Efficacy and Durability of Faricimab in Naïve Eyes with Neovascular Age-Related Macular Degeneration

Age-related macular degeneration (AMD) is one of the leading causes of blindness at the global level. It has been estimated that among adults aged over 50 years, 33.6 billion were blind in 2020 and 1.8 million cases were attributable to neovascular AMD (nAMD) [1]. It has also been reported that there was a decrease by almost 30% of the age-standardized prevalence of blindness due to AMD; this trend could be clearly associated with the introduction of anti-vascular endothelial growth factor (anti-VEGF) inhibitor treatment for the nAMD [1].

In this context, it is clear the role of new therapeutic agents that enable the appropriate treatment of nAMD patients is to stop the progression of their vision loss while keeping the treatment burden sustainable for them and the healthcare system. Faricimab is a novel bispecific antibody with a dual mechanism of action that simultaneously and independently inhibits both VEGFA and angiopoietin-2 (Ang2) [2]. It is a human IgG1-like CrossMAb, a bispecific molecule constituted by two antigen-binding domains for Ang2 and VEGFA and a fragment crystallizable (Fc) portion which has been engineered and optimized to reduce systemic exposure by eliminating binding interactions with neonatal receptors Fc and Fcγ [2, 3].

Faricimab has been approved by FDA and EMA for nAMD based on the results of TENAYA and LUCERNE phase 3 trials demonstrating efficacy in improving and maintaining visual acuity while reducing macular thickness up to 112 weeks compared to aflibercept 2 mg [4, 5]. The innovative study design allowed the patients in the faricimab 6.0 mg arm to be extended at 2, 3, or 4 months immediately after the four monthly injections of the loading phase, leading to almost 60% and 80% of patients treated every 4 months and ≥3 months, respectively, at 112 weeks. Faricimab also showed better fluid reductions compared to aflibercept 2 mg [5]. Faricimab is confirmed to be efficacious and safe even in the real-world setting, as shown by some studies conducted on naive and previously treated patients with nAMD [6‒9].

The aims of our study were to describe the first Italian real-world experience since its commercialization in Italy and report the efficacy, safety, and durability of faricimab. We herein report the early functional and anatomical changes after faricimab treatment administered as per label in nAMD patients and the percentage of patients assigned to q8w, q12w, or q16w right after the loading phase.

Eighteen eyes of 18 subjects with a diagnosis of AMD and active naïve macular neovascularization (MNV) referring to the Ophthalmology Clinic of University G. D’Annunzio, Chieti-Pescara, Italy, from February to May 2023 fulfilling defined inclusion and exclusion criteria were consecutively enrolled in this prospective interventional study. All patients were treated with intravitreal faricimab injections as per label.

Exclusion criteria were photodynamic therapy, intravitreal injections (IVIs) of anti-VEGF or other previous treatments for AMD-related MNV, concomitant ophthalmological diseases, autoimmune diseases, ocular media opacities preventing the optical coherence tomography (OCT) execution, poor collaboration, and low quality of the scans. The study complied with the tenets of the Declaration of Helsinki and was approved by “University G. D’Annunzio” of Chieti-Pescara institutional board. All patients provided written informed consent for participation in the study.

According to the standard treatment, all patients were injected with 0.05 mL solution of faricimab 6 mg. Patients received a loading dose of four initial monthly doses of faricimab, and then at week 20, a first disease assessment was performed by investigator (L.T.); thus, based on functional and anatomical criteria, patients were shifted toward an 8-week interval (Q8) in the presence of disease activity or a 12/16-week interval (Q12/16) in case of the absence of disease activity with a further evaluation at week 24 for Q12 assignment in case of disease activity or a Q16 assignment in case of the absence of active disease. Patients will be treated with the same interval defined for each patient at weeks 20, 24, and 28 up to 12 months, and after the first year, a treat and extended regimen will be adopted.

All patients underwent comprehensive ophthalmic examination including best-corrected visual acuity (BCVA) using the Early Treatment Diabetic Retinopathy Study (ETDRS) chart, Goldmann applanation tonometry, anterior segment examination through slit lamp examination, indirect fundus oculi examination through ophthalmoscopy, spectral domain OCT (SD-OCT) using Spectralis HRA + OCT (Heidelberg Engineering, Heidelberg, Germany). All examinations were performed at baseline (T0) and week 4 (T1), week 8 (T2), week 12 (T3), and week 20 (T4). In addition, fundus fluorescein angiography and indocyanine green angiography were performed at baseline. The main outcome measures were BCVA, central macular thickness (CMT), subfoveal choroidal thickness (SFCT), macular intraretinal fluid (IRF) presence, subfoveal subretinal fluid (SSRF) presence and thickness (SSRFT), SRF presence and maximum SRF thickness in the 6-mm ETDRS grid (MSRFT ETDRS grid), pigment epithelium detachment (PED) presence and maximum height (PED-MH) in the 6-mm ETDRS grid.

