Durable Treatment for Macular Degeneration

Innovations bring promise for more durable drugs, combination therapies, and sustained-release devices.


Anti-vascular endothelial growth factor (VEGF) treatments revolutionized the treatment of neovascular age-related macular degeneration, but the treatments put a significant burden on patients. Now, after years of disappointment, projects to relieve treatment burden look likely to succeed.

Improved regimens of single drugs, combinations of drugs, sustained-release devices, and new drugs are all still under investigation. But the most hopeful news came in June when I and other investigators reported encouraging results from the HAWK and HARRIER trials of brolucizumab (Novartis) at the American Academy of Ophthalmology (AAO) annual meeting.1 In these trials, brolucizumab outperformed aflibercept on disease activity assessments, including key measures of disease progression seen on optical coherence tomography (OCT).

These results were maintained on a 12-week dosing interval immediately following the loading phase through week 48. Brolucizumab is now the first anti-VEGF treatment for wet AMD to demonstrate robust visual gains with a majority of patients maintained on a treatment interval this long in randomized clinical trials.

That’s important because of the challenge that these treatments pose for our patients. Many are elderly and frail or live far from the offices of a retina specialist. Appointments can require assistance, often from a family member whose availability is limited. Intravitreal injections can be uncomfortable and pose risks of adverse events, including endophthalmitis. And the treatments are costly.

For all these reasons, our field has long sought ways to reduce the frequency of injections needed to improve our patients’ vision. But the durability of a treatment’s effect is far from the only measure of its success. The ideal treatment would also bring about even greater and faster improvement in visual acuity. These improvements would last not only from one treatment to the next, but also for the rest of the patient’s life. And all these benefits would be achieved while reducing the risk of adverse effects.


In the pivotal trials of 2 anti-VEGF agents now being used for wet macular degeneration, bevacizumab (Avastin; Genentech) and ranibizumab (Lucentis; Genentech), patients received monthly treatments after an initial loading phase. In the third, aflibercept (Eylea; Regeneron) patients received bimonthly injections.2 But in practice, most US clinicians are now treating their patients less frequently on average, some by using a PRN (pro re nata or “as needed”) regimen and more by using a “treat-and-extend” regimen.3 In PRN, patients typically begin with 3 monthly injections, then visit their physicians every 4 weeks and receive another injection only if there are signs of active disease, such as loss of visual acuity or evidence visible on OCT evaluation of intra- or subretinal fluid or hemorrhages.

In treat-and-extend, patients begin with 3 monthly injections, then receive an injection at every visit. While protocols vary, the interval between visits typically increases by 2 weeks when no signs of exudation are present and decreases by 2 weeks, to a minimum of 4 weeks, if recurrent disease is detected on the OCT images.

Studies comparing these approaches suggest that fixed interval protocols are more effective than PRN protocols. But they also suggest that treat-and-extend protocols have the potential to match the results of fixed-interval protocols. “It seems that the number of injections and visits can be reduced without negatively affecting visual function,” concluded the authors of a recent review.2

Questions remain, however, about treat-and-extend approaches. Long-term head-to-head comparisons of protocols are lacking, particularly with aflibercept. Also, treat-and-extend regimens may require more frequent OCT examinations than fixed-interval regimens, and these can be expensive and time consuming.4 And both treat-and-extend and PRN regimens still require far more office visits, examinations, and injections than would be ideal.


Until recently, combination drugs looked like the best hope for making wet macular degeneration treatments more durable as well as more effective. While VEGF plays a central role in choroidal neovascularization, suppressing it doesn’t cure the disease. Some patients don’t get any benefit from anti-VEGF treatment. And long-term studies suggest that patients may develop resistance to this treatment.

