FormalPara Key Summary Points

Faricimab is a dual inhibitor of vascular endothelial growth factor-A and angiopoietin 2, with the potential to fill a gap in current neovascular age-related macular degeneration (nAMD) and diabetic macula oedema (DMO) treatment options.

Angiopoietin 2 plays a key role in nAMD and DMO pathophysiology by inducing angiogenesis; vascular leakage; hypotensive, abnormal vessel structure; inflammation; and fibrosis of the retina.

Faricimab has demonstrated efficacy, durability, and acceptable safety in phase 3 trials in mostly treatment-naïve DMO or nAMD populations, as well as efficacy in retrospective case–control studies in treatment-resistant nAMD or DMO populations, and efficacy in a retrospective, observational study in a mostly previously treated nAMD population.

Although results so far seem promising, further research is required on faricimab and its ideal target demographics.

Introduction

Neovascular age-related macular degeneration (nAMD) is a leading cause of blindness and vision impairment worldwide [1,2,3]. Age is the main risk factor for nAMD, and its prevalence is expected to rise with the increasing ageing population [2, 4]. nAMD is characterised by choroidal neovascularisations (CNV) which, in combination with surrounding inflammation, may lead to retinal haemorrhage, oedema and fibrosis, which, in turn, impairs vision [2, 5,6,7].

Diabetic macular oedema (DMO) is another leading cause of blindness and vision impairment whose prevalence is expected to rise [8, 9]. DMO is characterised by the accumulation of fluid in the macula. In the healthy retina, the blood-retinal barrier (BRB), which consists of capillary endothelial cells, pericytes, the Müller glia and the retinal pigment epithelium [10], maintains a balance of fluid entry and fluid exit of the retina, so that the retina is kept dehydrated. In DMO, destabilisation of the BRB, in particular destabilisation of the endothelial cell-to-cell adhesions, increases fluid entry while decreasing fluid exit, causing macular oedema which, in turn, impairs vision [8, 10].

In both nAMD and DMO, angiogenic and inflammatory growth factors, such as vascular endothelial growth factor (VEGF) and angiopoietin-2 (ang-2), are upregulated. VEGF regulates angiogenesis, and its upregulation may cause CNV. Ang-2 regulates vascular stability and inflammation, and its effects seem to be many, although not fully understood. Upregulation of ang-2 promotes apoptosis and detachment of pericytes, which sensitises the retinal vasculature to VEGF and other pro-angiogenic and pro-inflammatory factors. Ang-2 also promotes changes in the actin cytoskeleton of endothelial cells, which destabilises endothelial cell-to-cell adhesions and cell-to-extracellular matrix adhesions, thus causing vascular instability. Ang-2 also promotes the adhesion and transmigration of leucocytes into inflamed tissues, thus promoting inflammation. In short, pathological upregulation of ang-2 induces angiogenesis, vascular leakage, hypotensive, abnormal vessel structure, inflammation and fibrosis of the retina [5, 6].

No cure exists for nAMD or DMO, but intravitreal injections of anti-VEGF agents have revolutionised the visual prognosis [2, 8]. Anti-VEGF therapy has demonstrated efficacy and safety in clinical trials. However, optimal outcomes have proven difficult to achieve and maintain in clinical practice [11,12,13,14,15,16] with real-world best-corrected visual acuity (BCVA) gains lower than those in clinical trials [11, 12, 15]. The treatment burden on the patient, caregivers and the healthcare system—due to the need for frequent injections and close monitoring—is considered to be a key underlying reason for these lower gains in the clinical setting. Many patients are undertreated, leading to worse visual prognosis [7, 17, 18]. Furthermore, intravitreal anti-VEGF monotherapy may fail to address the multifaceted pathophysiology of nAMD, which apart from VEGF, also includes pathological upregulation of other modulators, such as ang-2. Thus, there is an unmet need for treatments that: (1) achieve sustainable efficacy with less frequent injections, thus reducing treatment burden, and (2) address the multifaceted pathogenesis of nAMD and DMO.

