Ehsan Rahimy, MD; Elham Rahimy, BS; and Anthony Joseph, MD

The advent of intravitreal anti-vascular endothelial growth factor (VEGF) therapy has undoubtedly revolutionized the field of ophthalmology and how physicians manage many potentially debilitating retinal conditions. In the recently published 5-year outcomes with anti-VEGF therapy from the Comparison of Age-Related Macular Degeneration Treatment Trials, a remarkable 50% of follow-up participants had visual acuity of 20/40 or better; however, 20% of individuals had acuity of 20/200 or worse.¹ While this represents a significant improvement over previous natural history studies prior to the commencement of the anti-VEGF era, it is a reminder that much work remains to be accomplished.

In an effort to reduce the overall treatment burden of serial anti-VEGF injections on retinal specialists, patients, caregivers, and society at large, numerous targeted therapies are under investigation to further improve patient outcomes and decrease overall costs of treatment to the health care system. This chapter will serve to summarize some of the most pertinent programs in development, and is broken down based on the different routes of drug delivery: gene therapy, intravitreal injections, systemic therapy, and topical treatment.

Gene Therapy

The goal of gene therapy is to provide continual expression of a protein(s) of interest involved in the pathogenesis of a disease for sustained therapeutic benefit. Gene therapy for retinal degenerative diseases is a field receiving significant research and developmental focus recently, as a number of therapies are progressing to human clinical trials. In these individuals, a viral vector is used to carry the desired genetic information encoding a protein(s) of interest into the target cells. Successfully transduced vectors then use the host cell’s machinery to express the particular protein(s). Two main categories of vectors are being investigated: integrating vectors and nonintegrating vectors. Integrating vectors (ie, lentiviral vectors) insert themselves into the recipient’s genome, whereas nonintegrating vectors (ie, adeno-associated virus [AAV]) usually form an extrachro-mosomal genetic element.

The secreted extracellular domain of sFlt-1 is a soluble isoform of the VEGF receptor 1 and a naturally occurring protein antagonist of VEGF. Adeno-associated virus type 2 (AAV2)-sFLT01 (Genzyme) is a replication-deficient AAV vector with a plasmid that expresses a portion of the sFlt1 receptor termed sFlt01.2 The viral vector is injected intravitreally, after which it transfects native cells to produce the modified protein. It has been shown to be well tolerated in monkeys.³ Intravitreal AAV2-sFLT01 is currently being studied in phase I clinical trials in humans with neovascular age-related macular degeneration (AMD).⁴

Subretinal AVA-101 and Intravitreal AVA-102

Like Genzyme’s lead candidate, AAV2-sFLT01, Avalanche Biotechnologies developed a similar replication-deficient AAV2 vector containing the sFLT-1 plasmid (AVA-101). However, in contrast to the intravitreal approach, AVA-101 is injected into the subretinal space, requiring concurrent vitrectomy surgery.

Phase I studies demonstrated AVA-101 to be well tolerated in the 6 study participants with no significant drug-related safety concerns.⁵ Topline results from a phase IIa study involving 32 patients with advanced AMD became available in 2015.⁶ Patients were randomized to the AVA-101 treatment group (n = 21) or control (n = 11). Participants in both arms received 2 initial ranibizumab (Lucentis) injections at day 0 and week 4, followed by ranibizumab rescue therapy as early as week 8, according to prespecified criteria. Twenty-nine of 32 individuals received prior anti-VEGF therapy (average of 10 prior injections). The reported data met the primary endpoint in terms of ophthalmic and systemic safety measures; however, secondary endpoints assessing functional and anatomic outcomes were not as clear. In terms of visual improvement, the mean change from baseline showed a difference of 11.5 letters between the treatment (2.2 letters gained) and the ranibizumab control (9.3 letters lost) groups. The degree of vision loss in the control group was uncharacteristic, but was attributed to the refractory nature of disease displayed by many of the study’s participants as they received extensive previous anti-VEGF treatment prior to enrollment. A significant number of AVA-101-treated patients (43%) improved or maintained stable vision with 2 or fewer rescue injections, compared with individuals in the control group (9%). With regards to anatomic changes, the treatment group demonstrated a mean increase of 25 μm over baseline in central retinal thickness measurements on optical coherence tomography (OCT) compared to a mean decrease of 56 μm in the control group.⁶

