Results from Pivotal Clinical Trials Released at AAO Meeting
Specialists comment on their significance.
During Retina Subspecialty Day and the joint meeting of the American Academy of Ophthalmology and the European Society of Ophthalmology, held Oct. 22-26 in New Orleans, key results were presented from several posterior-segment clinical trials:
► groups B, H, and N of the Submacular Surgery Trials (SST)
► initial results from the C-01-99 trial of anecortave acetate 15 mg depot vs photodynamic therapy (PDT) with verteporfin in the treatment of predominantly classic lesions in wet age-related macular degeneration (AMD)
► 2-year data on pegaptanib sodium for treatment of subfoveal choroidal neovascularization (CNV) secondary to AMD. (Pegaptanib sodium was approved by the FDA on Dec. 17. For information on the approval, see page 9.)
► preliminary 2-year results from TTT4CNV, which was designed to determine whether transpupillary thermotherapy (TTT) laser treatment can reduce the risk of vision loss for patients with occult wet AMD.
Retinal Physician asked specialists, some involved in the trials and some not, to share their initial thoughts on the significance of the results. Their responses are presented here.
Neil M. Bressler, MD, chairman of the SST Research Group: Overall, eyes with relatively large predominantly hemorrhagic subfoveal choroidal neovascular lesions in study participants with AMD that were assigned to surgical removal of the lesion and associated blood (n = 168) did not benefit with respect to stabilization or improvement of visual acuity at 2 years compared with similar eyes (n = 168) that were assigned to observation: 44% of surgery eyes and 41% of observed eyes had improved or stable visual acuity compared with baseline. However, for the secondary outcome of severe visual acuity loss at 2 years, 21% of surgery eyes vs. 36% of observation eyes had severe visual acuity loss (loss of 6 or more lines) from baseline.
Rhegmatogenous retinal detachment developed in 27 surgery eyes (16%) compared with 3 observed eyes (2%). 44% of surgery eyes that had not had cataract surgery prior to SST had cataract surgery by 2 years after surgery.
Submacular surgery should be considered as a treatment option for eyes similar to those enrolled in the SST Group B Trial that do not have very poor vision or do not have very large lesions (since retinal detachment was more likely). Surgery for relatively large predominantly hemorrhagic choroidal neovascular lesions in patients with AMD reduced the risk of severe visual acuity loss but did not alter the chance of stable or improved visual acuity.
Stephen R. Russell, MD, Director, Vitreoretinal Service, Clinical Director, Center for Macular Degeneration, The University of Iowa, Carver School of Medicine: In a letter to the editor (Ophthalmology) more than a decade ago, two colleagues and I suggested that the available pilot studies of submacular surgery for subfoveal CNV in AMD lacked appropriate control groups and resulted in an "absence of demonstrated surgical effectiveness." The recent Group N study results support our earlier analyses and the beliefs of many retinal specialists regarding the advisability of subfoveal surgery. Given the marginal clinical visual outcomes of the SST, and the visually and anatomically significant side-effects and complications of surgery, the SST results should warn retinal surgeons concerning the procedure. The disappointing SST results will motivate the development of alternative pharmacological and biological approaches to managing CNV. By addressing the risks and benefits of subfoveal surgery in a systematic manner, the SST investigators have extended our understanding of this modality and its application to this recalcitrant disorder.
Neil M. Bressler, MD, chairman of the SST Research Group: Overall, submacular surgery to remove subfoveal CNV that was not associated with large amounts of hemorrhage was not effective in improving or maintaining visual acuity compared with observation among the 454 Group N study participants.
Rhegmatogenous retinal detachment occurred in 12 surgery eyes (5%) and 1 observation eye (0.4%). 39% of eyes assigned to submacular surgery that had not undergone cataract surgery by entry into the SST Group N Trial had cataract surgery by the 24-month examination compared to 5% of eyes in the observation arm.
These findings confirm clinical impressions among many retinal surgeons that submacular surgery is not helpful for AMD lesions of the type enrolled, even though some benefits of submacular surgery, as performed in the SST, were noted in the SST Group B and Group H trials. Despite better quality of life outcomes with surgery than with observation, submacular surgery as performed in the SST is not recommended for individuals similar to those in Group N because no visual acuity benefit was observed.
