ARVO 2013: Highlights From Eight Key Studies
Anti-VEGF under-treatment among surprising findings.
BY JERRY HELZNER, SENIOR EDITOR
As is the case every year, ARVO 2013 saw numerous presentations covering every area of the retina subspecialty. In reviewing a large number of these study abstracts, Retinal Physician editors were able to select eight key studies whose findings should interest — and perhaps surprise — our readers. Following are highlights from these important studies whose findings will be sure to resonate throughout the retina community.
1. ANTI-VEGF UNDER-TREATMENT
A major, large-scale study of half a million Medicare patients who received anti-VEGF injections over a five-year period suggested that the frequency of treatment for this huge cohort has been far less than the frequency of treatment in pivotal clinical trials for these drugs.
In a study led by Eleonora M. Lad, MD, of the Duke University Eye Center, researchers reviewed claims from more than 501,000 Medicare beneficiaries diagnosed with wet AMD from 2005 to 2010. This cohort constituted 100% of all claims made under the CPT code 67028 (intravitreal injection) with the supporting diagnostic code of neovascular AMD. Identification of the specific medication the patient received was determined by using procedure codes and payment amounts.
The purpose of the review was to identify treatment patterns for anti-VEGF therapy for one to four years following initial treatment. This review focused on the rate and frequency of subsequent treatment.
Researchers found that the mean number of injections for this entire cohort following the initial injection was 4.3 in the first year of treatment. However, the mean number of anti-VEGF injections surprisingly dropped to 1.9 in the second year, 1.6 in the third year and 1.1 in the fourth year. Among patients who received seven or more injections in the first year, only 29% received comparable high-frequency treatment in year two and 14% received no injections at all in year two. Of the patients who received only one to three injections in year one, 76% received no injections at all in year two and 18% continued to received low-frequency anti-VEGF treatment in year two.
The researchers concluded no evidence indicated the frequency of anti-VEGF injections per patient has increased since the therapy was introduced in 2005, and the mean frequency of anti-VEGF injections for wet AMD remained relatively stable through the years 2005 to 2009. However, the researchers also noted the frequency of anti-VEGF injections in the overall wet AMD population in Medicare is significantly less than in the CATT and HORIZON studies, suggesting general under-treatment of patients with neovascular AMD.
1. Lad EM, Hammill BG, Qualls LG, Fang W, Cousins SW, Curtis LH. Treatment patterns in neovascular age-related macular degeneration between 2005-2010. Invest Ophthalmol Vis Sci. 2013;54:ARVO E-Abstract 3819.
2. EYEDROP FOR CNV FAILS TO SHOW EFFICACY
Hope that the topical ocular pazopanib (Votrient, Glaxo SmithKline, Brentford, UK) could provide a more patient-friendly eyedrop treatment for choroidal neovascularization (CNV) associated with AMD received a setback when researchers led by Rishi Singh, MD, of the Cole Eye Institute, Cleveland Clinic, reported at ARVO on the results of a 19-patient, 12-week study.
The researchers set out to determine the potential of pazopanib, a tyrosine kinase inhibitor, to reduce retinal edema and improve visual acuity in patients with previously untreated CNV secondary to AMD. In a multicountry, open-label phase 2 study, the 19 patients were treated with pazopanib eyedrops in one eye four times daily for 12 weeks. The patients were regularly examined throughout the study, with primary assessments being changes from baseline in central retinal thickness and best-corrected visual acuity (BCVA) following four weeks of daily treatment. Safety was also assessed over the full 12 weeks of the study.
At week four, the researchers found no meaningful changes from baseline in either mean central retinal thickness or mean BCVA. Similarly, they found no meaningful changes in any of the additional parameters measured by OCT, or in CNV or total lesion size as measured by fluorescein angiography. Eight of the 19 subjects experienced nine ocular adverse events in the study eye; one event was severe. Nine patients reported non-ocular adverse events; one was severe. Nine patients received rescue medication after they met criteria for dropping out of the study.
