Prostaglandin Analogues for Ophthalmic Use: A Review of the Comparative Clinical Effectiveness and Cost-Effectiveness

Transcription

1 TITLE: Prostaglandin Analogues for Ophthalmic Use: A Review of the Comparative Clinical Effectiveness and Cost-Effectiveness DATE: 30 July 2015 CONTEXT AND POLICY ISSUES Glaucoma is an eye condition in which progressive damage to the optic nerve leads to visual impairment and may cause blindness. 1 The estimated prevalence of glaucoma is 60.5 million people worldwide. 2 In Canada, approximately 400,000 individuals are affected by glaucoma. 3 Mild glaucoma may be asymptomatic but as damage to the optic nerve increases, the individual experiences difficulty with peripheral vision, contrast sensitivity, glare, and light variations, and eventually this affects day to day activities and impacts quality of life. 2 In the healthy eye, the ciliary body secretes aqueous humour into the eye and this fluid is drained out through the trabecular meshwork. 2 If this process is disrupted, the intraocular pressure (IOP) may increase and increased IOP is thought to be associated with development of glaucoma. 2,4 Other factors that may play a role in the development and progression glaucoma include retinal ischemia, and reduced or deregulated blood flow. 4 As IOP is considered to be an important risk factor, most treatments involve lowering the IOP. 5 Treatments include medical intervention, laser treatment, and surgery. Medications include a variety of drug types such as α2-adrenergic agonist, β- adrenergic antagonist, carbonic anhydrase inhibitors and prostaglandin analogues. 6 The prostaglandin analogues include bimatoprost, latanaprost, travoprost and tafluprost. They are used as monotherapy or in combination with other drugs such as timolol, which is a β- adrenergic antagonist. Considering limited health care resources and variation in the costs of these prostaglandin analogues with the availability of generic brands for latanoprost and travoprost, an assessment of the clinical effectiveness and cost-effectiveness of these drugs is needed to assist with decision making with respect to the use of these drugs. A previous health technology assessment report 7 published by CADTH in 2007 included a comparison of prostaglandin analogues with alternative treatments. It reported that not all prostaglandin analogues were the same and that there could be cost savings with prostaglandin analogues depending on the alternatives used. The report concluded that as first line therapy, bimatoprost, travoprost, and latanoprost demonstrated statistically significant reduction in IOP compared with timolol, Disclaimer: The Rapid Response Service is an information service for those involved in planning and providing health care in Canada. Rapid responses are based on a limited literature search and are not comprehensive, systematic reviews. The intent is to provide a list of sources of the best evidence on the topic that CADTH could identify using all reasonable efforts within the time allowed. Rapid responses should be considered along with other types of information and health care considerations. The information included in this response is not intended to replace professional medical advice, nor should it be construed as a recommendation for or against the use of a particular health technology. Readers are also cautioned that a lack of good quality evidence does not necessarily mean a lack of effectiveness particularly in the case of new and emerging health technologies, for which little information can be found, but which may in future prove to be effective. While CADTH has taken care in the preparation of the report to ensure that its contents are accurate, complete and up to date, CADTH does not make any guarantee to that effect. CADTH is not liable for any loss or damages resulting from use of the information in the report. Copyright: This report contains CADTH copyright material and may contain material in which a third party owns copyright. This report may be used for the purposes of research or private study only. It may not be copied, posted on a web site, redistributed by or stored on an electronic system without the prior written permission of CADTH or applicable copyright owner. Links: This report may contain links to other information available on the websites of third parties on the Internet. CADTH does not have control over the content of such sites. Use of third party sites is governed by the owners own terms and conditions.

2 however, for bimatoprost the evidence was limited to one small low quality trial. The report also stated that from the standpoint of cost-effectiveness, timolol would be the preferred choice in patients for whom timolol is not contraindicated. Since then, several studies on the comparison of the different prostaglandin analogues have been published. This current rapid response report is focussed on prostaglandin analogues, including bimatoprost. The purpose of this report is to review the comparative clinical effectiveness and cost effectiveness of bimatoprost versus other prostaglandin analogues for the treatment of glaucoma and ocular hypertension. RESEARCH QUESTIONS 1. What is the comparative clinical effectiveness of bimatoprost versus other prostaglandin analogues for ophthalmic use? 2. What is the cost effectiveness of bimatoprost versus other prostaglandin analogues for ophthalmic use? KEY FINDINGS Overall, there was a suggestion that bimatoprost was better or equivalent to travoprost and latanoprost in reducing intraocular pressure (IOP), and was associated with numerically higher incidences of hyperemia, hence definitive conclusions could not be made. There were inconsistencies in the findings of the economic studies and definitive conclusions were not possible. METHODS Literature Search Strategy A limited literature search was conducted on key resources including Medline, Embase, PubMed, The Cochrane Library, University of York Centre for Reviews and Dissemination (CRD) databases, Canadian and major international health technology agencies, as well as a focused Internet search. No filters were applied to limit the retrieval by study type. Where possible, retrieval was limited to the human population. The search was also limited to English language documents published between January 1, 2010 and June 30, Selection Criteria and Methods One reviewer screened citations and selected studies. In the first level of screening, titles and abstracts were reviewed and potentially relevant articles were retrieved and assessed for inclusion. The final selection of full-text articles was based on the inclusion criteria presented in Table 1. Prostaglandin Analogues for Ophthalmic Use 2

