DOI: https://doi.org/10.21203/rs.3.rs-2715544/v1
We evaluated the response to faricimab in a retrospective series of neovascular age-related macular degeneration (nAMD) eyes with persistent edema or subretinal fluid on other existing anti-VEGF drugs. Inclusion criteria were at least 3 faricimab injections with 6 months follow-up. We also determined the subtypes of nAMD such as polypoidal choroidal vasculopathy (PCV) or non-PCV. Intervals between injections were extended as long as the macula remained dry. Thirteen eyes of thirteen patients were included. Mean age was 77.1 ± 9.3 years. Six were males and seven were females. All eyes had type 1 choroidal neovascularization (CNV) (CNV below the retinal pigment epithelium (RPE), and ten had PCV and three had occult CNV (non-PCV). Faricimab resolved the macula edema in 69.2% of anti-VEGF resistant eyes after three injections. Neither central retinal thickness nor visual acuity significantly changed 6 months after faricimab injections were started. The mean interval of injections before and after faricimab therapy did not change and was 5.3 ± 1.5 and 5.4 ± 0.8 weeks. Central retinal thickness, visual acuity, and the interval of injections were not significantly different between PCV and non-PCV subtypes of exudative AMD.
Age-related macular degeneration (AMD) is the leading cause of blindness [1–3], but therapeutic intravitreal anti-vascular endothelial growth factor (anti-VEGF) agents have resulted in dramatic improvements in the anatomic and visual outcomes in neovascular AMD (nAMD) [3]. Bevacizumab, ranibizumab, aflibercept, and brolucizumab have been in clinical use. More recently, faricimab was approved by U.S. Food and Drug Administration (FDA) for the treatments of nAMD on January 28, 2022.
Faricimab is a bi-specific antibody, which inhibits not only VEGF but also angiopoetin 2 (Ang-2) [4]. Anti-VEGF agents reduces leakage and inhibits choroidal neovascularization (CNV) in nAMD. Ang-2 results in vascular destabilization and inflammation. Anti-Ang-2 binding can restore vascular stability by reducing leakage, neovascularization and inflammation, which potentially allows a more significant clinical result than anti-VEGF alone.
Anti-VEGF resistance is seen when there is persistent subretinal fluid, macular edema or subretinal hemorrhage despite frequent anti-VEGF injections. This is more common in the polypoidal choroidal vasculopathy (PCV) subtype of nAMD [5]. The Tenaya and Lucerne studies have shown the benefit of faricimab in the management of treatment-naïve eyes with nAMD [4]. However, the results of management in treatment experienced eyes with anti-VEGF resistance has not been significantly reported.
Herein, we present a retrospective series of exudative AMD eyes with poor response to existing therapy prior to the FDA approval of faricimab. All eyes had indocyanine green angiography (ICGA), which allowed for evaluation of the presence of PCV.
This is a retrospective review of a consecutive series of nAMD with poor response to previous anti-VEGF agents, but then were treated faricimab for at least 6 months. All cases were evaluated at the Hawaii Macula and Retina Institute, and this study was deemed exempt from institutional review board approval due to its retrospective design (Western Copernicus Group IRB #1-1588221-1) and was performed in accordance with the Health Insurance Portability and Accountability Act and the tenets of the Declaration of Helsinki for research involving human subjects. Informed consent was obtained for publication of non-identifying medical details and photographs. The time interval for the study was from February 2022 to January 2023.
Patients with nAMD being treated with existing anti-VEGF injections including bevacizumab, ranibizumab, aflibercept or brolucizumab were included in the study if there was persistent disease activity (subretinal fluid, intraretinal fluid or subretinal hemorrhage). Following the FDA approval of faricimab in February of 2022 this series includes a consecutive series of patients treated with faricimab for nAMD with poor response to previous agents and often chronic frequent injections. The primary endpoint was the resolution of subretinal fluid, macular edema or subretinal hemorrhage after at least 3 faricimab injections with minimum follow up of six months. Secondary endpoints at 6 months after starting faricimab injections included increase in the interval between injections, improvement in vision, and decrease in central retinal thickness.
