In recent years, numerous retinal disease-related clinical trials have been completed, particularly concerning age-related macular degeneration (AMD) and retinal vein occlusion (RVO).
Today's “standard care” for treating AMD is based primarily on two phase 3, randomized clinical trials, MARINA and ANCHOR, that evaluated intravitreal ranibizumab (Lucentis, Genentech).
The primary endpoint of these studies was visual stabilization, defined in these trials as losing fewer than three lines of visual acuity. Ranibizumab was found to be far superior to sham injection and photodynamic therapy (PDT) in these studies.
The most interesting finding from these studies was that on average, patients experienced visual gains with ranibizumab. Prior to these data, we thought that the only way we could achieve good vision in patients with AMD was to prevent the occurrence of neovascularization.1,2 Clearly, from a pathologic standpoint there is a period of time when reversible changes are present and the timely correction can lead to vision improvement.
REDUCED DOSING FOR AMD
PIER. The PIER trial evaluated a fixed reduced schedule. Patients had three monthly treatments and then were treated on a quarterly basis out to the primary endpoint of 12 months. Although the results at month 12 were better than sham, the results were not as good as they were in MARINA and ANCHOR (Figure 1).3
PRONTO. The PRONTO trial evaluated as-needed (PRN) therapy with ranibizumab. In PRONTO, patients were treated with three mandated initial monthly injections, after which they received PRN injections based on the criteria of five or more letter loss of vision from the last visit, persistent or recurrent fluid on OCT, an increase in central retinal thickness (CRT) of more than 100 μm, or new hemorrhage. PRONTO set a relatively low threshold for retreatment, and the outcomes were good and comparable to what was seen in MARINA/ANCHOR at 1 year (Figure 2).4 It is important to note, however, that there were only 40 patients enrolled in PRONTO, compared with 716 in MARINA and 423 in ANCHOR.
HORIZON. What other data do we have for PRN treatment? The HORIZON trial was an extension trial of the patients in the MARINA, FOCUS, and ANCHOR trials.
The patients who did not receive ranibizumab in the original trials could now crossover to intravitreal ranibizumab treatment in HORIZON. Relatively few injections were given to patients during the 2 years in HORIZON. Sixty-two percent of patients in the initially treated group received zero to three injections and 58% of patients in the crossover group received zero to three injections. Through 24 months of HORIZON, the mean and median number of injections in the ranibizumab treated-initial group was 3.6 and 2.0, and for the treatedcrossover group, the mean and median was 4.2 and 3.0.
As you can see, when they were in the fixed period in the original trials, patients treated with ranibizumab did well. When they went into the PRN dosing phase in HORIZON, however, the initial visual acuity gains appeared to fade over time (Figure 3).5 The main question that these results raised was whether this is the natural course after 2 years of continued treatment or whether these patients received inadequate dosing. The results of the HORIZON trial did provide insight into the pathophysiology of patients who were not treated in the first 2 years of the initial trial. For the crossover patients, there was little visual improvement during their treatment in HORIZON, which suggests that the window of opportunity for reversible vision loss had passed. However, these patients may have continued to lose more vision had they not received this treatment.
SAILOR. SAILOR was another trial that evaluated PRN dosing. In this trial, over 1,000 patients in each arm were randomized to either 0.3 mg or 0.5 mg ranibizumab. Patients had three monthly injections and then PRN injections as determined at follow-up visits at intervals of 3 months. The bar for retreatment for SAILOR was set fairly high: at least one line of visual acuity loss or an increase in CRT of 100 µm.
During the monthly treatment period there were improvements in visual acuity but much of this was lost in them PRN dosing period (Figure 4).6 Whether a more liberal retreatment regimen with closer follow-up would have improved the results is probable. However, whether any PRN dosing regimen can equal the results of monthly dosing is uncertain and we await the results of the CATT trial, which should help clarify this issue.
A BETTER DRUG FOR AMD?
To date, the preponderance of data suggests that to guarantee good outcomes while treating neovascular AMD, monthly injections of ranibizumab is the most rational approach. I do think, however, that a PRN treatment schedule can produce similar results if the threshold for retreatment is low and patients are evaluated monthly. A treatand- extend approach may also be an alternative and is employed by many to identify subgroups of patients who may need less frequent treatment and follow-up. Investigational approaches to improving the duration of treatment will employ higher dosages of ranibizumab, extended release of the drug, higher affinity molecules, and the use of combination therapies.
