Long-Term Follow-Up of Faricimab Intravitreal Injections in Naïve Neovascular Age-Related Macular Degeneration

Abstract

Introduction

The aim of the study was to evaluate functional and anatomical changes at the 44-week follow-up in patients with naïve neovascular age-related macular degeneration (nAMD) treated with faricimab intravitreal injections (IVIs).

Methods

Fifty-four eyes of 54 patients with naïve active macular neovascularization and nAMD were enrolled at the Ophthalmology Clinic of University “G. d’Annunzio,” Chieti-Pescara, Italy. All patients were scheduled for faricimab IVI. Each patient underwent complete ophthalmic examination including best-corrected visual acuity (BCVA) using the Early Treatment Diabetic Retinopathy Study (ETDRS) charts and optical coherence tomography. All measurements were evaluated at baseline, at week 20 and then according to fixed retreatment interval up to week 44. Fluorescein angiography and indocyanine green angiography were also performed at baseline. Main outcome measures were changes in BCVA, central macular thickness (CMT), subfoveal choroidal thickness (SFCT), intraretinal fluid presence, subfoveal subretinal fluid presence and thickness, the presence of pigment epithelial detachments (PEDs), and its maximum height (PED-MH).

Results

BCVA improved and CMT reduced significantly from baseline to week 44 (p = 0.002 and p = 0.020, respectively) in the overall sample with a higher significant improvement from baseline to week 20 (p < 0.001 for both parameters) and no additional significant improvement from week 20 to week 44 in the overall sample (p = 0.348 and p = 0.146, respectively). At week 20, 72.3% of patients were in the Q12/Q16 interval. Patients in the Q8 interval showed significant improvement in BCVA (p < 0.001) and significant reduction of CMT (p = 0.008) from baseline to week 44, respectively. PED-MH as well showed a significant reduction from baseline to week 44 (p < 0.001). Patients in the Q12 interval showed significant improvement in BCVA (p = 0.030) and significant reduction of CMT, SFCT, and PED-MH from baseline to week 36 (p < 0.001). Patients in the Q16 interval showed significant reduction of BCVA during follow-up (p = 0.026). CMT, SFCT, and PED-MH were significantly reduced from baseline to week 44 (p < 0.001).

Conclusion

Faricimab showed efficacy in the treatment of naïve nAMD patients with an improvement of many anatomical and functional parameters at 44 weeks, allowing the maintenance of a treatment regimen for most patients equal to or greater than 12 weeks.

Introduction

Age-related macular degeneration (AMD) is a chronic and progressive condition that affects the macula. It has been estimated that AMD affects approximately 196 million people worldwide and that it is the leading cause of irreversible central vision loss in individuals over 65 years old [1]. AMD is classified into two different forms based on the presence or absence of neovascularization: neovascular AMD (nAMD) and dry AMD, respectively. Although nAMD accounts for only 10–15% of all cases, it is estimated to be responsible for more than 80% of all cases of severe AMD-related visual loss [2, 3]. The introduction of drugs targeting vascular endothelial growth factor (VEGF) has substantially improved visual outcomes for patients with nAMD. Despite the proven positive effect of these molecules, these drugs only target VEGF [4]. In the literature, it is reported that response to anti-VEGF therapy varies a lot among patients: approximately 30% of patients are considered non-responders [5]. For this reason, it would be desirable to search for new therapeutic targets.

Faricimab is a novel bispecific antibody characterized by a dual mechanism of action that concurrently inhibits vascular endothelial growth factor A (VEGF-A) and angiopoietin-2 (Ang2). It represents the first intraocular therapeutic agent administered via intravitreal injection (IVI) designed to target both pathways. Ang2 plays a pivotal role in vascular pathophysiology by modulating angiogenic activity and endothelial permeability. Based on the results of phase 3 regulatory trials demonstrating robust efficacy and a favorable safety profile, faricimab has received approval from both the FDA and the EMA for the treatment of nAMD. The present study aimed to assess the real-world efficacy and durability of faricimab over a 44-week follow-up period [6].