The acquisition protocol included 49 horizontal raster dense linear B-scans centered on the fovea, vertical and horizontal B-scans centered on the fovea with enhanced depth imaging mode. All acquisitions following the baseline were performed using the follow-up function.

CMT was measured using the central 1-mm-diameter circle of the ETDRS thickness map. SSRFT, defined as the vertical distance between the end of the outer segment and the RPE at the foveal center, was measured using the inbuilt manual caliper.

MSRFT ETDRS grid defined as the greatest vertical distance between the end of the outer segment and the RPE within the central 6-mm-diameter circle of the ETDRS grid was measured using the inbuilt manual caliper. PED was defined as the separation between the basal lamina of RPE and the inner collagenous layers of Bruch’s membrane and was manually measured in its PED-MH using the caliper tool using the scan revealing the most prominent lesion site within the central 6-mm-diameter circle of the ETDRS grid.

SFCT, measured vertically from the outer border of the RPE to the inner border of the sclera, was measured using the inbuilt manual caliper on enhanced depth imaging OCT scans. IRF, SSRF, and SRF in the central 6-mm-diameter circle of the ETDRS grid (SRF ETDRS grid) were defined as present, absent, or unreadable.

Images with poor signal strength (<25) were excluded and thus repeated. Two independent, experienced readers performed all quantitative measurements, and the average of measurements was considered for analysis.

For qualitative parameters, OCT images were reviewed independently by two graders (L.T. and F.F.). In case of disagreement, a third retinal specialist (R.M.) was consulted to resolve the discrepancies between graders by open adjudication.

Patients underwent OCTA imaging using Spectralis HRA-OCTA (version 6.12.4.0; Heidelberg Engineering, Heidelberg, Germany). In order to assure coverage of the entire lesion, 10° × 10° or 15° × 10° volume scans were performed according to the extension of the MNV. The most adequate slab to identify MNV was chosen in each case. The avascular slab and choriocapillaris (CC) slab were obtained using the default layer segmentation settings of the device; on the contrary, the outer retina to CC segmentation slab was set manually, extending from the outer boundary of the outer plexiform layer to 8 µm beneath Bruch’s membrane. The projection artifact removal algorithms were used. The images of all devices were exported from each device as JPEG files and then analyzed with Image J software version 1.52° (National Institutes of Health, Bethesda, MD, USA; available at http://rsb.info.nih.gov/jj/index.html). The NV lesions were manually circumscribed on en-face images by two independent retinal specialists (L.T. and F.F.), and the flow within this area was calculated as the number of pixels over a nonperfusion threshold and then converted in a comparable mm² area value. For NV flow area for statistical analysis, each subject was attributed the mean value between reader 1 and reader 2.

Assuming a statistical power of 80% and a significance level of 5% (two-sided), the study requires a sample size of 18 patients to detect an effect size of 0.74 in BCVA (logMAR) changes. The Shapiro-Wilk test was used for normality analysis. The descriptive analysis reports, by the distribution of the variables, the median (q1 = first; q3 = quartile) quartiles for continuous variables and the absolute frequency and percentage for categorical variables. The Pearson correlation coefficient (r) assessed the relationship between the continuous variables at baseline and week 20. The Friedman test for paired samples was used to compare the change over time of BCVA, CMT, SCT, and PED, followed by a pairwise post hoc test for multiple comparisons of rank sums according to Durbin and Conover. The Wilcoxon rank sum test compared FA values between baseline and week 20. The categorical variables were compared using the McNemar test to assess baseline and week 20 changes. For cells with zero, we applied a correction of +1. All tests were two-tailed, and the significance level (alpha) was set to 0.05. The analysis was carried out with the R software (A Language and Environment for Statistical Computing, https://www.R-project.org/).

Eighteen patients (18 eyes) with AMD and active MNV were included in this study. According to consensus nomenclature for reporting neovascular AMD by Spaide et al. [10], patients were classified as follows: 7 patients had type 1 MNV, 6 had polypoidal choroidal vasculopathy (PCV), 4 patients had mixed MNV type 1 and type 2, and 1 patient had type 3 MNV. The mean age was 70.72 ± 17.91 years (range from 50 to 86 years), and 9 out of 18 patients were men, 6 were right eyes, and 12 were left eyes.

