Exploring the comparative regressive effects of aflibercept and faricimab on pigment epithelial detachment

Abstract

Background

This study aimed to compare the regressive effects of aflibercept and faricimab on pigment epithelial detachment (PED) in patients with neovascular age-related macular degeneration.

Methods

In total, 41 eyes of 40 patients diagnosed with type 1 macular neovascularization were retrospectively analyzed using multimodal imaging. Of these, 23 eyes were treated with intravitreal aflibercept injections (IVA group), and 18 eyes were treated with intravitreal faricimab (IVFa group), with 3 consecutive injections administered as loading dose therapy. Before treatment and at 1, 2, and 3 months after the first treatment, the maximum height (MH) and maximum diameter (MD) of the PED were measured using optical coherence tomography in each treatment group.

Results

In the IVA group, the MH at baseline (215 ± 177 μm) was reduced to 141 ± 150 (P = 0.06), 119 ± 150 (P < 0.01), and 107 ± 150 µm (P < 0.0001) at 1, 2, and 3 months after treatment, respectively. Similarly, in the IVFa group, the MH decreased from 240 ± 195 µm before treatment to 165 ± 170 µm (P = 0.24), 139 ± 142 µm (P < 0.05), and 117 ± 112 µm (P < 0.01) at 1, 2, and 3 months after treatment, respectively. The reduction at 2 and 3 months was significant in both treatments. The mean changes of MH from baseline were -108 ± 142 µm in the IVA group and -124 ± 112 µm in the IVFa group, with no significant difference (P = 0.21). In both groups, the MD did not regress significantly.

Conclusions

The results suggested that the MH of the PED between the IVA and IVFa groups regressed similarly after each loading therapy.

Background

Age-related macular degeneration (AMD) is a significant global cause of blindness, affecting many individuals. Since 2000, the treatment of exudative AMD lesions has primarily involved the use of anti-vascular endothelial growth factor (anti-VEGF) drugs. Various formulations, such as bevacizumab [1], ranibizumab [2], aflibercept [3], brolucizumab [4, 5], and their biosimilars have been employed to stabilize the disease and improve vision. A notable addition to these treatment options is faricimab, a bi-specific antibody that targets VEGF and angiopoietin-2 (Ang-2) [6]. This innovative medication has shown promising results for reducing inflammation, hyperpermeability, and angiogenesis associated with neovascular AMD (nAMD). Following successful phase 3 international clinical trials (TENAYA and LUCERNE) for wet AMD, faricimab became commercially available worldwide in 2022 [7].

Regarding vascular integrity, angiopoietin-1 (Ang-1) plays a potential role in stabilizing healthy blood vessels by promoting pericyte coverage on the endothelium's outer surface. In contrast, Ang-2 acts as an antagonist of Ang-1; both factors competitively bind to a specific subfamily of growth factor tyrosine kinase (Tie2) receptors [6, 8]. Signaling between Ang-2 and Tie2 receptors facilitates inflammatory responses within the active microenvironment and regulates vascular integrity. However, the clinical manifestations of Ang-2 deactivation in patients with nAMD remain unclear.

Pigment epithelial detachment (PED) is closely associated with nAMD, particularly in older individuals [8]. Exudative changes in the retina accompanied by shallow PED indicate macular neovascularization (MNV). This condition can lead to various complications, including subretinal fluid (SRF), intraretinal fluid, subretinal pigment epithelial fluid, and subretinal or subretinal pigment epithelial (sub-RPE) hemorrhage, ultimately resulting in a decline in visual acuity [9]. Moreover, a large PED associated with MNV can increase the risk of RPE tear [10]. This study evaluated the regressive effect of faricimab on PED and compared its efficacy with those of intravitreal aflibercept (IVA) and intravitreal faricimab (IVFa) in a real-world setting.