Methods

Participants

Fifty-four subjects (54 eyes) with a diagnosis of AMD and active naïve macular neovascularization were referred to the Ophthalmology Clinic of University “G. d’Annunzio,” Chieti-Pescara, Italy from February to July 2023 and were enrolled in this study according to defined inclusion and exclusion criteria. All patients were treated with intravitreal faricimab injections as per the label of the drug. Exclusion criteria were IVIs of anti-VEGF or other previous treatments for nAMD, photodynamic therapy, other ophthalmological diseases, autoimmune diseases, ocular media opacities preventing the optical coherence tomography (OCT) execution and low quality of the scans. The study was conducted in accordance with the tenets of the Declaration of Helsinki and received approval from the Institutional Review Board of the “G. d’Annunzio” University of Chieti-Pescara on November 24, 2022. Written informed consent was obtained from all participants prior to enrollment. In accordance with the standard treatment protocol, all patients received a loading phase consisting of four consecutive monthly IVIs of faricimab (0.05 mL solution containing 6 mg of the drug).

Starting from week 20, a disease activity (DA) assessment based on functional and anatomical criteria (decrease of >5 letters of best-corrected visual acuity (BCVA) compared with week 12 assessment visit and presence of intraretinal fluid (IRF) and/or SRF, new neovascularization or new macular hemorrhage) was performed by investigator (L.T.). Patients with DA at week 20 were included in an 8-week interval (Q8), patients with DA at week 24 were included in a 12-week interval (Q12) and patients with no DA at 20 and 24 weeks were included in a 16-week interval (Q16). The fixed treatment intervals were maintained for each patient up to 44 weeks.

Following the first year of treatment, a treat-and-extend regimen was planned for all patients. Each participant underwent a comprehensive ophthalmic evaluation, including BCVA assessment using Early Treatment Diabetic Retinopathy Study (ETDRS) charts, Goldmann applanation tonometry, slit-lamp biomicroscopy of the anterior segment, and indirect ophthalmoscopy of the posterior pole. Structural retinal imaging was performed using spectral-domain optical coherence tomography (SD-OCT; Spectralis HRA+OCT, Heidelberg Engineering, Heidelberg, Germany).

All examinations were conducted at baseline (T0) and subsequently at week 20 (T1), week 28 (T2), week 36 (T3), and week 44 (T4). Fundus fluorescein angiography and indocyanine green angiography were additionally performed at baseline.

The main outcome measures included BCVA, central macular thickness (CMT), subfoveal choroidal thickness (SFCT), presence of IRF, presence and thickness of subfoveal subretinal fluid, presence and maximum thickness of subretinal fluid within the 6-mm ETDRS grid (MSRFT), and presence and maximum height of pigment epithelial detachment (PED-MH) within the same grid.

SD-OCT Analysis

The acquisition protocol included horizontal and vertical line B-scans centered on the fovea using enhanced depth imaging mode, 49 horizontal raster dense linear B-scans centered on the fovea. Post-baseline imaging acquisitions were conducted using the follow-up mode to ensure precise alignment with the initial scans. All the parameters (CMT; IRF; SRF; SRFT; MSRFT and PED-MH) were measured as reported in Toto et al. [7]. All quantitative measurements were independently performed by two experienced, certified graders. The intraclass correlation coefficient between the two readers was calculated for all quantitative measurements.

For qualitative parameters, OCT images were reviewed independently by two graders (L.T. and L.B.B.). In case of disagreement, a third retinal specialist (R.M.) was consulted.

Statistical Analysis

Descriptive statistics were reported as median and interquartile range (first quartiles; third quartiles) for continuous variables, while categorical data were summarized as absolute frequency and percentage. Statistical units were eyes, independently analyzed. The Shapiro-Wilk test was performed to verify the normal distribution of data. A linear mixed model for repeated measurements was employed to assess the impact of time on the primary outcome measures (BCVA, CMT, SFCT, PED-MH). A mixed model is an effective technique for examining data from longitudinal studies when each participant is measured more than once. This method considers the correlation between repeated measurements on the same subject and enables explicit modeling of within-person and between-person variance in the outcome. Baseline and different follow-up measures were regressed on the fixed-effect components, assuming an unstructured covariance matrix, using a linear mixed model for repeated measurements. In addition, the Durbin-Conover post hoc test correction for multiple comparison tests was applied. Spearman’s rank correlation coefficient was applied, stratified for the treatment interval, to evaluate at baseline and, at the end of follow-up, the relations between parameters were analyzed. All tests were two-tailed, with a significance level set at p ≤ 0.05. Statistical analyses were performed using R software environment for statistical computing and graphics (version 4.2; https://www.r-project.org/).