This need may be met by faricimab, a VEGF-A and ang-2 dual inhibitor, which has been approved by both the US Food and Drug Administration and the European Medicine Agency [19, 20]. Faricimab is a humanised, bispecific, IgG monoclonal antibody, administered by intravitreal injection. Dual inhibition of VEGF and ang-2 has been hypothesised to provide superior durability, and perhaps even superior efficacy, compared to anti-VEGF monotherapy, as it addresses not only one but two key pathways in nAMD and DMO pathogenesis.

This review aims to assess the efficacy, durability and safety of faricimab in nAMD and DMO. It summarises the findings of current studies on faricimab and discusses whether faricimab may fill a gap in current treatment options.

This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.

Studies on Faricimab: Methods and Results

Faricimab in nAMD

The phase 2 trial AVENUE [21] was a randomised, active comparator–controlled, double-masked clinical trial that was conducted at 58 sites in the USA. The study included 263 eyes in 263 patients with treatment-naïve nAMD. Patients were randomised 3:2:2:2:3 to receive ranibizumab 0.5 mg every 4 weeks (group A); faricimab 1.5 mg every 4 weeks (group B); faricimab 6.0 mg every 4 weeks (group C); faricimab 6.0 mg every 4 weeks until week 12, then faricimab 6.0 mg every 8 weeks (group D); and ranibizumab 0.5 mg every 4 weeks until week 8, then faricimab 6.0 mg every 4 weeks (group E). The primary outcome was change in BCVA from baseline. At week 36, adjusted mean change in BCVA when compared to group A was + 1.6 for group B (p = 0.52), − 1.6 for group C (p = 0.53), − 1.5 for group D (p = 0.53) and –1.7 for group E from week 12 onwards (p = 0.30). Thus, faricimab bordered on inferiority, although this inferiority was not statistically significant.

The phase 2 trial STAIRWAY [22] was a randomised, active comparator–controlled, double-masked clinical trial that was conducted at 25 sites in the USA between January and March 2017. The study included 76 eyes in 76 patients with treatment-naive nAMD. Patients were randomised 1:2:2 to receive intravitreal ranibizumab 0.5 mg every 4 weeks, faricimab 6.0 mg every 12 weeks or faricimab 6.0 mg every 16 weeks. Patients receiving faricimab every 16 weeks were assessed for disease activity at week 24. Patients with active disease were changed to faricimab every 12 weeks, while patients without active disease continued on 16-week dosing interval. The primary outcome was change in BCVA from baseline. At week 52, BCVA gains were comparable between groups. At week 52, patients had received a mean (SD) of 12.9 (0.25), 6.7 (0.91) and 6.2 (0.93) total injections in the ranibizumab group, faricimab every 12 weeks group, and faricimab every 16 weeks group, respectively.

The identically designed phase 3 trials TENAYA and LUCERNE [7] were randomised, active comparator-controlled, double-masked clinical trials, conducted at 271 different clinical locations in 24 different countries from February 2019 to October 2020. Patients aged ≥ 50 years were eligible for inclusion. Ocular inclusion criteria included CNV secondary to nAMD. Patients had to be treatment-naïve; BCVA had to be 24–78 Early Treatment Diabetic Retinopathy Study (ETDRS) letters; and the CNV lesion had to be: (1) active, (2) ≤ 9 disc areas in size, (3) with exudation, (4) with a CNV component area sized ≥ 50% of the total lesion size and (5) be either subfoveal or have a subfoveal component related to the CNV activity.

TENAYA included 671 eyes in 671 patients, while LUCERNE included 658 eyes in 658 patients. Patient baseline characteristics (Table 1) were deemed comparable between groups. Patients were randomised 1:1 to aflibercept 2.0 mg every 4 weeks up to week 8 (3 injections in total) followed by injections every 8 weeks, or to faricimab 6.0 mg every 4 weeks up to week 12 (4 injections). At week 20, faricimab-treated patients were assessed for disease activity, and those patients with active disease received one injection, followed by injections every 8 weeks until week 60 (6 injections). At week 24, patients (excluding those already on 8-week doing intervals) were re-assessed for disease activity. Patients with active disease received one injection, followed by injections every 12 weeks until week 60 (4 injections). Patients without active disease received injections every 16 weeks until week 60 (3 injections).