3 Population Intervention Comparator Outcomes Study Designs Table 1: Selection Criteria Adults with ophthalmic conditions requiring treatment for increased intraocular pressure (i.e., glaucoma) Bimatoprost (e.g., Lumigan) as monotherapy or combination therapy Other prostaglandin analogues for ophthalmic use, in monotherapy or combination therapy: Latanoprost (e.g., Xalatan) Tafluprost (e.g., Zioptan) Travoprost (e.g., Travatan) Q1: comparative clinical effectiveness (clinical benefit, harm, safety) Q2: cost effectiveness Health technology assessment (HTA), systematic review (SR), metaanalysis (MA), and randomized controlled trial (RCT), non-randomized study (NRS), and economic studies Exclusion Criteria Studies were excluded if they did not satisfy the selection criteria, if they were duplicate publications, or were published prior to Non-randomized studies were not considered if there were several RCTs identified as the RCTs are considered as a higher level of evidence than non-randomized studies. RCTs comparing monotherapy with monotherapy or combination therapy with combination therapy were included and RCTs comparing monotherapy with combination therapy were excluded. Comparison of monotherapy with combination therapy would not give a true comparison between two prostaglandins due to the presence of the additional drug in the combination therapy. For studies with multiple publications, only the most recent or most comprehensive one was included. RCTs that were included in a selected SR were excluded. SRs, that included studies that were already included in a more recent or comprehensive SR, were excluded. Economic studies reporting only on direct costs were excluded. Critical Appraisal of Individual Studies Critical appraisal of a study was conducted based on an assessment tool appropriate for the particular study design. The AMSTAR checklist 8 was used for systematic reviews; the Downs and Black checklist 9 for RCTs, and the Drummond checklist 10 for economic studies. For the critical appraisal, a numeric score was not calculated. Instead, the strength and limitations of the study were described narratively. SUMMARY OF EVIDENCE Quantity of Research Available A total of 331 citations were identified in the literature search. Following screening of titles and abstracts, 287 citations were excluded and 44 potentially relevant reports from the electronic search were retrieved for full-text review. One potentially relevant publication was retrieved from the grey literature search. Of these potentially relevant articles, 27 publications were excluded for various reasons, while 18 publications met the inclusion criteria and were included in this Prostaglandin Analogues for Ophthalmic Use 3