Results

Fifty-four eyes from 54 treatment-naïve patients diagnosed with nAMD and active macular neovascularization were included in the study. The mean age of the participants was 70.7 ± 17.9 years (range, 50–86 years). Of the 54 patients, 27 were men; 18 study eyes were right and 36 were left. According to consensus nomenclature for nAMD by Spaide et al. [8] patients were classified as having: type 1 MNV (21 patients), mixed MNV type 1 and type 2 (12 patients), type 3 MNV (3 patients), and polypoidal choroidal vasculopathy (PCV) (18 patients) [9]. None of the patients included in the study were lost during follow-up. All patients were evaluated at baseline and at week 20.

BCVA improved (median value at T0: 0.35 LogMAR: median value at T4: 0.10 LogMAR). The intraclass correlation coefficient between the two readers for all quantitative OCT parameters was >0.98.

CMT reduced (median value at T0: 332.50 μm: median value at T4: 264.50 μm) significantly from baseline to week 44 (p = 0.002 and p = 0.020, respectively) in the overall sample with a higher significant improvement from baseline to week 20 (p < 0.001 for both parameters) and no additional significant improvement from week 20 to week 44 in the overall sample (p = 0.348 and p = 0.146, respectively). In addition, SFCT did not significantly modify from baseline to week 44 (p = 0.054).

At week 20 after the disease activity assessment, 27.7% of patients were in the Q8 interval, 11.2% were in the Q12 regimen, and 61.1% were in the Q16 regimen. The various weeks of assessment are shown for each group in Figure 1. Patients in the Q8 interval showed significant improvement in BCVA (p < 0.001) and significant reduction of CMT (p = 0.008) from baseline to week 44, respectively (Fig. 2, 3). PED-MH as well showed a significant reduction from baseline to week 44 (p < 0.001) (Fig. 4). Patients in the Q12 interval showed significant improvement in BCVA (p = 0.030) and significant reduction of CMT, SFCT, and PED-MH from baseline to week 36 (p < 0.001) (Fig. 2–5). Patients in the Q16 interval show significant improvement in BCVA during follow-up (p = 0.026). CMT, SFCT, and PED-MH were significantly reduced from baseline to week 44 (p < 0.001) (Fig. 2–5). Patients with intraretinal and/or subretinal fluid in different treatment regimens at different follow-up points are reported in Table 1.

Fig. 1.

Time points of patients’ assessment displayed for each treatment interval.

Fig. 2.

BCVA changes displayed for each treatment regimen during different follow-up periods.

Fig. 3.

CMT changes displayed for each treatment regimen during different follow-up periods.

Fig. 4.

PED-MH changes displayed for each treatment regimen during different follow-up periods.

Fig. 5.

SFCT changes displayed for each treatment regimen during different follow-up periods.

Table 1.

Number of patients (percentages) with any type of fluid at different follow-up points

​	Q8	Q12	Q16
Baseline	15 (100.0)	6 (100.0)	33 (100.0)
Week 20	15 (100.0)	0 (0)	0 (0)
Week 24	​	6 (100.0)	0
Week 28	15 (100.0)	​	3 (9.0)
Week 36	15 (100.0)	6 (100.0)	​
Week 44	15 (100.0)	​	9 (33.3)

Ordinal regression analysis shows that baseline values of functional and anatomical parameters are predictors of treatment-interval group (shown in Table 2). In particular, patients with higher BCVA values and greater anatomical values alias higher CMT, SFCT and PED-MH values have high probability to fall in the Q8 interval than in the Q16 interval. No safety issues were reported in the overall sample.

Table 2.

Ordinal logistic regression analysis results

​	OR (95% CI)	Q12-treatment interval vs. Q16-treatment interval	Q16-treatment interval vs. Q8-treatment interval
BCVA, logMAR	3.16 (0.93–10.67); p = 0.050	OR = 0.25; p = 0.005	OR = 3.31; p = 0.011
CMT, μm	1.01 (1.00–1.02); p = 0.040	OR = 2.22; p = 0.546	OR = 30.96; p = 0.015
SFCT, μm	1.03 (1.00–1.05); p = 0.032	OR = 1.34; p = 0.779	OR = 19.68; p = 0.010
PED-MH, μm	1.01 (1.00–1.02); p = 0.020	OR = 0.37; p = 0.072	OR = 5.41; p = 0.003

Treatment-interval group was defined as the outcome and baseline parameters as predictor.

OR, odds ratio; 95% CI, 95% confidence interval.