The researchers concluded that the pazopanib eyedrops did not appear to demonstrate efficacy in decreasing central retinal thickness or improving BCVA. No meaningful change from baseline was observed in CNV size, total lesion size or retinal morphology. The eyedrops were generally safe and well tolerated when instilled four times daily over 12 weeks.
Singh R, Wurzelmann JI, Ye L, et al. Topical pazopanib for the treatment of previously untreated choroidal neovascularization due to age-related macular degeneration. Invest Ophthalmol Vis Sci. 2013;54:ARVO E-Abstract 2178.
3. INITIAL WET AMD DATA: INTEGRIN PEPTIDE THERAPY
Integrin peptide therapy (Allegro Ophthalmics) for DME and wet AMD has created great interest because it is designed to shut off VEGF production at its source. After a highly promising phase 1 study of ALG-1001 integrin peptide monotherapy in 15 patients with DME, a phase 1b/2a trial using integrin peptide monotherapy for patients with wet AMD has produced very early, but encouraging data. The data was presented at ARVO by a team of researchers led by Peter K. Kaiser, MD, of the Cole Eye Institute, Cleveland Clinic. Dr. Kaiser also serves as editor-in-chief of Retinal Physician.
To date, 15 patients with wet AMD have been enrolled in this ongoing open-label safety study. A combination of treatment-naïve patients and patients previously treated with anti-VEGF therapy were enrolled. No significant adverse events have been reported to date.
The researchers reported that “while this is an ongoing study, there are already clear improvements in BCVA and macular anatomy (as measured) by OCT in this monotherapy study in approximately 40% of the study subjects, with clinical benefits lasting 60 days off treatment after three monthly injection loading doses.”
The researchers found consistency in the lack of toxicity across all study metrics at doses up to 4 mg. In addition, despite the small study size, “clinically relevant indicators of efficacy are apparent, with improvements in BCVA alongside anatomic improvements in OCT-measured central macular thickness that persist at least 60 days past the last intravitreal treatment.” Further studies are planned.
Kaiser PK, Boyer DS, Campochiaro PA, et al. Integrin peptide therapy: the first wet AMD experience. Invest Ophthalmol Vis Sci. 2013;54:ARVO E-Abstract 2177.
4. TELEMEDICINE EFFECTIVE FOR DR SCREENING
Primary care providers’ offices equipped to screen for diabetic retinopathy (DR) and using telemedicine may be the most effective method of providing such screening, especially in communities with a large percentage of minorities, including native Americans.
This was the finding of a research team led by Steven Mansberger, MD, MPH, of the Devers Eye Institute in Portland, Ore. The researchers compared the percentage of diabetic patients screened for DR in community health clinics with a non-mydriatic camera to the percentage of patients screened by a traditional eye care provider. In the first year, 91.9% of the 296 diabetic patients seen in community health clinics were screened, as opposed to 43.9% of the 271 diabetic patients who were enrolled to see traditional eye care providers.
In the second year, the primary care group had 53% of patients screened vs. 33.2% of the patients enrolled with the eye-care providers. After two years, all participants were offered telemedicine screening and the gap between the two groups soon closed.
Of all the patients screened for DR in the first year, 8.3% were referred to an ophthalmologist for treatment using the criterion of “moderate diabetic retinopathy or worse.” In the second year, the percentage of those screened and referred for treatment dropped to 6.2 %. In the final three years of the study, the percentage referred for treatment were 3.9% in year three, 6.4% in year four and 5.2% in year five.
The researchers concluded that telemedicine in the primary medical clinic setting using non-mydriatic cameras increased the proportion of diabetic participants who received DR screening when compared to traditional surveillance. They also concluded that most patients who were screened did not require referral to an ophthalmologist for treatment. The researchers noted that the number of patients screened did drop in the later years of the study when telemedicine was available to both groups. They said this could indicate that other barriers to diabetic screening exist even when telemedicine is available.
Mansberger SL, Gardiner SK, Demirel S, Becker T. Long-term comparative effectiveness of telemedicine to detect diabetic retinopathy. Invest Ophthalmol Vis Sci. 2013;54:ARVO E-Abstract 1549.
