The novel anti-VEGF conbercept is a fusion protein with a relative molecular mass of 143,000 composed of immunoglobulin-like region 2 (FLt-1) of human VEGFR1 and immunoglobulin-like regions 3 and 4 of VEGFR2 and fused to the Fc segment of human IgG1 to give it a high affinity and long half-life; the extracellular domains of the VEGFR2 of conbercept can significantly reduce neovascularization and, at the same time, enhance the binding of conbercept to VEGF (7, 43, 44). In the phase 2 clinical trial of conbercept for the treatment of nARMD (AURORA), visual acuity and anatomical restoration were maintained or further improved after different dosing regimens with different dosages (CONBERCEPT 0.5 versus 2.0 mg monthly or PRN) (59). In the phase III clinical trial (PHOENIX), the conbercept injection group received a monthly intravitreal injection (0.5 mg) for 3 months followed by quarterly injections until month 12, while the control group received a monthly sham injection for 3 months, followed by a quarterly injection of conbercept (0.5 mg) until month 12. At 3 months, the mean increase in best corrected visual acuity (BCVA) was 9.20 letters in the conbercept group versus 2.02 letters in the control group (p < 0.001); at 12 months, the mean increase in BCVA was 9.98 letters in the conbercept group versus 8.81 letters in the control group (p = 0.64). These findings indicate that monthly intravitreal conbercept injection for 3 months, followed by administration every 3 months, resulted in improved and sustained visual acuity and is effective for the treatment of AMD, whereas the other anti-VEGF drugs did not maintain a similar clinical effect with quarterly administration (7). A meta-analysis of the efficacy and safety of conbercept for the treatment of nARMD found no difference in clinical efficacy between conbercept and ranibizumab and that serum VEGF levels were lower in the conbercept group than in the ranibizumab group (44, 60). Furthermore, conbercept was well tolerated intravitreally with no serious adverse events (AEs) reported in the PHOENIX study: the most common ocular AEs were elevated intraocular pressure and subconjunctival hemorrhage, and the large molecular size of conbercept limits the permeability of the blood-ocular barrier and reduces systemic exposure (7). As a new generation drug, conbercept has the advantages of multi-targeting, high affinity, long half-life, and low levels in serum. Large-sample, multi-center clinical studies are needed to further evaluate the long-term efficacy and safety of conbercept and to develop individualized patient regimens, testing the need to maximize visual gain and minimize treatment frequency.

Aflibercept is a recombinant protein comprising two protein sequences of the VEGF receptor (VEGFR-1 and VEGFR-2) and the Fc portion of immunoglobulin G1. Aflibercept binds with high affinity to all isoforms of VEGF-A as well as to VEGF-B and placental growth factor (PLGF), inhibiting the downstream signaling mediated by these receptors. In vitro, the equilibrium dissociation constants (KD) of aflibercept with human VEGF-A121 and -A165 isoforms are 0.36 and 0.49 pmol /L, respectively (45). Since its approval by the U.S. FDA in November 2011 (61), aflibercept has become the initial therapeutic drug of choice for nARMD with studies showing that intravitreal injection of aflibercept effectively improves BCVA and reduces central macular thickness (CMT) in nARMD patients (62–66).

Currently, the mainstream treatment regimens are 3 + PRN (pro re nata) and treatment-and-extend (T&E). The 3 + PRN regimen consists of one injection per month for the first 3 months of treatment, followed by monthly follow-up and re-injections as indicated (67). The T&E regimen consists of a gradual extension of treatment and follow-up to determine the longest interval between treatment and follow-up for individual patients (68). The T&E regimen has been shown to be effective at reducing the number of injections per month during the first 3 months of treatment. Both the PRN and T&E regimens are clinically designed to reduce the burden of anti-VEGF therapy and to stabilize BCVA. The T&E regimen is the preferred option among Spanish retina specialists who believe that, because of the longer duration of drug action in the body of aflibercept and fitch, the regimen is suitable for routine clinical use (69). A UK panel of experts also recommends the T&E regimen, setting specific criteria based on visual acuity and retinal morphologic features shown by optical coherence tomography to achieve longer injection intervals, fewer injections, and maintenance of stable visual acuity. The panel suggests that the T&E regimen has a stronger long-term effect than the PRN regimen for predicting future demand for nARMD treatment (70). The T&E regimen has become a standard treatment option for ARMD with more experts choosing this regimen for the treatment of nARMD with aflibercept (Table 2).