Discussion

The present study aimed to evaluate functional and anatomical changes during the 44-week follow-up in patients with nAMD treated with faricimab IVIs. After the loading dose, 39 out of 54 eyes (72.3%) were in a treatment interval equal or greater than 12 weeks. There was an improvement of BCVA in patients with all treatment intervals with higher statistical significance for the Q8 interval compared to the other two intervals. For what concerns anatomical parameters, our study showed a statistically significant reduction of CMT and PED-MH in all groups. SFCT was significantly reduced in the 12- and 16-week treatment intervals from baseline to 36 and 44 weeks, respectively. TENAYA and LUCERNE are two phase 3 randomized clinical trials in which 1,329 eyes were treated with faricimab showing non-inferiority to aflibercept, meeting the primary endpoints. Moreover, faricimab treatment resulted in a reduction in CMT similar to the aflibercept group. In addition, 78% of patients in both studies achieved dosing intervals of 12 weeks or more and 45% of patients reached the maximum dosing interval of 16 weeks by Year 1. For what concerns safety, the occurrence of side effects was similar across all groups and consistent with expected effects in people receiving intravitreal nAMD treatment: no cases of retinal vasculitis or occlusive vascular events were reported [8]. Mori et al. [10] analyzed 133 eyes of Japanese patients, a subgroup of the phase 3 TENAYA trial including a greater proportion of PCV, showing an increase of +7.1 letters for the faricimab group and +7.7 letters for the aflibercept group. At week 48, the faricimab group had 66.1% of patients on a 16-week dosing interval. It also showed a reduction of mean CMT at 1 year follow-up that was maintained during year 2 of follow-up [10]. Different real-world studies also confirmed the results of registrational clinical trials concerning the efficacy and safety of faricimab in AMD-related MNV. Leung et al. [11] analyzed 190 eyes that received at least 3 injections of intravitreal faricimab with at least 3 months of follow-up showing an improvement of BCVA from 0.33 ± 0.32 logMAR to 0.27 ± 0.32 logMAR, CMT also reduced from 312 ± 87 μm to 287 ± 71 μm and 24% of patients had no subretinal/IRF on OCT at the last visit. The average dosing interval between the last two consecutive faricimab injections was 7.64 ± 6.2 weeks, which was significantly longer compared to ranibizumab (5.16 ± 2.0 weeks) or aflibercept (5.57 ± 3.6 weeks) [11]. The TRUCKEE study, a real-world multicenter retrospective study, evaluated 376 eyes treated with faricimab IVIs at the 6-month follow-up. All eyes showed a BCVA improvement of +1.1 letters (p = 0.035) and a CST reduction of 31.3 μm (p < 0.001). Treatment-naïve eyes (n = 39) demonstrated a BCVA improvement of +4.9 letters (p = 0.076) and a CST reduction of 84.5 μm (p < 0.001) [12]. Mukai et al. [13] evaluated 63 eyes treated with faricimab IVIs, showing that 82% of the eyes achieved a dry macula at month 3 (defined as the absence of intraretinal or subretinal fluid). In addition, complete regression of polypoidal lesions was observed in 52% of eyes with PCV [13]. Similarly to these reported studies, we observed a statistically significant functional improvement during follow-up in all patients. Interestingly, for patients in the Q12 and Q16 regimen of treatment, there was a significant improvement after the loading dose and a stability of the result during the maintenance phase, while in the Q8 group, there was a continuous improvement of BCVA during all follow-up with statistical significance versus previous time point up to week 36. We hypothesize that this result could be explained by the fact that patients included in Q12 and particularly Q16 regimen had the highest visual acuity at baseline, which may have induced to less gain of BCVA probably related to a ceiling effect. It is reported in the literature that CMT and, particularly, SRF and IRF are good biomarkers to monitor AMD’s activity and to evaluate the response to anti-VEGF IVIs. Also, PED-MH is a good biomarker of NV activity [14]. Saint-Geniez et al. [15] reported that RPE is implicated in the development and maintenance of the choroidal vasculature by producing VEGF that bonds to VEGF receptors that are expressed in the RPE-choroid complex. Otani et al. [16] suggested that Ang2 and VEGF are co-upregulated in patients with CNV and that Tie2 is overexpressed in many cell types of CNV further supporting a crucial role of the interaction between VEGF and Ang2 in pathologic angiogenesis. From this evidence, faricimab could regulate choroidal vessel permeability and reduce intra- and/or subretinal fluid production. Regarding PED height modification, Cho et al. [17] showed a reduction in PED height at 12 months follow-up of 202 eyes affected by naïve nAMD treated with IVIs of anti-VEGF. After 12 months of treatment, the mean PED height decreased from 453 ± 261 μm at baseline to 230 ± 142 μm (p = 0.002). Also, Veritti et al. [18] showed a reduction in PED-MH in twenty-two eyes of 22 patients affected by naïve nAMD and treated with 4 IVIs of faricimab. PED-MH reduced in 84% treated eyes at 4 months. We observed a significant improvement of CMT and PED-MH in all the three groups of patients during follow-up. CMT mainly reduced post-loading in Q12 and Q16 groups and continuously during follow-up in the Q8 group. We also demonstrate a statistically significant reduction of mean PED-MH in all three regimens of treatment from baseline to week 44 for Q16/Q8 regimens of treatment (p < 0.001) and to week 36 for Q12 regimens of treatment (p < 0.001). We hypothesized that faricimab, targeting two different molecules (VEGF-A and Ang2), could be more effective in controlling neovascular phenomena and vascular permeability, leading to fluid accumulation in the sub-RPE space. SFCT has also been found to reduce after anti-VEGF treatment in several clinical studies [19].