4 report. These 18 publications were composed of three systematic reviews, 2,11,12 12 RCTs, and three economic studies Appendix 1 describes the PRISMA flowchart of the study selection. Additional references that did not meet the inclusion criteria but may be of potential interest are included in Appendix 2. Summary of Study Characteristics Characteristics of the included systematic reviews (SRs), randomized controlled trials (RCTs), and economic studies are summarized below and details are provided in Appendices 3 and 4. Systematic review (SR) Three relevant SRs 2,11,12 were identified. One SR 11 of RCTs was published in 2015 from China, one SR 12 which was a SR of SRs was published in 2014 from Croatia and one SR 2 which included SRs as well as RCTs and observational studies, was published in 2012 from USA. One SR 11 assessed the comparative efficacy and tolerability of prostaglandins and timolol combination therapies, one SR 12 assessed the comparative efficacy and tolerability of monotherapies with prostaglandins as well as other drug classes, and one SR 2 had a broad objective and assessed the comparative efficacy and safety of medical (including prostaglandins and other drug classes) and surgical treatments for patients with ocular hypertension or glaucoma. Outcomes reported in all three SRs included intra ocular pressure and hyperemia. Randomized controlled trial Twelve relevant RCTs, were identified. Two RCTs were published from Canada in 2010, 22 and Three RCTs were published from USA in 2010, 14,24 and Two RCTs were published from India in 2013, 18 and One RCT each was published from UK in 2010, 22 from Brazil in 2013, 17 from Egypt in 2010, 20 from Malaysia in 2012, 13 and from Turkey in The RCTs include patients with glaucoma and ocular hypertension (OH) and who were treatment naïve, 15-17,21 or had received prior medication, 13,14,19,23,24 or were a mixed population. 18,20,22 Mean ages of the patients ranged between 46 years and 68 years, and the proportion of females ranged between 37% and 64%. The study duration ranged between 6 weeks and 6 months except for one RCT 16 in which the duration was 12 months. Six RCTs 15,16,19,21-23 compared bimatoprost, travoprost and latanoprost monotherapies, four RCTs 14,18,20,24 compared bimatoprost and travoprost monotherapies and two RCTs 13,17 compared bimatoprost plus timolol and travoprost plus timolol combination therapies. All the RCTs reported on intraocular pressure (IOP). Types of adverse events reported varied among the RCTs. Adverse events included hyperemia, ocular itching, and effects on tear function and ocular surface. Hyperemia was the most commonly reported adverse event. Economic studies Three relevant economic studies were identified. One study 26 was a cost-effectiveness study and was published in 2012 from UK. It used a Markov model and assessed treatments with bimatoprost, travoprost or latanoprost in patients with mild to moderate glaucoma or OH. The Markov model was based on the assumption that treatment switching occurred due to three triggers: lack of tolerance (using hyperthermia as a Prostaglandin Analogues for Ophthalmic Use 4

5 proxy), not achieving target IOP, and progression of glaucoma. The perspective was that of the UK National Health Services (NHS) payer, the time horizon was 10 years and discounting was considered to be 3.5%. Costs considered included drug costs, costs for follow up visits and other costs such as costs for surgery if needed, and long-term costs related to low vision. Considering life expectancy and assuming that the average age of the patient at initiation of first line therapy was 65 years, the 10 year timeframe was selected. The study reported on average cumulative cost per patient and quality adjusted life years (QALY). One study 27 conducted a cost offset analysis and was published in 2011 from USA. It used a Markov cohort model and assessed treatments exclusively with bimatoprost, travoprost or latanoprost in patients with primary open angle glaucoma. The perspective was that of the US payer, the time horizon was seven years and discounting was considered to be 3%. Costs considered included drug costs, costs for follow up visits and other costs such as costs for laser therapy or surgery if needed. Details as to why a seven year timeframe was selected, was not provided. The study reported on cost savings due to delayed or avoided progression. One study 25 was a cost consequence analysis and was published in 2011 from France. It assessed treatments with bimatoprost followed by bimatoprost plus timolol (BBT), travoprost followed by travoprost plus timolol (TTT), and latanoprost followed by latanoprost plus timolol (LLT) in newly diagnosed patients with either glaucoma or OH. The perspective was that of the UK NHS payer, the time horizon was 36 months. Costs considered included drug costs, costs for medical visits, and costs for laser therapy or surgery, if needed. This economic analysis was based on resource utilization during successful treatment. A particular treatment was considered a failure if there was a prescription change or if a patient needed to have laser therapy or surgery. This was a retrospective study and there was no discounting. It reported on treatment persistence and the average monthly glaucoma costs according to treatment sequence. Summary of Critical Appraisal The strengths and limitations of the included SRs, RCTs, and economic studies are summarized below and details are provided in Appendix 5 Systematic review Three relevant SRs 2,11,12 were identified. In all three SRs the objective and the inclusion criteria were stated, a comprehensive literature search using multiple databases was conducted, and the study selection was done in either duplicate or triplicate. Data extraction was done by one author and checked by another author in one SR, 2 done in duplicate in one SR, 11 and in triplicate in one SR. 12 Two SRs 2,12 provided lists of excluded studies and one SR did not. 11 In all three SRs, quality assessments were conducted. In one SR, 11 all the included RCTs were assessed to be of adequate quality and were pooled, In one SR, 12 the quality of the included SRs were assessed to be of low to moderate quality and the findings from only the SRs of moderate quality were presented. In one SR, 2 the quality of the included RCTs and observational studies were assessed to be of fair to good quality. However, for all the SRs, the quality of the evidence was not explicitly stated when the conclusion was presented. In two SRs 11,12 publication bias was explored and there appeared to be no concerns and in one SR 2 it was unclear if publication bias was explored. In all three SRs, the authors declared there was no conflict of interest. Prostaglandin Analogues for Ophthalmic Use 5

