 |
|
Twice-Daily 0.2% Brimonidine–0.5% Timolol Fixed-Combination Therapy vs Monotherapy With Timolol or Brimonidine in Patients With Glaucoma or Ocular Hypertension
A 12-Month Randomized Trial
Mark B. Sherwood, MD;
E. Randy Craven, MD;
Connie Chou, PhD;
Harvey B. DuBiner, MD;
Amy L. Batoosingh, BA;
Rhett M. Schiffman, MD, MS, MHSA;
Scott M. Whitcup, MD; for the Combigan Study Groups I and II
Arch Ophthalmol. 2006;124:1230-1238.
ABSTRACT
| |
Objective To evaluate the intraocular pressure (IOP)–lowering efficacy and safety of a fixed combination of 0.2% brimonidine tartrate and 0.5% timolol maleate (fixed brimonidine-timolol) compared with the component medications.
Methods In 2 identical, 12-month, randomized, double-masked multicenter trials, patients with ocular hypertension or glaucoma were treated with fixed brimonidine-timolol twice daily (n = 385), 0.2% brimonidine tartrate 3 times daily (n = 382), or 0.5% timolol maleate twice daily (n = 392).
Main Outcomes Measures Mean change from baseline IOP and incidence of adverse events.
Results The mean decrease from baseline IOP during 12-month follow-up was 4.4 to 7.6 mm Hg with fixed brimonidine-timolol, 2.7 to 5.5 mm Hg with brimonidine, and 3.9 to 6.2 mm Hg with timolol. Mean IOP reductions were significantly greater with fixed brimonidine-timolol compared with timolol at all measurements (P .002) and brimonidine at 8 AM, 10 AM, and 3 PM (P<.001) but not at 5 PM. The incidence of treatment-related adverse events in the fixed-combination group was lower than that in the brimonidine group (P = .006) but higher than that in the timolol group (P<.001). The rate of discontinuation for adverse events was 14.3% with the fixed combination, 30.6% with brimonidine, and 5.1% with timolol.
Conclusions Twice-daily fixed brimonidine-timolol therapy provides sustained IOP lowering superior to monotherapy with either thrice-daily brimonidine or twice-daily timolol and is better tolerated than brimonidine but less well tolerated than timolol.
Application to Clinical Practice Fixed brimonidine-timolol is an effective and convenient IOP-lowering therapy.
INTRODUCTION
Intraocular pressure (IOP) is a primary risk factor for the progression of visual field loss in patients with glaucoma. Lowering of IOP is effective in delaying or preventing the development of glaucoma in patients with ocular hypertension1 and in reducing the progression of established disease.2-4 Powerful IOP-lowering medications are currently available that allow many patients to achieve their IOP-lowering goals with a single medication,5-6 but not all patients can use these drugs. Despite the substantial IOP lowering possible with monotherapy, many patients may need to use 2 or more medications to reach a target IOP sufficiently low to halt further visual deterioration.1, 7 In the Ocular Hypertension Treatment Study,1 by year 5, almost 40% of the patients required 2 or more medications to achieve their target IOP. In the medical arm of the Collaborative Initial Glaucoma Treatment Study,7 more than 75% of the patients required 2 or more medications after 2 years.
Two commonly used classes of IOP-lowering drugs are the  -adrenergic antagonists (-blockers) and the  -adrenergic agonists. The nonselective -blocker timolol lowers IOP by decreasing aqueous humor production,8 and the -adrenergic agonist brimonidine lowers IOP by decreasing aqueous humor production and increasing aqueous humor outflow through the uveoscleral pathway.9 These 2 drugs are widely used, and each was shown to be effective as adjunctive therapy with other classes of IOP-lowering medications.10-14 When brimonidine and timolol are used concomitantly, IOP is reduced more than when either drug is used alone.15-18
For patients who require more than 1 medication for IOP control, fixed combinations of drugs are more convenient than the concomitant use of the separate components and may improve compliance. Use of a combination product can also decrease the patient's daily ocular exposure to preservatives. The purpose of this study was to compare the efficacy and safety of a fixed combination of 0.2% brimonidine tartrate and 0.5% timolol maleate with that of either component administered as monotherapy.
METHODS
STUDY DESIGN
We conducted a prospective, 12-month, phase 3, randomized, double-masked, parallel-group study comparing the fixed combination of 0.2% brimonidine and 0.5% timolol twice daily with 0.2% brimonidine 3 times daily or 0.5% timolol twice daily at 53 sites throughout the United States. The study was divided into 2 identical, separate trials for regulatory purposes. Because the protocols were identical, the results were similar, and the same conclusions were drawn from each trial, the data were pooled to create a larger cohort for the present analysis, as has been done in studies of other ophthalmic therapeutic drugs.19
The studies were conducted in compliance with the Declaration of Helsinki (1996) and in accordance with institutional review board regulations. Each participating investigator received institutional review board approval and obtained written informed consent from patients before initiation of the study.