Methods

This retrospective study was approved by Fukushima Medical University Certified Review Board (1 Hikarigaoka, Fukushima-city, Fukushima, Japan, CRB2200002) and adhered to the Declaration of Helsinki. The requirement for informed consent was waived due to the retrospective nature of the study by the Fukushima Medical University Certified Review Board (1 Hikarigaoka, Fukushima-city, Fukushima, Japan, CRB2200002). All patients with a clinical diagnosis of type 1 MNV and previously untreated nAMD at the Department of Ophthalmology of Fukushima Medical University Hospital between June 2017 and December 2022 were included in this study. All participants underwent fundus ophthalmoscopy, fluorescein angiography, indocyanine green angiography, and optical coherence tomography (OCT) (Heidelberg Engineering, Heidelberg, Germany). The retinal images were obtained by the HRA-OCT, as described in details in our previous report [11]. The methodology is briefly summarized as follows: This study assessed the effects of treatment on PED over 3 months. We measured the maximum height (MH) and maximum diameter (MD) of the PED using the OCT images before and at 1, 2, and 3 months after the first treatment. Additionally, we examined best-corrected visual acuity (BCVA), central macular thickness (CMT), and subfoveal choroidal thickness (SCT) at the start and 3 months post-treatment. MH was the distance between RPE and Bruch’s membrane, while MD was the PED's expansion measured through OCT images. CMT was defined as the distance between the internal limiting membrane and RPE at the fovea; SCT was defined as the distance between Bruch’s membrane and the margin of the choroid and sclera under the fovea. Two blinded examiners took all measurements using computer-based tools to ensure impartial recording. The diagnostic criteria for nAMD were based on a previous study. Three monthly injections of aflibercept (Eylea; 20 mg/0.05 mL; Bayer) or faricimab (Vabysmo; 6 mg/0.05 mL; Roche) were administered as loading dose therapy, depending on when the participants had visited the hospital: IVA was administered in 2017, and IVFa was administered in 2022. Moreover, we assessed retinal hemorrhage and retinal pigment epithelial tears using OCT, fundus, and autofluorescence images.

Statistical analysis

The Mann–Whitney U test was used to compare the mean age between the IVA and IVFa groups. An one-way analysis of variance was used to assess the changes in MH and MD in each group. An unpaired t-test was used to assess the changes in MH and MD between the two groups. Fisher’s exact test was performed to analyze the odds ratio and p-value for the dominance of male or female patients and the prevalence of RPE tears in each group. Data analyses were performed using GraphPad Prism version 9 (GraphPad Software, La Jolla, CA, USA). Statistical significance was set at P < 0.05. All data are presented as mean ± standard deviation.

Results

Table 1 presents the demographic characteristics at baseline, incidence of RPE tear, and other clinical profiles of patients in the IVA and IVFa groups.

Table 1 Demographic characteristics at baseline and prevalence of RPE tear and reduction ratio of SCT after intravitreal aflibercept (IVA) and intravitreal faricimab (IVFa) at 3 months

The two groups showed no significant differences in age or sex distribution. No significant difference was observed in the mean MHs of each group at baseline (P = 0.66). In the IVA groups, the mean MHs of the PED at baseline and 1, 2, and 3 months after the first treatment were 215 ± 177, 141 ± 150 (P = 0.06), 119 ± 150 (P < 0.01), and 107 ± 150 μm (P < 0.0001), respectively. MH showed significant regression 2 and 3 months after the first treatment compared with that at baseline. In contrast, the MHs in the IVFa group at baseline and 1, 2, and 3 months after the first treatment were 241 ± 195, 165 ± 170 (P = 0.24), 139 ± 141 (P < 0.05), and 117 ± 113 μm (P < 0.01), respectively, indicating that the MH significantly decreased at 2 and 3 months after the first treatment (Fig. 1). At 1, 2 and 3 months, the mean decreases of the MH in the IVA group were -74.6, -96.5, and 108.4 μm; those in the IVFa group were -75.8, -102.2, and -12.4 μm, respectively, without statistically significant differences at those periods (P = 0.97, 0.87, and 0.68, respectively) (Fig. 2).

Fig. 1

Changes in the maximum height (MH) of pigment epithelial detachment measured in the intravitreal aflibercept (IVA) and intravitreal faricimab (IVFa) groups before and at 1, 2, and 3 months after the first treatment. *: P < 0.05, **: P < 0.01, ****: P < 0.0001

The MDs before and at 1, 2, and 3 months after the first treatment in the IVA group were 2544 ± 1285, 2148 ± 1133, 1956 ± 1056, and 1844 ± 1093 μm, respectively, and 2405 ± 1343, 2205 ± 1360, 1859 ± 928, and 1765 ± 922 μm in the IVFa group, respectively. In both groups, MD before and after treatment did not change significantly (Fig. 3).

Fig. 2

Mean changes from baseline in the maximum height (MH) of pigment epithelial detachment measured in the intravitreal aflibercept (IVA) and intravitreal faricimab (IVFa) groups before and at 1, 2, and 3 months after the first treatment

The incidence of RPE tears was 0/23 (0%) in the IVA group and 2/18 (11%) in the IVFa group, without significant difference between the groups (P = 0.19).

Changes in CMT, SCT, and BCVA are summarized in Additional File 1.

The representative case of PED regression after IVFa is shown in Fig. 4.