Avalanche reported in late 2015 that it will not proceed with a phase IIb trial on AVA-101 and its focus has shifted toward developing its next-generation candidate, AVA-201. In contrast to AVA-101, which uses a wild-type AAV2 vector, AVA-201 employs a novel vector that has been optimized for delivery via intravitreal injection. Because of the less-invasive route of administration and potential for 1-time dosing, the company is exploring AVA-201 as a potential prevention therapy for patients with high risk of progression from non-neovascular to neovascular AMD.⁷

RetinoStat (Oxford BioMedica) is an equine infectious anemia viral lentiviral vector expressing the genes of 2 naturally occurring inhibitors of angiogenesis, endostatin and angiostatin. Subretinal injection of RetinoStat in monkey and rabbit models have been shown to be capable of persistent, localized gene expression.8,9 On the basis of preclinical data, it is anticipated that RetinoStat may require only a single administration if it proves to be safe and effective in human trials.

A phase I clinical trial enrolling 21 patients with advanced neovascular AMD to receive escalating doses of subretinally injected RetinoStat was recently completed.¹⁰ The study met the primary endpoints of safety and tolerability at 6-month postsurgical transfection.¹¹ In addition, patients showed signs of clinical benefit, with stabilization of vision and an anatomic reduction in vascular leakage. Successful retinal transduction was demonstrated by a substantial increase in expression and secretion of endostatin and angiostatin proteins, as measured in the anterior chamber of the study’s eyes.¹¹ Longer-term protein expression has been sustained for up to one year post-treatment, and preliminary data show a dose response, with the higher dose levels yielding a proportional increase in average protein expression.¹¹

Intravitreal Therapy

RTH258

RTH258 (formerly ESBA1008) is a novel, humanized single-chain antibody fragment that inhibits all isoforms of VEGF-A. With a molecular weight of 26 kDa, RTH258 is smaller than any of the commercially available VEGF inhibitors. For comparison, the molecular weights of ranibizumab, aflibercept, and bevacizumab are 48 kDa, 115 kDa, and 150 kDa, respectively.¹² The smaller size of RTH258 may offer the following several potential advantages:

• Greater drug penetration into the desired ocular tissue¹³
  
• Delivery of a higher molar dose of medication in a given volume (RTH258 can be concentrated up to 120 mg/ml, allowing the administration of 6 mg in a single 50 μl intravitreal injection)^12,13
  
• Formulation within a sustained release platform in the future
  
• Reduced risk of systemic side effects owing to the rapid systemic clearance^12,13

The phase II OSPREY trial was a randomized, double-masked, controlled study that enrolled 89 patients with neovascular AMD to receive repeated doses of 6 mg RTH258 or aflibercept.¹⁴ Both therapies were dosed upfront every 8 weeks through week 32, and then the dosing interval was increased to 12 weeks, with follow-up continuing to week 56. Throughout the course of the study, patients were reassessed monthly and could receive rescue treatment injections as needed. The results demonstrated promising visual acuity gains as well as fluid reduction in the RTH258 group that were noninferior to aflibercept, meeting the study’s primary endpoint. Additionally, after the treatment interval was switched to every 12 weeks, patients in the RTH258 group continued to require fewer rescue treatments (60 vs 94 in the aflibercept arm), potentially leading to a reduced overall treatment burden (approximately half of patients in the RTH258 arm were successfully maintained on quarterly injections). Both treatments were well tolerated with no new safety concerns reported during the study.

On the heels of these positive phase II results, Alcon initiated a large phase III clinical study (HAWK) to evaluate the efficacy and safety of RTH258 vs aflibercept in neovascular AMD.15 A target enrollment is set at approximately 1700 patients and is due for completion in 2018. The primary objective of the phase III study is to compare the efficacy of RTH258 (3 mg and 6 mg doses) vs aflibercept, with the mean visual change from baseline to week 48 as the primary endpoint. Participants will be dosed every 3 months with RTH258, while a bimonthly dosing regimen will be followed for those patients considered unsuitable for a quarterly dosing schedule because of their level of disease activity.