PATIENTS
Patients 18 years or older who required bilateral treatment for glaucoma or ocular hypertension were eligible to enroll. Eligible patients had to have a baseline IOP (after washout) between 22 and 34 mm Hg in each eye, with no more than a 5-mm Hg difference between eyes and a best-corrected visual acuity of 20/100 or better OU.
Key exclusion criteria included active ocular disease, functionally significant or progressive visual field loss in the previous year, abnormally low or high blood pressure or pulse rate, contraindications or sensitivity to any component of the study treatments, use of other ocular medications or other therapies that might have a substantial effect on IOP, and ocular surgery in the past 3 months. Women who were not using an effective means of contraception or who were pregnant or nursing were excluded. Continuation of long-term systemic therapy that could have a substantial effect on IOP was allowed, when considered necessary for the patient's welfare, as long as the dosages remained constant throughout the trial.
INTERVENTION
All recruited patients were required to undergo washout of any IOP-lowering medications before the baseline visit. Washout periods were 4 days for parasympathomimetics and carbonic anhydrase inhibitors, 2 weeks for sympathomimetics and -agonists, and 4 weeks for -blockers (alone or in combination) and topical prostaglandin analogues. After the washout period, a single-day baseline visit was performed before the instillation of study medication. Intraocular pressure was measured at 8 AM, 10 AM, 3 PM, and 5 PM. In addition, at the baseline visit, patients were randomly assigned to 1 of 3 treatment groups using a 1:1:1 allocation: 0.2% brimonidine tartrate–0.5% timolol maleate fixed-combination ophthalmic solution (Combigan) twice daily (fixed brimonidine-timolol), 0.5% timolol maleate twice daily (timolol), or 0.2% brimonidine tartrate (Alphagan) 3 times daily (brimonidine). All drugs were from Allergan Inc (Irvine, Calif). The randomization sequence was generated by the study sponsor using the PLAN procedure in SAS version 6.12 (SAS Institute Inc, Cary, NC). Medications were shipped to each investigational site for patient enrollment. Before initiation of study treatment, each patient who qualified for entry was assigned a patient randomization number, which was used on all documentation. The investigator was instructed to use the bottles of study medication that corresponded to the assigned patient randomization number.
Patients received their first dose of medication (1 drop in each eye) after all evaluations at the baseline visit. Thereafter, patients self-instilled 1 drop of the appropriate study medication into both eyes in the morning between 7 and 9 AM, in the afternoon between 1 and 3 PM, and in the evening between 7 and 9 PM. To maintain masking, patients in the fixed brimonidine-timolol group and the timolol group administered study medication in the morning and evening and a vehicle solution in the afternoon. Patients in the brimonidine group administered study medication in the morning, afternoon, and evening. The afternoon dose in all 3 groups (vehicle solution or brimonidine) was provided in a separate, smaller bottle to ensure that all medications were given at the correct time. Follow-up study visits were scheduled at weeks 2 and 6 and at months 3, 6, 9, and 12.
The fixed combination was preserved using benzalkonium chloride because of the improved stability of timolol in this preservative. Other formulations used in the study had the same concentration of active component and were preserved using benzalkonium chloride to allow them to serve as appropriate controls.
OUTCOME MEASURES
The primary efficacy measure was mean change from baseline IOP at each follow-up time point. Intraocular pressure was measured at 8 AM (before instillation of study drug), 10 AM, 3 PM (before instillation of study drug), and 5 PM at weeks 2 and 6 and at months 3, 6, and 12. At the 9-month visit, IOP was measured only at 8 and 10 AM. The 5 PM IOP measurement, at peak effect after the afternoon dose in the brimonidine group, was included in this study to determine whether the added dose of brimonidine in the afternoon provided additional IOP lowering compared with timolol or the fixed combination, which were administered only twice daily. Two consecutive IOP measurements of each eye were obtained using a Goldmann applanation tonometer. If the 2 measurements differed by more than 2 mm Hg, a third measurement was obtained. For each eye, IOP was analyzed as the mean of 2 measurements or as the median of 3 measurements. For each patient, change from baseline IOP was determined by averaging the values for each eye.
The following secondary efficacy measures were prospectively planned for the study: mean IOP, the percentage of patients reaching a mean daytime IOP (defined as the mean of the 8 AM, 10 AM, 3 PM, and 5 PM values from a given visit) of less than 18 mm Hg at all follow-up visits, the percentage of patients with a mean follow-up IOP (average of IOP from all 22 follow-up time points) within specified target pressure ranges (<14, 14-17.5, and >17.5 mm Hg), and the percentage of patients achieving a mean daytime decrease from baseline IOP (defined as the mean decrease from baseline IOP across all 4 diurnal measurements at a given visit) of greater than 20% at all the follow-up visits. Each analysis used the mean of IOP measurements from both eyes as the IOP value for the individual patient.