Eckardt et al. [20] showed a reduction of SFCT after the loading dose of faricimab in patients switched from other anti-VEGF agents (p = 0.029). Takahashi and colleagues [21] in patients switched to faricimab from brolucizumab showed a significant reduction of subfoveal thickness after also in patients with polypoidal choroidal vasculopathy. In our study, mean SFCT reduced from 304 μm (228–334.5 μm) at baseline to 277 μm (180.5–311.0 μm) (p < 0.001) at week 44 in patients in Q16 regimen, from 241 μm (115–367 μm) at baseline to 188.5 μm (70–307 μm) (p < 0.001) at week 36 in patients in Q12 regimen, and from 342 μm (281–360 μm) at baseline to 329 μm (222–397 μm) (p = 0.780) at week 44 in patients in Q8 regimen. This reduction suggests an effect of faricimab in reducing the diameter of dilated choroidal vessels and the choroidal vascular permeability and, thus, with a decrease in nAMD activity. In fact, it is known that VEGF receptors and Ang1/Tie2 receptors are expressed in the choroid [22, 23]. Ordinal regression analysis showed a correlation between functional and anatomical baseline values and a treatment regimen with higher logMAR BCVA values and CMT, SFCT, and PED-MH values, leading to a higher probability to fall in the Q8 interval than in the Q16 interval.

Conclusions

Faricimab is the first approved bispecific antibody for nAMD treatment. The 44-week follow-up performed on our patients showed a statistically significant improvement of both functional (BCVA) and anatomical parameters (CMT, SFCT, and PED-MH) in different regimens of treatment after the loading phase (Q8 and Q12/Q16). The optimal effect in reducing intra- and/or subretinal fluid after the loading dose (4 faricimab IVIs) let us to extend the regimen of treatment to a Q12/Q16 for the majority of patients. In this way, faricimab demonstrates to hold the potential to alleviate the treatment burden associated with nAMD, allowing us to keep satisfying results (both anatomical and functional) also in a long follow-up period (44 weeks). Other studies will be necessary to evaluate even longer term effects as well as eventual adverse effects of this new drug.

Statement of Ethics

The study adhered to the principles outlined in the Declaration of Helsinki and was approved by the Institutional Review Board of the University “G. d’Annunzio” of Chieti-Pescara (2/2023). Written informed consent was obtained from all participants prior to their inclusion in the study.

Conflict of Interest Statement

The authors declare that they have no known financial or personal relationships that could have appeared to influence the work reported in this paper.

Funding Sources

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author Contributions

L.B.B. conceptualized and designed the study. L.T. and L.B.B. were responsible for the analysis of all quantitative and qualitative data. R.M. served as the adjudicator to resolve discrepancies among the graders. M.D.N. performed the statistical analysis. L.T., L.B.B., F.F., M.P., M.L.R., R.A., and A.Q. carried out image acquisition (OCT and OCTA). L.B.B., L.T., and R.M. primarily drafter the manuscript and also critically reviewed it for intellectual content. All co-authors approved the final version of the manuscript.

Funding Statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data Availability Statement

All data analyzed in this study are included in this article. Further inquiries can be directed to the first author or to the corresponding author.