This study investigated the efficacy of intravitreal Faricimab treatment in controlling inward and outward exudation in T1 MNV, particularly focusing on PED volumes. The dual inhibition of VEGF-A and Ang-2 of Faricimab offers advantages over existing anti-VEGF mono-therapies, especially in cases where resistance or inadequate response to conventional treatments is encountered.
The primary objective was to assess the 12-month efficacy of intravitreal Faricimab on PED volumes, in both treatment-Naïve (Fig. 3) and previously treated non-responder patients (Fig. 4). Our findings align with previous post hoc analysis of pooled data from the TENAYA and LUCERNE trials, which demonstrated that Faricimab was associated with a more rapid and greater reduction in fluids and PEDs compared to Aflibercept during the head-to-head dosing phase of the studies [13]. Overall, this study shows that treatment with Faricimab positively affected retinal fluid control in both Naïve and Switch patients presenting with T1 MNV lesions, and resulted in a significant reduction of AI-assisted quantification of PED Volumes. These findings support the inclusion of Faricimab in the armamentarium of treatments for T1 MNV, particularly in cases complicated by PED.
We observed a rapid decrease in PED volume at the 3-month follow-up in both groups, particularly in Naïve eyes (p = 0.049). This may be attributed to the pronounced exudation typically associated with a recently developed lesion, compared to a chronic, previously treated lesion. However, even in the Switch eyes, there was a reduction in PED volume, sustained throughout the follow-up period (Fig. 2). This reduction not only suggests a decrease in fluid accumulation in the sub-RPE compartment, which is isolated and appears to exhibit a limited therapeutic response to anti-VEGF mono-therapy [20], but also indicates a potential anti-fibrotic effect on regression of the fibrovascular component contributing to the PED formation, due to the dual-target inhibition with Faricimab. These findings underscore the multifaceted impact of Faricimab treatment on the structural and functional aspects of MNV pathology, highlighting the need for comprehensive assessment of both fluid and fibrovascular components in evaluating treatment response.
The first real-world study on Faricimab by Stanga et al. [21](11 eyes of 9 patients) demonstrated that SRF, IRF, and PED at baseline were present in 8/11, 3/11 and 8/11 eyes, respectively. After the first Faricimab injection, SRF and IRF were completely resolved in 75% (6/8) and 33% (1/3) of eyes, respectively, and morphological changes such as PED flattening were observed in all affected eyes (8/8). Schneider et al. [22] reported a series of 50 eyes from 46 nAMD patients previously treated for at least 3 times with Aflibercept q4W with persistent fluids that were switched to Faricimab. At 1 month, 84% of eyes showed fluid decrease (responder group), with 33% demonstrating complete fluid resolution and both central retinal thickness (median: −31 μm, p-value < 0.01) and PED height (median: −21 μm, p-value < 0.01) reduction. A recent paper by Szigiato et al. [15], reported that switching to Faricimab resulted in a reduction in mean central subfield thickness (-11.6 μm, p-value = 0.01) and PED height (-44.2 μm, p-value = 0.01) after 3 injections, with stable VA and at a similar treatment interval to prior anti-VEGF therapy. Veritti et al. [23] recently published a study demonstrating that Faricimab has an early and substantial efficacy in reducing PED volumes and fluids in treatment-naïve eyes affected by both type 1 and type 3 MNV. The study included 22 eyes (16 with Type 1 MNV and 6 with Type 3 MNV) with a relatively short follow-up period of 120 days and demonstrated a significant and rapid reduction in both mean subretinal fluid (SRF) and intraretinal fluid (IRF). On average, SRF decreased by 14.4% on Day 1, 59.7% on Day 7, and 91.2% on Day 14, while IRF showed a reduction of 23.5% on Day 1, 71.9% on Day 7, and 90.7% on Day 14. These findings suggest that Faricimab may offer substantial and rapid morphological improvements, with enhanced visual acuity outcomes and a favorable safety profile in patients with PEDs associated with nAMD. We agree with the observed trend of early reduction in fluid and PED volumes following Faricimab injections, which, in our experience, is sustained over a period of 12 months.
Our study is not directly comparable to these recent Faricimab studies due to differences in patient inclusion criteria. Unlike these studies, which likely encompassed a broader range of nAMD neovascular lesions, our cohort specifically focused on patients with T1 MNV with PED. This targeted approach allowed us to investigate the efficacy of Faricimab in managing both inward and outward retinal exudation components over 12 months follow-up period, as evidenced by the decrease in IRF and SRF volumes. This reduction was consistent across both groups, indicating the efficacy of Faricimab in controlling disease activity in both Naïve and previously treated Switch eyes.
In our study, BCVA improved at the 12-month follow-up in Naïve eyes (+ 6.9 ETDRS letters from baseline, p = 0.053) and remained stable in Switch eyes, despite the Switch group presented a higher baseline BCVA. This difference may be attributed to prior treatments in the Switch group, which had partially managed the exudative component but failed to achieve complete anatomical resolution. In contrast, the naïve group exhibited lower baseline BCVA due to the more active exudative lesions but demonstrated greater improvement throughout the study period with a smaller number of injections.
Indeed, if we evaluate the number of injections performed in the two groups, we found that it was significantly higher in pretreated switch patients (p = 0.009). As reported in the literature, in chronically active nAMD, there may be changes in retinal structure and inflammation that alter pharmacokinetics and pharmacodynamics. These include higher concentrations of VEGF, upregulation of other growth factors, vascular endothelial cell mutations, chronic inflammation and even neutralizing antibodies against anti-VEGF [24] leading to a more difficult response to treatment. On the other hand, despite the high number of previous treatments and the chronic and mature stage of the disease, the Switch group showed positive outcomes, supporting the efficacy of Faricimab also in these hard to treat patients.
In our cohort (65 eyes of 65 patients), Faricimab was well tolerated with no instance of IOI episodes recorded during the follow-up period. However, we emphasize the importance of proactive screening to monitor for potential acute IOI episodes during follow-up visits after any type of intravitreal injection, including Faricimab, to ensure overall treatment safety.
AI-driven retinal image analysis is transforming ophthalmology, especially in managing AMD [17, 18]. AI models, particularly those using deep learning (DL) and machine learning (ML), are proving highly effective in measuring biomarkers, forecasting disease progression, and supporting treatment decisions [25]. Recent research has increasingly focused on the quantitative analysis of OCT images using AI algorithms. Schlegl et al. developed a deep learning network capable of automatically quantify retinal fluids [26], , Erfurth et al. used a deep learning algorithm to identify and quantify retinal exudations, and explored the relationship between the fluid volume and visual function following intravitreal injections in AMD patients [27]. Moraes et al. further advanced the field by incorporating additional biomarkers like hyperreflective foci, reinforcing AI’s clinical relevance [28].
Aligned with recent literature, our study underscores the pivotal role of AI in assessing anti-VEGF treatment for AMD through precise measurement of IRF, SRF, and PED volumes. By facilitating accurate quantification of volumetric data, AI provides clinicians with valuable insights into disease management that surpass those offered by traditional methods [29].
This study has several limitations that warrant discussion. Its retrospective design may introduce inherent biases related to data collection and analysis. The relatively small patient cohort limits the statistical power and generalizability of the findings. Additionally, the Switch group had a high mean number of prior injections, which may have influenced treatment outcomes. Furthermore, while the study focuses on PED volume reduction in Type 1 MNV, it does not explore the potential correlation between specific anatomic features, such as intraretinal fluid changes, and vision recovery—a topic that could be addressed in future investigations.
Leave a Reply