The safety evaluation included an assessment of reported adverse events, biomicroscopy, tests of visual acuity and visual fields, ophthalmoscopy, cup-disc ratio, heart rate, blood pressure, complete blood cell count, serum chemistry, and urinalysis. Investigators elicited adverse events using nondirected questioning (with questions such as "How are you feeling today?") and rated their severity and probable relationship to the study medication. For analysis, all adverse events were coded using industry standard COSTART (Coding Symbols for Thesaurus of Adverse Reaction Terms) nomenclature. The vitreous and optic nerve head were evaluated through a dilated pupil after the 5 PM IOP measurement.
DATA ANALYSIS AND STATISTICS
Efficacy analyses were performed for the intention-to-treat (ITT) population (all patients randomized to treatment) using last observation carried forward (LOCF) for missing values. Results of efficacy analyses for the per-protocol population (patients with no major protocol violations) with no imputation for missing values confirmed the ITT results; only the ITT analyses are reported herein.
Nominal categorical variables were analyzed using the Fisher exact test or the Pearson  2 test. Ordinal categorical variables were analyzed using the Wilcoxon rank sum test. Within-group changes from baseline for categorical variables were analyzed using the Wilcoxon signed rank test. Continuous variables were analyzed using analysis of variance or analysis of covariance. Within-group changes from baseline for continuous variables were analyzed using paired t tests.
A 2-way analysis of variance model with treatment and investigator as fixed effects was used to analyze baseline IOP measurements. An analysis of covariance model with baseline IOP as covariate and treatment and investigator as fixed effects was used to analyze mean IOP and mean change from baseline IOP for the follow-up measurements. Comparisons were made between fixed-combination therapy and each of the 2 monotherapies in a pairwise manner using contrasts from the corresponding analysis of variance or analysis of covariance model. The significance level was set at = .05. Point estimates and 2-sided 95% confidence intervals of the difference between treatments were based on least-squares means.
A power calculation estimated that a sample size of 150 in each treatment group would provide a power of 98% to detect a 1.5-mm Hg difference and a power of 76% to detect a 1.0-mm Hg difference in mean change from baseline IOP between groups, using an estimated common standard deviation of 3.2 mm Hg.
RESULTS
PATIENT POPULATION
Of 1159 patients enrolled between December 1999 and December 2000, 385 were randomized to receive fixed brimonidine-timolol twice daily, 382 to receive brimonidine tartrate 3 times daily, and 392 to receive timolol maleate twice daily. A total of 833 patients (71.9%) completed 12 months of treatment, and 326 discontinued treatment before the end of the study. Reasons for discontinuation included adverse events (192 patients;16.6%), lack of efficacy (53 patients; 4.6%), and other reasons (81 patients; 7.0%) (Figure 1).
|
|
|
|
Figure 1. Patient flow through the study. Completion rates for study visits in the fixed-combination, brimonidine, and timolol groups, respectively, were as follows: week 2: 97.7%, 94.2%, and 96.9%; week 6: 94.8%, 87.2%, and 93.1%; month 3: 89.9%, 77.2%, and 90.0%; month 6: 83.6%, 66.5%, and 87.2%; month 9: 77.7%, 59.2%, and 86.0%; and month 12: 74.3%, 55.8%, and 85.2%. Patients who discontinued owing to both ocular and nonocular adverse events are included in both groups. "Other reasons" for discontinuation include protocol violations, administrative issues, and other reasons. Brimonidine was given as 0.2% brimonidine tartrate; timolol as 0.5% timolol maleate. BID indicates twice daily; TID, 3 times daily.
|
|
|
Demographic characteristics were generally similar among treatment groups, but there were slightly more female patients in the brimonidine group than in the other treatment groups (among-group P = .046) (Table 1). Patients ranged in age from 23 to 89 years (mean age, 62.6 years). The patient population was primarily white (75.8%; n = 879), but a substantial proportion of the patients were self-reported African Americans (16.1%; n = 187). The primary ophthalmic diagnosis was open-angle glaucoma in both eyes (65.7%; n = 762), and 68.6% of patients (n = 795) required a washout of ocular hypotensive medication before study entry (Table 1).
|
|
|
|
Table 1. Patient Demographics and Baseline Characteristics by Treatment Group*
|
|
|
IOP-LOWERING EFFICACY
There were no statistically significant between-group differences in baseline mean IOP at the 8 AM, 10 AM, or 3 PM measurements (Table 1). Mean baseline IOP in the 3 treatment groups ranged from 24.7 to 25.0 mm Hg at 8 AM, from 23.3 to 23.5 mm Hg at 10 AM, and from 22.1 to 22.5 mm Hg at 3 PM. At 5 PM, mean IOP was slightly lower in the fixed brimonidine-timolol group (21.8 mm Hg) than in the timolol group (22.4 mm Hg; P = .01) or the brimonidine group (22.2 mm Hg; P = .06) but was within 0.6 mm Hg of both groups.