Aflibercept, like other anti-VEGF drugs, can cause subconjunctival hemorrhage, intraocular infection, ocular pain, increased intraocular pressure, cataracts, and posterior vitreous detachment. A randomized, double-blind intervention study found that both treatments significantly increased BCVA and decreased CMT in patients, and that the 1.25 mg/0.05 mL and 2 mg/0.08 mL aflibercept regimens had similar safety profiles without any major unexpected AEs (71). Other studies have reported that tachycardia occurs in a small number of patients with nARMD after repeated aflibercept injection, particularly those with subretinal pigment epithelial lesions and no intraretinal edema (72). In a study of aflibercept for the treatment of nARMD, 8 patients (12.7%) developed early tachycardia or shortness of breath, which is an important clinical problem in patients receiving long-term anti-VEGF therapy (73). Patients’ macular function continued to improve during intravitreal injections of aflibercept and peripapillary retinal cone cell function declined after multiple treatments, suggesting that aflibercept may have adverse effects on normal peripapillary retinal function in the macular area (74). From the perspective of current clinical applications, aflibercept is safe. Although adverse reactions can be treated and recovered within a short period of time, caution is warranted when using aflibercept and effort should be made to increase the number of patients in trials and carry out longer-term follow-up.

Ranibizumab is a second-generation recombinant humanized monoclonal antibody to Fab fragments with a relative molecular mass of 48,000 and high binding affinity to various isoforms of VEGF-A (VEGF165, VEGF121, and VEGF110) (46–48). Ranibizumab has been proven to be safe and effective in several clinical trials, including MARINA, ANCHOR, FOCUS, PIER, and PrONTO. Ranibizumab can effectively improve retinal morphology, reduce choroidal neovascularization, reduce retinal thickness, and improve patients’ visual acuity (75–78). However, a 7-year follow-up study found that vision stabilized in about half 1/2 of eyes compared to baseline vision, but that 1/3 of eyes had a loss of 15 letters or more and 98% of eyes had macular atrophy and photoreceptor cell damage (16). There are rare reports of localized adverse reactions including endophthalmitis, uveitis, vitreous hemorrhage, primary retinal detachment, retinal tears, and lens damage, but more common reports of transient elevation of intraocular pressure. Systemic adverse reactions include aortic aneurysm, atrial fibrillation, carotid artery stenosis, coronary artery disease, falls, femur fracture, gastrointestinal bleeding, and systemic immune system reactions (79). For current clinical treatment of nARMD, ranibizumab is the main first-line anti-VEGF drug.

Brolucizumab is a humanized monoclonal single-chain antibody fragment (scFV) produced by Escherichia coli through DNA recombination technology that shows a high affinity for all three major isoforms of VEGF-A (VEGF110, VEGF121, and VEGF165) and prevents their interactions with the VEGF-1 and VEGF-2 receptors, thus inhibiting neovascularization and decreasing vascular permeability (Fifure 2) to play a role in the treatment of nARMD (21, 49, 50).