Barbara S. Hawkins, PhD, SST Research Group: Overall, eyes with idiopathic CNV or CNV secondary to ocular histoplasmosis syndrome that were assigned to surgical removal (n = 112) did not benefit with respect to stabilization or improvement of visual acuity at 2 years after enrollment compared to similar eyes that were assigned to observation (no treatment): 55% of surgery eyes and 46% of observed eyes had improved or stable visual acuity compared to baseline. However, in the prespecified subgroup of eyes with baseline visual acuity worse than 20/100 (Snellen equivalent), 76% of 41 surgery eyes vs. 50% of 40 observation eyes had visual acuity that was better than or no more than 1 line (7 letters) worse than at baseline.
Rhegmatogenous retinal detachment developed in 4 surgery eyes and no observed eyes. Recurrent CNV developed in 58% of surgery eyes, but active CNV was present in a similar proportion of study eyes in both treatment arms at 2 years and later. Cataract surgery was reported for 24% of surgery eyes but no observed eyes.
Submacular surgery should be considered as a treatment option, at least for eyes with visual acuity worse than 20/100 that are similar to those enrolled in the SST Group H Trial. Many ophthalmologists may try photodynamic therapy first and recommend surgery if CNV is present at the first post-PDT examination. However, eyes already treated with PDT were not eligible for the SST Group H Trial and this treatment sequence (PDT first, then surgery) has not been evaluated in a controlled study of any design.
Submacular surgery is the only treatment proven to be effective for treating any group of eyes with subfoveal CNV due to ocular histoplasmosis or with subfoveal idiopathic CNV.
Donald J. D'Amico, MD, Professor of Ophthalmology, Harvard Medical School, Associate Chief of Ophthalmology, Massachusetts Eye and Ear Infirmary: Anecortave acetate is an angiostatic cortisene that has been chemically modified to eliminate glucocorticoid effects, but it exhibits potent activity against neovascularization in many experimental models. In patients, it is delivered by posterior juxtascleral deposition, utilizing a unique curved cannula that allows placement outside the globe directly under the macula, and it is administered at 6-month intervals. A preliminary trial (C-98-03) in patients with neovascular AMD had demonstrated efficacy up to 2 years for preservation of visual acuity, prevention of severe visual loss, and control of neovascularization utilizing a 15 mg dose, but a 30 mg dose was not effective.
Results of the C-01-99 trial directly comparing anecortave acetate and PDT with verteporfin for predominantly classic lesions in AMD are the latest to be released. This trial was specifically designed to establish noninferiority of anecortave acetate, and, in sharp contrast with previous results, the study failed to demonstrate noninferiority for these lesions. Forty-five percent of patients treated with anecortave acetate experienced less than a 3-line visual acuity loss at 1 year, compared with 49% of verteporfin-treated patients, and precisely defined statistical testing to demonstrate non-inferiority of anecortave acetate did not result in significance.
Several factors emerged in data analysis that offered a possible explanation of these results. During some injections, reflux of the drug from the conjunctival incision was noted, and the presence of reflux was carefully documented prospectively as part of the trial. Also, the interval to second administration of drug, intended to be at 6 months, varied somewhat in the actual performance of the trial, and an exploratory analysis that controlled for both drug reflux and prolonged interval to second drug administration yielded statistical significance for anecortave acetate. Both of these additional factors--drug reflux and treatment interval--are potentially controllable, and a pharmacokinetic study has been initiated to demonstrate the efficacy of reflux control with a specially-designed counter pressure device.
An additional factor in the explanation of the negative trial results may have involved the nature of the patients and lesions enrolled. While the C-98-03 study had an average lesion size of 8.0 mm and a lesion duration before enrollment of 237 days, the recent C-01-99 trial had an average lesion size of 3.7 mm and a duration of 37 days, suggesting a greater inclusion of smaller, more aggressive lesions with a poorer prognosis. This is consistent with the lower demonstrated efficacy of the photodynamic therapy comparison group in the anecortave acetate pivotal trial compared with initial vertepporfin studies, and indicates that the trial landscape for AMD patient enrollment itself may be changing as a result of new therapies, patient/physician awareness, and many other factors.
Given the broad and consistent efficacy of anecortave acetate in previous models and studies, its novel mechanism of action, and its convenient 6-monthly extraocular administration, strong interest continues in this treatment for AMD and other retinal diseases, either alone or in combination with other drugs and modalities. Careful study of every promising agent and therapy is vital to enhance the visual results and treatment options for patients burdened with neovascular AMD.
Alcon, Inc. has submitted the third and final reviewable unit of its New Drug Application (NDA) for anecortave acetate 15 mg for depot suspension to the FDA.