Table 1 Overview of the patient characteristics in the TENAYA/LUCERNE phase 3 trials
Full size table

The primary outcome was change in BCVA from baseline. At week 48, BCVA was comparable between the groups. The faricimab groups gained + 5.8 (TENAYA) and + 6.6 (LUCERNE) ETDRS letters, while the aflibercept groups gained + 5.1 (TENAYA) and + 6.6 (LUCERNE) ETDRS letters. Secondary outcomes included proportion of patients on 16-week, 12-week and 8-week dosing intervals. At week 48, 80% of faricimab-treated patients were on ≥ 12-week dosing intervals, and 44.9% (TENAYA) and 45.7% (LUCERNE) of patients were on 16-week dosing intervals. Total adverse events, as well as serious ocular adverse events, were comparable between the faricimab and aflibercept arms.

Rush and Rush [23] conducted a retrospective, case-controlled study of 55 patients with treatment-resistant nAMD (defined as macular oedema despite ≥ 6 anti-VEGF treatments over 1 year) treated with either faricimab or aflibercept from February to May 2022 at a private practice in Texas, USA. Inclusion criteria were: (1) the patient had received intravitreal aflibercept (IVA) injections for DMO) prior to the study period; (2) optical coherence tomography (OCT)-guided treat-and-extend management protocol was being used; (3) the patient had received ≥ 6 IVA injections the past 12 months and ≥ 4 IVA injections the past 6 months; and (4) central macular thickness (CMT) ≥ 300 um with intraretinal and/or subretinal fluid at baseline. Exclusion criteria were: (1) Snellen BCVA of 20/200 or worse; (2) ocular treatment other than anti-VEGF in the past 6 months or during the study period; and (3) a decrease in BCVA of ≥ 2 lines due to a condition other than DMO. Patients were allocated to the case cohort if they had changed from IVA to intravitreal faricimab (IVF) between February and March 2022 and received three IVF injections within 4 months. Patients were allocated to the control cohort if they stayed on IVA between February and May 2022 and received three IVA injections within 4 months.

In total, 55 eyes in 55 patients were included, of which 28 patients were cases, and 27 patients were controls. Patient baseline characteristics (Table 2) were deemed comparable. An OCT-guided treat-and-extend management protocol was used: patients underwent monthly injections until the retinal oedema had resolved, after which patients’ dosing intervals were extended in 1- to 2-week increments until recurrence of oedema. Dosing intervals were then adjusted accordingly. At the end of the 4-month study period, BCVA gains of + ≥ 2 lines were achieved by 35.7% of faricimab-treated patients and 7.4% of aflibercept-treated patients (p = 0.008). Durability was not assessed.

Table 2 Overview of the patient characteristics in Rush and Rush’s study on neovascular age-related macular degeneration
Full size table

Stanga et al. [24] conducted a retrospective, observational study of 11 eyes in nine patients with nAMD treated with faricimab between May and November 2022 at an independent clinic in London, UK. Three eyes (27.27%) were treatment-naïve, while eight eyes had previously received anti-VEGF injections with persistent intraretinal/subretinal fluid and/or subretinal haemorrhage. Patients received ≥ 1 injection with faricimab with varying treatment intervals. BCVA was assessed at baseline and 1 month after the first faricimab injection. Mean (± SD) BCVA was 0.55 ± 0.73 logMAR at baseline and 0.34 ± 0.32 logMAR at the 1-month follow-up. At the 1-month follow-up, five eyes (45.45%) gained ≥ 1 line of vision, six eyes (54.54%) remained stable and one eye (9.09%) lost 1 line of vision.