We also determined the subtypes of nAMD and classified them as PCV or typical nAMD. Intervals between injections were extended only if the macula was completely dry without edema or subretinal fluid. All patients had a complete medical history, Snellen vision testing, slit lamp examination, fundus biomicroscopy, intraocular pressure measurement and spectral domain optical coherence tomography (SD-OCT) utilizing the Cirrus HD-OCT (Carl Zeiss Meditec, Dublin, CA). All eyes had prior ICGA with scanning laser ophthalmoscope (SLO) (Spectralis HRA + OCT, Heidelberg Engineering, Heidelberg, Germany) prior to entry into the study.
Thirteen eyes of thirteen patients were included in this study. Mean age was 77.1 ± 9.3 years. Six were males and seven were females. There were five Caucasian and eight Asian patients. All eyes had type 1 CNV (CNV below the retinal pigment epithelium (RPE) and above Bruch’s membrane.). Ten eyes had PCV and three eyes had typical nAMD without PCV. Previous treatments were brolucizumab in five eyes and aflibercept in eight eyes. Mean number of injections before faricimab therapy was 49.5 ± 27.4, and mean interval of anti-VEGF injections before faricimab therapy was 5.3 ± 1.5 weeks.
Anatomic results showed that one injection of faricimab resolved the macula edema completely in five of 13 eyes (4 PCV and 1 typical nAMD), and two or three injections resolved the macular edema in four eyes (3 PCV and 1 typical nAMD). However, persistent edema or subretinal fluid was noted in four eyes (3 PCV and 1 typical nAMD), in which faricimab was stopped and another anti-VEGF agent was started. The subtypes of exudative AMD (PCV or typical nAMD) did not also result in a significant difference in response to faricimab after six months of faricimab including CRT, the logMAR VA, and the interval of injections, but the numbers are very small. The mean CRT before faricimab was 262 ± 81 µm and after 6 months of faricimab was 242 ± 83 µm. The CRT did not significantly improve 6 months after injections of faricimab were started (Figure 1A). The mean logMAR VA prior to faricimab was 0.36 ± 0.19 and that after six months of faricimab was 0.45 ± 0.25. The VA also did not significantly improve 6 months after injections of were faricimab started. The mean interval of injections after 6 months was 5.4 ± 0.8 weeks and this was not increased after starting faricimab (Figure 1B).
A case demonstrating a good response to faricimab involved a 75-year-old Caucasian woman who had the PCV subtype of nAMD with a large central polypoidal complex in her left eye on treatment for eight years. Fluorescence angiography (FA) showed superior occult leakage (Figures 2A and 2B), and the recent ICGA showed micropolyps and a thick double layer sign (Figures 2C and 2D). OCT angiography shows the large branch vascular network (BVN) (Figure 2E). She had received 3 bevacizumab, 66 aflibercept and 2 brolucizumab, and 2 photodynamic therapies (PDTs) before faricimab. There was persistent subretinal fluid despite every 4-week aflibercept injections (Figure 2F). The subretinal fluid completely resolved just after one injection of faricimab (Figure 2G). The subretinal fluid remains resolved with extension of the interval to 5 weeks (Figure 2H). The interval was extended to 5.5 weeks once but slight subretinal fluid recurred indicating that every 5 weeks treatment was the maximum interval on faricimab.