Oncology research has put some of the blame on pericytes, a type of cell that covers and protects new vessels as they develop and mature. Pericytes supply VEGF and other cell survival factors to proliferating endothelial cells and confer anti-VEGF resistance.5 Platelet-derived growth factor (PDGF) recruits pericytes and influences their survival and maturation.6

With these findings in mind, Ophthotech developed a PEGylated DNA aptamer as a PDGF inhibitor. It tested this agent, pegpleranib (Fovista), in combination with bevacizumab, ranibizumab, and aflibercept. But after encouraging phase 2 results, none of these combinations demonstrated superiority over monotherapy with the anti-VEGF drugs in phase 3.

Regeneron faced a similar letdown in a trial of its own PDGF inhibitor, rinucumab. At 12 weeks, best-corrected visual acuity (BCVA) in patients receiving rinucumab in combination with aflibercept improved 5.8 letters. Patients getting aflibercept alone had a 7.5-letter improvement. And ocular adverse events were more common in the combination groups compared to the group that received aflibercept alone.7

Another disappointment came when Regeneron added the angiopoietin-2 (Ang2) antibody nesvacumab to aflibercept. Preclinical data showed that angiopoietins act in tandem with VEGF. The company decided after phase 2 trials that the new molecule did not provide enough of a boost to warrant phase 3 development.8

Adding to the frustration, Ohr Pharmaceuticals reported bad news in its trial of topical squalamine in combination with ranibizumab monthly injections. A calmodulin inhibitor, squalamine prevents growth-factor receptor activation, interfering with PDGF, as well as VEGF and basic fibroblast growth factor. The 119 patients receiving the combination gained a mean of 8.33 letters while the 118 patients on ranibizumab alone gained 10.58 letters. The results left the company re-evaluating this approach.9

The results of these trials on new agents added to existing anti-VEGF therapies have come as a shock to many in the field who had seen such combinations as the most promising approach on the horizon. They illustrated the high bar that new treatments have to clear before they can improve on the results the existing anti-VEGFs are now achieving.

Not everyone has given up on adding new drugs to the existing anti-VEGFs. The existing anti-VEGF treatments target VEGF-A. Opthea designed its OPT-302 to inhibit VEGF-C/D and is combining it with ranibizumab in a phase 2b trial. A phase 1/2a trial in 51 patients suggested the combination could be more effective than ranibizumab alone.10 Santen, meanwhile, is conducting a phase 2a trial of intravitreal injections of carotuximab (DE-120), a small molecule that inhibits tyrosine kinase, a key signaling pathway for both VEGF and PDGF, in combination with ranibizumab.11

Combinations of anti-VEGF therapy with older treatments have also yielded mixed results. A few studies have combined anti-VEGF with photodynamic therapy, which was the most effective treatment before the advent of anti-VEGFs.12 But there is little research that can be used to compare this approach with anti-VEGF monotherapy alone, and photodynamic therapy carries a risk of loss of visual clarity, among other adverse effects.

Other researchers have examined a combination of anti-VEGF treatment with steroids on the theory that inflammation and fibrosis also contribute to wet macular degeneration. In one study, investigators implanted dexamethasone intravitreal implants (Ozurdex; Allergan) in some patients who had choroidal neovascularization that had not responded adequately to ranibizumab alone. A similar group of patients received sham implants. All the patients continued to receive ranibizumab injections. The patients implanted with the dexamethasone needed ranibizumab injections a median of every 34 days vs 29 days for those who received the sham implants. Only those who received the dexamethasone had statistically significant improvements in central retinal subfield thickness. However, patients in neither group showed a sustained, clinically significant improvement in BCVA. And conjunctival hemorrhage and increased intraocular pressure were significantly more common in the group receiving dexamethasone.13


If dexamethasone can be administered through an implant, why not take the same approach with an anti-VEGF? In theory, this could reduce the need for patients to return for frequent injections. While several companies have worked on this approach, Genentech may have progressed the farthest in research, with the refillable rigid port delivery system for ranibizumab it acquired from ForSight Vision4.

