Purpose
To investigate whether cataract surgery was beneficial in patients with neovascular age-related macular degeneration (AMD) receiving monthly ranibizumab injections in the ANCHOR (Anti-VEGF Antibody for the Treatment of Predominantly Classic Choroidal Neovascularization in AMD) and MARINA (Minimally Classic/Occult Trial of the Anti-VEGF Antibody Ranibizumab in the Treatment of Neovascular AMD) phase 3 trials.
Design
Retrospective analysis.
Methods
Patients were identified who underwent cataract surgery during the 2 pivotal trials. For this analysis, the best-corrected visual acuity (VA) just prior to cataract surgery was referred to as the redefined baseline VA. For the period after cataract surgery, endpoints included change in VA, time to first postsurgery injection, and total number of injections. Monthly follow-up visits after surgery were defined at 30-day intervals ± 15 days.
Results
Three subgroups were identified: study eyes of ranibizumab-treated patients (758 eyes [23 undergoing surgery]), fellow eyes of ranibizumab-treated patients (758 eyes [28 undergoing surgery]), and eyes of non-ranibizumab patients (762 [16 undergoing surgery]). Three months postsurgery, the VA of ranibizumab-treated eyes improved by a mean of 10.4 (± 3.4) letters compared to the redefined baseline (n = 20; 95% confidence interval +3.3 letters to +17.5 letters). The mean VA change from redefined baseline VA was not significantly different between the 3 groups at any of the evaluated time points postsurgery ( P > .44 for all comparisons between each pair of the 3 groups at 1, 2, 3, and 4 months following surgery).
Conclusions
In the phase 3 trials, cataract surgery appeared to be safe and beneficial for all eyes with AMD, including ranibizumab-treated eyes with neovascular AMD. An average VA improvement of more than 2 lines was typically observed.
Since the 1970s, the association between cataract surgery and the progression of age-related macular degeneration (AMD) has been debated. This debate is complicated by 1) the presence of both cataract and AMD in an aging population, 2) the question of whether cataract surgery is responsible for the progression of AMD, and 3) the question of whether comorbidities (such as cardiovascular disease, hypertension, and nutritional deficiencies) contribute to the progression of both diseases and confound the interpretation of the data. Evidence supporting a relationship between cataract surgery and the progression of AMD includes small, uncontrolled, retrospective case reports; large, population-based cohort studies; and a recent review of all the published reports.
Evidence against a relationship between cataract surgery and the progression of dry AMD includes larger retrospective reviews, prospective studies, and a study of autopsy eyes. However, in 2008, the authors of a comprehensive literature review could not conclude whether cataract surgery was beneficial or harmful in people with AMD. Subsequent to that report, compelling evidence from the Age-Related Eye Disease Study (AREDS) was published demonstrating that cataract surgery is safe in the setting of dry AMD. AREDS, a prospective, randomized trial designed to investigate vitamin supplementation in dry AMD, was originally designed as an observational trial of both AMD and cataract.
This debate is further complicated by the evolution of the surgical technique for cataract removal since the 1970s, when the first reports of disease progression were published. Older methods such as extracapsular surgery have been largely replaced by small-incision phacoemulsification techniques, resulting in shorter surgical times with presumably less light-induced phototoxicity, more rapid recovery, and less postoperative inflammation.
While it may now appear that cataract surgery is not associated with the progression of dry AMD, the debate continues over the safety of cataract surgery in the setting of neovascular AMD. With the availability of effective treatments that preserve central visual acuity (VA) in neovascular AMD patients, it seems likely that cataract surgery will be needed to preserve the VA gains achieved in phakic eyes undergoing therapy with an inhibitor of vascular endothelial growth factor (VEGF). In a recent report by Furino and associates, 20 treatment-naïve eyes with AMD and predominantly classic choroidal neovascularization (CNV) underwent a single intravitreal injection with bevacizumab (Avastin; Genentech, Inc, South San Francisco, California, USA) immediately following cataract surgery. One month after the surgery, average VA improved and macular thickness decreased as determined by optical coherence tomography (OCT) imaging. Longer follow-up was not reported. While these results are encouraging, it seems unlikely that most cataract surgeons would operate on an eye with neovascular AMD that had not received at least 1 prior treatment with an anti-VEGF drug.