At present, brolucizumab is the smallest molecular weight (26 kDa) VEGF conjugate and has a molecular weight less than 1/4 that of aflibercept (115 kDa), less than 1/5 that of conbercept (143 kDa), and approximately 1/2 that of ranibizumab (48 kDa). Brolucizumab binds to VEGF-A at a ratio of 2:1. The amount injected into the intravitreal cavity in 0.05 mL can reach 6 mg. ml of vitreous cavity to achieve a volume of 6 mg (51), the molar concentration of brolucizumab is 12, 66, and 22 times higher than that of aflibercept, conbercept, and ranibizumab, respectively. It is because of brolucizumab’s low molecular weight and high molar concentration that it can effectively penetrate the retina and choroid. In studies with rabbits, brolucizumab was exposed at concentrations 2.2-fold higher than that of ranibizumab in the retina and 1.7-fold higher than that of ranibizumab in the retinal pigment epithelial/choroidal layer (80). Ranibizumab has a maximum peak time of 6 h in the monkey vitreous cavity and a half-life of 2.6 days, aflibercept has a maximum peak time of 24 h in a rabbit model and a half-life of 3.63 days, conbercept has a maximum peak time of 6–12 h in the rabbit vitreous cavity and a half-life of 3.7 days, and brolucizumab has a maximum peak time of 1–6 h in the monkey retina and a half-life of 2.4 days (81–83). Due to these properties and the absence of the Fc domain, brolucizumab is cleared rapidly and systemically (5.6 ± 1.5 h). As retinal exposure is maintained at a high level, this reduces the risk of systemic adverse effects while maintaining a durable therapeutic effect. The mean half-lives of brolucizumab, ranibizumab, and aflibercept in the vitreous cavity are reported to be 56.8, 62, and 53 h, respectively (84).

The incidence of aseptic intraocular inflammation (IOI) after brolucizumab treatment is up to 4.6% (85), which was slightly higher than that of other anti-VEGF agents (0.3–2.9%) (86). Pre-existing IOI and simultaneous bilateral injections have been suggested as risk factors for the development of IOI after brolucizumab injection (87). Mukai et al. (88) reported that advanced age, female sex, and diabetes are also risk factors for IOI after injection. Other studies have found that IOI is somewhat self-limiting, with some cases resolving on their own without treatment (86). The cause of IOI occurrence may be the small molecular weight of brolucizumab or that the epitope is not recognized by the immune system after intraocular exposure. Secondary IOI may be due to a type III hypersensitivity reaction. Early recognition of inflammation due to an immunogenic reaction and prompt glucocorticoid treatment are thus essential when applying brolucizumab to patients with concomitant endophthalmitis. Brolucizumab has been widely used for the treatment of nARMD since it was approved by the FDA in 2019. The available clinical data indicate that brolucizumab has a favorable safety profile and provides a new pathway for clinical treatment of nARMD compared with other anti-VEGF drugs.

Co-expression of Ang-2 and VEGF-A has been shown to accelerate CNV formation. On the one hand, Ang-2 competitively inhibits the Ang-1/Tie2 pathway, disrupting the interaction between vascular endothelial cells and pericytes, returning endothelial cells to and activated state and, simultaneously, attracting further aggregation of angiogenic factors, such as VEGF, to the endothelial cells to cause endothelial cell outgrowth, neovascularization, and vascular exudate (52). On the other hand, unlike the stable expression of Ang-1, Ang-2 is inducible and VEGF stimulates Ang-2 secretion by Weibel Palade vesicles, which reduces the stability of the vessels and ultimately leads to the formation of CNV (53). Faricimab is a humanized bispecific IgG1 antibody that acts through dual pathway inhibition by binding to and neutralizing VEGF-a and Ang-2. It has been show recently that Ang2, play a crucial component of the Ang/Tie pathway, plays a multifaceted role in vascular homeostasis, influencing vascular permeability and participating in neoangiogenic and proinflammatory processes. For these reasons, Faricimab may have a specific influence on choroidal flow signal (89, 90). Studies of mouse CNV models have shown that, although both anti-Ang-2 and anti-VEGF-A/Ang-2 treatments reduce CNV count and vascular exudation, combined anti-VEGF-A/Ang-2 treatment is more effective (54). Significant reduction in macular retinal thickness from baseline levels after treatment with faresimab has been reported in a clinical trial of nARMD patients (91, 92). Intravitreal faricimab is generally well tolerated in clinical trials of patients with nAMD or DME with an adverse event and safety profile comparable to that of aflibercept (91, 92).