Surgeons implant the device through a scleral incision, then refill it in their offices using a proprietary needle. The company estimates that the device will need to be refilled every 4 to 6 months. Genentech is conducting the 220-patient phase 2 trial, and has received fast-track designation from the US Food and Drug Administration. Meanwhile, Regeneron is developing its own sustained-release device for aflibercept.14

Taking a similar approach, Replenish has developed a small, refillable, implantable ocular drug pump, the MicroPump. It can be programmed to dispense nanoliter-sized doses of drugs every hour, day, or month, as needed. It is refilled using a disposable proprietary needle tubing kit. It features a 1-way check valve to prevent backflow leakage, a fluid flow sensor, a bidirectional telemetry system for wireless programming and battery recharges, a programmable microcontroller with a calendar to “wake up” when it’s time for medication, and an optional cannula with a pars plana clip to direct medication in the same location as intravitreal injections.

In a pilot study, researchers implanted the device in 11 patients with diabetic macular edema, where it delivered the programmed dose of ranibizumab in 7 of them. The other 4 patients received a lower dose than programmed, requiring additional intravitreal injections of the drug.14

Many other devices under investigation would elute tyrosine kinase inhibitors. Ocular Therapeutix is working on a sustained-release hydrogel-based drug delivery depot that can be formulated with both small-molecule pharmaceuticals as well as anti-VEGF drugs. It has reported promising preclinical data on the OTX-TKI, which releases tyrosine kinase inhibitor particles for 6 months.15-17

pSivida, which developed the implant vehicles for both the diabetic macular edema drug Iluvien (Alimera Sciences) and the uveitis drug Retisert (Bausch + Lomb), is at work on a sustained-release approach to retinal diseases. In preclinical trials, it has used its Durasert, an injectable, sustained-release drug delivery system for small molecules that can last up to 3 years, as a delivery vehicle for tyrosine kinase inhibitors.

Neurotech Pharmaceuticals has abandoned NT-503 ECT, a cell line that synthesizes ciliary neurotrophic factor within a semipermeable membrane with the goal of protecting photoreceptor cells.

The future could bring a gene-therapy approach to wet macular degeneration. Some proof-of-concept trials have already taken place. In one, researchers at Nightstarx used an adeno-associated viral vector to deliver the gene-encoding soluble Fms-related tyrosine kinase-1. They hope that transduced cells will produce sufficient amounts of this natural VEGF inhibitor to diffuse into the vitreous and inactivate VEGF. While there was no control group, the researchers measured a reduction in the need for standard ranibizumab injections that suggested a treatment effect.18


Brolucizumab (RTH258, formerly ESBA1008) could change the future of wet macular degeneration treatment for several reasons. Besides improving visual acuity on a 12-week treatment interval, brolucizumab is now the first anti-VEGF treatment for wet AMD to demonstrate robust visual gains with most patients maintained on a less-frequent 12-week treatment interval immediately following the loading phase in randomized clinical trials.

A humanized single-chain antibody fragment that inhibits all isoforms of VEGF-A, it is the smallest of the anti-VEGF antibodies. It has a molecular weight of 26 kDa, compared with 115 kDa for aflibercept and 48 kDa for ranibizumab. Its design allows it to concentrate up to 120 mg/mL. As a result, 6 mg can be administered in a single 50-mL intravitreal injection.

Six milligrams of brolucizumab equals approximately 12 times the 2.0-mg dose of aflibercept and 22 times the 0.5-mg dose of ranibizumab. Its small molecular weight and high drug concentration gradient between the vitreous and retina may allow its distribution into the retina. If 2 drugs have comparable half-lives, higher molar doses may be cleared more slowly from the eye, thus prolonging brolucizumab’s duration of action.19

The HAWK and HARRIER trials, in which brolucizumab was compared to aflibercept, demonstrated what a difference this can make. The primary objective of these 2 trials was to show that brolucizumab could improve mean BCVA at least as much as aflibercept, and brolucizumab did achieve that. But a key secondary goal was to measure how many patients could enjoy significant improvements on a 12-week dosing interval. To that end, the researchers used an innovative regimen that assigned the majority of the patients to a 12-week schedule immediately following the loading phase.