The mean decrease from baseline IOP at the 22 measurements across all follow-up visits ranged from 4.4 to 7.6 mm Hg in the fixed brimonidine-timolol group, from 2.7 to 5.5 mm Hg in the brimonidine group, and from 3.9 to 6.2 mm Hg in the timolol group (Figure 2). The mean decrease in IOP from baseline was significantly greater in the fixed brimonidine-timolol group than in the timolol group at all 22 follow-up measurements throughout the study (P.002 at 8 AM; P<.001 at 10 AM, 3 PM, and 5 PM). The mean decrease in IOP from baseline was significantly greater in the fixed brimonidine-timolol group than in the brimonidine group at all 8 AM, 10 AM, and 3 PM follow-up measurements throughout the study (P<.001) but not at the 5 PM measurements. Note that the 5 PM measurement was 9 hours after administration of the most recent dose of active medication in the fixed brimonidine-timolol and timolol groups but only 2 hours after the most recent dose in the brimonidine group. The only significant difference between these 2 groups at the 5 PM measurement was at week 6, when the mean decrease from baseline IOP was significantly larger in the fixed-combination group than in the brimonidine group (P = .04). The mean of the absolute difference in IOP lowering between the 2 eyes ranged from 1.3 to 1.7 mm Hg across all measurement times and treatment groups. In addition, the median difference in IOP lowering between the 2 eyes was 1 mm Hg or less at 21 of 22 measurement times. These results suggest that for most patients, the IOP-lowering response to treatment was similar in both eyes.
|
|
|
|
Figure 2. Mean change from baseline intraocular pressure (IOP) at 8 AM (A), 10 AM (B), 3 PM (C), and 5 PM (D). Analysis is based on the intention-to-treat patient population with last observation carried forward for missing values. Similar results were obtained using observed values in the per-protocol patient population. Brimonidine was given as 0.2% brimonidine tartrate; timolol as 0.5% timolol maleate. BID indicates twice daily; TID, 3 times daily. Error bars represent SEM.
|
|
|
Daytime mean IOP was lower in the fixed brimonidine-timolol group than in either of the monotherapy groups (Figure 3). The differences between the fixed-combination group and the timolol group were significant at all follow-up measurements (P.002). The differences between the fixed-combination group and the brimonidine group were significant at all 8 AM, 10 AM, and 3 PM measurements (P<.001) and at the 5 PM measurement at week 6 (P = .04).
|
|
|
|
Figure 3. Mean intraocular pressure (IOP) at 8 AM, 10 AM, 3 PM, and 5 PM at month 12. Analysis is based on the intention-to-treat patient population with last observation carried forward for missing values. Similar results were obtained using observed values in the per-protocol patient population. Arrows indicate administration of the morning and afternoon doses. Brimonidine was given as 0.2% brimonidine tartrate; timolol as 0.5% timolol maleate. BID indicates twice daily; TID, 3 times daily. Error bars represent SEM.
|
|
|
A significantly greater percentage of patients in the fixed brimonidine-timolol group than in either monotherapy group maintained a mean IOP during the day (IOP averaged across all diurnal measurements at a given visit) of less than 18 mm Hg throughout the study (P<.001). Mean daytime IOP was consistently below this level in 39.5% of patients in the fixed brimonidine-timolol group, 14.9% of patients in the brimonidine group, and 21.7% of patients in the timolol group. Similar results were obtained in analyses that used data from left or right eyes only rather than mean data from both eyes. The percentage of left eyes with a mean daytime IOP consistently less than 18 mm Hg was 40.0%, 14.9%, and 21.7% in the fixed brimonidine-timolol, brimonidine, and timolol groups, respectively (P<.001 for fixed-combination therapy vs each monotherapy). For right eyes, the percentages were 38.4%, 14.7%, and 20.4%, respectively (P<.001 for fixed-combination therapy vs each monotherapy).
A significantly greater percentage of patients in the fixed brimonidine-timolol group than in either monotherapy group achieved a mean daytime decrease from baseline IOP of greater than 20% at every follow-up visit (P<.001). The daytime mean decrease from baseline IOP was greater than 20% in 42.1% of patients in the fixed brimonidine-timolol group, 13.1% in the brimonidine group, and 27.3% in the timolol group.
