmaz, Murat Arıcı, Mehmet Egemen Karataş, Nursal Melda Yenerel, Halil Özgür Artunay, Ecem Önder-Tokuç, Veysel Levent Karabaş, Melih Ünal, Ziya Kapran, Hakan Özdemir, Akın Çakır Published: 2026 ABSTRACT AbstractIntroductionThe aim of this study was to evaluate the real-world efficacy and safety of intravitreal faricimab for macular edema (ME) secondary to retinal vein occlusion (RVO) across a multicenter cohort with up to 52 weeks of follow-up.MethodsData from 17 centers across Türkiye were retrospectively analyzed between 22 November 2024 and 1 March 2026. Patients with branch RVO (BRVO) or central RVO (CRVO) complicated by ME who received at least one intravitreal faricimab injection and had complete ophthalmic and optical coherence tomography (OCT) data with a minimum 4-week follow-up were included. Demographics, diagnosis, prior treatments, best-corrected visual acuity (BCVA; decimal converted to logMAR), central macular thickness (CMT), intraocular pressure (IOP), injection number, follow-up duration, and ocular/systemic adverse events were recorded at baseline, day 7, monthly after each of the first three injections, and at the final visit.ResultsA total of 70 patients (37 BRVO [52.9%], 33 CRVO [47.1%]) were included, with a mean follow-up of 23.69 ± 8.54 weeks; 44 (62.8%) were treatment-naïve and 26 (37.2%) were switch patients. When comparing baseline and final visits, treatment-naïve patients showed a mean CMT reduction of 306.9 ± 231.3 µm (from 594.1 ± 206.9 µm at baseline to 287.2 ± 89.9 µm at final visit; p < 0.001), while switch patients demonstrated a mean CMT reduction of 290.6 ± 114.3 µm (from 562.1 ± 107.3 to 271.5 ± 90.7 µm; p < 0.001). The mean logMAR BCVA gain was 0.77 (p < 0.001) in naïve and 0.38 (p < 0.001) in switch patients. IOP remained stable throughout, and no serious ocular or systemic adverse events were recorded.ConclusionsFaricimab demonstrated rapid anatomical and functional improvements in the RVO cohort, evident as early as day 7 after the initial injection. These findings support faricimab as a potent and reliable therapeutic option across both RVO subtypes in routine clinical practice.Supplementary InformationThe online version contains Supplementary Material available at 10.1007/s40123-026-01436-8.Keywords: Angiopoietin-2, Anti-VEGF, Branch retinal vein occlusion, Central retinal vein occlusion, Faricimab, Macular edema, Multicenter study, Real-world study, Retinal vein occlusion Key Summary Points Why carry out this study?Real-world evidence on faricimab for retinal vein occlusion (RVO)-associated macular edema is limited, as clinical trial outcomes may not fully reflect routine practice outcomes in heterogeneous patient populations.This multicenter study evaluated short-to-mid-term functional and anatomical outcomes of faricimab in 70 patients with branch RVO (BRVO) or central RVO (CRVO) across 17 centers in Türkiye, comparing treatment-naïve and previously treated (switch) patients.What was learned from the study?Faricimab produced significant improvements in best-corrected visual acuity and central macular thickness from as early as day 7 post-injection in both patients with BRVO and CRVO, regardless of prior anti-VEGF exposure.After adjustment for baseline differences, functional gains did not differ significantly between treatment-naïve and switch patients, supporting comparable effectiveness across treatment histories; no serious adverse events were observed, though rare inflammatory events cannot be adequately assessed in a cohort of this size and follow-up duration. Introduction Macular edema (ME) represents the most common and vision-threatening complication of retinal vein occlusion (RVO), the second most prevalent retinal vascular disorder after diabetic retinopathy. RVO is broadly classified into branch (BRVO) and central (CRVO) forms according to the anatomical site of venous obstruction, and both subtypes can lead to significant visual impairment through the accumulation of intraretinal and subretinal fluid [1–3]. The pathophysiology of RVO-associated ME is multifactorial, involving venous stasis, retinal ischemia, inflammation, and the breakdown of the blood–retinal barrier [3]. While vascular endothelial growth factor-A (VEGF-A) has long been recognized as a central mediator of vascular permeability, increasing evidence highlights the complementary role of angiopoietin-2 (Ang-2). Ang-2 promotes endothelial destabilization, inflammation, and vascular leakage via the disruption of the Tie2 signaling pathway [4, 5]. Intravitreal anti-VEGF therapy has transformed the management of ME secondary to RVO and remains the current standard of care. Agents such as ranibizumab and aflibercept have consistently demonstrated significant improvements in best-corrected visual acuity (BCVA) and reductions in central macular thickness (CMT), thereby substantially decreasing the risk of vision loss [2, 6]. Nevertheless, these therapies are associated with a considerable treatment burden, requiring frequent injections and intensive follow-up. In real-world settings, this burden often leads to suboptimal adherence, contributing to a well-documented efficacy gap between randomized controlled trials and routine clinical practice [7, 8]. Furthermore, both mechanistic and clinical data suggest that targeting VEGF alone may be inadequate to fully resolve macular edema in all eyes [4]. Many patients show residual disease activity despite ongoing anti-VEGF therapy, highlighting the need for agents capable of addressing both VEGF-mediated permeability and Ang-2-driven vascular instability through the Tie2 pathway [9, 10]. To address these limitations, therapeutic strategies targeting multiple pathogenic pathways have emerged. Faricimab is a bispecific monoclonal antibody designed to simultaneously inhibit VEGF-A and Ang-2, thereby providing a more comprehensive blockade of the inflammatory and permeability cascades underlying ME [5, 7]. By modulating these dual pathways, faricimab is hypothesized to provide enhanced and more durable disease control compared with anti-VEGF monotherapy. This is supported by the phase III BALATON and COMINO trials, which demonstrated that faricimab achieves visual and anatomical outcomes comparable to aflibercept while allowing for extended dosing intervals in a substantial proportion of patients [9, 10]. Despite these promising trial results, outcomes from highly controlled settings may not fully translate to real-world effectiveness owing to strict inclusion criteria and the limited representation of complex or previously treated cases. In routine clinical practice, patients frequently present with heterogeneous baseline characteristics, variable adherence, and diverse treatment histories. Consequently, real-world studies are essential to validate the effectiveness, durability, and safety of faricimab across broader patient populations, particularly comparing treatment-naïve individuals with those switched from prior therapies [11, 12]. In this context, following the approval of faricimab for clinical use in Türkiye, the present multicenter study conducted by the FARTURK study group aims to evaluate the short-to-mid-term real-world anatomical and functional outcomes of intravitreal faricimab in patients with RVO-associated ME. By incorporating data from multiple tertiary referral centers and reflecting routine clinical practice patterns, this study seeks to provide comprehensive evidence on the efficacy, durability, and safety of faricimab, comparing outcomes between treatment-naïve patients and those switched from prior anti-VEGF therapies, thereby contextualizing our findings within the