CLEAR-IT 2. VEGF-trap is one drug that might provide increased duration of effect and efficacy over ranibizumab, as it has higher affinity binding for VEGF. VEGF-trap also blocks placental growth factor, which may provide additional benefits.
The phase 3 data for VEGF-trap are still under evaluation, but the phase 2 data (CLEAR-IT 2) suggested increased durability over ranibizumab (Figure 5).7 The two phase 3 studies, VIEW 1 and VIEW 2, are currently fully enrolled and under way. In these trials, patients are randomized 1:1:1:1 to either 0.5 mg VEGF-trap every 4 weeks, 2.0 mg VEGF-trap every 4 weeks, 2.0 mg VEGF-trap every 8 weeks, or 0.5 mg ranibizumab every 4 weeks.
COMBINATION THERAPIES
If A works and B works and A and B work differently, is it reasonable to conclude that A+B will work best? This is the reasoning behind any combination therapy and is well documented in oncology and infectious disease management. Photodynamic therapy (PDT) was one of the first adjunctive therapies that we evaluated in combination, as it had already proved to be safe for patients. Additionally, PDT appears to have a different mechanism of action than ranibizumab or steroids. In theory, it selectively targets and accumulates in abnormal vessels. Activation with laser leads angio-occlusion of the choroidal neovascularization. Anti- VEGF therapy works primarily by reducing vascular leakage and closing early neovascularization. PDT offers the possibility of closing more mature larger caliber neovascularization, perhaps reducing the chance for recurrence.
MONT BLANC. There are a number of clinical trials evaluating ranibizumab on combination with PDT. The MONT BLANC randomized patients to combination therapy (ranibizumab plus standard fluence PDT) or monotherapy with ranibizumab. Patients had a fixed dosing for the first 3 months in both groups received three ranibizumab injections. One group also received standard fluence PDT and then was treated on a PRN basis during the maintenance period. There was no statistically significant difference in visual outcome between the two groups (Figure 6),8 although the combination group seemed to do slightly worse. Interestingly, there was no significant difference in the number of retreatments and the number of injections, suggesting that PDT had no effect on reducing the number of treatments with ranibizumab.
RADICAL. The RADICAL trial investigated triple therapy (steroid plus anti-VEGF plus PDT), as the preliminary work by Albert Augustin, MD, suggested that this might lead to similar results as monotherapy with anti-VEGF but with many fewer treatments. The addition of the initial intravitreal steroid at the time of the PDT therapy might reduce the initial negative effects of PDT.
RADICAL also investigated the possibility that the toxic effects of PDT could be reduced by lowering the fluence. Patients were randomized to one-quarter fluence PDT plus ranibizumab plus dexamethasone; one-half fluence PDT plus ranibizumab plus dexamethasone; one-half fluence PDT plus ranibizumab; and monotherapy with ranibizumab.
Although no statistically significant differences in visual outcomes were seen between the groups (Figure 7),9 the group that seemed to do the best was the triple therapy group receiving one-half fluence PDT group, followed by the monotherapy group. Triple therapy seemed to reduce retreatments slightly: the ranibizumab monotherapy group had the highest number of retreatments and the triple therapy group with one-half fluence PDT had the fewest.
Epimacular Brachytherapy. Another option that is being evaluated is epimacular brachytherapy, which uses a device (Vidion ANV Therapy System , Neovista, Inc.) to deliver 24 Gy of strontium 90 via a 20-gauge cannula to the retinal surface over 4 minutes after after 23- or 25-gauge pars plana vitrectomy in conjunction with anti-VEGF agents. In theory this device might permit selective treatment to the neovascularization while minimizing the collateral damage to surrounding ocular tissues.
In the nonrandomized, multicenter feasibility study that enrolled 34 patients in Mexico and Brazil predominantly classic, minimally classic or occult (with no classic) CNV, patients received a single treatment with 24 Gy beta radiation using the Vidion device and two injections of bevacizumab 1.25 mg. Group 1 received a bevacizumab injection 10 days (±4 days) prior to surgery and group 2 received an injection the time of surgery, postradiation delivery. Both groups received a second injection at 1 month postsurgery.
The mean change in visual acuity at month 18 was a gain of 6.6 letters. Seventy-three percent of patients received no additional injections, nine total additional injections were delivered by 18 months with seven patients receiving one additional injection and 1 patient receiving two.10
Whether this approach is safe and efficacious as an alternative to reduce the treatment burden of monotherapy anti-VEGF agents is being investigated in a large randomized multicenter trial in the United States, CABERNET.
