Trying to keep pace with 80-year-olds ...

... and their AMD. The Big Three will have issues handling demand from our aging society.

By Evan N. Dunn, MD, and Veeral S. Sheth, MD, MBA, FACS

The prevalence of age-related macular degeneration (AMD), diabetic macular edema (DME) and retinal vein occlusion (RVO) are projected to increase considerably over the coming decades. If current population modeling estimates are correct, the Centers for Medicare & Medicaid Services will spend $20 billion over the next decade on bevacizumab and ranibizumab (Avastin, Lucentis, Genentech).¹ New indications for intravitreal injections and innovative intravitreal drug development will only further increase demand. The tremendous costs to our health-care system, now and in the future, highlight the need for data-driven approaches to the utilization of anti-VEGF agents.

DME — which anti-VEGF?

What about steroid implants?

Diabetic retinopathy and DME are a leading cause of blindness among working-age adults.² Ranibizumab, bevacizumab and aflibercept (Eylea, Regeneron) have all demonstrated efficacy in the treatment of DME through randomized prospective clinical trials, although only ranibizumab and aflibercept are FDA-approved for DME.

The Diabetic Retinopathy Clinical Research Network’s (DRCR) Protocol I evaluated the effect of 0.5 mg ranibizumab with prompt or deferred (eligible at the 6-month visit) focal/grid laser and 4 mg triamcinolone (IVT) with a prompt laser for DME. The BCVA in pseudophakic eyes receiving IVT with prompt laser was comparable to the ranibizumab groups at year 1.³ The 5-year results demonstrated that eyes in the ranibizumab + deferred laser group had a BCVA of 2.6 letters greater than ranibizumab + prompt laser.⁴

The retail cost of ranibizumab is $1,200 per injection, bevacizumab is less than $50 and aflibercept is $1,950. DRCR Protocol T was designed to provide a comparison of the efficacy of these three intravitreal therapies. DRCR recently released the 2-year data from Protocol T, a randomized prospective multicenter clinical trial, which enrolled adult patients with center-involving DME who had received no anti-VEGF agents in the past 12 months.⁵ Patients were randomized to 0.3 mg ranibizumab, 1.25 mg bevacizumab, and 2.0 mg aflibercept with q4 week injections for the first 24 weeks. Patients received focal/grid laser to persistent or nonimproving DME on or after the 24-week visit. Exams were performed every 4 weeks during year 1 and every 4 to 16 weeks during year 2, depending on the treatment course. The mean number of anti-VEGF treatments was approximately 10 at year 1; at year 2 it was 6 (the aflibercept cohort received one less at years 1 and 2).

Protocol T concluded that at 1 year aflibercept has greater efficacy in improving BCVA than ranibizumab or bevacizumab in patients with visual acuity of 20/50 or worse. However, at 2 years Protocol T concluded that in this subgroup, aflibercept was more effective at improving visual acuity than bevacizumab, but not ranibizumab. There was little difference in the visual acuity outcomes among the three agents if the BCVA was better than 20/50.

Persistent macular edema (optical coherence tomography central subfield thickness of 250 um or greater) was present in 25% of patients in the aflibercept group, 34% in the ranibizumab group, and 54% in the bevacizumab group at year 2.5. The presence of persistent macular edema, despite protocol driven treatment with anti-VEGF agents, highlights the need for adjuvant therapy, such as corticosteroids.

Two intravitreal sustained-release implants, dexamethasone 0.7 mg (Ozurdex) and fluocinolone acetonide 0.19 mg (Iluvien) are FDA-approved for the treatment of DME. The MEAD study demonstrated the long-term efficacy of the Ozurdex implant with 22.7% of patients gaining >=15 letters from baseline with a mean of 4.1 treatments over 3 years.⁶ The incidence of cataracts was 67.9%, cataract surgery in 59.2% of phakic eyes, 27.7% had an IOP elevation of >=10 points from baseline, and 0.6% required trabeculectomy.

The FAME trial of Iluvien demonstrated 27.8% of patients gained >= 15 letters from baseline with an average of one implant per 3 years.⁷ Cataracts progressed in nearly all phakic patients and incisional glaucoma surgery was performed in 4.8%. Iluvien is FDA-approved for the treatment of DME in patients who have not had a clinically significant IOP rise with previous intravitreal corticosteroid injections. Subgroup analysis in these studies indicated a possible advantage to using intravitreal steroid implants in cases of chronic DME.

What is our pharmacologic approach to treating DME?

Generally, we initiate treatment with an anti-VEGF for three months, then choose further treatment based on improvement in visual acuity and presence or absence of macular edema. We reserve focal/grid laser for noncenter-involving macular edema following six months of treatment with intravitreal injections. If edema remains, we consider switching agents within the anti-VEGF category or to the corticosteroid category. Approximately 30% of patients are resistant to anti-VEGF agents and may need supplementation with corticosteroids. Our first-line corticosteroid is Ozurdex, given its lower risk of cataracts or glaucoma compared to Iluvien and the opportunity to decrease retreatment burden compared to anti-VEGF injections. Our treatment criteria for Iluvien include previous Ozurdex with no significant IOP rise.

A meta-analysis evaluating the systemic risk of receiving bevacizumab or ranibizumab compared to control reported no significantly increased risk of overall mortality, stroke, myocardial infarction, venous thromboembolic events or hypertension.8 Given that studies and meta-analyses thus far have not been powered enough to adequately assess these risks, we use relative caution in using anti-VEGF injections in patients who have had a recent cardiovascular event.