Harry W. Flynn, Jr., MD, Professor of Ophthalmology, Bascom Palmer Eye Institute, University of Miami School of Medicine: In the trial comparing anecortave acetate 15 mg depot vs. verteporfin PDT in the treatment of wet AMD, only patients with predominantly classic CNV were treated. The percentage of patients who maintained vision (defined as less than a 3-line loss in logmar visual acuity) was 45% when treated with anecortave acetate and 49% using PDT (following study protocol). These outcomes indicate that the 2 therapies are not statistically different from each other, but the anecortave acetate treatment arm did not meet the primary noninferiority endpoint set up for the clinical study. It remains to be seen whether the effect will be different for other forms of CNV.
In looking back over the study, there may be controllable factors that adversely influenced the anecortave acetate results. These include potential drug reflux from the injection site and delayed treatment interval. The first issue of drug reflux occurs when some of the medication leaks out from its juxtascleral position through the conjunctival incision. The second issue refers to the ideal timeframe (within 6 months) during which retreatment should occur. By modifying these variables, significantly improved outcomes may be possible, but it was these variables 1) ease of administration and 2) 6-month dose interval that made anecortave acetate such an attractive option.
The recent "head-to-head" study demonstrated that there were no clinically relevant safety issues from anecortave acetate during the course of the study. Both the drug itself and the posterior juxtascleral delivery appear to be safe and well-tolerated by patients.
Also keep in mind that PDT with verteporfin has the risk of severe visual decrease in up to 4% of patients (not present with juxtascleral anecortave acetate).
Where does anecortave acetate fit in to the scheme of treatments for AMD? This question is yet to be decided. "An ounce of prevention is worth a pound of cure," the old saying goes. In the management of AMD, we are all looking for this ounce of prevention. It may be that anecortave acetate is less effective in treatment of active CNV but may be more effective in prophylaxis against progression in high risk fellow eyes. Anecortave acetate, administered as a juxtascleral depot, has the potential to offer prophylactic treatment to high-risk fellow eyes when 1 eye has had severe visual loss from advanced AMD. In spite of the recently reported data, anecortave acetate remains a promising therapy for AMD.
Abdhish R. Bhavsar, MD, Director of Clinical Research, Retina Center, and Vice Chairman of the Phillips Eye Institute, Minneapolis: Pegaptanib sodium is a pegylated aptamer, which binds selectively to VEGF165. In the combined 12-month data from the 2 phase 3 pegaptanib sodium clinical trials. 1053 patients were randomized to usual care with sham injection or to treatment with pegaptanib. The treatment groups received intravitreal injections of pegaptanib every 6 weeks.There were 3 treatment groups: 0.3 mg (n=294), 1.0 mg (n=300), 3.0 mg (n=296), and a sham group (n=296). Verteporfin PDT was permitted for a maximum of 1 treatment prior to entry into the study and could be performed if clinically indicated during the course of the study.
Efficacy at 1 year was < 3 line loss of visual acuity in 70% of patients at the 0.3 mg dose, 71% at the 1.0 mg dose and 65% at the 3.0 mg dose vs 55% in the usual care/sham group. At 2 years with the 0.3 mg dose, 59% of pegaptanib treated patients experienced < 3 line loss of visual acuity vs 45% in the usual care/sham group.
It also appears that there is a benefit in favor of pegaptanib treatment for 2 years vs. 1 year.
The rate of endophthalmitis was 0.18 % prior to a protocol change to reinforce aseptic technique and 0.03% after the protocol change.
No serious systemic concerns were noted.
I believe that pegaptanib sodium will add to our armamentarium for treating exudative AMD. Now that pegaptanib has been approved by the FDA, I predict that we will attempt monotherapy in some patients, but (if allowed by CMS) we will also attempt combination treatment with verteporfin PDT in other patients, in an attempt to improve outcomes. Unfortunately, we do not yet have any solid data from a well-designed clinical trial to know what to expect from combination treatment using PDT and pegaptanib. We also do not know what an appropriate dosing/interval regimen would be. A number of clinical trials that may help to answer some of these questions are underway or in the planning stages.
Michael J. Cooney, MD, Assistant Professor of Ophthalmology and Director, Vitreoretinal Service, Duke University Eye Center: The 2-year study of pegaptanib sodium had two objectives: to investigate if 2 years of treatment with pegaptanib sodium are superior to 2 years of usual care, and to investigate if 2 years of pegaptanib sodium injection are superior to 1 year. The 2-year pegaptanib sodium data continue to demonstrate that pegaptanib sodium is superior to standard of care in slowing the rate of vision loss in patients with neovascular AMD. The data also appear to demonstrate that 2 years of pegaptanib sodium therapy is superior to 1 year of therapy.