5. TIMING OF VITRECTOMY FOR RETAINED LENS FRAGMENTS
When confronted with retained lens fragments during cataract surgery, traditional advice has been for the surgeon to refer the patient to a retina specialist for pars plana vitrectomy as quickly as possible. However, a team of researchers from Bascom Palmer Eye Institute in Miami led by Yasha S. Modi, MD, conducted a retrospective consecutive case series over a 22-year period from 1990 through 2011 and arrived at a somewhat different conclusion.
The team looked at 569 eyes of 568 patients with a median age of 74 years and a median follow-up of eight months. Of the overall cohort, 117 eyes (21%) received same-day vitrectomy, 131 eyes (23%) received vitrectomy within one week, and 321 eyes (56%) received delayed vitrectomy with a median delay of 22 days.
The researchers found no statistically significant differences in visual acuity outcomes between groups at three months, six months, and at the last examination following vitrectomy. At the last examination, 61%, 63% and 59% of patients receiving same-day, one week and greater than one week vitrectomy achieved VA of approximately 20/40, while 16%, 15% and 21% had VA of approximately 20/200, respectively. The researchers found no significant differences between the groups in the incidence of cystoid macular edema, retinal detachment, IOP, vitreous hemorrhage or macular edema.
The most common cause of reduced VA was cystoid macular edema, followed by retinal detachment, pre-existing ocular disease, AMD and corneal edema. When the researchers excluded individuals with pre-existing disease, advanced AMD, prior retinal detachment and advanced glaucoma, they found no signiicant diferences in VA and complication rates between the three groups.
The researchers concluded that patients undergoing same-day, one-week and greater than one week vitrectomy for retained lens fragments have similar outcomes and complication rates. They noted that same-day vitrectomy, while logistically attractive, was not associated with better outcomes.
Modi YS, Epstein AE, Smiddy WE, Feuer WJ, Flynn HW Jr. Retained lens fragments after cataract surgery: outcomes of same-day versus delayed pars plana vitrectomy. Invest Ophthalmol Vis Sci. 2013;54:ARVO E-Abstract 5768.
6. AFLIBERCEPT AS RESCUE THERAPY
Two ARVO studies found generally encouraging and similar results in the use of the anti-VEGF drug aflibercept (Eylea, Regeneron Pharmaceuticals, Tarrytown, NY) as rescue therapy for patients who were not responsive to repeated injections of either ranibizumab (Lucentis, Genentech, South San Francisco, CA), or bevacizumab (Avastin, Genentech).
One study was conducted by researchers at the University of Mississippi led by Benjamin Guidry, MD, and the other by a team from the Retina Institute of California headed by Vincent Hau, MD.
In the Mississippi study, 52 patients (60 eyes) with active wet AMD who had not responded well to no fewer than five consecutive ranibizumab or bevacizumab injections were chosen to receive aflibercept rescue therapy — three consecutive monthly injections followed by assessments every two months to determine if as-needed injections were required. All the patients in the rescue study were regularly examined for VA, with OCT imaging, fundus exam and IOP measurement at each visit.
The researchers found 15 of the 60 eyes (25%) showed complete resolution of CNV activity at or within the initial three-month treatment period. Only a single additional eye showed CNV resolution after the initial three-month period, making the final total 26.7% of eyes showing complete resolution of CNV activity at some point during treatment. The researchers noted 28 of the 60 eyes demonstrated improved visual acuity following three aflibercept injections, but 18 eyes showed reduced visual acuity after three aflibercept injections. Fourteen eyes displayed no changed in VA following the three injections.
The researchers concluded “in the setting of noresponsiveness to Avastin and or Lucentis, Eylea injections are a prudent alternative.” They also noted that “CNV activity is unlikely to cease if it does not do so during the initial three rescue injections.”
The Retina Institute of California Eylea rescue study involved 35 patients (41 eyes) with recalcitrant wet AMD despite monthly Lucentis or Avastin injections. Patients were treated with Eylea initially and then again on an asneeded basis when OCT, fluorescien angiography or clinical examination showed evidence of CNV activity.