framework of pivotal clinical trials and emerging real-world literature. We hypothesized that intravitreal faricimab would produce significant improvements in BCVA and CMT compared with baseline in both treatment-naïve and switch patients with RVO-associated macular edema in a real-world setting, and that treatment-naïve patients would demonstrate greater functional gains than those previously exposed to anti-VEGF therapy. Methods Study Design and Ethical Considerations This was a retrospective, multicenter, observational real-world study assessing the anatomical and functional outcomes of intravitreal faricimab 6 mg/0.05 mL (Vabysmo®, Roche/Genentech) for ME secondary to RVO after its approval and introduction into routine clinical practice in Türkiye. The study was conducted by the FARTURK study group, a collaborative network formed by 38 centers across Türkiye that administer faricimab treatment. Specifically, 17 of these centers participated in the current study, contributing data on patients treated with faricimab for RVO-related ME following its approval for this indication. This study was conducted in accordance with the Declaration of Helsinki and its later amendments. Ethical approval was granted by the Istanbul Medipol University Ethics Committee (approval ID: E-10840098-50.04-2136). Written informed consent for intravitreal treatment and the secondary use of anonymized clinical data was obtained from all participants prior to inclusion. This study was conducted and reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for observational cohort studies. Study Population The inclusion criteria were defined as follows: patients ≥ 18 years of age with a diagnosis of BRVO or CRVO and ME, who received at least one intravitreal faricimab injection between 22 November 2024 and 1 March 2026, and had a follow-up of at least 4 weeks after the first injection, along with complete baseline and follow-up ophthalmic examination and optical coherence tomography (OCT) imaging data available. Both treatment-naïve eyes and eyes switched from prior anti-VEGF therapy were included. Patients with concurrent ocular conditions capable of independently reducing BCVA, including advanced glaucoma, optic neuropathy, or significant media opacities, were excluded at the discretion of the treating ophthalmologist at each center, on the basis of clinical judgment. No formal, prespecified exclusion criterion was applied for these conditions; accordingly, the potential influence of undetected coexisting optic pathology on BCVA outcomes cannot be entirely excluded and represents a limitation of the retrospective design. Patients with insufficient or recalcitrant response to prior treatment were included, reflecting real-world clinical practice and the heterogeneous population typically seen in RVO clinics. Exclusion criteria were the following: concomitant macular/retinal disease (e.g., diabetic retinopathy or neovascular age-related macular degeneration), history of pars plana vitrectomy, significant media opacities precluding reliable imaging, irregular follow-up, or incomplete medical records. Data Collection and Clinical Assessments Data were collected retrospectively from institutional electronic medical records using a standardized case report form. To ensure data accuracy and consistency, data extraction was performed by the treating ophthalmologists at each participating center. The extracted variables were categorized into baseline characteristics and longitudinal clinical parameters. Baseline data encompassed patient demographics (age and sex), ocular characteristics (lens status [phakic/pseudophakic] and RVO subtype [BRVO/CRVO]), and prior anti-VEGF treatment history (specific agent and number of prior injections). For both the baseline and each subsequent follow-up visit, clinical parameters retrieved from routine ophthalmic examinations included BCVA, slit-lamp biomicroscopy findings, intraocular pressure (IOP) measured by Goldmann applanation tonometry, and dilated fundus examination results. BCVA values, initially recorded in decimal notation in the clinical charts, were converted to the logarithm of the minimum angle of resolution (logMAR) scale for statistical analysis. Finally, the total number and precise timing of faricimab injections administered throughout the follow-up period were recorded to evaluate treatment patterns. To evaluate longitudinal clinical outcomes, data regarding BCVA, CMT, and IOP were extracted from the medical records at specific intervals. These time points for analysis included: baseline, approximately 7 days after the initial injection (if such data were available from routine practice), 1 month after each of the first three injections, and at the final follow-up visit. Additionally, clinical notes from all available visits were retrospectively reviewed to identify any documented ocular or systemic adverse events. Optical Coherence Tomography Imaging Macular imaging was performed using Heidelberg Spectralis spectral-domain optical coherence tomography (SD-OCT; Heidelberg Engineering, Heidelberg, Germany) at all participating centers. Central macular thickness (CMT) values were obtained automatically from the device’s proprietary software (HEYEX), on the basis of standard macular scanning protocols used in routine clinical practice. In BRVO cases with predominantly parafoveal or paramacular edema distribution, CMT was used as the primary anatomical outcome measure, consistent with routine clinical practice at all participating centers. Additional OCT parameters, such as macular volume, subfield thickness maps, or qualitative assessment of edema distribution, were not systematically recorded across all centers and were therefore not reported. This reliance on CMT alone in BRVO may underestimate the true extent of macular involvement in cases with eccentric edema, and represents a methodological limitation. Statistical Analysis All statistical analyses were performed using SPSS software (version 26.0; IBM Corp., Armonk, NY, USA). The distribution of continuous variables was evaluated visually (histograms, Q–Q plots) and with Kolmogorov–Smirnov and Shapiro–Wilk tests. Continuous data are presented as mean ± standard deviation (SD) or median (interquartile range) as appropriate; categorical variables are presented as counts and percentages. Within-group changes in BCVA and CMT over time were assessed using linear mixed-effects models (LMM), implemented in R using the lme4 and lmerTest packages (version 4.4.3). Each model included time point as a fixed effect (with baseline as reference), random intercepts per patient to account for the correlation of repeated measurements, and the following covariates: baseline BCVA (logMAR), baseline CMT, age, RVO subtype, treatment group, and number of prior anti-VEGF injections. This approach accommodates missing observations at intermediate time points without imputation and provides valid inference under unbalanced follow-up. Estimated marginal means and contrasts versus baseline at each time point, with 95% confidence intervals, were derived using the emmeans package. Pairwise comparisons between baseline and each follow-up time point were additionally performed with paired t-tests or Wilcoxon signed-rank tests applying Bonferroni correction, as a complementary analysis. Between-group comparisons (BRVO versus CRVO) used independent samples t tests or Mann–Whitney U tests as appropriate. A two-sided p-value < 0.05 was considered statistically significant. To address potential confounding in the comparison between treatment-naïve and switch patients, an analysis of covariance (ANCOVA) was