To determine whether cataract surgery is beneficial in eyes with neovascular AMD undergoing injections with ranibizumab (Lucentis, Genentech, Inc), we performed a retrospective review of all eyes undergoing cataract surgery in the pivotal ranibizumab phase 3 trials MARINA (Minimally Classic/Occult Trial of the Anti-VEGF Antibody Ranibizumab In the Treatment of Neovascular AMD) and ANCHOR (Anti-VEGF Antibody for the Treatment of Predominantly Classic Choroidal Neovascularization in AMD). In these trials, study eyes with neovascular AMD received monthly ranibizumab injections or control treatments for 2 years, during which time some eyes underwent cataract surgery at the discretion of the investigator.
Patients and Methods
Patients in MARINA and ANCHOR were randomized to receive 1 of 2 ranibizumab doses (0.3 mg or 0.5 mg) or a sham injection over 2 years. In the ANCHOR trial, which investigated eyes with predominantly classic CNV, the eyes randomized to sham injection also received verteporfin photodynamic therapy (PDT) while the eyes randomized to ranibizumab injections also received sham PDT with saline infusion at baseline and then every 3 months as needed. In the MARINA trial, eyes with minimally classic and occult CNV could also receive PDT if the neovascular lesion progressed to predominantly classic CNV or for certain active minimally classic or occult CNV lesions. Only 1 eye (the study eye) received ranibizumab injections and the 0.3-mg and 0.5-mg ranibizumab dose groups were pooled for these analyses. In the current retrospective analysis, 3 subgroups from the MARINA and ANCHOR trials were reviewed: study eyes receiving ranibizumab, fellow eyes of patients receiving ranibizumab injections, and eyes of non–ranibizumab-treated patients (fellow eyes of sham-injected patients as well as any eyes in the ANCHOR trial control group receiving PDT). There were no cataract surgeries in study eyes from the MARINA trial sham-injected control group.
Eyes undergoing cataract surgery were identified from the clinical trial databases using the following ocular procedure names: (a) cataract extraction, (b) cataract removal, (c) cataract surgery, (d) lens implant, and (e) lensectomy. Detailed information about the cataract surgery was not available from the databases. All VA measurements were best corrected and obtained monthly from day 0 through month 24 using the standard protocol Early Treatment Diabetic Retinopathy Study (ETDRS) charts and procedures as previously described. For this analysis, the VA just prior to cataract surgery was referred to as the redefined baseline VA. For the period between the true baseline (study day 0) and cataract surgery, endpoints include the time from the true baseline until surgery, the number of injections, and VA at the true and redefined baselines. For the period after cataract surgery, the endpoints include the change in VA up to 4 months after the surgery, the time until the first postsurgery injection, and the total number of injections. A follow-up period of 4 months was chosen because it included the usual 3-months-postoperative global period following routine cataract surgery when maximal VA improvement is observed in uncomplicated surgical cases.
Monthly follow-up visits after surgery were defined at 30-day intervals ± 15 days. If 2 study visits fell within the same ±15-day period on either side of a presumed monthly postsurgery follow-up visit, then the study visit closest to the defined monthly postsurgery follow-up visit was used in this analysis. If 2 study visits were the same number of days away from the defined monthly postsurgery follow-up visit, then the study visit following the defined postsurgery follow-up visit was used.
Mean VA change for each month following the redefined baseline was compared between each pair of the 3 groups using a 2-tailed t test. The nonparametric Wilcoxon rank sum test ( t approximation) was used as confirmation given that the number of cataract surgeries in each group was small. Ninety-five percent confidence intervals of the mean change were calculated from the t distribution. Statistical significance was assumed for P < .05. All analyses were performed on observed data without imputation for missing values, using SAS v9.1 (SAS Institute Inc., Cary, North Carolina, USA).
Results
A total of 758 study eyes were randomized to ranibizumab therapy, 478 from MARINA and 280 from ANCHOR. A total of 381 study eyes were randomized to control, 238 to sham injections in MARINA and 143 to sham injections plus verteporfin PDT in ANCHOR. There were 758 fellow eyes from the ranibizumab-randomized patients and 381 fellow eyes from the sham-randomized patients.
