Considering the Options for RVO Therapy

JOHN A. WELLS, III, MD ∙ JOHN J. WROBLEWSKI, MD

The era of pharmacotherapy in retinal diseases that produced such enormous improvements in our treatment of AMD is now impacting retinal vascular diseases as well. Central retinal vein occlusion has long been resistant to therapy, and a meta-analysis of published treatments for CRVO found that no effective therapy existed as of 2007.

Clinicians now find themselves with at least four viable options. Randomized clinical trials of two VEGF inhibitors (pegaptanib sodium and ranibizumab) and two corticosteroids (triamcinolone acetonide and dexamethasone) have been reported to have positive effects on vision and macular edema in CRVO and BRVO. This paper will review these results and discuss their implications for clinical practice.

PEGAPTANIB SODIUM

The publication of the results of a phase 2 randomized trial of pegaptanib sodium for treatment of macular edema secondary to CRVO in April 2009 was the first randomized trial of any treatment for CRVO to show an improvement in vision compared to sham controls.

Ninety-nine patients with CRVO of less than six months duration, vision from 20/40 to 20/320, and OCT edema > 250 microns were randomized to receive either pegaptanib 1 mg, 0.3 mg, or sham injections every six weeks from baseline to week 24 for a total of five injections. Patients with signs of ischemic CRVO, such as a brisk afferent pupillary defect, were excluded. The primary outcome was 15 or more letters gained at week 30. Secondary outcomes were mean letters gained, proportion of subjects losing 15 or more letters of vision, proportion of subjects with vision better than or equal to 20/200, and reduction in macular edema on OCT.

At week 30, the percentage of patients gaining 15 or more letters was 39%, 36% and 28% in the 1 mg, 0.3 mg and sham groups, respectively. There was no statistically significant difference between the treatment groups and the sham group. However, the mean number of letters gained was +9.9, +7.1 and −3.2 for the 1mg, 0.3 mg and sham groups, respectively. This equals a +13.1 and +10.3 letter improvement over the sham group for the 1 and 0.3 mg treatment groups, which was statistically significant for the 1 mg group only. Less than 10% (6% and 9% for the 1 mg and 0.3 mg groups, respectively) versus 31% of sham eyes lost 15 or more letters, which was statistically significant. There was also a significant difference in proportion of patients with 20/200 or better vision at week 30 favoring the treatment groups: 90% of treated eyes versus 63% of the sham group.

OCT-measured CME was reduced by −243 μm, −179 μm, and −148 μm in the 1 mg, 0.3 mg and sham groups, respectively; however, there was no statistically significant difference between the groups. Finally, two of 66 eyes (3%) receiving pegaptanib developed ocular neovascularization versus three of 32 eyes (9%) receiving sham treatment.

While the study failed to meet its primary endpoint, there were significant gains in mean VA compared to sham. The fact that over one quarter of the sham group gained three or more lines and that the reduction in macular edema was similar in all groups illustrates the need for randomized trials when examining interventions for CRVO, particularly when entry criteria are established to exclude eyes likely to have severe ischemia and a poor visual outcome. The natural history of non-ischemic CRVO is that about a third of eyes will recover without treatment, and the trial results reflect that. However, treated eyes did gain more vision than untreated eyes despite similar reductions in macular edema, suggesting that other mechanisms for vision loss may be a factor in this condition. Treated eyes were less likely to lose three lines and more likely to have better than 20/200 vision.

One other factor in the trial design may have contributed to the vision results. Measurement of vision and OCT edema was done at week 30, six weeks after the last injection when the intravitreal pegaptanib levels would be at a therapeutic trough. Measurements taken at 1-3 weeks of injection, when levels would be at a therapeutic peak, may have given improved results.

RANIBIZUMAB

The CRUISE study of ranibizumab for CRVO randomized 392 patients to 0.3 or 0.5 mg ranibizumab or sham injections monthly for six months. The entry criteria included vision of 20/40-20/320, OCT edema > 250 μm and duration of CRVO less than 12 months. The primary outcome was mean change in vision at six months. Secondary outcomes were proportions of eyes gaining or losing 15 or more letters at six months, and reduction in macular edema.

Statistically significant gains in mean vision were realized in both treatment groups relative to sham. Mean letters gained were +14.9, +12.7,and +0.8 letters for the 0.5 mg, 0.3 mg and sham groups, respectively. This equals a +14.1 and +11.9 letter improvement over sham for the 0.5 mg and 0.3 mg groups. Additionally, 48% and 46% of eyes in the 0.5 and 0.3 mg groups gained 15 or more letters at six months versus 17% of sham-treated eyes, again statistically significant. Reductions in edema were also significant, with a mean reduction of −452 μm, −435 μm and −167 μm in the 0.5 mg, 0.3 mg and sham groups, respectively.

Three of 262 treated eyes (1%) developed ocular neovascularization vs 12 of 130 (9.2%) of sham treated eyes.

These results are similar to the pegaptanib results in mean letters gained relative to sham, but the absolute gains were greater and the reductions in macular edema were more substantial than in the pegaptanib study. The placebo group also fared worse in the proportion of eyes gaining 15 letters at six months compared to the placebo group in the pegaptanib trial, suggesting that there was a selection bias toward more severe cases in the CRUISE trial than in the pegaptanib trial and perhaps the potential for larger improvements. However, mean baseline vision was similar at about 48 ETDRS letters in all groups in both trials, so the difference in the sham results may reflect the larger number of patients in the CRUISE trial and shows the importance of a larger phase 3 trial to detect statistically significant differences when comparing treatment to a condition that frequently has a good natural history.