The distribution of patients across the target pressure ranges of less than 14.0, 14.0 to 17.5, and greater than 17.5 mm Hg favored the fixed brimonidine-timolol group over each of the monotherapy groups (Figure 4). There was a significant shift toward the lower target-pressure range in the fixed brimonidine-timolol group compared with the monotherapy groups (P<.001) (Figure 4).
|
|
|
|
Figure 4. Percentage of patients with mean follow-up intraocular pressure (IOP) (IOP averaged across all 22 measurements) in specified target pressure ranges. Analysis is based on the intention-to-treat patient population with last observation carried forward for missing values. Similar results were obtained using observed values in the per-protocol patient population. There was a significant shift toward the lower-pressure ranges in the fixed brimonidine-timolol group compared with the monotherapy groups (P<.001). Brimonidine was given as 0.2% brimonidine tartrate; timolol as 0.5% timolol maleate. BID indicates twice daily; TID, 3 times daily.
|
|
|
SAFETY
Compared with patients receiving brimonidine monotherapy, patients receiving the fixed combination had a lower incidence of 1 or more treatment-related adverse events (53.0% vs 62.8%; P = .006), discontinuations due to adverse events (14.3% vs 30.6%; P<.001), and at least a 1-grade increase in severity in biomicroscopic findings (54.8% vs 62.8%; P = .02). Compared with patients receiving timolol monotherapy, patients receiving the fixed combination had a higher incidence of 1 or more treatment-related adverse events (53.0% vs 40.8%; P<.001) and discontinuations due to adverse events (14.3% vs 5.1%; P<.001) and a similar incidence of at least a 1-grade increase in severity in biomicroscopic findings (54.8% vs 51.5%; P = .36). The only treatment-related individual adverse events that occurred in at least 10% of patients in any group and showed a significant difference among groups were conjunctival hyperemia and eye pruritus (Table 2). The incidence of these events in the fixed-combination group was significantly lower than that in the brimonidine group (P.005) but higher than that in the timolol group (P.06). Overall, the treatment-related incidence of adverse events related to conjunctival allergy and inflammation, which are typically associated with brimonidine and included any combination of conjunctival hyperemia, eye pruritus, follicular conjunctivitis, conjunctival folliculosis, allergic conjunctivitis, conjunctivitis, chemical conjunctivitis, conjunctival edema, and blepharoconjunctivitis, was significantly lower in the fixed-combination group than in the brimonidine group (26.0% vs 39.8%; P<.001). Significantly fewer fixed brimonidine-timolol–treated patients than brimonidine-treated patients exhibited at least a 1-grade increase from baseline in the severity of conjunctival erythema, conjunctival follicles, and conjunctival edema. Ocular stinging was the only treatment-related adverse event that occurred more frequently in the fixed-combination group than in the brimonidine group (P = .03), but the incidence was comparable between the fixed-combination group (6.2%) and the timolol group (6.6%; P = .82) (Table 2).
|
|
|
|
Table 2. Treatment-Related Adverse Events by Treatment Group*
|
|
|
There were only 2 patients with treatment-related serious adverse events, and both were in the timolol monotherapy group. One patient was hospitalized for respiratory distress secondary to emphysema and the other for tachycardia, sweating, and nausea. Four patients died during the study (2 in the brimonidine group and 1 each in the fixed-combination and timolol groups), but the causes of death were unrelated to treatment.
There were no significant between-group differences in changes in the cup-disc ratio or in the visual field mean deviation in the worse eye during the study. There were no clinically relevant differences between the fixed brimonidine-timolol group and either monotherapy group in the mean change from baseline to month 12 for any laboratory values (hematologic, chemical, or urinalysis testing).
Small but statistically significant mean changes in heart rate (2-3 beats per minute) and blood pressure were found in the fixed brimonidine-timolol and timolol groups during the study, but none of these mean changes were considered clinically relevant. There were no statistically significant differences in mean heart rate between the fixed-combination and timolol groups. Bradycardia was reported as a treatment-related adverse event for 2 patients (0.5%) in the fixed-combination group.