The researchers found that mean Log Mar VA improved from 0.46 prior to Eylea treatment to mean 0.37 at six months. Anatomically, mean central thickness and mean total volume improvement from baseline was 11.21 μm and 0.44 mm3 respectively at one month and 74.15 μm and 0.24 mm3 at six months. Foveal thickness at six months was statistically significant, while other endpoints were not but showed a (positive) trend. The researchers identified no unexpected adverse events were identified during the follow-up period.
The researchers concluded that the study supports the use of aflibercept “as an effective treatment option” for patients with recalcitrant wet AMD unresponsive to (other) anti-VEGF intravitreal injections.
Guidry B, Chen CJ. Eylea rescue therapy in eyes with proven non-response to other anti-VEGF molecules. Invest Ophthalmol Vis Sci. 2013;54:ARVO E-Abstract 3806.
Hau V, Samuel M, Davis M, Lin R, Chang T. Aflibercept effect on macula thickness and visual acuity in exudative AMD patients recalcitrant to ranibizumab and bevacizumab. Invest Ophthalmol Vis Sci. 2013;54:ARVO E-Abstract 6270.
7. RE-EVALUATING THE AREDS FORMULATION
Clinical studies have shown that the progression of macular degeneration can be reduced by as much as 25% by taking supplements based on the AREDS formulation of the antioxidant vitamins C and E, beta carotene and zinc.
However, after evaluating the clinical data in the AREDS2 study, the AREDS team, directed by Emily Chew, MD, of the National Institute of Health, and presenting at ARVO, believes that the formulation can be improved by dropping beta-carotene, which has been proven to increase the risk of lung cancer for smokers, and adding the carotenoids lutein and zeaxanthin, plus the omega-3 polyunsaturated fatty acids DHA and EPA. When compared for their effectiveness against AMD, zeaxanthin and lutein provided 20% more protection than beta carotene.
Age-Related Eye Disease Study 2 Research Group. Lutein + zeaxanthin and omega-3 fatty acids for age-related macular degeneration: the Age-Related Eye Disease Study 2 (AREDS2) randomized clinical trial. JAMA. 2013;309:2005-2015.
8. ENDOGENOUS ENDOPHTHALMITIS CAUSED BY MOLD
Endogenous endophthalmitis caused by mold is a rather rare disease. Its major risk factors are current or recent hospitalization, iatrogenic immunosuppression and cardiac disease. Visual outcomes are generally very poor.
Researchers at Bascom Palmer Eye Institute led by Jayanth Sridhar, MD, retrospectively were able to identify 15 eyes with culture-proven unilateral or bilateral endogenous endophthalmitis to determine methods of diagnosis, treatment patterns and outcomes. The 13 individuals studied had all been patients at Bascom Palmer.
The most common presenting symptom was decreased vision (93.3%), followed by redness (66.7%) and pain (40%). None of the patients were diagnosed on routine screening for fungemia. Diffuse anterior and posterior inflammation was the most common finding upon examination of the affected eyes (66.7%). Hypopyon was seen in one-third of the affected eyes.
The most common organisms cultured by vitreous aspirate or vitrectomy specimen was aspergillus fumigatus (46.2%), fusarium oxysporum (15.4%) and aspergillus glaucos (15.4%).
Initial treatment was pars plana vitrectomy (PPV) in 46.7% of affected eyes. In the course of treatment, 93.3% of eyes received either oral or intravenous antifungal therapy, 60% of eyes received intravitreal injection of an antifungal agent and 80% of eyes underwent PPV, with or without lensectomy.
The researchers concluded that despite early and appropriate management, endogenous endophthalmitis caused by molds generally results in poor visual outcomes, with acuity of approximately 20/200 at final follow-up seen in 21.4% of the affected eyes (the only final VA results reported for eyes that were not enucleated). In this study, retinal detachment occurred in 13.3% of eyes and enucleation was performed in 26.7% of eyes. RP
Sridhar J, Flynn HW Jr, Miller D, Albini TA. Endogenous endophthalmitis caused by molds: risk factors, management strategies, and visual acuity outcomes. Invest Ophthalmol Vis Sci. 2013;54:ARVO E-Abstract 1126.