A case demonstrating a poor response to faricimab involved a 67-year-old Asian woman who had the PCV subtype of nAMD in her right eye requiring treatment for 9 years (Figure 3). FA showed occult leakage (Figure 3A and 3B) and ICGA showed large polypoidal lesions and an extensive branching vascular network (Figure 3C). B-scan OCT showed an inverted U-shaped polypoidal lesion typical of a polypoidal lesion as shown in Figure 3D. She had received 43 aflibercept injections with persistent subretinal fluid despite treatment every 5 weeks, and persistent subretinal fluid after 15 brolucizumab with a minimum interval of 4 weeks and a maximum interval of 7 weeks. However, ocular inflammation with floaters and mild vitritis without vasculitis resulted in immediate stoppage of brolucizumab. Faricimab injections were started when subretinal fluid recurred (Figure 3E). After four injections of faricimab there was an increase in subretinal hyperreflective material (SHRM) above the RPED and intraretinal fluid (Figure 3F). Faricimab was discontinued and treatment with aflibercept was restarted.
Although the response of treatment naïve nAMD in the TENAYA and LUCERNE studies was shown to have excellent results with extended fixed treatment regimens of every-8-week to every-16-weeks [4], the response of anti-VEGF resistant nAMD eyes to faricimab has not been well reported. In this series there was a significant improvement in the main endpoint of anatomic resolution of subretinal and intraretinal fluid in 69% of eyes, but 31% of eyes did not show a beneficial effect of faricimab anatomically at 6 months over previous anti-VEGF injections. Unlike in treatment naïve eyes there was no evidence of increased interval between anti-VEGF injections in this short-term study. However, a completely dry macula was required in this study before extension of the injection interval, whereas interval extension was allowed in Tenaya and Lucerne if there was no increase ≥ 50 µm in CRT, no decrease ≥ 5 letters in best corrected VA, and no new macular hemorrhage. This study had only small numbers, but no difference was detected between the subgroups of typical nAMD or PCV.
There is one previous paper of case control study that compared effects of 3 injections of faricimab to those of aflibercept in treatment-resistant exudative AMD subjects previously treated with intravitreal aflibercept [6]. They reported no difference in vision but just decrease in central retinal thickness. Different from our study, they did neither require resolution of edema or subretinal fluid, nor evaluate to determine PCV or typical nAMD. Our study did not show difference in central macular thickness.
Aflibercept has been shown to have significantly more effect on the anatomic results of exudative AMD than ranibizumab [7]. Aflibercept is a full-length antibody with a normal Fc component, whereas ranibizumab is an antibody fragment without an Fc component. In vitro studies have shown that agents with a Fc component are actively transported into the subretinal space, where agents without a Fc component move into the subretinal space through passive diffusion. Faricimab has a modified Fc fragment which limits attachment to receptors to allow more prolonged drug levels in the eye, decreased systemic levels, and decreased inflammation. We speculate that the effect of faricimab is affected by the alterations in the Fc component of the drug, which may affect active transport of faricimab by the RPE into the sub-RPE space.
This study has several limitations. This is a retrospective study, but the case series was consecutive. The number of subjects is small, but because of only recent FDA approval the response in anti-VEGF resistant eyes to faricimab is limited, and this case series provides useful preliminary data. The follow-up is limited at 6 months, but this includes a response to at least 3 faricimab injections. Anti-VEGF resistance is more common in PCV, and this is the only study which evaluates the response to faricimab in anti-VEGF resistant eyes with the diagnosis of PCV or typical nAMD confirmed on ICG angiography. Further investigations including longer follow-up and larger numbers are needed.
In the treatment of anti-VEGF resistant eyes with nAMD faricimab was effective at resulting in a dry macula without subretinal fluid or macular edema in 69% of eyes, but 31% of eyes did not meet the primary endpoint of resolution of edema and subretinal fluid at 6 months after starting faricimab injections. Injection interval was not increased after starting faricimab. Faricimab can result in resolution of edema and subretinal fluid in some but not all eyes with poor response to previous anti-VEGF injections.
Acknowledgements: None
Author Contributions: SF and GTK both attest that they have made substantial contributions to the conception/design of the work, the acquisition/analysis of data, drafted the work and have approved the submitted version.
Data Availability Statement: The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.
Additional Information:
Competing Interests Statement:Dr. Kokame has received support for research, unrelated to the submitted work, from Genentech. Dr. Fujimoto has no competing interests.