COMMENT
Overall, the fixed combination of brimonidine and timolol provided significantly better IOP-lowering efficacy than did either component used as monotherapy. This was seen in the evaluation of mean change from baseline IOP and the evaluation of mean IOP. Efficacy analyses for the per-protocol patient population confirmed the results for the ITT patient population, which are presented herein. At the 8 AM (morning trough) measurements, the differences in mean IOP lowering between the fixed-combination group and the brimonidine group ranged from 2.8 to 3.4 mm Hg, whereas the differences between the fixed-combination group and the timolol group ranged from 0.7 to 1.1 mm Hg. At the 10 AM time of peak effect, the differences in mean IOP lowering between the fixed combination and brimonidine monotherapy ranged from 1.9 to 2.5 mm Hg, whereas the differences between the fixed combination and timolol monotherapy ranged from 2.0 to 2.3 mm Hg. Although there is no agreement on the clinical relevance of changes in mean IOP in patient populations, data from the Early Manifest Glaucoma Trial4 have shown that even 1 mm Hg of additional IOP lowering reduces the risk of glaucoma progression by 10%. In addition, it is important to look at responder analyses when clinically relevant reductions in IOP are assessed. Patients randomly assigned to receive brimonidine-timolol therapy were more apt to achieve each clinically relevant end point (such as IOP <18 mm Hg and IOP decrease >20%) compared with patients in either monotherapy group. These findings are consistent with previous studies13, 17 that showed successful efficacy of brimonidine in concomitant therapy with timolol.
The observation that 40% of patients taking the fixed combination consistently achieved a mean diurnal IOP less than 18 mm Hg throughout follow-up compared with 15% of patients taking brimonidine and 22% of patients taking timolol may be important because it was noted in the Advanced Glaucoma Intervention Study that patients with glaucoma who consistently achieved pressures less than 18 mm Hg had a reduced risk of progressive glaucomatous visual field loss.3 Mean diurnal IOP, rather than IOP at a particular hour, was used in this analysis because IOP was measured only 1 time at each visit in the Advanced Glaucoma Intervention Study and the time of day was not specified. For our analysis to be similar to the Advanced Glaucoma Intervention Study analysis, we chose to use all data for all patients (rather than picking a specific time) and averaged data from the 4 diurnal measurements for each patient. Patients taking fixed brimonidine-timolol were also significantly more likely to achieve a greater than 20% reduction from baseline in daytime IOP. A decrease from baseline IOP of 20% to 30% has been recommended as an initial goal of glaucoma therapy in the American Academy of Ophthalmology preferred practice guidelines for glaucoma,20 and a 20% IOP reduction was the target used in the Ocular Hypertension Treatment Study, which showed that IOP lowering in ocular hypertensive individuals reduces the risk of glaucoma.1
The overall safety profile of fixed brimonidine-timolol therapy in this study was consistent with the results of previous studies21-22 of brimonidine and timolol as monotherapies. No safety concerns arose with fixed brimonidine-timolol therapy that had not been observed with the individual component drugs. Adverse events related to conjunctival allergy and inflammation were reported significantly less often for patients taking the fixed combination than for patients receiving brimonidine monotherapy, and fewer patients treated with the fixed combination discontinued from the study early owing to adverse events. This improvement in tolerability with fixed-combination therapy might be expected because patients receiving brimonidine monotherapy had a dosage of 3 times daily, whereas patients receiving the fixed combination had a dosage of twice daily and had reduced exposure to brimonidine.
In this study, 28.1% of the randomized patients discontinued treatment before year 1. The overall study completion rate of 71.9% was slightly lower than the 79.4% study completion rate in a 1-year trial23 that compared a fixed combination of latanoprost and timolol with the individual component medications used as monotherapy. It is important to examine the reasons for study discontinuation in any clinical trial to ensure that a difference between groups in discontinuation rates for lack of efficacy does not confound the results. In this trial, discontinuation for lack of efficacy was 3% for patients receiving the fixed combination or timolol alone. The most common reason for early termination of study participation was adverse effects. In the fixed-combination group, 14.3% of patients discontinued because of adverse events compared with 5.1% in the timolol monotherapy group, suggesting that the discontinuation rate for the fixed combination in clinical practice could be higher than that for timolol alone.
We used 2 different methods to manage the missing observations in this study: LOCF (in the ITT analysis) and observed data only (with no imputation, in the per-protocol analysis). The analyses presented were based on data from all patients randomized to treatment (the ITT population), and they used LOCF to estimate missing data. This is the most frequently used method to deal with missing data and is required by the Food and Drug Administration. Other potential ways to handle missing data are to censor it or to estimate it by extrapolation from a line drawn through observed data. Each method has associated assumptions, and each is subject to bias. For example, if missing data are ignored, analyses at later visits in the study will omit data from patients who prematurely drop out of a study because of lack of efficacy or drug intolerance, and this may inflate the apparent efficacy of treatment. On the other hand, use of LOCF could lead to erroneous conclusions if the effect of treatment would be expected to change across time. We believe that missing data in this study do not invalidate our conclusions for 2 reasons. First, efficacy at month 3 in each study arm was stable during the next 9 months, and the conclusions drawn from the data at month 3, when completion rates were similar among groups (fixed combination, 89.9%; brimonidine, 77.2%; and timolol, 90.5%), were the same as those drawn from the data at month 12. Second, analyses using LOCF in the ITT population and analyses using observed values in the per-protocol population gave similar results, with the fixed combination being superior to each component medication in each analysis.