performed with BCVA gain (logMAR) and CMT reduction (μm) as dependent variables, treatment group (naïve versus switch) as the fixed factor, and the following covariates: baseline BCVA (logMAR), baseline CMT (μm), age, RVO subtype (BRVO versus CRVO), and number of prior anti-VEGF injections. Estimated marginal means (adjusted group means) and pairwise contrasts with 95% confidence intervals were derived using the emmeans package in R (version 4.4.3). As this was a retrospective real-world study, a formal a priori sample size calculation was not performed. The study is therefore considered exploratory and hypothesis-generating in nature. Post hoc power analysis indicated that with n = 70 patients and an observed mean BCVA gain of 0.58 logMAR (SD approximately 0.45), the study had greater than 99% power to detect within-group improvement at α = 0.05 using a two-sided paired t test, confirming that the primary outcome analysis was adequately powered. The naïve versus switch comparison (n = 44 versus 26) was also adequately powered for the observed effect size; however, this comparison was not adjusted for baseline differences between groups and should be interpreted with caution on grounds of confounding rather than statistical power. Subgroup analyses comparing BRVO and CRVO outcomes (n = 37 versus 33) were exploratory and may have been underpowered to detect modest between-subtype differences, as reflected by the nonsignificant p-value of 0.094 for the BRVO versus CRVO visual acuity comparison. The final visit was defined as the last available clinical assessment for each individual patient during the study period, irrespective of the total number of injections received or follow-up duration; accordingly, all 70 patients contributed data to the final visit analysis. Missing intermediate data were not imputed; analyses at each time point were performed on the available case set, and the varying sample sizes at each interval are reported transparently in Tables 2 and 3. As a post hoc attrition analysis, baseline characteristics of patients with data available at all intermediate time points (completers; n = 43) were compared with those with incomplete intermediate follow-up (n = 27). While baseline BCVA (p = 0.061), age (p = 0.133), sex (p = 0.678), RVO subtype (p = 0.704), and treatment group (p = 0.073) did not differ significantly between groups, completers had significantly higher baseline CMT than incomplete patients (641.9 ± 161.3 versus 487.2 ± 158.8 μm; p < 0.001), and longer follow-up duration (27.1 ± 7.7 versus 18.2 ± 6.9 weeks; p < 0.001). Covariate-adjusted estimated marginal means from the LMM at each time point are provided in Supplementary Table S1. Full attrition analysis results are presented in Supplementary Table S2. Table 2Longitudinal best-corrected visual acuity (BCVA) outcomes by groupTime pointOverall (n = 70)BRVO (n = 37)CRVO (n = 33)Treatment-naïve (n = 44)Switch (n = 26)Baselinen = 70n = 37n = 33n = 44n = 26Mean ± SD (logMAR)0.76 ± 0.420.74 ± 0.490.79 ± 0.320.79 ± 0.460.72 ± 0.3595% CI0.66 to 0.860.57 to 0.900.68 to 0.900.65 to 0.930.58 to 0.86p-Value versus baselineReferenceReferenceReferenceReferenceReferenceDay 7 (post-injection 1)†n = 63n = 34n = 29n = 41n = 22Mean ± SD (logMAR)0.39 ± 0.310.31 ± 0.280.49 ± 0.320.36 ± 0.310.44 ± 0.3095% CI0.31 to 0.470.21 to 0.410.36 to 0.610.27 to 0.460.31 to 0.57p-Value versus baseline< 0.001< 0.001< 0.001< 0.001< 0.001Month 1 (post-injection 1)n = 52n = 28n = 24n = 28n = 24Mean ± SD (logMAR)0.46 ± 0.340.39 ± 0.350.55 ± 0.320.43 ± 0.340.50 ± 0.3595% CI0.37 to 0.560.25 to 0.520.42 to 0.680.30 to 0.560.35 to 0.65 p-Value versus baseline< 0.001< 0.001< 0.001< 0.001< 0.001Month 1 (post-injection 2)n = 48n = 25n = 23n = 26n = 22Mean ± SD (logMAR)0.49 ± 0.430.34 ± 0.320.65 ± 0.490.51 ± 0.550.47 ± 0.2595% CI0.36 to 0.620.21 to 0.480.44 to 0.860.28 to 0.730.36 to 0.58p-Value versus baseline< 0.001< 0.0010.0600.002< 0.001Month 1 (post-injection 3)n = 46n = 25n = 21n = 26n = 20Mean ± SD (logMAR)0.30 ± 0.260.21 ± 0.210.40 ± 0.270.28 ± 0.240.32 ± 0.2995% CI0.22 to 0.370.12 to 0.300.28 to 0.530.18 to 0.380.19 to 0.45p-Value versus baseline< 0.001< 0.001< 0.001< 0.001< 0.001Final visitn = 70n = 37n = 33n = 44n = 26Mean ± SD (logMAR)0.26 ± 0.260.18 ± 0.200.35 ± 0.300.21 ± 0.240.34 ± 0.2995% CI0.20 to 0.320.11 to 0.240.24 to 0.460.14 to 0.280.22 to 0.46p-Value versus baseline< 0.001< 0.001< 0.001< 0.001< 0.001Mean BCVA gain0.58 ± 0.400.56 ± 0.450.44 ± 0.340.77 ± 0.420.38 ± 0.36All values expressed as logarithm of the minimum angle of resolution (logMAR); lower values indicate better visual acuity. p-Values are from two-tailed paired t tests versus baseline within each group†Day-7 data available for 63/70 patients (90.0%); day-7 visits were performed as part of routine clinical practice and not available for all patients. Mean BCVA gain calculated as baseline minus final visit logMAR value (positive = improvement). Between-group differences were not adjusted for covariates; formal adjusted analyses are reported separatelyBRVO, branch retinal vein occlusion; CI, confidence interval; CRVO, central retinal vein occlusion; SD, standard deviation Longitudinal best-corrected visual acuity (BCVA) outcomes by group All values expressed as logarithm of the minimum angle of resolution (logMAR); lower values indicate better visual acuity. p-Values are from two-tailed paired t tests versus baseline within each group †Day-7 data available for 63/70 patients (90.0%); day-7 visits were performed as part of routine clinical practice and not available for all patients. Mean BCVA gain calculated as baseline minus final visit logMAR value (positive = improvement). Between-group differences were not adjusted for covariates; formal adjusted analyses are reported separately BRVO, branch retinal vein occlusion; CI, confidence interval; CRVO, central retinal vein occlusion; SD, standard deviation Table 3Longitudinal central macular thickness (CMT) outcomes by groupTime pointOverall (n = 70)BRVO (n = 37)CRVO (n = 33)Treatment-naïve (n = 44)Switch (n = 26)Baselinen = 70n = 37n = 33n = 44n = 26Mean ± SD (µm)582.2 ± 176.3596.9 ± 132.9565.8 ± 215.9594.1 ± 206.9562.1 ± 107.395% CI (µm)540.2 to 624.2552.5 to 641.2489.2 to 642.3531.2 to 657.0518.7 to 605.4p-Value versus baselineReferenceReferenceReferenceReferenceReferenceDay 7 (post-injection 1)†n = 63n = 34n = 29n = 41n = 22Mean ± SD (µm)345.0 ± 118.6345.6 ± 118.4344.2 ± 120.9342.7 ± 116.9349.1 ± 124.595% CI (µm)315.1 to 374.8304.2 to 386.9298.2 to 390.2305.8 to 379.6293.9 to 404.3p-Value versus baseline< 0.001< 0.001< 0.001< 0.001< 0.001Month 1 (post-injection 1)n = 52n = 28n = 24n = 28n = 24Mean ± SD (µm)324.4 ± 100.3318.6 ± 109.7331.1 ± 89.9315.1 ± 82.7335.2 ± 118.595% CI (µm)296.4 to 352.3276.1 to 361.2293.1 to 369.0283.0 to 347.1285.2 to 385.3p-Value versus baseline< 0.001< 0.001< 0.001< 0.001< 0.001Month 1 (post-injection 2)n = 48n = 25n = 23n = 26n = 22Mean ± SD (µm)356.3 ± 211.0311.8 ± 85.1404.7 ± 287.3402.7 ± 260.7301.5 ± 113.795% CI (µm)295.0 to 417.6276.7 to 346.9280.4 to 528.9297.4 to 508.0251.1 to 351.9p-Value versus baseline< 0.001< 0.001< 0.001< 0.001< 0.001Month 1 (post-injection 3)n = 45n = 24n = 21n = 25n = 20Mean ± SD (µm)261.2 ± 99.4236.6 ± 74.9289.3 ± 117.2275.5 ± 88.3243.3 ± 111.595% CI (µm)231.3 to 291.1205.0 to 268.2236.0 to 342.7239.0 to 311.9191.2 to 295.5p-Value versus baseline< 0.001< 0.001< 0.001< 0.001< 0.001Final visitn = 70n = 37n = 33n = 44n = 26Mean ± SD (µm)281.4 ± 89.8280.8 ± 102.4282.0 ± 74.8287.2 ± 89.9271.5 ± 90.795% CI (µm)259.9 to 302.8246.6 to 314.9255.5 to 308.6259.9 to 314.5234.8 to 308.1p-Value versus baseline< 0.001< 0.001< 0.001< 