Fig. 3

Changes in the horizontal maximum diameter (H-MD) of pigment epithelial detachment in the intravitreal aflibercept (IVA) and intravitreal faricimab (IVFa) groups before and at 1, 2, and 3 months after the first treatment

Fig. 4

A male in his 80 s in the intravitreal faricimab (IVFa) treatment group at baseline (A-E). A Fundus photographs showing pigment epithelial detachment (PED) at the macula. B The middle phase of fluorescein angiography detecting occult macular neovascularization (MNV) at the macula. C The early phase of indocyanine green angiography identifying MNV at the upper side of PED. D, E Optical coherence tomography (OCT; horizontal (D) and vertical images (E) revealed PED with subretinal fluid at baseline. F, G At 1 month after IVFa treatment, the PED is dramatically regressed. H, I At 2 months, the PED is gradually reduced. J, K At 3 months, the PED almost disappeared at 3 months

The incidence of RPE tear and another clinical profile are shown in Table 1.

Discussion

IVA and IVFa treatments for type 1 MNV achieved comparable timing and extent of PED regression. Additionally, the rate of occurrence of RPE tears and changes in CMT and SCT did not differ between the two groups in this study.

Serous PED in nAMD develops first because of MNV invasion between the RPE and Bruch’s membrane, secondary to high static pressure from the MNV [12]. Thus, the regression of PED in in patients with nAMD after the administration of anti-VEGF could be strongly associated with the reaching of anti-VEGF at the MNV beneath the RPE and the extent of MNV regression. During loading therapy, aflibercept and faricimab cause approximately 85% [13, 14] and 84% [15] reduction in choroidal thickness, respectively. Thus, the PED regression after each treatment seemed consistent with this effect. Fluid absorption after loading therapy can indicate the effectiveness of each drug in MNV. In polypoidal choroidal vasculopathy (PCV), a subtype of type 1 MNV, aflibercept reduces SRF by 85% [13], whereas faricimab achieves dry macula in 82% [15] of PCV cases. These effects also seemed to account for the similar regression of the PED for each drug.

Previously, we compared the extent of PED between brolucizumab and aflibercept during loading therapy and found that brolucizumab regresses PED more rapidly than aflibercept [11]. Collectively, brolucizumab may exhibit a superior effect on PED regression compared with aflibercept and faricimab. Recently, Maruyama-Inoue et al. have compared the functional and morphological changes induced by brolucizumab and faricimab in patients with nAMD. In their study, the decrease in SCT tended to be greater with brolucizumab treatment than with faricimab [16], suggesting that brolucizumab potentially has a stronger effect on PED than faricimab. In the preceding investigation, a comparative analysis of structural and visual outcomes between IVA and ranibizumab was conducted. The findings revealed that IVA exhibited a superior reduction in PED height and yielded improved visual acuity outcomes at the one-year mark [17]. Anti-VEGF has emerged as the primary treatment of nAMD with PED; however, in some instances, photodynamic therapy (PDT) is employed in conjunction with anti-VEGF agents for cases resistant to frequent anti-VEGF treatments or presenting with PCV. A study involving the combination of PDT and IVA demonstrated complete resolution of PED in 4 out of 7 cases. A noteworthy improvement was observed in the remaining 3 cases, accompanied by visual enhancements [18]. In cases of PED resistant to anti-VEGF therapies, the application of PDT in conjunction with anti-VEGF agents may be considered.

In this study, RPE tears developed in 2/17 (11%) eyes treated with IVFa. The MH of PED at baseline for these two cases was 504 µm and 829 µm. In contrast, although 0/23 eyes were treated with IVA, the two treatment groups showed no significant difference in the development of RPE tears (P = 0.19). We previously reported a similar comparison of RPE tears between aflibercept and brolucizumab. In the previous report, we observed RPE tears in 5 of 49 (10%) eyes after IVA treatment. Generally, RPE tears develop in patients with a higher PED (> 400 μm) [19]. Relatively thinner mean MH (215 ± 177 μm) in the IVA group in this study, compared with that (225 ± 169 μm) in the previous study [11], may contribute to the lower incidence of RPE tear in the IVA group in this study.

This study has some limitations, including the relatively small sample size and short duration of follow-up. In addition, the retrospective nature of this study introduces inherent constraints. No significant difference existed in the distribution of serous, fibrovascular, and hemorrhagic PED between the IVA and IVFa groups. However, we hypothesize that the higher frequencies of hemorrhagic PED in the IVFa group in this cohort could be closely related to underestimating the reduced efficacy of the IVFa. Additional investigations involving larger cohorts and an extended follow-up period are imperative to mitigate these limitations and acquire more robust insights.

Conclusions

The results suggested that the MH of the PED between the IVA and IVFa regressed similarly after each loading therapy. Furthermore, IVA and IVFa may contribute to the stability of sub-RPE lesions.