SD-OCT images were acceptable for all eyes, and qualitative and quantitative parameters were scored in all cases. Median BCVA improved significantly over time from 0.4 [0.1; 0.7] logMAR at T0 to 0.1 [0.1; 0.4] logMAR at T4 (p < 0.001). Median CMT changed significantly from 332.0 [298.0; 392.0] μm at T0 to 264.0 [244.0; 286.0] μm at T4 (p < 0.001), and median SFCT reduced significantly from 308.0 [220.0; 345.0] μm at T0 to 266.0 [224.0; 297.0] μm (p = 0.031). In addition, median PED-MH reduced not significantly over time from 176.5 [93.0; 284.0] μm at T0 to 106.5 [61.8; 168.0] μm at T4 (p = 0.228) (Table 1). It reduced in 58.8% of patients at T4. Median SSRFT reduced from 61 μm at T0 to 22 μm at T3 and slightly increased to 40 μm at T4, whereas MSRFT ETDRS grid reduced from 93 μm at T0 to 75.5 μm at T3 and 75 μm at T4 (shown in Fig. 1). SSRF presence reduced ranging from 55.6% at T0 to 16.7% at T4 (p = 0.045); SRF ETDRS grid presence modified from 61.1% at T0 to 27.8% at T4 (p = 0.077); IRF presence changed from 50% at T0 to 22.2% at T4 (p = 0.074) (shown in Fig. 2). The median FA modified significantly from 0.021 mm² to 0.010 mm² (p = 0.002) (shown in Fig. 3, 4).

Pearson correlation coefficients (r) and p value for continuous variables at baseline and week 20 are shown in Figure 5. At baseline, the correlation analysis showed a significant correlation between SSRFT and CMT (r = 0.51) and MSRFT ETDRS grid (r = 0.58). At week 20, the MSRFT ETDRS grid, CMT, and PED-MH positively correlated with SSRFT (r = 0.74, r = 0.65, r = 0.56). In addition, a significant positive correlation was found between PED-MH and CMT (r = 0.73, p < 0.001).

Five eyes out of 18 eyes were in the q8 interval when evaluated at week 20, and 13 eyes were in the q12/q16 interval. Faricimab showed an overall well-tolerated safety profile. Neither adverse serious ocular events (endophthalmitis, retinal tear, rhegmatogenous retinal detachment, increase in intraocular pressure, traumatic cataract, intraocular inflammation, retinal vasculitis, retinal vascular occlusion), nor adverse systemic events including Antiplatelet Trialists’ Collaboration events were observed. Conjunctival hemorrhage was the only ocular adverse events reported by the investigators at the site of IVI.

Faricimab is a novel intravitreal drug for neovascular AMD with dual mechanism of action suppressing VEGF- and Ang-2-modulated pathology and thus has been demonstrated to provide superior disease control. This study evaluates the effect of anti-VEGF therapy with faricimab administered as per label in patients with treatment-naïve AMD-related MNV.

In our study, there was a reduction of CMT related to the resolution/reduction of IRF and subretinal fluid and a reduction of MNV flow area. The anatomical outcomes were paralleled by a significant improvement of visual acuity. Also, significant changes were observed for SFCT showing a reduction over time, while PED maximal height did not reveal a significant reduction. Thirteen out of 18 eyes (72.3%) were in a treatment interval equal or greater than 12 weeks.

TENAYA and LUCERNE phase 3 trials including treatment-naïve AMD patients with MNV demonstrated significant improvement of visual acuity meeting the primary endpoint of non-inferiority to aflibercept and a significant reduction of central retinal thickness similar to the aflibercept group. In addition, 78% of patients had a treatment interval of 12 weeks or more at 1 year [4, 5]. In the Japanese subgroup of the phase 3 TENAYA including a greater proportion of patients with PCV compared with patients in the pooled global TENAYA/LUCERNE trials, there was a higher proportion of patients (66%) on Q16W treatment intervals compared with patients in the pooled global TENAYA/LUCERNE trials (45%) [11].

Real-world studies in naïve patients or previously treated patients with other anti-VEGF confirmed the results of registrative clinical trials concerning the efficacy and safety of faricimab in AMD-related MNV [6‒9, 12‒15]. A study by Rush et al. [14] in eyes previously treated with aflibercept with persistent IRF and/or subretinal fluid on OCT showed at the 4-month follow-up after a loading dose of 3 IVI of faricimab an anatomical and functional improvement with 39.3% of patients without retinal fluid.