Faricimab in DMO

The phase 2 trial BOULEVARD [25] was a randomised, active comparator-controlled, double-masked clinical trial that was conducted at 59 sites in the USA. The trial included 229 eyes in 229 patients with DMO, of whom 168 were treatment-naïve and 61 had been previously treated with anti-VEGF. Treatment-naïve patients were randomised 1:1:1 to ranibizumab 0.3 mg, faricimab 1.5 mg (group B) or faricimab 6.0 mg; previously treated patients were randomised 1:1 to ranibizumab 0.3 mg or faricimab 6.0 mg. Patients received injections every 4 weeks until week 20, followed by an observation period up to week 36 to assess durability. The primary outcome was change in BCVA from baseline at week 24. In treatment-naïve patients, BCVA gains were + 10.3 for ranibizumab 0.3 mg, + 11.7 for faricimab 1.5 mg and + 13.9 for faricimab 6.0 mg. Thus, faricimab 6.0 mg demonstrated superiority when compared to ranibizumab 0.3 mg (p = 0.03). In previously treated patients, BCVA gains were + 8.3 for ranibizumab 0.3 mg and + 9.6 for faricimab 6.0 mg. During the observation period, faricimab 6.0 mg demonstrated superior durability compared to ranibizumab 0.3 mg.

The identically designed phase 3 trials YOSEMITE and RHINE [26] were randomised, active comparator-controlled, double-masked clinical trials conducted at 353 different clinical locations in 31 different countries from September 2018 to October 2020. Patients aged ≥ 18 years were eligible for inclusion. Ocular inclusion criteria included centre-involving macular oedema secondary to diabetes mellitus (type 1 or 2). Most participants were anti-VEGF treatment-naïve, while 24% (YOSEMITE) and 20% (RHINE) were previously treated (with the latest treatment administered ≥ 3 months prior to day 1 of the study). Other ocular inclusion criteria were BCVA of 25–73 ETDRS letters and central subfield thickness (CST) ≥ 325 μm.

YOSEMITE included 940 eyes in 940 patients, while RHINE included 951 eyes in 951 patients. Patient baseline characteristics (Table 3) were deemed comparable between groups. Patients were randomised 1:1:1 to aflibercept every 8 weeks, faricimab every 8 weeks or faricimab with personalised treatment intervals (PTI). The aflibercept group received injections with aflibercept 2.0 mg every 4 weeks up to week 16 (5 injections in total), then every 8 weeks up to week 96. The faricimab every 8 weeks group received injections with faricimab 6.0 mg every 4 weeks up to week 20 (6 injections), then every 8 weeks up to week 96. The faricimab PTI group received injections with faricimab 6.0 mg every 4 weeks until they reached CST < 325 at/after week 12. Once achieved, dosing intervals were extended to 8 weeks for one injection. From thereon, dosing intervals were adjusted based on CST and BCVA.

Table 3 Overview of the patient characteristics in the YOSEMITE/RHINE randomised, active comparator-controlled, double-masked clinical trials
Full size table

The primary efficacy outcome was change in BCVA from baseline. At year 1, BCVA was comparable between the groups. Faricimab groups gained + 10.7–11.6 (YOSEMITE) and + 10.8–11.8 (RHINE) ETDRS letters, while aflibercept groups gained + 10.9 (YOSEMITE) and + 10.3 (RHINE) ETDRS letters.

Secondary outcomes included proportion of patients on 16-week, 12-week, and 8-week dosing intervals in the PTI group. At year 1, > 70% of patients in the PTI group were on ≥ 12-week dosing intervals, and 53% (YOSEMITE) and 51% (RHINE) received 16-week dosing intervals. Total adverse events were comparable between faricimab and aflibercept. The rate of serious ocular adverse events was higher in faricimab-treated patients than in aflibercept-treated patients (1.9%/3.1% in faricimab-treated patients vs. 0.6%/1.9% in aflibercept-treated patients [YOSEMITE/RHINE]). However, the authors assessed that serious ocular adverse events were comparable between faricimab and aflibercept. Serious ocular adverse events that were observed in faricimab-treated patients included endophthalmitis, chorioretinitis, keratouveitis, viral keratouveitis, uveitis, uveitic glaucoma, glaucoma, increased intraocular pressure, ocular hypertension, narrow anterior chamber angle, retinal tear, rhegmatogenous retinal detachment, vitreous haemorrhage, retinal vein occlusion, cataract, diabetic retinal oedema, diabetic retinopathy, retinal neovascularisation, visual acuity reduced transiently and visual impairment.