An implantable, refillable Posterior MicroPump is also being developed to allow for pulsatile intravitreal drug delivery of RTH258 over many months, which may help significantly reduce injection burden. In a separate phase II microvolume study evaluating RTH258 administered via Posterior MicroPump, clinical outcomes were compared in 13 patients randomly assigned to 1.2 mg of a 10-μL injection of RTH258 and 13 patients treated with an infusion of 1 mg in 8.3 μL RTH258.¹⁶ For the primary endpoint, patients had to meet at least 3 of the following 4 criteria: a gain in best-corrected visual acuity (BCVA) of at least 4 letters at day 14 or at day 28, and a decrease in central subfield foveal thickness of at least 80 μm at day 14 or at day 28. This was met by 70% of patients in the injection group and 60% in the infusion group. The results of the microvolume study showed promise that a small volume of RTH258 can be successfully delivered via micropump. Further development is ongoing.

Abicipar

Another potential anti-VEGF agent under development is based on designed ankyrin repeat proteins, or DARPins for short. DARPins are genetically engineered proteins, typically with high-affinity binding to their respective target sites, that can be made into custom-designed therapeutics with optimized properties: small molecular size, high stability, high solubility, and low immunogenicity. Compared to monoclonal antibodies, DARPins appear to bind with similar high affinity and specificity to their targets, but in addition, show increased potency and longer ocular pharmacokinetics.¹⁷

The lead candidate in this class is abicipar pegol (previously MP0112, Allergan, and Molecular Partners), a small-molecular weight (34 kDa) DARPin with highly potent activity against all VEGF-A isoforms. In a rabbit model of VEGF-induced vasculopathy, abicipar demonstrated a higher binding affinity and longer vitreous half-life compared with ranibizumab, and provided superior duration of action compared with ranibizumab in equimolar dosages.^17,18

A phase II clinical study of abicipar for the treatment of neovascular AMD (REACH) assessed the safety and treatment effects of 2 different doses of abicipar every 4 weeks in treatment-naïve patients. Sixty-four patients in total were randomized to 2 mg abicipar (n = 23), 1 mg abicipar (n = 25), or ranibizumab (n = 16) and were followed for 20 weeks.¹⁹All patients received intravitreal injections at day 1, week 4, and week 8. Patients in the ranibizumab arm of the study received additional doses at weeks 12 and 16. The results demonstrated that after 16 weeks (8 weeks after the last abicipar injection), mean visual acuity improvement from baseline was 8.2 letters for abicipar 2 mg, 6.3 letters for abicipar 1 mg, and 5.3 letters for ranibizumab. After 20 weeks (12 weeks after the last abicipar injection and 4 weeks after the last ranibizumab injection), mean visual acuity improvement from baseline was 9.0 letters for abicipar 2 mg, 7.1 letters for abicipar 1 mg, and 4.7 letters for ranibizumab. Anatomic improvements on OCT imaging were also supportive of the functional visual gains. While the study was not powered to show statistically significant differences between treatment groups, the data suggested that abicipar was at least as effective as monthly ranibizumab, with a longer duration of action.

Of note, during preclinical and early clinical studies, episodes of intraocular inflammation (uveitis, vitritis, choroiditis) related to abicipar were encountered.17 Specifically, during the REACH trial, inflammatory events occurred in 2 patients treated with abicipar 2 mg, in 3 patients treated with abicipar 1 mg, and none receiving ranibizumab.¹⁹ Believing this adverse event to be due to the purification process, Allergan has since reformulated abicipar to remove potential proinflammatory impurities.