0.001< 0.001Mean CMT reduction300.8 µm316.1 µm283.8 µm306.9 µm290.6 µmAll values expressed in micrometers (µm). p-Values are from two-tailed paired t tests versus baseline within each group†Day-7 data available for 63/70 patients (90.0%). Mean CMT reduction calculated as baseline minus final visit CMT value. Between-group differences were not adjusted for covariatesBRVO, branch retinal vein occlusion; CI, confidence interval; CRVO, central retinal vein occlusion; SD, standard deviation Longitudinal central macular thickness (CMT) outcomes by group All values expressed in micrometers (µm). p-Values are from two-tailed paired t tests versus baseline within each group †Day-7 data available for 63/70 patients (90.0%). Mean CMT reduction calculated as baseline minus final visit CMT value. Between-group differences were not adjusted for covariates BRVO, branch retinal vein occlusion; CI, confidence interval; CRVO, central retinal vein occlusion; SD, standard deviation The primary outcome of this study was change in BCVA (logMAR) from baseline to the final visit. Secondary outcomes were change in CMT and change in IOP from baseline to the final visit. As this is a retrospective real-world study, outcomes were not prespecified prospectively; this designation was made post hoc for clarity of reporting. In accordance with the Sex and Gender Equity in Research (SAGER) guidelines, sex was recorded and reported for all participants (38 male [54.3%], 32 female [45.7%]). Based on the existing literature, no clinically meaningful sex differences in anti-VEGF treatment response for RVO-associated macular edema are consistently reported; therefore, sex-disaggregated outcome analysis was not performed as a primary analysis. This decision is further addressed in the Limitations. Results Baseline Characteristics Overall, 70 patients with RVO were included in the study. Race and ethnicity data were not collected, as all patients were treated in Türkiye where formal race/ethnicity classification was not part of routine clinical documentation at the participating centers. The mean age of the patients was 59.9 ± 10.4 years (range 40–88 years), comprising 38 male (54.3%) and 32 female patients (45.7%). Regarding RVO subtype, 37 patients (52.9%) presented with BRVO and 33 patients (47.1%) with CRVO. Lens status analysis revealed that 42 eyes (60.0%) were phakic, whereas 28 eyes (40.0%) were pseudophakic. Comprehensive baseline demographic and clinical characteristics are summarized in Table 1. Table 1Baseline patient characteristicsVariableOverall (n = 70)BRVO (n = 37)CRVO (n = 33)Treatment-naïve (n = 44)Switch (n = 26)AgeMean ± SD (years)60.0 ± 10.559.5 ± 10.060.5 ± 11.157.6 ± 10.164.0 ± 9.9Range (years)40–8841–8840–8440–8245–88SexMale, n (%)38 (54.3%)21 (56.8%)17 (51.5%)22 (50.0%)16 (61.5%)Female, n (%)32 (45.7%)16 (43.2%)16 (48.5%)22 (50.0%)10 (38.5%)RVO subtypeBRVO, n (%)37 (52.9%)––23 (52.3%)14 (53.8%)CRVO, n (%)33 (47.1%)––21 (47.7%)12 (46.2%)Lens statusPhakic, n (%)42 (60.0%)24 (64.9%)18 (54.5%)30 (68.2%)12 (46.2%)Pseudophakic, n (%)28 (40.0%)13 (35.1%)15 (45.5%)14 (31.8%)14 (53.8%)Baseline visual acuityBCVA, mean ± SD (logMAR)0.76 ± 0.420.74 ± 0.490.79 ± 0.320.79 ± 0.460.72 ± 0.3595% CI0.66 to 0.860.57 to 0.900.68 to 0.900.65 to 0.930.58 to 0.86Baseline CMTCMT, mean ± SD (µm)582.2 ± 176.3596.9 ± 132.9565.8 ± 215.9594.1 ± 206.9562.1 ± 107.395% CI (µm)540.2 to 624.2552.5 to 641.2489.2 to 642.3531.2 to 657.0518.7 to 605.4Baseline IOPIOP, mean ± SD (mmHg)16.4 ± 3.117.0 ± 3.415.6 ± 2.616.8 ± 3.315.5 ± 2.695% CI (mmHg)15.6 to 17.115.9 to 18.214.7 to 16.515.8 to 17.914.5 to 16.6Faricimab treatmentTotal injections, mean ± SD2.49 ± 1.052.51 ± 1.042.47 ± 1.072.38 ± 1.052.69 ± 1.05Follow-up duration, mean ± SD (weeks)23.7 ± 8.523.4 ± 7.924.0 ± 9.223.4 ± 8.724.1 ± 8.3Prior treatment (switch group only)Prior injections, mean ± SD––––7.12 ± 4.81Prior agent: aflibercept, n (%)––––25 (96.2%)Prior agent: bevacizumab, n (%)––––1 (3.8%)Switch group prior agent breakdown: aflibercept 25/26 (96.2%), bevacizumab 1/26 (3.8%). Treatment-naïve patients (n = 44) received no prior anti-VEGF therapy“–” indicates not applicableBCVA, best-corrected visual acuity; BRVO, branch retinal vein occlusion; CI, confidence interval; CMT, central macular thickness; CRVO, central retinal vein occlusion; IOP, intraocular pressure; logMAR, logarithm of the minimum angle of resolution; RVO, retinal vein occlusion; SD, standard deviation Baseline patient characteristics Switch group prior agent breakdown: aflibercept 25/26 (96.2%), bevacizumab 1/26 (3.8%). Treatment-naïve patients (n = 44) received no prior anti-VEGF therapy “–” indicates not applicable BCVA, best-corrected visual acuity; BRVO, branch retinal vein occlusion; CI, confidence interval; CMT, central macular thickness; CRVO, central retinal vein occlusion; IOP, intraocular pressure; logMAR, logarithm of the minimum angle of resolution; RVO, retinal vein occlusion; SD, standard deviation Based on prior treatment history, 44 patients (62.8%) were treatment-naïve, whereas 26 patients (37.2%) had prior anti-VEGF exposure (switch cohort). Among the switch cohort, 25 had previously been treated with aflibercept and 1 had received bevacizumab. Prior to the switch, patients had received a mean of 7.12 ± 4.73 injections (range 1–17). No standardized injection protocol was mandated across the 17 participating centers; treatment decisions, including injection intervals, use of loading doses, and subsequent treat-and-extend or pro re nata scheduling, were made at the discretion of each treating physician according to local clinical practice. The mean number of faricimab injections administered during the study period was 2.49 ± 1.05 (range 1–4). The mean follow-up duration was 23.69 ± 8.54 weeks (range 10–52 weeks). Functional Outcomes: Visual Acuity Outcomes Faricimab therapy resulted in rapid and statistically significant functional improvements across the entire study cohort. Mean BCVA improved from 0.84 ± 0.45 logMAR at baseline (95% CI 0.74 to 0.95) to 0.44 ± 0.27 logMAR by the first week post-injection in 63 of 70 patients for whom day-7 data were available (90.0%; 95% CI 0.37 to 0.51). This functional recovery was sustained throughout the follow-up period, with mean BCVA values of 0.48 ± 0.33 logMAR at 1 month following the first injection (n = 52; the remaining 18 patients received only one injection during the study period and therefore had no subsequent monthly visit data), 0.45 ± 0.38 logMAR at 1 month following the second injection, and 0.29 ± 0.25 logMAR at 1 month following the third injection, respectively. At the final visit, the overall mean BCVA reached 0.26 ± 0.24 logMAR (95% CI 0.20 to 0.32), representing a mean gain of 0.58 logMAR (95% CI 0.46 to 0.70; p < 0.001). When functional outcomes were stratified by RVO subtype, both groups demonstrated comparable baseline visual acuity, measuring 0.86 ± 0.56 logMAR in the BRVO group and 0.82 ± 0.27 logMAR in the CRVO group. Following treatment, both cohorts exhibited consistent functional gains. At the final visit, the BRVO group achieved a mean BCVA of 0.19 ± 0.19 logMAR (95% CI 0.12 to 0.25), while the CRVO cohort stabilized at 0.35 ± 0.27 logMAR (95% CI 0.25 to 0.45). Although the BRVO group achieved a numerically better final visual acuity, comparative analysis did not reach statistical significance (p = 0.094). Nevertheless, both RVO subtypes demonstrated substantial and clinically meaningful functional recovery. The greater functional gain observed in the BRVO subgroup is consistent with the generally more favorable natural history and lower ischemic burden of branch compared with central