Rush and Rush [27] conducted a retrospective, case-controlled study of 51 patients with treatment-resistant DMO (defined as macular oedema despite ≥ 6 anti-VEGF treatments over 1 year) treated with either faricimab or aflibercept from February to May 2022 at a private practice in Texas, USA. Inclusion criteria for both case and control cohorts were: (1) the patient had received IVA injections for DMO prior to the study period; (2) an OCT-guided treat-and-extend management protocol was being used; (3) the patient had received ≥ 6 IVA injections the past 12 months and ≥ 4 IVA injections the past 6 months; and (4) CMT ≥ 300 um with DME prior to the study period. Exclusion criteria werek: (1) Snellen BCVA of 20/200 or worse; (2) ocular treatment other than anti-VEGF the past 6 months; and (3) a decrease in BCVA of ≥ 2 lines due to a condition other than DME.

Patients were allocated to the case cohort if they were changed from IVA to IVF between February and March 2022 and received three IVF injections within 4 months. Patients were allocated to the control cohort if they stayed on IVA between February and May 2022 and received three IVA treatments within 4 months. In total, 51 eyes in 51 patients were included, of which 24 patients were cases, and 27 patients were controls. Patient baseline characteristics (Table 4) were deemed comparable. An OCT-guided treat-and-extend management protocol was used: patients underwent monthly injections until the retinal oedema had resolved, after which s’ dosing intervals were extended in 1–2 week increments until recurrence of oedema. Dosing intervals were then adjusted accordingly.

Table 4 Overview of the patient characteristics in Rush and Rush’s diabetic macula oedema study
Full size table

At the end of the 4-month study period, BCVA gains of +  ≥ 2 lines were achieved by 41.7% of faricimab-treated patients and 11.1% of aflibercept-treated patients (p = 0.01). Durability was not assessed.

Discussion

In summary, faricimab demonstrated non-inferior to superior efficacy compared to anti-VEGF monotherapy, strong durability and acceptable safety in the mostly treatment-naïve patient populations with nAMD and in the mostly treatment-naïve patient populations with DMO.

Dual inhibition of VEGF and ang-2 has been hypothesised as the better therapy to address the multifaceted pathogenesis of nAMD and DMO, with the potential to achieve superior efficacy. In all of the studies included in this review, the primary efficacy outcome was change in BCVA. In the treatment-naïve nAMD populations of the STAIRWAY [22], AVENUE [21], TENAYA and LUCERNE [7] trials, faricimab demonstrated non-inferior BCVA gains. In the previously treated nAMD population of Rush and Rush’s study [23], faricimab demonstrated superior BCVA gains. In the mostly previously treated nAMD population of Stanga et al.’s study [24], faricimab demonstrated BCVA gains in some patients. In the treatment-naïve DMO population of the BOULEVARD [25], YOSEMITE and RHINE [26] trials, faricimab demonstrated non-inferior to superior BCVA gains. In the previously treated DMO population of the BOULEVARD trial and the Rush and Rush case–control study [27], faricimab demonstrated superior BCVA gains.

Previous attempts have been made to inhibit VEGF and ang-2 through co-formulated injections with nevascumab (an ang-2 inhibitor) and aflibercept. However, phase 2 trials showed no benefit compared to anti-VEGF monotherapy, and development was abandoned [28, 29]. The success of faricimab may suggest that part of this new drug’s advantage is due to the molecular linkage between its two binding domains, which would make faricimab an ‘obligate bispecific antibody’ [29, 30].

As previously stated, the efficacy of aflibercept in clinical trials has not fully translated to clinical practice. Faricimab has only recently been approved for clinical use and, consequently,, it is unknown if the efficacy of faricimab in clinical trials will translate to clinical practice. However, as treatment burden is considered a key reason for the relatively reduced efficacy of aflibercept in clinical practice, and faricimab has demonstrated strong durability, it is possible that faricimab will demonstrate superior efficacy in clinical practice compared to aflibercept.

In the real-world setting of Rush and Rush’s studies on treatment-resistant nAMD and DMO, faricimab demonstrated superior efficacy to aflibercept. In the real-world setting of Stanga et al.’s study, faricimab demonstrated efficacy in 45.45% of mostly previously treated patients. These results seem promising for the real-world efficacy of faricimab. However, these studies had small populations and short study periods, and in the case of Stanga et al.’s study, the study lacked a control group. For these reasons, results should be interpreted with caution.