Another retrospective study by Leung et al. [13] showed a significant reduction of central retinal thickness and an increase of visual acuity after IVI of faricimab in eyes previously treated with other anti-VEGF with an extension of the treatment interval treatment with a faricimab dosing interval being longer than for the other anti-VEGF agents. Stanga et al. [15] described the efficacy of faricimab IVI in previously treated and naïve eyes (11 eyes) with nAMD. A complete resolution of retinal fluid was observed in 75% of the naïve eyes after the first injection.

A multicenter study in USA showed after a loading phase of faricimab a resolution of IRF in 25% of eyes and of SRF in 40% of eyes and a resolution of PED in 41% of eyes [9]. Matsumoto et al. [8] investigated the efficacy of faricimab in naïve eyes (40 eyes) with nAMD treated with a loading dose of three IVIs of faricimab showing a significant reduction of CMT with complete resolution of IRF and SRF in 79.5% of the eyes, a corresponding improvement of visual acuity, and additionally a decrease of central choroidal thickness. The authors also detected a regression of polypoidal lesions evaluated with indocyanine green angiography in patients with PCV.

Another study in Japan including naïve eyes with nAMD treated with a loading dose of 3 faricimab demonstrated an improvement of CMT and BCVA and a dry macula with complete resolution of retinal fluids in 82% of the eyes and a regression of PCV lesion in 52% of the eyes with associated reduction of choroidal thickness [12]. CMT and retinal fluid are important parameters, and OCT biomarkers are used to monitor nAMD activity and are frequently used to assess efficacy of anti-VEGF drugs [16].

Similarly to registrative trials and other real-world studies, we observed a significant reduction of CMT and a related improvement of BCVA. The anatomical improvement was related to retinal fluid control both intraretinal and subretinal during treatment. At week 20, our patients showed a percentage of fluid-free eyes of 72.3%, which was comparable to the percentage reported in the registrative trials, leading to the majority of patients being included in a 12 or more treatment interval. This percentage was also similar to the percentage reported in other real-world studies, although most of these studies had a loading dose of 3 IVIs.

PED vertical diameter is another OCT biomarker of disease activity in nAMD, and its modification is considered related to the exudative state of the disease [17]. In our study, similarly to other studies on faricimab in nAMD, we found a reduction of PED-MH from baseline values in 58.8% of patients; the median value was reduced, although it was not statistically significant.

We also found a significant reduction of FA from baseline to week 20 demonstrating the effect of the bispecific molecule on neovessels and MNV of different types. Several clinical reports described a reduction of flow area after anti-VEGF treatment of nAMD [16, 18, 19].

Similarly to other real-world studies of faricimab treatment in nAMD and other anti-VEGF drugs for nAMD, we observed a significant decrease of SFCT [8, 12, 20‒22]. It is known that VEGF is secreted from the RPE with VEGF receptors being expressed on the choroidal endothelium facing the RPE layer. The VEGF secreted at this level is essential for retinal function and maintenance of the CC [23]. In addition, Ang-1-Tie 2 receptors are expressed in the choroid, where they maintain the stability of choroidal vessels. It is possible to hypothesize that faricimab targeting these receptors could modify choroidal vessels and this could contribute to retinal fluid control [24].

In conclusion, faricimab is efficient in improving anatomical parameters such as macular thickness related to resolution or reduction of IRF and subretinal fluid, MNV flow area, choroidal thickness, and PED height. The anatomical improvement is paralleled by functional improvement. The drying effect of the bispecific antibody leads to a Q12/Q16 treatment interval in the majority of patients.

A longer follow-up is needed to better assess long-term effects as well as eventual complications of this new anti-VEGF agent. Another limitation of this study is the relatively small sample, which did not allow to stratify the subjects into different subgroups of MNV according to MNV types of nAMD and consequently describe eventual differences of anatomical and functional results in different MNV types.

The study observed the tenets of the Declaration of Helsinki and was approved by an Institutional Review Board of the University “G. D’Annunzio” of Chieti-Pescara (2/2023). Patients provided written informed consent for participation in the study.

The authors of this paper have no conflict of interest.

The authors of this paper have no study funding or sponsorship to declare.

L.T. was responsible for designing the study. L.T., R.M., and M.R.L. were responsible for writing the paper and reviewing it critically. L.T., F.F., A.R., C.D.N., A.Q., and L.B.B. were responsible for acquiring OCT and OCTA images. L.T. and F.F. were responsible for analyzing all quantitative measurements and all qualitative parameters. R.M. was responsible for solving the discrepancies between graders by open adjudication. A.P. and M.D.N. were responsible for performing statistical analysis. Final manuscript was approved by all coauthors.

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.