TRIAMCINOLONE

The SCORE Study Report 5 details a randomized trial comparing 1 mg and 4 mg intravitreal triamcinolone injections every four months to observation for one year in 271 eyes. Entry criteria included vision of 20/40 to 20/400, OCT edema >250 μm, and no prior history of glaucoma or steroid related IOP elevation. No restriction on duration of CRVO was defined. Eyes received treatment every four months based on pre-specified retreatment criteria, as opposed to the pegaptanib and ranibizumab trials where treatment was given on a regular dosing schedule.

Specifically, retreatment after the initial injection was deferred if edema was < 225 μm, the vision was equal to or better than 20/25, there was a significant IOP rise that contraindicated further treatment in the opinion of the investigator, or additional treatment was judged to be futile in the opinion of the investigator. The primary endpoint was the proportion of eyes gaining 15 or more letters at 12 months. Secondary outcomes included mean change in vision and mean change in OCT measured edema at 12 months.

The treatment groups received a mean of 2.0 and 2.2 treatments in the 4 mg and 1 mg arms. At 12 months, 25.6%, 26.5% and 6.8% of eyes in the 4 mg, 1 mg and observation groups gained 15 or more letters of vision, which was statistically significant in favor of the treatment arms. However, the mean change in vision was −1.2 letters in both treatment groups and −12.1 letters in the observation group, a +10.9 letter gain for both treatment doses relative to observation. Reductions in edema were −261 μm, −196 μm and −277 μm in the 4 mg, 1 mg and observation groups, respectively, which was not significantly different.

Twenty-one of 183 (11.5%) of treated eyes developed signs of ocular neovascularization versus 10 of 88 (11.3%) of observation eyes. This suggests, but does not prove, that steroids are less effective than pegaptanib and ranibizumab in preventing neovascular complications of CRVO.

As opposed to treatment with anti-VEGF agents, there were significant side effects of therapy with triamcinolone. Thirty-five percent of eyes in the 4 mg group and 20% of eyes in the 1 mg group required IOP-lowering medication during the 12-month study versus 8% of the observation group. Thirty-three percent and 26% of the 4 mg and 1 mg group eyes had new lens opacity or progression of baseline opacity at 12 months versus 18% of the observation group. Interestingly, no cases of non-infectious endophthalmitis occurred in the treatment groups.

Based on the similar efficacy and lower side effect profile of the 1 mg dose vs the 4 mg dose, the study recommended that 1 mg of triamcinolone be considered to treat vision loss secondary to macular edema from CRVO.

DEXAMETHASONE INTRAVITREAL IMPLANT

In 2009, the FDA approved a dexamethasone intravitreal implant to treat macular edema due to CRVO and BRVO. Ozurdex (Allergan) is a dexamethasone 0.7 mg sustained-release intravitreal solid polymer drug delivery system that has been studied in 2 clinical trials, one including patients with CRVO and one in patients with BRVO. In the CRVO study, 450 patients were randomized to either the 0.7 mg implant (n=226) or sham (n=224) injection. Treated eyes had a statistically significant greater percentage of eyes gaining 15 or more letters of vision through three months of follow-up, but by six months the difference between the treated eyes and the sham eyes was no longer significant.

Similar to the SCORE results, a maximum of 30% of treated eyes showed this 15-letter gain at month two, and this had declined to 21% by month three. Sham-treated eyes showed proportions for three-line gainers more similar to the SCORE and CRUISE data than the pegaptanib trial, with 14% of sham eyes improving at three months and a maximum of 17% gaining at six months. Twenty-five percent of treated eyes vs 1% of sham eyes developed significant IOP elevation by six months, but cataract formation was relatively low at 4% and 1% in treated and control eyes, respectively.

DISCUSSION

Each of the aforementioned trials demonstrated improvements in visual acuity and central retinal thickness. However, anti-VEGF treatment with pegaptanib and ranibizumab seem to yield greater improvements in vision relative to sham injections than steroids, both in mean letters gained and percentage of patients gaining three or more lines of vision, and were not associated with side effects such as glaucoma and progression of cataract as in the steroid trials.

Disadvantages of anti-VEGF treatment include more frequent administration and cost. Pegaptanib and ranibiz umab require much more frequent administration than triamcinolone and Ozurdex. Pegaptanib ($995 per dose) and ranibizumab ($1950 per dose) are also significantly more expensive than the triamcinolone used in clinical practice. While the triamcinolone preparation used in the SCORE study is not yet commercially available, a similar preservative-free triamcinolone (Triescence, Alcon) costs $130 per dose while Ozurdex costs $1295 per dose. Even so, the much less frequent dosing of both corticosteroid preparations significantly reduces the cost by minimizing the costs associated with multiple injections and the associated inconvenience to the patient, the burden on the physician's office and the small but real risks of complications associated with injections.

Finally, all of these trials present relatively short-term results, with six-month vision data for the pegaptanib, ranibizumab and Ozurdex trials, and 12 month vision data for the SCORE study. The long-term effect of chronic VEGF inhibition in RVO is unknown, but concerns about an effect on neuronal health exist. Similarly, with longer follow-up in the steroid studies, the rates of glaucoma and cataract formation are likely to increase. Additionally, it is unknown whether treatment will be required indefinitely to maintain the vision gains seen in the first 6-12 months of therapy, as is so often the case with treatment of wet AMD. The development of a sustained delivery anti-VEGF device or formulation would seem to offer the advantages of a low side effect profile and less frequent injections.

In summary, these trial results are the most significant advance in our treatment of CRVO since the proof of effectiveness of PRP to prevent the neovascular complications of CRVO almost 20 years ago. While current reimbursement policies likely preclude payment for pegaptanib and ranibizumab for CRVO, it is likely that clinicians will continue to use bevacizumab as a low cost alternative. However, clinicians can feel comfortable in recommending triamcinolone and the dexamethasone implant for CRVO given the available scientific evidence supporting their use. RP