In the ongoing TRUCKEE study, patients with nAMD treated with faricimab will be evaluated in the real-world setting [31, 32]. So far, patients have been evaluated via a retrospective chart review after one and three injections with faricimab. After one injection, previously treated (n = 337) and treatment-naïve (n = 39) eyes demonstrated a + 0.7 letter (p = 0.196) and a + 4.9 letter (p = 0.076) improvement in BCVA, respectively. After three injections, previously treated (n = 81) and treatment-naïve (n = 13) eyes demonstrated a + 2.7 letter (p = 0.045) and a + 8.1 letter (p = 0.437) improvement in BCVA, respectively. Thus, faricimab has so far demonstrated efficacy in these patients, particularly in those previously treated, who continued to show improvements after three injections, which, as the authors note, may indicate a benefit of loading doses or multiple injections in these patients. However, these are still early outcomes. The TRUCKEE study is ongoing as a prospective chart review study and future data will provide further insight into the long-term efficacy of faricimab. The lack of a control group is also a limitation to this study [31, 32].

Further research in real-world populations is needed. It would also be interesting to investigate the efficacy of faricimab in treatment-naïve versus treatment-resistant populations, as the above findings may suggest that faricimab is particularly efficient in treatment-resistant populations. Unfortunately, the short study periods of the included studies limit our understanding of BCVA gains over the long term. Further research on long-term BCVA gains is needed.

Dual inhibition of VEGF and ang-2 has also been hypothesised to provide superior durability compared to anti-VEGF monotherapy, which would reduce treatment burden. In both nAMD and DMO patient populations, faricimab demonstrated strong durability and was the first injectable ophthalmic drug to achieve 16-week dosing intervals in a phase 3 study of nAMD and DMO [33]. However, an important limitation to most of these trials is the lack of PTI for the anti-VEGF monotherapy groups, which means that the trials lack a direct comparison of the durability of faricimab compared to the durability of anti-VEGF monotherapy. Thus, technically speaking, it has yet to be determined whether faricimab would demonstrate superior durability. Although clinical trials of aflibercept and ranibizumab favour fixed dosing intervals, most clinical practitioners favour a PTI plan with variable dosing intervals based on the disease activity [17]. Thus, trials comparing the real-world use of aflibercept to faricimab are needed. Although Rush and Rush’s studies took place in a real-world setting with PTI for both faricimab and aflibercept, there is no available information regarding the durability of these drugs.

Hypothetically speaking, the more signalling pathways a drug interferes with, the higher the risk of unwanted adverse effects. For this reason, dual inhibition of VEGF and ang-2 has been hypothesised to increase the risk of adverse effects compared to VEGF monotherapy, which in turn would reduce safety. In the nAMD and DMO populations of AVENUE, STAIRWAY, BOULEVARD, TENAYA and LUCERNE, faricimab demonstrated non-inferior safety. However, in YOSEMITE and RHINE, the rate of serious ocular adverse events was higher in faricimab-treated patients (1.9–3.1% in faricimab-treated patients vs. 0.6–1.9% in aflibercept-treated patients).

The short study periods limit our understanding of long-term adverse effects. Fortunately, the TENAYA/LUCERNE and YOSEMITE/RHINE trials are ongoing, and year 2 data will provide a better understanding of the safety profile of faricimab. Another limitation of the TENAYA/LUCERNE and YOSEMITE/RHINE trials is that the populations only included hundreds of patients, which limits our understanding of rare adverse effects.

In the HAWK and HARRIER phase 3 trials [34], brolucizumab (a VEGF-A inhibitor, which, like faricimab, has been studied for increased durability) demonstrated safety comparable to aflibercept. However, after receiving approval, severe adverse effects, such as retinal vasculitis and retinal artery occlusion, were reported [34]. Thus, although results on the safety of faricimab seem promising so far, it is important to exercise caution with regards to this new drug, and further research is needed to assess the rate of serious ocular adverse effects.

As faricimab has demonstrated efficacy, durability, and acceptable safety in the TENAYA/LUCERNE and YOSEMITE/RHINE populations, populations with similar clinical characteristics may be a suitable target demographic.