RETINAL VEIN OCCLUSION
In 2009, data from five large randomized trials for RVO were released.
SCORE. The SCORE (BRVO and CRVO) trials evaluated the standard of care for branch retinal vein occlusion (BRVO) and central retinal vein occlusion (CRVO), laser photocoagulation (BRVO) and observation (CRVO), to a noncommercially available formulation of 4 mg triamcinolone acetonide (Trivaris, Allergan, Inc.). The BRVO trial found no significant benefit to triamcinolone over laser photocoagulation (Figure 8).11 In fact, patients treated with steroids had a significantly higher risk of cataract and glaucoma. The CRVO trial, however, demonstrated that patients may benefit from injections of triamcinolone over observation (Figure 9).12
Intravitreal Dexamethasone Implant. The intravitreal dexamethasone implant (Ozurdex, Allergan, Inc.) trial compared the effects of the intravitreal dexamethasone implant 0.7 mg and 0.35 mg to sham in patients with BRVO and CRVO, using an endpoint of visual acuity better than or equal to 15 letters and patients were pooled together in this study. The intravitreal dexamethasone implant is currently US Food and Drug Administration (FDA)-approved for treating macular edema secondary to RVO based on the 6-month data. The duration of action of dexamethasone in this sustained-delivery vehicle is 6 months, with a peak effect occurring between 1 to 3 months.
The percentage of patients in the 0.35 mg dexamethasone group who gained equal to or better than 15 letters at 6 months was 40%; the percentage in the 0.7 mg group was 41%; and the percentage in the sham group was 23%. The mean change in best corrected visual acuity (BCVA) for patients with BRVO at the peak effect point (2 months) was 9.8 letters in the dexamethasone groups, compared to 3.1 letters in the sham group (Figure 10A). The mean change in BCVA for patients with CRVO at 2 months was 8.7 letters in the dexamethasone groups, compared to -0.5 letters in the sham group (Figure 10B).13
BRAVO and CRUISE. In the BRAVO trial, BRVO patients were randomized 1:1:1 to 0.3 mg ranibizumab, 0.5 mg ranibizumab, or sham. Patients received monthly injections and all patients were eligible for rescue laser beginning at month 3 during the 6-month treatment period if they met the following criteria: BCVA of 20/40 or worse or mean central subfield thickness equal or greater than 250 μm. During the 6-month observation period, all patients were eligible for PRN ranibizumab (the sham group received 0.5 mg, and the 0.3 mg and 0.5 mg groups received their assigned dose) and rescue laser, if eligible.
At 6 months, the proportion of patients who gained 15 or more letters from baseline BCVA was 61.1% in the 0.5 mg ranibizumab group, 55.2% in the 0.3 mg ranibizumab group, and 28.8% in the sham group (Figure 11A). Foveal thickness was also markedly decreased in the ranibizumab groups (Figure 11B).14
The CRUISE study design was essentially the same as BRAVO, except that CRUISE CRVO patients did not receive rescue laser, as previous studies have demonstrated no role for laser photocoagulation in the treatment of CRVO macular edema. During the observation period, all subjects were eligible for PRN ranibizumab treatment (the sham group received 0.5 mg ranibizumab, and the 0.3 mg and 0.5 mg groups received their assigned dose) if they met the following criteria: BCVA of 20/40 or worse or mean central subfield thickness equal or greater than 250 µm.
At 6 months, the proportion of patients who gained 15 or more letters from baseline BCVA was 47.7% in the 0.5 mg group, 46.2% in the 0.3 mg group, and 8.5% in the sham group (Figure 12A). As with BRAVO patients, the patients in CRUISE also experienced a significant decrease in mean excess foveal thickness at 6 months (Figure 12B).15
As a result of the positive results of these studies, ranibizumab was approved for BRVO and CRVO by the FDA in July 2010.
SUMMARY
VEGF is an important pathogenic mechanism in AMD and RVO and its inhibition has been proved an effective and safe treatment for patients with these diseases. In AMD and RVO, treatments may be protracted with anti-VEGF agents or steroids. Newer agents, higher dosages, or combination therapies may offer similar outcomes with a reduced treatment burden.
These new treatments for RVO and AMD are major medical advances in the treatment of our patients. These new therapies can result in visual results that would have been unthinkable only a few years ago, and it is likely that we will look back at this past decade as one of unparalleled progress.
_1773249222.png?auto=compress,format&w=75)