The studies included more than 1,800 patients at 400 centers worldwide. The study randomly assigned patients to 6-mg and (in the case of HAWK) 3-mg doses of brolucizumab or 2 mg of aflibercept. Immediately following the 3-month loading phase, patients in the brolucizumab arms switched to a 12-week dosing interval with an option to adjust to an 8-week interval based on disease activity assessments at defined visits. Patients taking aflibercept followed the bimonthly interval according to its label.

Fifty-seven percent of the patients on brolucizumab in HAWK and 52% in HARRIER stayed on a 12-week dosing interval immediately following the loading phase through week 48. Compared to the patients taking aflibercept, 35% fewer patients on the 6-mg dose of brolucizumab in HAWK, and 33% fewer in HARRIER, had intraretinal or subretinal fluid at week 16. The results were statistically significant (P<.0001 for both). And this wide difference between the 2 treatments persisted in both trials through week 48. This suggests the potential for a long-lasting effect and decreased treatment need.

Patients taking the 6-mg dose of brolucizumab also had bigger reductions in central subfield thickness (CST) as measured by OCT, a key indicator of abnormal fluid accumulation in the retina. Significantly improved CST reductions were evident at weeks 16 and 48 in both trials.

The 2 drugs in these trials appeared equally safe. The most frequent ocular adverse event for patients in the brolucizumab 6-mg group was reduced visual acuity, which affected 6.9% in HAWK, compared to 8.9% of patients taking aflibercept in that trial. It was followed by conjunctival hemorrhage, which affected 4.4% of the 6 mg brolucizumab group and 8.9% of the aflibercept group. Vitreous floaters affected 5.0% of the 6-mg brolucizumab group and 3.1% of the aflibercept group. Eye pain affected 4.4% of the 6-mg brolucizumab group and 4.2% of the aflibercept group. The incidence of adverse events in the HARRIER trial was similar.

Brolucizumab may have the best results so far, but it won’t necessarily get the last word. Other molecules are finding their way through the pipeline. Among the most interesting is RG7716. Roche/Genentech developed this molecule using its patented CrossMab technology that allows one antibody molecule to bind to 2 different targets, in this case, VEGF and Ang2. In addition, its central arm has 2 mutations — one to increase systemic clearance, and the other to decrease inflammation. In a phase I trial, BCVA increased by a median of 7 letters in 28 days.20

Also coming down the pike is Allergan/Molecular Partners’ abicipar pegol, a designed ankyrin repeat protein (DARPin). DARPins are very small and soluble proteins, allowing the delivery of large amounts of the agent in a low volume with the potential for long action. In a phase 2 study, researchers pitted the new drug against 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 pegol 2 mg, 7.1 letters for abicipar pegol 1 mg, and 4.7 letters for ranibizumab. Abicipar pegol has now moved into phase 3.

And that’s not all. Panoptica’s PAN-90806 targets inhibiting tyrosine kinase. Panoptica reported positive results in a phase 1/2 trial with PAN-90806, a topical anti-VEGF. About half of treated patients experienced improvements including in vision, vascular leakage, and lesion morphology.