retinal vein occlusion [1–3]. Ischemic status was not systematically assessed in this retrospective cohort; accordingly, the relative contributions of ischemic versus non-ischemic disease to the observed between-subtype differences in visual recovery cannot be determined from the available data. Both treatment-naïve and switch patients achieved statistically significant visual acuity improvements following faricimab treatment (p < 0.001 for both groups). In unadjusted analysis, treatment-naïve patients demonstrated greater mean BCVA gain compared with switch patients (0.77 ± 0.45 versus 0.41 ± 0.33 logMAR; p = 0.006); however, the two groups differed substantially in baseline visual acuity (0.98 versus 0.70 logMAR) and disease chronicity. After adjustment for baseline BCVA, baseline CMT, age, RVO subtype, and number of prior injections using ANCOVA, the estimated marginal means for BCVA gain were 1.971 logMAR (95% CI 1.939 to 2.003) for naïve patients and 1.950 logMAR (95% CI 1.903 to 1.998) for switch patients, with an adjusted between-group difference of 0.021 logMAR (95% CI −0.049 to +0.090), which did not reach statistical significance. Similarly, adjusted CMT reduction did not differ significantly between groups (naïve: 309.5 μm [95% CI 273.6 to 345.3] versus switch: 285.3 μm [95% CI 232.5 to 338.2]; difference: 24.1 μm [95% CI −52.2 to +100.5]). These findings indicate that the apparent superiority of visual gains in treatment-naïve patients in unadjusted analyses was largely attributable to baseline differences between groups, particularly the greater baseline visual impairment in naïve eyes, rather than a true differential treatment effect. Longitudinal BCVA values across all groups and time points are presented in Table 2 and shown in Fig. 1. Longitudinal changes in mean best-corrected visual acuity (BCVA) following intravitreal faricimab treatment. Panel A shows BCVA trajectories for the overall cohort and by retinal vein occlusion (RVO) subtype (branch RVO [BRVO] and central RVO [CRVO]). Panel B shows BCVA trajectories stratified by prior treatment history (treatment-naïve and switch patients). Data points represent mean logMAR values; error bars indicate 95% confidence intervals. Lower logMAR values indicate better visual acuity. Day-7 data were available for 63 of 70 patients (90.0%). Dotted horizontal reference lines indicate 1.0 logMAR (20/200; legal blindness threshold) and 0.3 logMAR (20/40; driving standard). All within-group improvements from baseline were statistically significant at all time points (p < 0.001). The between-group comparison of treatment-naïve versus switch patients was unadjusted for baseline differences. BCVA, best-corrected visual acuity; BRVO, branch retinal vein occlusion; CI, confidence interval; CRVO, central retinal vein occlusion; logMAR, logarithm of the minimum angle of resolution; RVO, retinal vein occlusion Anatomical Outcomes: Central Macular Thickness Consistent with the functional gains, morphological parameters demonstrated a progressive reduction in ME following faricimab treatment. Mean CMT decreased from a baseline of 582.2 ± 176.3 µm (95% CI 540.2 to 624.2) to 345.0 ± 118.6 µm at 7 days post-injection among the 63 patients with available day-7 data (90.0%; 95% CI 315.1 to 374.8; p < 0.001). This anatomical improvement was sustained throughout the follow-up period, with mean CMT values of 330.1 ± 107.1 µm at 1 month following the first injection (n = 52), 348.2 ± 173.4 µm at 1 month following the second injection, and 254.8 ± 96.8 µm at 1 month following the third injection, respectively. By the final visit, the mean CMT was 281.4 ± 89.8 µm (95% CI 259.9 to 302.8). A transient increase in mean CMT was observed in the CRVO subgroup at 1 month following the second injection compared with 1 month post-injection 1, followed by a subsequent decline at 1 month post-injection 3. This transient fluctuation may reflect natural disease variability, heterogeneous treatment intervals between centers, the relatively small number of patients with CRVO contributing data at this time point (n = 23), and the potential influence of a small subset of patients with more refractory disease. No formal subgroup analysis was performed to determine whether this pattern was driven by individual outliers, given the limited sample size at this time point. All reductions in CMT compared with baseline were statistically significant across all evaluated time points (p < 0.001). Subgroup analysis by etiology revealed comparable baseline ME profiles, with mean CMT values of 596.9 ± 132.9 µm and 565.8 ± 215.9 µm in the BRVO and CRVO cohorts, respectively. At the final visit, CMT decreased to 266.7 ± 85.4 μm in BRVO and 275.1 ± 65.0 μm in CRVO eyes. Ultimately, both RVO subtypes achieved comparably favorable anatomical outcomes, with marked resolution of ME at the final visit. Longitudinal CMT values across all groups and time points are presented in Table 3 and shown in Fig. 2. Longitudinal changes in mean central macular thickness (CMT) following intravitreal faricimab treatment. Panel A shows CMT trajectories for the overall cohort and by RVO subtype (BRVO and CRVO). Panel B shows CMT trajectories stratified by prior treatment history (treatment-naïve and switch patients). Data points represent mean CMT values in micrometers (µm); error bars indicate 95% confidence intervals. Day-7 data were available for 63 of 70 patients (90.0%). The dashed green horizontal line indicates the approximate normal CMT (~260 µm). All within-group reductions from baseline were statistically significant at all time points (p < 0.001). BRVO, branch retinal vein occlusion; CI, confidence interval; CMT, central macular thickness; CRVO, central retinal vein occlusion; RVO, retinal vein occlusion; µm, micrometers To complement the available-case analyses, longitudinal BCVA and CMT outcomes were additionally evaluated using linear mixed-effects models adjusting for baseline BCVA, baseline CMT, age, RVO subtype, treatment group, and prior injection burden. The LMM confirmed significant improvement in BCVA at all post-baseline time points (F[5,301] = 520.9, p < 0.001). Covariate-adjusted estimated marginal means for BCVA were 0.407 logMAR (95% CI 0.340 to 0.474) at day 7, 0.444 logMAR (95% CI 0.372 to 0.516) at month 1 post-injection 1, 0.457 logMAR (95% CI 0.381 to 0.532) at month 1 post-injection 2, and 0.281 logMAR (95% CI 0.204 to 0.357) at month 1 post-injection 3, compared with a baseline adjusted mean of 0.778 logMAR (95% CI 0.715 to 0.842). The adjusted BCVA change from baseline at the final visit was −1.964 logMAR (95% CI −2.072 to −1.855; p < 0.001). Among covariates, baseline BCVA (F[1,49] = 83.6, p < 0.001) and RVO subtype (F[1,69] = 8.8, p = 0.004) were significant predictors, while treatment group (p = 0.449), age (p = 0.261), and prior injection burden (p = 0.129) were not, consistent with the ANCOVA findings. For CMT, the LMM similarly confirmed significant reductions at all time points, with adjusted CMT of 358.8 μm (95% CI 325.9 to 391.6) at day 7, 319.0 μm (95% CI 283.4 to 354.6) at month 1 post-injection 1, and 291.0 μm (95% CI 259.9 to 322.1) at the final visit, representing an adjusted reduction of −300.8 μm (95% CI: −353.8 to −247.9; p < 0.001) from the baseline adjusted mean of 591.8 μm (95% CI 560.7 to 622.9). To supplement mean logMAR outcomes, the proportion of eyes achieving clinically meaningful visual thresholds was calculated. A gain of ≥ 0.3 logMAR (approximately equivalent to 15 Early Treatment Diabetic Retinopathy Study [ETDRS] letters) was achieved by 51 of 69 patients (73.9%) overall, including 72.7% of treatment-naïve