A large phase III program for abicipar in the treatment of neovascular AMD is underway, composed of 2 trials (Colour Contrast Sensitivity for the Early Detection of Wet Age-Related Macular Degeneration [CEDAR] and Safety and Efficacy of Abicipar Pegol [AGN-150998] in Patients With Neovascular Age-Related Macular Degeneration [SEQUOIA]), with each expected to recruit 900 patients.²⁰ Beyond the indications for AMD, abicipar is also in development for diabetic macular edema (DME), with a phase II study ongoing. Furthermore, Allegan and Molecular Partners are collaborating on a combination VEGF and platelet-derived growth factor (PDGF) DARPin for neovascular AMD that is currently in preclinical testing.

OPT-302

Currently available VEGF inhibitors target VEGF-A; however, OPT-302 (formerly VGX-300, Opthea, South Yarra) is a soluble form of human VEGF receptor-3 (VEGFR-3) that specifically functions by trapping VEGF-C and VEGF-D. VEGF-C and VEGF-D can stimulate angiogenesis and leakage either through the same pathway as VEGF-A (via VEGFR-2) or through an alternative mechanism independent of VEGF-A (via VEGFR-3). VEGF-C levels have additionally been demonstrated to be significantly elevated in the plasma of AMD individuals compared to healthy controls, potentially implicating this isoform in the pathogenesis of the disease process.²¹

In a preclinical mouse model of neovascular AMD, OPT-302 was able to significantly inhibit choroidal neovascularization (CNV) and vascular leakage to a comparable extent as aflibercept.^21–23 When used in combination in the same model, an additive benefit was observed, with more effective inhibition of CNV lesions than using either agent alone.^21–23 This observed synergistic effect may be due to a more complete and effective blockade of the VEGF family of molecules.

In a phase I study that enrolled 20 patients with neovascular AMD, OPT-302 administered by intravitreal injection as a monotherapy (2.0 mg) or at 3 escalating doses (0.3, 1.0, or 2.0 mg) in combination with ranibizumab was safe and well tolerated at all dose levels.^24,25 A phase IIa study is currently ongoing.

PF582

PF582 (Pfenex) is a humanized monoclonal antibody fragment targeted against VEGF-A that is biologically similar to ranibizumab. Referred to as biosimilars, these agents are comparable to generic drugs, and are intended to be a more affordable and cost-effective alternative to their United States Food and Drug Administration (FDA)-licensed reference counterparts.26 The difference from generics is that they are derived from biologics (ie, monoclonal antibodies) rather than pharmaceuticals, which are made from synthetic chemicals. A phase I/II trial in New Zealand is comparing 3 monthly intravitreal injections of PF582 to ranibizumab in patients with neovascular AMD to demonstrate equal safety and efficacy of PF582 over a 12-month period.²⁷

DE-120

DE-120 (Santen Pharmaceuticals) is a dual tyrosine kinase receptor inhibitor of VEGF and PDGF. Currently, patients with treatment-naïve neovascular AMD are being recruited to participate in a multicenter, randomized, open-label, phase IIa study (VAPOR1) assessing the efficacy, safety, and the duration of effect of intravitreal injections of DE-120 as monotherapy and with a single aflibercept injection.²⁸

Fovista (Pegpleranib)

In addition to VEGF, a key modulator in the pathogenesis of neovascular AMD is PDGF. Through the oncology literature, the role of PDGF in the proliferation and maturation of new vessels has become better elucidated, and these findings are now being extrapolated to retinovascular diseases. Chronic anti-VEGF therapy for neovascular AMD is known to induce CNV remodeling whereby it becomes covered and protected by pericytes as new vessels form and mature within the complex. These pericytes provide VEGF and other proliferative factors to the endothelial cells in which they house.^29–31 Sprout cells, also known as “tip” cells, lead the growth of the CNV membrane, and are the only endothelial cells associated with the neovascular complex not covered by pericytes. Therefore, it is believed that the tip cells are the most vulnerable to anti-VEGF therapy, while the remainder of the CNV tissue is conferred relative resistance to current treatments by the pericytes.^32,33