The combination of brolucizumab’s promising phase 3 results and so many devices and molecules under development leaves me optimistic that patients of the future will enjoy better efficacy and long-term results, as well as a lighter burden of treatment. RP


  1. Novartis. Novartis brolucizumab (RTH258) demonstrates superiority versus aflibercept in key secondary endpoint measures of disease activity in nAMD, a leading cause of blindness. News release. Available at
  2. Gemenetzi M, Patel PJ. A Systematic review of the treat and extend treatment regimen with anti-VEGF agents for neovascular age-related macular degeneration. Ophthalmol Ther. 2017;6(1):79-92.
  3. Stewart MW. Treat and extend for chorioretinal vascular conditions. Retinal Phys. 2017;14(3):28,30,32-35.
  4. Peden MC, Hammer ME, Suner IJ. Fixed-interval dosing of anti-VEGF for wet AMD. Retinal Phys. 2015;12(5):23,28-31.
  5. Reinmuth N1, Liu W, Jung YD, et al. Induction of VEGF in perivascular cells defines a potential paracrine mechanism for endothelial cell survival. FASEB J. 2001;15(7):1239-1241.
  6. Benjamin LE, Hemo I, Keshet E. A plasticity window for blood vessel remodelling is defined by pericyte coverage of the preformed endothelial network and is regulated by PDGF-B and VEGF. Development. 1998;125(9):1591-1598.
  7. Regeneron announces phase 2 study of aflibercept co-formulated with rinucumab (anti-PDGFR-beta) shows no benefit over aflibercept alone in neovascular age-related macular degeneration. News release. Available at: .
  8. Regeneron provides update on Eylea (aflibercept) injection and nesvacumab (Ang2 antibody) combination program. News release. Available at: .
  9. Ohr pharmaceutical announces efficacy results from the MAKO study in wet-AMD. News release. Available at: .
  10. Opthea doses first patient in phase 2b study of OPT-302 for wet AMD. News release. Available at: .
  11. Santen and Tracon announce initiation of a phase 2a study of wet age-related macular degeneration. News release. Available at: .
  12. Schmidt-Erfurth U, Wolf S; PROTECT Study Group. Same-day administration of verteporfin and ranibizumab 0.5 mg in patients with choroidal neovascularisation due to age-related macular degeneration. Br J Ophthalmol. 2008;92(12):1628-1635.
  13. Kuppermann BD, Goldstein M, Maturi RK, et al. Dexamethasone intravitreal implant as adjunctive therapy to ranibizumab in neovascular age-related macular degeneration: a multicenter randomized controlled trial. Ophthalmologica. 2015;234(1):40-54.
  14. Humayun M, Santos A, Altamirano JC, et al. Implantable micropump for drug delivery in patients with diabetic macular edema. Transl Vis Sci Technol. 2014;3(6):5.
  15. Jarrett PK, Elhayek RF, Jarrett T, et al. Tolerability of a 6 month sustained hydrogel delivery system for tyrosine kinase inhibitors in Dutch belted rabbits. Paper presented at: The Association for Research in Vision and Ophthalmology annual meeting; May 2017; Baltimore, MD.
  16. Elhayek RF, Jarrett T, Lattrell Z, et al. Efficacy of a 6 month sustained hydrogel delivery system for tyrosine kinase inhibitors in a VEGF induced retinal leakage model. Paper presented at: The Association for Research in Vision and Ophthalmology annual meeting; May 2017; Baltimore, MD.
  17. Elhayek RF, Jarrett T, Lattrell S, et al. Tolerability of a shape changing intravitreal hydrogel depot. Paper presented at: The Association for Research in Vision and Ophthalmology annual meeting; May 2017; Baltimore, MD.
  18. MacLaren RE. Comment: Gene Therapy for Age-related Macular Degeneration. Lancet. 2015;386(10011):2369-2370.
  19. Dugel PU1, Jaffe GJ2, Sallstig P, et al. Brolucizumab versus aflibercept in participants with neovascular age-related macular degeneration: a randomized trial. Ophthalmology. 2017;124(9):1296-1304.
  20. Chakravarthy U, Bailey C, Brown D, et al. Phase I trial of anti-vascular endothelial growth factor/anti-angiopoietin 2 bispecific antibody RG7716 for neovascular age-related macular degeneration. Presented at: Association for Research in Vision and Ophthalmology Annual Meeting; May 1-5, 2016; Seattle, WA.