and 76.0% of switch patients, and 80.6% of BRVO versus 66.7% of CRVO eyes. A gain of ≥ 0.5 logMAR was achieved by 26 patients (37.7%) overall, with a notably higher proportion in the naïve group (50.0%) than the switch group (16.0%). At the final visit, 44 of 69 patients (63.8%) achieved a BCVA of ≤ 0.3 logMAR (20/40; driving standard), including 70.5% of naïve and 52.0% of switch patients. Of the 24 patients (34.8%) presenting with legally blind visual acuity (≥ 1.0 logMAR; 20/200) at baseline, 21 (87.5%) recovered functional vision below this threshold by the final visit. These threshold-level outcomes are presented visually across treatment history and RVO subtype subgroups in Fig. 3, illustrating the clinical breadth of response achieved in this real-world cohort. Proportion of eyes achieving clinically meaningful visual thresholds by treatment history and retinal vein occlusion subtype. Panel A shows the proportion achieving a gain of ≥ 0.3 logMAR (≈15 ETDRS letters) from baseline. Panel B shows the proportion achieving a gain of ≥ 0.5 logMAR (substantial gain). Panel C shows the proportion achieving a final best-corrected visual acuity (BCVA) of ≤ 0.3 logMAR (20/40; driving standard). Panel D shows the proportion of patients presenting with legally blind visual acuity (≥ 1.0 logMAR; 20/200) at baseline who recovered functional vision (< 1.0 logMAR) by the final visit. Bars show the proportion of eyes achieving each threshold history (naïve versus switch) and RVO subtype (BRVO versus CRVO). Dashed horizontal lines indicate the overall cohort proportion for each threshold. †Denominator for panel D restricted to patients with baseline BCVA ≥ 1.0 logMAR (n = 24 overall; n = 14 naïve, n = 10 switch, n = 8 BRVO, n = 16 CRVO). BCVA, best-corrected visual acuity; BRVO, branch retinal vein occlusion; CRVO, central retinal vein occlusion; ETDRS, Early Treatment Diabetic Retinopathy Study; logMAR, logarithm of the minimum angle of resolution Safety Profile The safety profile of intravitreal faricimab was favorable, with IOP remaining stable throughout the entire follow-up period. From a mean baseline of 16.4 ± 3.1 mmHg, IOP measurements exhibited no clinically meaningful fluctuations or ocular hypertensive spikes. Specifically, post-treatment mean IOP measured 16.5 ± 2.6 mmHg at day 7, with subsequent values of 16.5 ± 2.4 mmHg, 16.4 ± 2.7 mmHg, and 16.5 ± 2.5 mmHg at 1 month post-first, second, and third injections, respectively. This stability was maintained through the final follow-up visit (16.5 ± 2.7 mmHg) and was consistently observed across both the BRVO and CRVO cohorts. Furthermore, the treatment was well tolerated by all patients. There were no reports of any drug-related ocular or systemic adverse events during the study period. Specifically, no cases of intraocular inflammation, endophthalmitis, retinal tear or detachment, or any systemic thromboembolic events (such as myocardial infarction or stroke) were observed following faricimab injections, underscoring a robust short-to-mid-term safety profile. Discussion The results of the FARTURK-RVO study indicate that the short-term efficacy and safety profile of faricimab in patients with RVO is largely consistent with the findings reported in the pivotal BALATON and COMINO trials. However, while phase III randomized controlled trials provide essential safety and efficacy data under standardized protocols, they do not always fully reflect the complexities of routine clinical practice. Historical evidence across retinal vascular diseases treated with various anti-VEGF agents has frequently shown a disparity between outcomes achieved in controlled research settings and those observed in more heterogeneous, real-world populations [13, 14]. In this context, our study sought to evaluate the performance of faricimab across a diverse national cohort in Türkiye shortly after its clinical approval. A notable finding in our analysis was the early therapeutic response observed at day 7 following the first injection (Tables 2 and 3 and Figs. 1 and 2). Day-7 data were available for 63 of 70 patients (90.0%) and were collected as part of routine clinical practice rather than a predefined protocol; accordingly, this subset may not be fully representative of the overall cohort, and these findings should be interpreted with caution. To the best of our knowledge, structured assessment at this very early time point has received limited attention in previous landmark trials, and these observations may offer preliminary insights into the early treatment dynamics of faricimab in a real-world setting. We observed that mean BCVA improved from 0.84 ± 0.45 logMAR at baseline to 0.44 ± 0.27 logMAR within the first week, while CMT decreased from 582.2 ± 176.3 to 345.0 ± 118.6 μm during the same interval. These early improvements are consistent with the known pharmacodynamic profile of anti-VEGF agents in RVO-associated macular edema. The landmark BALATON and COMINO trials focused on primary outcomes at week 24, with the earliest structured assessments occurring at week 4; direct comparison with these trials is limited by substantial differences in study design, treatment intensity, and follow-up protocols. The clinical relevance of the day-7 response as a prognostic indicator warrants evaluation in larger, prospective studies with predefined early assessment time points. The early anatomical response observed in our study may reflect the rapid onset of VEGF-A inhibition; whether the additional Ang-2 blockade provided by faricimab contributes to an accelerated early drying effect compared with anti-VEGF monotherapy cannot be determined from the present observational data in the absence of a comparator arm. This remains a hypothesis that requires prospective evaluation. The consistency of early functional gains across both BRVO and CRVO subtypes is noted, though given the absence of a control group, these findings should be interpreted as descriptive rather than mechanistically inferential. Furthermore, the consistency of these early functional gains across both BRVO and CRVO subtypes in our cohort reinforces the broad applicability of dual-pathway inhibition in RVO. While BRVO and CRVO differ in their underlying venous anatomy and systemic risk profiles, the common final pathway of macular edema, along with retinal ischemia, represents a major determinant of visual impairment in both BRVO and CRVO [15]. There was no statistically significant difference in overall visual gain between the BRVO and CRVO groups, indicating a robust response regardless of the subtype. Our findings suggest that the initial response to faricimab is not limited to a specific RVO subtype but rather reflects a generalized potent effect on the retinal microenvironment. These observations provide a foundation for future longitudinal studies to determine whether this robust early response serves as a predictor for long-term treatment durability or the achievement of extended dosing intervals, as suggested by the 72-week results of the BALATON and COMINO trials [16]. Our cohort included both treatment-naïve patients (n = 44, 62.8%) and those switched from prior anti-VEGF agents (n = 26, 37.2%), predominantly aflibercept. Despite a mean of 7.12 ± 4.73 prior injections, switch patients demonstrated meaningful improvements following faricimab initiation, achieving a mean visual gain of 0.38 logMAR. However, unadjusted analysis suggested greater visual gains in treatment-naïve patients (0.77 logMAR versus 0.41 logMAR; p = 0.006); however, after ANCOVA adjustment for baseline BCVA, CMT, age, RVO subtype, and prior injection burden, the adjusted