In all cases, we believe it is important to educate patients on the potential risks of anti-VEGF agents. Also warranted is a strong consideration for use of a corticosteroid implant for the treatment of DME in high-risk cases. Ozurdex and Iluvien are both contraindicated in patients with an active or suspected ocular or periocular infection and in patients with a torn or ruptured posterior capsule (due to concern for migration into the anterior chamber). Previous laser posterior capsulotomy is not a contraindication. Ozurdex and Iluvien are also contraindicated in patients with glaucoma and a cup-disc ratio greater than 0.8.

Table. Annual total burden to the U.S. economy of AMD, Cataract, Diabetic Retinopathy, Glaucoma, Refractive Errors, Visual Impairment and Blindness. (2006)
Direct medical costs: $16.2 billion; Direct non-medical costs: $11.2 billion; Productivity losses: $8 billion; Total: $35.4 billion
Illustration from: Richman EA, Netrabile S, Sekulich K. The economic impact of vision problems: The toll of major adult eye disorders, visual impairment, and blindness on the U.S. economy. Chicago: Prevent Blindness America, 2007.
Found on: Vision Health Initiative Report. Centers for Disease Control and Prevention. http://www.cdc.gov/visionhealth/publications/economic_impact.htm

Neovascular AMD — which treatment regimen?

The landmark ANCHOR study — anti-VEGF antibody for the treatment of predominantly classic choroidal neovascularization in AMD — was a phase 3 multicenter clinical trial comparing the efficacy of ranibizumab with photodynamic therapy (PDT); it established the superiority in BCVA outcomes comparing ranibizumab to PDT in the treatment of predominately classic neovascular AMD.⁹ Patients received a fixed monthly dose of ranibizumab. The HORIZON study, an open-label extension of ANCHOR, and other neovascular AMD trials, were designed to evaluate the effect of PRN treatment at physician discretion with 0.5 mg ranibizumab on BCVA.¹⁰ Study participants previously receiving monthly injections lost a mean of 5.3 letters from baseline. Importantly, the PRN strategy did not decrease the frequency of clinic visits.

The TREX-AMD study (n=60) compared monthly injections of ranibizumab to a treat-and-extend regimen and found that at 12 months, the BCVA gain was 9.2 and 10.5 letters (P=0.60), respectively.¹¹ The treat-and-extend group required three fewer office visits and treatments compared to the monthly cohort. Despite a small sample size, the TREX-AMD study provided level 1 evidence in favor of treat-and-extend.

A 2014 survey by ASRS reported 78% of providers use a treat-and-extend protocol.¹² We use such a protocol. The treat-and-extend protocol begins with three initial monthly injections. Upon absence of macular hemorrhage and an OCT demonstrating no subretinal fluid with monthly injections, the patient is asked to return in six weeks. Visual acuity, OCT and fundus exam are then repeated.

The patient is administered an intravitreal anti-VEGF regardless of the presence or absence of hemorrhage or fluid. In the absence of disease recurrence, the interval is increased by one to two weeks. In the presence of disease recurrence, we shorten the interval until fluid and hemorrhage resolve. We then begin to extend the interval again, but not often past the interval of last disease recurrence.

Which anti-VEGF?

While Level 1 evidence exists for all three anti-VEGF therapies for the treatment of neovascular AMD, only ranibizumab and aflibercept are FDA-approved for this condition. The Comparison of AMD Treatment Trials (CATT) was a landmark noninferiority trial comparing ranibizumab to bevacizumab. It found at two years that bevacizumab and ranibizumab had a similar efficacy and safety profile when administered according to the same dosing schedule.¹³

The View-1 study (VEGF Trap-Eye: Investigation of Efficacy and Safety in Wet AMD) was a phase 3 noninferiority study comparing ranibizumab 0.5 mg every four weeks to aflibercept 0.5 mg every four weeks, aflibercept 2 mg every four weeks and aflibercept 2 mg every eight weeks (after three initial monthly injections).¹⁴ Aflibercept was FDA-approved for 2 mg monthly injections for the first three months followed by every eight weeks (with the option to dose every four weeks if needed). No comparative efficacy trials exist comparing aflibercept to bevacizumab.

We utilize the treat-and-extend protocol. If the patient is not initially improving as anticipated, we switch anti-VEGF agents.

Non-FDA approved indications for intravitreal injections

DRCR protocol S compared ranibizumab to panretinal photocoagulation (PRP) for the treatment of proliferative diabetic retinopathy and established that ranibizumab is as effective as PRP for the treatment of proliferative diabetic retinopathy.¹⁵ There is a growing consensus that anti-VEGF therapy for neovascular glaucoma leads to regression of neovascularization of the iris and angle, and a decrease in intraocular pressure when parts of the angle remain open.¹⁶

Final thoughts

There is a large and growing need for more efficacious drugs with greater durability for patients with DME and AMD. Exciting new agents for intravitreal use are being evaluated in clinical trials. One such example is an antiplatelet-derived growth factor-BB pegylated aptamer (Fovista), which is being evaluated in combination with ranibizumab, bevacizumab and aflibercept for patients with neovascular AMD.¹⁷

Dramatic changes in the prevalence of AMD and DME and increasing indications for intravitreal therapies will cause a substantial rise in the demand for intravitreal injections. We must continue to innovate and use data-driven therapy to provide the most efficacious care in the most cost-effective manner. OM

REFERENCES

About the Authors
	Evan N. Dunn, MD, is a vitreoretinal surgery fellow at University Retina/University of Illinois College of Medicine.
	Veeral S. Sheth, MD, MBA, FACS, is director, Scientific Affairs at University Retina and Macula Associates, and clinical assistant professor at University of Illinois at Chicago. Contact him at vsheth@gmail.com.