To address the possibility of patient selection bias in the study, we evaluated the number of patients who required washout of study medication. Approximately 70% of enrolled patients required washout of 1 or more medications before study enrollment. One third of enrolled patients washed out a nonselective -blocker or the timolol-dorzolamide fixed combination before study entry, and 15.5% washed out an -adrenergic agonist. These percentages of patients are consistent with current prescribing patterns and do not suggest preferential enrollment of patients who were previously successfully treated with the study drugs. Moreover, supplemental analysis found no notable difference in the incidence of adverse events or in the safety conclusions drawn from the study when patients who underwent washout of -blockers or -agonists were excluded from the analysis.
To meet regulatory requirements in this phase 3 study, timolol and the fixed combination were administered twice daily and brimonidine was administered 3 times daily. The plot of mean IOPs throughout the day at month 12 suggests that the difference in dosing schedules did not affect the conclusions drawn from the study (Figure 4). The twice-daily fixed combination provided significantly lower mean IOP compared with brimonidine given 3 times daily at all 8 AM, 10 AM, and 3 PM measurements, and even at 5 PM when it was near trough effect and brimonidine was at peak effect, the fixed combination was at least as efficacious as brimonidine alone, despite the added afternoon dose of brimonidine in the monotherapy arm.
The primary analyses of IOP used mean values from both eyes, a method of data analysis accepted by the Food and Drug Administration. When the analysis of consistent achievement of a mean diurnal IOP less than 18 mm Hg was repeated using data from only left or right eyes, similar results were obtained. It is possible that one eye of a patient may have achieved a mean diurnal IOP consistently less than 18 mm Hg, whereas the other eye did not, but physicians treat eyes individually, and these results demonstrate that the percentage of eyes that were successful in meeting the clinically relevant secondary outcome measure of IOP less than 18 mm Hg was higher in the fixed-combination treatment group than in the monotherapy groups.
Use of the combination of brimonidine and timolol in a single solution for topical ocular application is simpler and more convenient than administration of the 2 medications sequentially. Because fixed brimonidine-timolol is preserved with 0.005% benzalkonium, whereas 0.5% timolol is preserved with 0.01% benzalkonium and 0.2% brimonidine is preserved with 0.005% benzalkonium, use of the combination product twice daily results in a daily ocular exposure to preservative that is one third of that associated with the use of both component drugs twice daily. This is important because benzalkonium, the preservative most commonly used in ophthalmic medications, may have dose-related harmful effects on the corneal surface.24 The concentration of benzalkonium in the fixed brimonidine-timolol combination (0.005%) is slightly less than that in the fixed dorzolamide-timolol combination (0.0075%). A formulation that allows the delivery of 2 ocular drugs in a single drop also ensures full exposure to both medications by preventing dilution of the first drug by application of a second drop before the first has been absorbed.25
CONCLUSIONS
Fixed brimonidine-timolol administered twice daily provides an effective and convenient therapeutic alternative for the treatment of glaucoma and ocular hypertension. The fixed combination of brimonidine and timolol twice daily provided greater IOP-lowering efficacy than either 0.5% timolol twice daily or 0.2% brimonidine 3 times daily used as monotherapy. The fixed combination was better tolerated than brimonidine monotherapy but less well tolerated than timolol monotherapy.
AUTHOR INFORMATION
Correspondence: Mark B. Sherwood, MD, Department of Ophthalmology, University of Florida College of Medicine, PO Box 100284, Gainesville, FL 32610-0284.
Submitted for Publication: September 2, 2005; final revision received March 27, 2006; accepted March 31, 2006.
Author Contributions: Dr Sherwood had independent full access to all the data in the study and was involved in the development of the manuscript. Dr Whitcup had access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Combigan Study Group I Investigators: John Brennan, MD; David Brodstein, MD; Robert Caine, MD; Neil Choplin, MD; Richard Evans, MD; David Gendelman, MD; Hebert Gould, MD; Donald Kellum, MD; Wallace Landholm, MD; Robert Noecker, MD; Scott Pastor, MD; Aron Rose, MD; Michael Rotberg, MD; Michael Savitt, MD; Howard Schenker, MD; Elizabeth Sharpe, MD; John Sheppard, MD; Mark B. Sherwood, MD; Robert Shields, MD; Steven Simmons, MD; Alfred Solish, MD; Richard Sturm, MD; Sidney Weiss, MD; Jeffrey Whitsett, MD; Robert Williams, MD; and Brandon Wool, MD.