between-group difference was 0.021 logMAR (95% CI −0.049 to +0.090) and did not reach statistical significance, suggesting the unadjusted difference was largely attributable to regression to the mean in naïve eyes rather than a true differential pharmacological effect. Nevertheless, the meaningful recovery observed in the switch group is consistent with the concept that incomplete responders to VEGF-A monotherapy may harbor ongoing Ang-2-mediated vascular pathology that is insufficiently addressed by conventional agents, as suggested by preclinical and clinical data on dual Ang-2/VEGF-A inhibition [5, 17, 18]. The dual-pathway inhibition of faricimab may therefore offer a meaningful therapeutic advantage in patients who are refractory or suboptimally responsive to prior anti-VEGF monotherapy [19]. The present study included both patients with BRVO (n = 37, 52.9%) and CRVO (n = 33, 47.1%), reflecting the real-world spectrum of RVO encountered in clinical practice. While both subtypes demonstrated significant functional and anatomical improvements following faricimab treatment, the BRVO subgroup achieved numerically superior final visual acuity (0.19 versus 0.35 logMAR), consistent with the well-established prognostic difference between these entities. BRVO is generally associated with a more limited extent of retinal nonperfusion, less severe macular ischemia, and a more favorable visual prognosis compared with CRVO, in which venous obstruction at the level of the optic disc produces more extensive retinal involvement and greater ischemic burden [1–3]. The LMM confirmed RVO subtype as a significant independent covariate (F[1,69] = 8.8, p = 0.004) after adjusting for baseline BCVA, age, treatment history, and prior injection burden. Importantly, ischemic status, a well-recognized determinant of visual prognosis in RVO, particularly in CRVO, was not systematically assessed in this retrospective cohort, as fluorescein angiography and OCT angiography were not uniformly performed across all 17 centers. This represents a meaningful limitation, as the proportion of ischemic versus non-ischemic cases within each subtype may substantially influence the observed outcomes and their generalizability. Future prospective studies should incorporate standardized ischemia assessment to allow more granular interpretation of subtype-specific treatment responses. Beyond the subgroup-level observations, the overall functional and anatomical trajectory across the entire cohort remained consistently favorable throughout the follow-up period. Overall mean BCVA demonstrated a statistically and clinically significant improvement from baseline to the final visit, with a mean gain of 0.58 logMAR, a magnitude of functional recovery that is directionally aligned with, though not directly comparable to, the visual acuity gains reported in both the BALATON and COMINO faricimab arms at week 24, given the substantial differences in study design, treatment intensity, follow-up duration, and patient composition between controlled trial and real-world settings [9]. Anatomically, overall CMT declined from 582.2 ± 176.3 to 281.4 ± 89.8 μm at the final visit. In treatment-naïve patients, CMT decreased from 594.1 ± 206.9 to 287.2 ± 89.9 μm (reduction: 306.9 ± 231.3 μm), while switch patients showed a reduction from 562.1 ± 107.3 to 271.5 ± 90.7 μm (reduction: 290.6 ± 114.3 μm). When examined by RVO subtype, the BRVO subgroup demonstrated a mean CMT reduction of 316.1 μm, which is broadly consistent with the −311.4 μm reported in BALATON. In the CRVO subgroup, the reduction of 283.8 μm was somewhat more modest than the −461.6 μm documented in COMINO. Importantly, direct comparison with these trial figures must be interpreted with caution: BALATON and COMINO employed fixed monthly loading injections over 6 months, whereas patients in the present study received a mean of only 2.49 ± 1.05 injections over approximately 24 weeks, reflecting the lower treatment intensity typical of real-world practice. This difference in injection frequency is a primary driver of the gap in anatomical outcomes and should be considered before drawing comparative conclusions. Taken together, these parallel improvements in both visual acuity and macular thickness suggest that faricimab delivers clinically meaningful functional and structural benefits across a heterogeneous real-world RVO population, in a manner that is broadly aligned with the therapeutic efficacy established in the pivotal phase III trials. Regarding safety, no serious ocular or systemic adverse events were observed during the study period, and mean IOP remained stable across all time points (baseline: 16.4 ± 3.1 mmHg; final visit: 16.5 ± 2.7 mmHg). These findings are consistent with the safety profiles reported in the pivotal BALATON and COMINO trials and available real-world studies. However, safety conclusions from the present study must be explicitly limited to the observed follow-up period of approximately 24 weeks in a cohort of 70 patients. Rare but clinically important inflammatory events associated with faricimab, including retinal vasculitis and occlusive retinal vasculitis, have been reported in postmarketing surveillance and pharmacovigilance analyses at estimated rates that are too low to be reliably detected in a cohort of this size [20, 21]. The absence of such events in our cohort should therefore not be interpreted as confirmation of safety beyond what the present sample size and follow-up duration permit. Continued real-world pharmacovigilance with larger cohorts and longer observation periods remains essential. This study has several important limitations that must be acknowledged. First, its retrospective and observational design introduces inherent risks of selection bias and confounding, while the absence of a comparator arm precludes direct efficacy comparisons with other anti-VEGF agents. Importantly, the between-group comparison of visual outcomes between treatment-naïve and switch patients was unadjusted for key confounders, including baseline BCVA, CMT, age, RVO subtype, and number of prior injections; these groups were not comparable at baseline, and the observed superiority of visual gains in naïve patients should be interpreted with caution. Second, although our cohort of 70 patients provides meaningful real-world data, the sample size may constrain the statistical power for more extensive subgroup analyses. Third, the heterogeneity of prior treatment history, which included both treatment-naïve patients and those who had received as many as 17 prior injections, introduces clinical variability that may influence outcome interpretation. Fourth, while some patients reached 52 weeks of follow-up, the mean duration of 23.69 weeks limits definitive conclusions regarding long-term durability and the potential for extended-interval dosing that was a hallmark of the BALATON and COMINO trials. Fifth, the absence of a standardized injection protocol across the 17 centers introduces variability in treatment intensity that may affect between-center comparability of outcomes and limits direct comparison with landmark trials such as BALATON and COMINO, which employed fixed monthly loading regimens. Sixth, in routine clinical practice, BCVA was recorded in decimal notation and subsequently converted to logMAR for statistical analysis, rather than being prospectively measured using standardized ETDRS letter scores. While clinically practical, this presents challenges in direct comparison with clinical trial data and may reduce precision in detecting subtle changes in visual function. Additionally, the absence of systematic fluorescein