The recruitment of pericytes is driven by locally produced PDGF.^34–36 Furthermore, VEGF antagonism causes upregulation of PDGF, so that more pericytes are recruited and the CNV membrane undergoes maturation.³⁷ Essentially, serial anti-VEGF injections prune off the exposed tip cells and mature the underlying neovascular tissue. It is postulated that an anti-PDGF agent would be able to strip the pericyte armor from the neovascular lesion, exposing the underlying endothelial cells to the effects of subsequent anti-VEGF therapy. In this regard, combination treatment with both anti-VEGF and anti-PDGF injections holds significant promise to potentially induce CNV regression and prevent fibrosis.³⁸

Numerous companies are working on anti-PDGF agents, but the one furthest along is Fovista (pegpleranib, formerly E10030, Ophthotech). In the largest phase IIb trial conducted in retina to date (449 patients), participants were randomized to receive 1 of the following treatment regimens administered every 4 weeks for 24 weeks:

• 0.3 mg Fovista in combination with 0.5 mg ranibizumab
  
• 1.5 mg Fovista in combination with 0.5 mg ranibizumab
  
• Sham in combination with 0.5 mg ranibizumab^39,40

The primary endpoint was mean change in visual acuity from baseline to week 24. The results showed that the combination arm with the higher dose of Fovista led to statistically significant greater visual improvements than ranibizumab alone. At the higher Fovista dose, BCVA improved by 62% more from baseline (10.6 letters gained) than with ranibizumab monotherapy (6.5 letters gained), with a classic dose-response curve that continued to outpace monotherapy over the 6-month course.39,40 There were no concerning safety events reported.

In addition to fluid reduction on OCT imaging, an area of particular interest was the significant regression of subretinal hyperreflective material (SRHM) present, thought to represent the neovascular lesion itself. Resolution of SRHM at week 24 demonstrated a clear dose-related response.^39,40 This effect on SRHM has not been previously described in clinical trials of anti-VEGF therapy. The dose-response effect was also seen in patients who gained 3 lines of visual acuity or more. Furthermore, for patients who lost vision in the study, eyes treated with ranibizumab monotherapy had developed fibrosis and disciform scarring, whereas eyes receiving combination therapy showed little, if any, scarring.^39,40 Because PDGF is a potent fibrotic agent, it is feasible that its blockade has a direct impact on the degree of fibrosis and scar formation, as seen in this subgroup.

Unfortunately, preliminary phase III results released in 2016 demonstrated that Fovista (1.5 mg) in combination therapy with ranibizumab did not result in any added visual acuity benefit compared to ranibizumab monotherapy alone after 12 months, failing to meet the study’s primary endpoint. In the combined analysis from the two studies, patients receiving combination therapy with Fovista gained a mean of 10.24 ETDRS letters compared to 10.01 letters on ranibizumab monotherapy, which did not achieve statistical significance. Ophthotech is further analyzing this data to better understand the results and the pathophysiology of neovascular AMD.⁴¹

REGN2176-3

REGN2176-3 (Regeneron Pharmaceuticals) is a combination product comprising an antibody to PDGF receptor-b (rinucumab) co-formulated together with aflibercept. In a phase I study, 4 cohorts of 3 patients each received REGN2176-3 at baseline and at 4 weeks. Visual acuity remained stable or increased in the majority of patients, and the central retinal thickness on OCT decreased in all 4 cohorts. No dose-limiting toxicities, intraocular inflammation, or treatment-related serious adverse events were encountered.⁴²

A phase II, multicenter, double-masked, randomized controlled study (CAPELLA) evaluating 4 different dosing regimens of REGN2176-3 is ongoing to investigate its efficacy and safety compared to aflibercept monotherapy in patients with neovascular AMD.⁴³

Nesvacumab

Angiopoietin-2 (Ang2) belongs to a family of proangiogenic factors selectively expressed during the angiogenesis process, and is an antagonistic ligand for the vascular endothelial cell receptor tyrosine kinase, Tie2.^44,45 Unlike some of the damage-inducing pathways in retinovascular diseases that are targeted by most pharmacotherapy agents, Tie2 is protective and improves the stability of retinal vasculature. Drugs that interfere with Ang2/Tie2 signaling may help regulate vascular permeability, which may be of particular interest in treating conditions such as diabetic retinopathy and DME. Nesvacumab (REGN910-3, Regeneron Pharmaceuticals) is a fully human immunoglobulin G1 (IgG1) monoclonal antibody that selectively binds Ang2 with high affinity and blocks it from binding to the Tie2 receptor.