Table 1 shows the number of eyes from each group that underwent cataract surgery during the trials. Of the 758 pooled ranibizumab patients, cataract surgery was performed on 23 study eyes and 28 fellow eyes. Sixteen eyes from non–ranibizumab-treated patients also underwent cataract surgery, composed of 1 study eye (ANCHOR PDT-treated sham-injected eye) and 15 fellow eyes. No other sham-injected study eye underwent cataract surgery. Compared with study entry baseline VA, mean redefined baseline VA was similar in ranibizumab-treated patient study eyes but was decreased in ranibizumab-treated patient fellow eyes and non–ranibizumab-treated patient eyes ( Table 1 ).
| Group a | Phakic Eyes at Baseline (n) | Eyes Undergoing Cataract Surgery (n) | Mean Baseline VA (SD) | Mean Redefined Baseline VA b (SD) |
|---|---|---|---|---|
| Ranibizumab-treated patients, study eye (N=758) | 417 c | 23 | 49.8 (15.5) | 46.0 (17.5) |
| Ranibizumab-treated patients, fellow eye (N=758) | 451 c | 28 | 59.5 (17.8) d | 46.9 (21.9) |
| Non–ranibizumab-treated patients (PDT-treated sham-injected eye and fellow eyes) e (N=762) | 437 | 16 | 62.7 (18.4) | 52.6 (22.4) |
a Ranibizumab-treated patients were pooled 0.3-mg and 0.5-mg dose cohorts from the MARINA and ANCHOR trials. In the MARINA trial, control patients received sham injections; in the ANCHOR trial, control patients received PDT plus sham injections.
b Redefined baseline VA is the VA documented just pior to cataract surgery.
c Baseline lens status missing in 2 patients.
e Includes 1 study eye from a PDT-treated patient and 15 fellow eyes from sham-/PDT-treated patients.
For patients undergoing cataract surgery, the AMD type in the fellow eye by study eye treatment group (based on the last classification prior to the cataract surgery) is shown in Table 2 . The percentages of patients with neovascular AMD, with geographic atrophy, or without either feature were 42.8% (12/28), 3.6% (1/28), and 53.6% (15/28) in ranibizumab-treated patient fellow eyes, respectively, and 33.3% (5/15), 6.7% (1/15), and 60.0% (9/15) in non–ranibizumab-treated patient fellow eyes, respectively ( Table 2 ).
| AMD Type a | Study Eye Treatment Group for Fellow Eyes With Cataract Surgery, n (%) | |||
|---|---|---|---|---|
| Ranibizumab | Sham | PDT | Pooled Sham and PDT | |
| Geographic atrophy | 1 (3.6) | 1 (10.0) | 0 | 1 (6.7) |
| Neovascular AMD | 12 (42.8) | 5 (50.0) | 0 | 5 (33.3) |
| Absence of geographic atrophy or neovascularization | 15 (53.6) | 4 (40.0) | 5 (100) | 9 (60.0) |
a Based on the last available classification prior to the cataract surgery.
Table 3 summarizes the time from study entry to the time of cataract surgery and the average number of injections for each group. Among the study eyes of ranibizumab-treated patients, the average time from the first injection to cataract surgery was 431 days, following an average of 13.8 injections ( Table 3 ).
| Group a | Eyes (n) | Time From Initial Injection to Cataract Surgery (Days), Mean (SD) | Number of Study Eye Injections Prior to Cataract Surgery, Mean (SD) | Time From Cataract Surgery to Next Study Eye Injection (Days), Mean (SD) | Number of Study Eye Injections After Cataract Surgery to End of Study, Mean (SD) |
|---|---|---|---|---|---|
| Ranibizumab-treated patients, study eye | 23 | 431 (152) | 13.8 (4.7) | 36 (13) b | 7.8 (5.0) |
| Ranibizumab-treated patients, fellow eye | 28 | 350 (212) | 11.6 (6.6) | 15 (9) c | 10.6 (6.7) |
| Non–ranibizumab-treated patients (PDT-treated sham-injected eye and fellow eyes) d | 16 | 339 (197) | 11.8 (6.5) | 18 (13) | 11.6 (6.9) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