angiography or OCT angiography data across all participating centers precluded classification of patients by ischemic status, which is a recognized determinant of visual prognosis in RVO, particularly in CRVO, where ischemic cases carry substantially worse visual outcomes [1–3]. The inability to stratify outcomes by ischemic versus non-ischemic disease represents an important limitation of this retrospective multicenter study and may have introduced heterogeneity within the BRVO and CRVO subgroups that is not fully captured by the reported analyses. Finally, as a multicenter study involving 17 centers, variability in follow-up schedules and treatment decisions may introduce center-specific bias despite standardized data collection procedures. Additionally, the attrition analysis revealed that patients contributing data at all intermediate time points had significantly higher baseline CMT compared with those with incomplete intermediate follow-up (641.9 versus 487.2 μm; p < 0.001), indicating that completers had more severe disease at baseline and consequently received more injections. While this pattern reflects a clinically plausible relationship rather than differential dropout due to poor response, it introduces a potential source of bias at intermediate time points, and trends observed at those visits should be interpreted accordingly. Although repeated-measures ANOVA and Wilcoxon tests were used as the primary within-group analytic approach for consistency with the prespecified analysis plan, supplementary linear mixed-effects models were fit to accommodate the irregular follow-up intervals and missing intermediate observations inherent to this real-world dataset; the LMM findings were consistent with the primary analyses, supporting the robustness of the reported outcomes. No adjunctive treatments (including macular laser or intravitreal corticosteroids) were administered to any patient during the follow-up period. Data regarding the duration of RVO prior to faricimab initiation and the specific clinical reasons for switching from prior anti-VEGF agents were not systematically collected in this retrospective dataset and therefore could not be reported; these represent additional contextual variables that future prospective studies should capture. The absence of a formal exclusion criterion for concurrent ocular conditions such as glaucoma or optic neuropathy, which could independently reduce BCVA, means that residual confounding from undetected coexisting pathology cannot be entirely excluded. Additionally, systemic comorbidity data, including hypertension, diabetes mellitus, cardiovascular disease, and cerebrovascular disease, were not systematically collected in this retrospective dataset and were therefore not evaluated as potential predictors of visual or anatomical outcomes. These factors are recognized risk modifiers in RVO, and their absence from the analysis represents a further contextual limitation. In addition, although sex was recorded for all patients, sex-disaggregated outcome analyses were not performed, as no clinically meaningful differences in anti-VEGF treatment response by sex have been consistently demonstrated in RVO. Future prospective studies should evaluate potential sex-based differences in faricimab response. Conclusions The FARTURK-RVO study provides the first multicenter, real-world evidence on faricimab use for RVO in Türkiye, suggesting that its efficacy and safety profile in routine clinical practice are largely consistent with the pivotal BALATON and COMINO trials. A key finding of our study is the robust and rapid therapeutic effect of faricimab, with significant visual and anatomical improvements observed as early as 7 days after the first injection, underscoring the potential clinical importance of early intervention in preventing irreversible visual loss in RVO. Our data indicate that both treatment-naïve patients and those suboptimally responsive to prior anti-VEGF-A monotherapy appear to benefit from the dual inhibition of VEGF-A and Ang-2, which effectively reduces ME and stabilizes the retinal vasculature. The favorable safety profile within the observed study period, characterized by stable IOP and the absence of serious adverse events, is consistent with the known tolerability of faricimab, though definitive conclusions regarding rare inflammatory events require larger cohorts and longer follow-up. While these short-term results are highly encouraging, longer-term follow-up is warranted to confirm sustained efficacy and to evaluate the potential for extended dosing intervals in a real-world setting. Supplementary Information Below is the link to the electronic supplementary material. Supplementary file1 (PDF 197 KB) Supplementary file2 (XLSX 26 KB) Acknowledgments Medical Writing, Editorial, and Other Assistance No medical writing or editorial assistance was used in the preparation of this manuscript. The authors declare that no artificial intelligence (AI) tools were used in the drafting, writing, or preparation of this manuscript. Figures 1 and 2 were generated using Python (version 3.12) with the Matplotlib library (open-source MIT license), which permits unrestricted use and publication. Author Contributions Study conceptualization and design were contributed by Fevzi Akkan, Akın Çakır, Hakan Özdemir, Veysel Levent Karabaş, Ziya Kapran, and Melih Ünal. Study supervision was provided by Akın Çakır, Hakan Özdemir, Veysel Levent Karabaş, Melih Ünal, and Ziya Kapran. Material preparation and data collection were performed by Fevzi Akkan, Ebru Görgün, Burak Erden, Yavuz Kâmil Bardak, Nigâr Şerif, Mahmut Öztürk, Erkan Çelik, Selim Bölükbaşı, Şefik Can İpek, Tahsin Uzundede, Gamze Karataş, Murat Karapapak, Ece Özal, Osman Murat Uyar, Doğukan Cömerter, Mehmet Numan Alp, Serhat Ermiş, Anıl Korkmaz, Murat Arıcı, Mehmet Egemen Karataş, Nursal Melda Yenerel, Halil Özgür Artunay, Ecem Önder Tokuç, Veysel Levent Karabaş, and Akın Çakır. Statistical analysis was performed by Akın Çakır. The first draft of the manuscript was written by Ebru Görgün, Fevzi Akkan, and Akın Çakır. All authors commented on previous versions of the manuscript. All authors critically reviewed and approved the final version of the manuscript and agreed to be accountable for all aspects of the work. Funding No funding or sponsorship was received for this study or for the publication of this article. The Rapid Service Fee was funded by the authors. Data Availability The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request. Declarations Conflict of Interest Fevzi Akkan, Ebru Görgün, Burak Erden, Yavuz Kâmil Bardak, Nigâr Şerif, Mahmut Öztürk, Erkan Çelik, Selim Bölükbaşı, Şefik Can İpek, Tahsin Uzundede, Gamze Karataş, Murat Karapapak, Ece Özal, Osman Murat Uyar, Doğukan Cömerter, Mehmet Numan Alp, Serhat Ermiş, Anıl Korkmaz, Murat Arıcı, Mehmet Egemen Karataş, Nursal Melda Yenerel, Halil Özgür Artunay, Ecem Önder Tokuç, Veysel Levent Karabaş, Melih Ünal, Ziya Kapran, Hakan Özdemir and Akın Çakır declare that they have no competing interests. Ethical Approval This study was conducted in accordance with the Declaration of Helsinki and its later amendments. Ethical approval was granted by the Istanbul Medipol University Ethics Committee (approval ID: E-10840098-50.04-2136). Written informed consent for intravitreal treatment and the secondary use of anonymized clinical data was obtained from all participants prior to inclusion. References Associated Data Supplementary Materials Supplementary file1 (PDF 197 KB) Supplementary file2 (XLSX 26 KB) Data Availability Statement The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.