The study showed high compliance among the 1053 participating patients over the 2 years, with the mean number of injections being 15.6. The treatment effect that was seen at year 1 continued at year 2 with a 45% pegaptanib sodium treatment benefit (p< 0.01) over usual care. There appeared to be a marginal benefit (p<0.05) of continuing pegaptanib sodium injections for the full 2 years of the study rather than discontinuing them after year 1. At 2 years, patients receiving pegaptanib sodium had a more favorable visual acuity distribution than patients receiving usual care. There were no changes in the systemic or ocular safety profiles at year 2 compared to year 1.
All in all, the trials continue to demonstrate that the efficacy of pegaptanib sodium is comparable to that of PDT with verteporfin. In clinical practice, the main advantage of pegaptanib sodium might be for occult and minimally classic lesions that are greater than 4 MPS disc areas in size where PDT with verteporfin has not been proven to be efficacious. I believe that retinal physicians who use pegaptanib sodium will use fluorescein angiography and OCT in determining the need for the next injection rather than injecting every 6 weeks as was done in the clinical trials.
Preliminary 2-year results from the TTT4CNV clinical trial showed that patients treated with the TTT laser protocol didn't derive any significant benefit compared with patients who were given sham treatment.
Elias Reichel, MD, associate professor of ophthalmology at the New England Eye Center, Tufts University School of Medicine, who chaired the 303-patient study, reported that at 2 years, 47% of eyes treated with TTT avoided modest or severe vision loss, compared with 43% of eyes that received sham treatment, which was not considered statistically significant.
It was also reported that 11% of patients who were treated with TTT showed vision improvement of 2 lines or greater after 1 year, compared with 3% of patients who showed similar improvement after the sham treatment. This secondary outcome is considered statistically significant.
"Further analysis of this subgroup of patients with improved vision may yield information regarding those patients who respond to TTT," said Dr. Reichel. "Specifically, this subgroup of patients may have certain characteristics that result in a high likelihood of improvement or stabilization over time."
Fareed Ali, MD, Director of Clinical Research, Canadian Centre for Advanced Eye Therapeutics, Mississauga, Ontario, Canada: Although I perform TTT treatments, I had no involvement in the trial and first heard the results at the meeting. My first concern is that for such a major trial, the trial executive committee had only about 2 weeks to analyze the data, and the principal investigator had only 7 minutes to present the results.The bottom line results were negative, but some important points were missed due to the short time available to present. I attended a subsequent talk of more than 20 minutes by the principal investigator that put the results in a much different light. One major possible problem with the trial is that 45% of the subjects did not in fact meet the entry criteria. This was due to a discrepancy between the angiogram interpretation of the reading center and the treating MD. So 45% of the subjects had a CNV that was too large. All investigators used the same treatment spot size, which means 45% of the subjects had a CNV that was only partially treated. We know from MPS and PDT trials that partially treating a CNV can be worse than not treating it.
The data will be reanalyzed with this in mind and hopefully we'll see something more positive. Of course, the data as a whole will be looked at with a focus on subgroups that did well, as was the case for other trials in the past, most notably those for PDT. One would hope that the retinal community is as patient with the TTT trial as we were with the PDT trials, which also had mediocre initial results until reanalysis showed benefits in certain subgroups.
Along the same lines, I have heard little mention of another very significant trial result that was presented: the 1-year VIO data for PDT with verteporfin. This trial was supposed to confirm the efficacy of PDT for minimally classic, small CNV. Meta-analysis of data from an earlier PDT trial showed these groups did well, which was the basis for expanded CMS coverage for PDT. Now the full prospective trial shows no benefit at 1 year. What happens to CMS coverage (and standard of care) now?
I think the bottom line from the meeting is that with no really positive results from anecortave acetate, TTT, or VIO/PDT, we are still without a proven treatment for nonclassic CNV. I personally am hopeful that the TTT4CNV committee will expedite reanalysis of the data excluding the 45% who should have not been in the trial, and release this data very soon which, if positive, will finally validate a rigorously proven therapy for nonclassic CNV.
Michael Ip, MD, Associate Professor, Department of Ophthalmology and Visual Sciences, University of Wisconsin: I was hoping for a positive trial because TTT is a low-cost and simple treatment to administer. However, given the fact that the primary outcome of the trial did not achieve statistical significance, it is unlikely that this treatment will be used widely as a first-line or sole treatment for CNV due to AMD.
Much credit should go to the group that organized the trial as I think that even the negative results of this trial are important. They are important because the treatment was used quite often by clinicians in the past without data from a well-designed trial. Future trials of TTT could evaluate different subgroups of patients with exudative AMD or use TTT as an adjunct to another therapy.