Two separate phase II clinical studies are underway to assess combination therapy of nesvacumab plus aflibercept as a co-formulated single intravitreal injection in patients with neovascular AMD (ONYX, Anti-angiOpoeitin 2 Plus Anti-vascular eNdothelial Growth Factor as a therapY for Neovascular Age Related Macular Degeneration: Evaluation of a fiXed Combination Intravitreal Injection)46 and DME (RUBY, Anti-vasculaR Endothelial Growth Factor plUs Anti-angiopoietin 2 in Fixed comBination therapY: Evaluation for the Treatment of Diabetic Macular Edema).⁴⁷

RG7716

RG7716 (Hoffmann-La Roche) is a unique bispecific monoclonal antibody that targets both VEGF and Ang2. In a phase I study evaluating RG7716 in patients with neovascular AMD who demonstrated suboptimal therapeutic response after multiple previous anti-VEGF injections, promising improvements in BCVA and key OCT parameters were observed.⁴⁸ Additionally, investigators found RG7716 to be well tolerated and exhibiting an overall favorable safety profile. Larger phase II studies are currently underway to evaluate RG7716’s potential to further enhance anti-VEGF monotherapy efficacy and safety.

ICON-1

ICON-1 (formerly hI-con1, Iconic Therapeutics) belongs to a class of novel human immunoconjugate proteins. It is a 160 kDa chimeric, IgG-like homodimeric protein composed of a targeting domain (mutated, inactive factor VIIa) fused to an effector domain (fragment crystallizable portion of IgG) with an intact hinge region.⁴⁹ ICON-1 targets tissue factor (TF), which is a promoter of inflammation and angiogenesis.⁵⁰ A positive feedback loop exists between VEGF and TF such that TF can induce angiogenesis by upregulating VEGF and increased VEGF can, in turn, increase TF expression.⁵⁰ By binding to cells that aberrantly overexpress TF (ie, in CNV tissue) the effector domain of ICON-1 triggers the immune destruction of these targeted cells, inducing natural killer cell-mediated cytotoxicity of the CNV complex.^51,52

The results of a multicenter phase I study evaluating a single intravitreal injection of 60 mg, 150 mg, or 300 mg ICON-1 in 18 eyes with neovascular AMD (6 eyes per cohort) revealed the therapy to be well tolerated and led to increased visual acuity, reduced retinal thickness, and CNV regression.⁵³ Furthermore, no drug-related serious ocular or systemic adverse events were observed. In 2015, Iconic initiated a phase II randomized, double-masked, multicenter, active-controlled study (EMERGE) to evaluate the safety and biological activity of repeated intravitreal administration of ICON-1 in patients with CNV secondary to AMD. Three treatment arms are being investigated in the 6-month study: ICON-1 monotherapy, ICON-1 with ranibizumab, and ranibizumab alone as a comparator.

Luminate

Integrins are cell surface receptors involved in angiogenesis and inflammation, acting both upstream and downstream of the VEGF pathway. There are 27 types of integrin receptors; however, there are no anti-integrin drugs currently available for commercial ophthalmic use.⁵⁴ Luminate (formerly ALG-1001, Allegro Ophthalmics) is an integrin antagonist that blocks 3 different integrin receptor sites (αvβ3, αvβ5, and α5β1) that mediate a number of angiogenic processes, including endothelial cell migration, proliferation, differentiation, and maturation (Figure 4-1).^54–56 Since Luminate inhibits all 3 of these receptors associated with angiogenesis, early clinical evidence suggests that it is effective in preventing the formation of new blood vessels as well as shrinking and stopping leakage from existing blood vessels.

Stay updated, free articles. Join our Telegram channel

Join