Combigan Study Group II Investigators: Mark Abelson, MD; Richard Bennion, MD; James Branch, MD; David Cooke, MD; E. Randy Craven, MD; Andrew Dahl, MD; Douglas Day, MD; Jon Dietlein, MD; Monte Dirks, MD; Harvey B. DuBiner, MD; Efraim Duzman, MD; David Godfrey, MD; Leonard Gurevich, MD; Melvyn Koby, MD; Martin Leopold, MD; Norman Levy, MD; Jeffrey Lozier, MD; David McGarry, MD; James McManus, MD; Thomas Mundorf, MD; Arnold Prywes, MD; Patrick Riedel, MD; Kenneth Sall, MD; Joseph Sokol, MD; George Spaeth, MD; Dara Stevenson, MD; Thomas Walters, MD; Stephan Whiteside, MD; and David Wirta, MD.
Financial Disclosure: Dr Sherwood has consultant agreements with Allergan Inc, Alcon Laboratories Inc, and Pfizer Inc and has received grant support for clinical trials from Allergan Inc and Alcon Laboratories Inc. Dr Craven receives speaker fees from Allergan Inc, Alcon Laboratories Inc, and Pfizer Inc and has participated in clinical trials for Allergan Inc, Alcon Laboratories Inc, Pfizer Inc, and Merck & Co Inc. Dr DuBiner has been a speaker for Allergan Inc and Alcon Laboratories Inc and has received grant support for clinical trials from Allergan Inc and Alcon Laboratories Inc. Drs Chou, Schiffman, and Whitcup and Ms Batoosingh are employees of Allergan Inc.
Funding/Support: This study was funded by Allergan Inc.
Role of the Sponsor: Allergan Inc had a primary role in the design of the study and in the management and analysis of the data. Allergan Inc monitored the conduct of the study and was involved with Dr Sherwood in the interpretation of the data and in the preparation, review, and approval of the manuscript.
Acknowledgment: We thank Kate Ivins, PhD, for her assistance with the writing of this manuscript.
Author Affiliations: Department of Ophthalmology, University of Florida College of Medicine, Gainesville (Dr Sherwood); Glaucoma Consultants of Colorado, Littleton (Dr Craven); Allergan Inc, Irvine, Calif (Drs Chou, Schiffman, and Whitcup and Ms Batoosingh); and Clayton Eye Center, Morrow, Ga (Dr DuBiner).
REFERENCES
| |
1. Kass MA, Heuer DK, Higginbotham EJ, et al. The Ocular Hypertension Treatment Study: a randomized trial determines that topical medication delays or prevents the onset of primary open-angle glaucoma. Arch Ophthalmol. 2002;120:701-713.
FREE FULL TEXT
2. Collaborative Normal-Tension Glaucoma Study Group (CNTG). Comparison of glaucomatous progression between untreated patients with normal-tension glaucoma and patients with therapeutically reduced intraocular pressures [published correction appears in Am J Ophthalmol. 1999;127:120]. Am J Ophthalmol. 1998;126:487-497.
FULL TEXT
|
ISI
| PUBMED
3. AGIS Investigators. The Advanced Glaucoma Intervention Study (AGIS), 7: the relationship between control of intraocular pressure and visual field deterioration. Am J Ophthalmol. 2000;130:429-440.
FULL TEXT
|
ISI
| PUBMED
4. Heijl A, Leske C, Bengtsson B, Hyman L, Bengtsson B, Hussein M, Early Manifest Glaucoma Trial Group. Reduction of intraocular pressure and glaucoma progression: results from the Early Manifest Glaucoma Trial. Arch Ophthalmol. 2002;120:1268-1279.
FREE FULL TEXT
5. Noecker RS, Dirks MS, Choplin NT, Bernstein P, Batoosingh AL, Whitcup SM. A six-month randomized clinical trial comparing the intraocular pressure-lowering efficacy of bimatoprost and latanoprost in patients with ocular hypertension or glaucoma. Am J Ophthalmol. 2003;135:55-63.
FULL TEXT
|
ISI
| PUBMED
6. Parrish RK, Palmberg P, Sheu WP, XLT Study Group. A comparison of latanoprost, bimatoprost, and travoprost in patients with elevated intraocular pressure: a 12-week, randomized, masked-evaluator multicenter study. Am J Ophthalmol. 2003;135:688-703.
FULL TEXT
|
ISI
| PUBMED
7. Lichter PR, Musch DC, Gillespie BW, et al, CIGTS Study Group. Interim clinical outcomes in the Collaborative Initial Glaucoma Treatment Study comparing initial treatment randomized to medications or surgery. Ophthalmology. 2001;108:1943-1953.
FULL TEXT
|
ISI
| PUBMED
8. Neufeld AH, Bartels SP, Liu JH. Laboratory and clinical studies on the mechanism of action of timolol. Surv Ophthalmol. 1983;28(suppl):286-292.
|