As previously stated, treatment-resistant DMO or nAMD may be particularly well-suited for faricimab. Other populations which may, in theory, be particularly well-suited for faricimab include populations with higher-than-average activity of ang-2 (e.g. patients with comorbid diabetes mellitus and nAMD), or populations with risk factors associated with nonadherence to anti-VEGF monotherapy (e.g. patients with other illnesses, fear of/discomfort after injections, or worse BCVA at baseline) [18].

Analysis of the Japanese subgroup of the TENAYA population [35] has shown that a higher proportion of patients in this subgroup had polypoidal choroidal vasculopathy (PCV) compared to patients in the pooled global TENAYA/LUCERNE population. PCV is a subtype of nAMD, which is more prevalent in Asian populations, with recent studies finding a PCV prevalence of 22–62% in Asian nAMD patients [36, 37]. Different subtypes of nAMD have been shown to respond differently to treatment, and findings from one study have suggested that a relation may exist between ang-2 and PCV. Interestingly, a higher proportion of patients in the Japanese subgroup (66%) were on 16-week treatment intervals compared to patients in the pooled global TENAYA/LUCERNE population (45%). The cause of this difference is unclear. However, it is possible that PCV patients may be particularly well-suited for treatment with faricimab. Further research is needed to evaluate the relation between nAMD subtypes and faricimab treatment response. Hopefully, the ongoing SALWEEN study, which will evaluate the efficacy, durability, and safety of faricimab at PTI in PCV patients, will provide further insight into the relationship between PCV and faricimab treatment response [35, 38].

In both the TENAYA/LUCERNE and YOSEMITE/RHINE trials, it would be interesting to see results stratified by clinical characteristics and disease subtypes to assess whether certain clinical characteristics or disease subtypes were associated with changes in efficacy, durability or safety. With ranibizumab, a model-based meta-marker was created to predict ranibizumab response in phase 3 trials based on clinical characteristics (including BCVA, age, CNV size), achieving a model accuracy of 0.86 [39, 40]. Similar efforts may be made in the TENAYA/LUCERNE and YOSEMITE/RHINE populations.

In the future, personalised medicine models may be used to identify populations suited for faricimab therapy. Biomarkers, such as cytokines, ang-2 or leukocyte telomere length, may be investigated as predictive factors of faricimab response, as they have been with other anti-VEGF agents [39].

In nAMD patients, as per phase 3 studies TENAYA/LUCERNE and Medscape guidelines, we propose a treatment regimen of 6 mg (0.05 ml) faricimab by intravitreal injection every 4 weeks for the first 4 doses, followed by OCT and BCVA assessments at week 20 and 24 to evaluate disease activity, which will inform whether patients should receive injections every 8 weeks, 12 weeks or 16 weeks until week 44 or 48.

In DMO patients, as per phase 3 studies YOSEMITE/RHINE and Medscape guidelines, we propose a treatment regimen of 6 mg (0.05 ml) faricimab by intravitreal injection every 4 weeks for the first four doses. If oedema resolves on OCT, dosing may be modified by extending intervals in up to 4-week increments or reducing intervals in up to 8-week increments based on OCT and BCVA assessments through week 52.

The strengths of this review include its systematic approach, whereas limitations include the relatively low number of clinical studies conducted on faricimab, which limits our ability to assess the efficacy, durability and safety of this drug.

Conclusions

In conclusion, faricimab is a newly approved dual inhibitor of VEGF-A and ang-2 with the potential to fill a gap in current treatment options due to its strong durability. So far, faricimab has demonstrated efficacy, durability and acceptable safety in phase 2 and 3 trials in mostly treatment-naïve nAMD or DMO populations, as well as efficacy in retrospective case–control studies in treatment-resistant nAMD or DMO populations, and efficacy in a retrospective, observational study in a mostly previously treated nAMD population. However, there is still much to learn about this new drug. Further research is needed to assess the efficacy of faricimab in real-world patients (treatment-naïve, treatment-resistant and treatment-responsive), directly compare the durability of faricimab to aflibercept, investigate long-term and rare adverse effects of faricimab and identify the ideal target demographic of faricimab. However, so far, there is reason to be cautiously optimistic about the potential of faricimab in nAMD or DMO.