6 Randomized controlled trial Twelve relevant RCTs, were identified. Objectives were stated, inclusion criteria were stated, and patient characteristics, interventions and outcomes were described in all the RCTs. Exclusion criteria was described in all RCTs except one. 22 Randomization was described in six RCTs and appeared to be appropriate in five RCTs, 13,15,17,23,24 inappropriate in one RCT, 20 where the two drugs were alternately given to the patients and randomization details were lacking in six RCTs. 14,16,18,19,21,22 Sample size calculations were described in six RCTs, 13,18,19,21,22,24 lacked details in one RCT, 17 and were not described in five RCTs ,20,23 The RCTs that conducted sample size calculations used the appropriate sample size to detect a difference at the desired power level. Numbers of patients who discontinued, withdrew, or were lost to follow up, were mentioned in nine RCTs, 13-15,17,19-21,23,24 and not mentioned in three RCTs. 16,18,22 The number lost to follow up or withdrawals were generally low except for three RCTs. 17,20,21 where withdrawal or lost to follow up, were somewhat high and not balanced between the treatment arms. This could impact the results either positively or negatively. Intention-to-treat (ITT) analysis was conducted in six RCTs, 13,14,18,19,21,24 and was unclear in six RCTs ,20,22,23 Study authors stated that there were no conflicts of interest in five RCTs, 13,15-17,20 conflict of interest was not mentioned in three RCTs 18,22 and the study was either sponsored by the drug manufacturer or some of the study authors were employees of the drug company in four RCTs. 14,19,21,23,24 All the RCTs had restrictive inclusion and exclusion criteria hence generalizability was limited to the study population. Economic studies Three relevant economic studies were identified. All three studies stated the objective, treatment strategies, perspective, time horizon, and sources of clinical and cost data and conclusions were consistent with results. In two studies, 26,27 discounting was considered, and sensitivity analysis was conducted but no incremental analysis was conducted. One study 25 was a retrospective study. One study 27 used a seven year time frame but did not provide a rationale for it. In all the studies it was unclear if the impact of side effects and adherence to drug use and the associated costs were considered. Summary of Findings The findings are summarized below and the details of SRs and RCTs are presented in Appendix 6 and details of economic studies are presented in Appendix 7. What is the comparative clinical effectiveness of bimatoprost versus other prostaglandin analogues for ophthalmic use? Systematic reviews (SR) Three relevant SRs 2,11,12 were identified and are described below. One SR 11 compared prostaglandin and timolol fixed combinations for treating patients with glaucoma and included a meta-analysis. It reported a statistically significantly greater reduction in IOP with bimatoprost and timolol fixed combination (BTFC) compared with travoprost and timolol fixed combination (TTFC) or latanoprost and timolol fixed combination (LTFC). For diurnal IOP reduction, the weighted mean difference (WMD) and the corresponding 95% confidence interval (CI) was 1.94 (0.19 to 3.68) for BTFC compared with TTFC and 0.88 (0.42 Prostaglandin Analogues for Ophthalmic Use 6

7 to 1.33) for BTFC compared with LTFC. The 95% CI does not include zero, which indicates a statistically significant difference. Incidence of conjunctival hyperemia was higher in the BTFC group compared with both TTFC and LTFC groups, but did not achieve statistical significance in case of BTFC compared with TTFC (OR 1.65; 95% CI 0.48 to 5.70) but did for BTFC compared with LTFC (OR 1.85; 95% CI 1.09 to 3.13). One SR 12 of SRs compared monotherapies for the treatment of patients with open angle glaucoma (OAG) or OH. Two included SRs reported greater reduction in IOP with bimatoprost compared with latanoprost or travoprost. However, the difference was between the limits of -1.0 and +1.0 mm Hg which was considered as reasonable limits of therapeutic equivalence. One SR reported no difference in IOP between bimatoprost and travoprost. One SR reported approximately 5-fold and 3-fold higher odds of occurrence of hyperemia with bimatoprost and travoprost respectively compared with latanoprost. Similar estimates were reported by another SR and two other SRs reported smaller odds (1.5 to 2.0- fold).this overview of SRs, based on moderate level evidence suggested that latanoprost had the best trade-off between efficacy and harms for treating patients with glaucoma or OH. One SR 2 with a broad objective, evaluating medical, laser, and surgical treatments for OAG, included a section on prostaglandins. This SR included SRs as well as individual RCTs and observational studies that were not already included in the included SRs. In this overview, one SR reported greater efficacy with bimatoprost compared with latanoprost and travoprost whereas one SR reported that the three drugs were equivalent and one SR reported greater efficacy with bimatoprost compared with travoprost. In this overview, two SRs reported a greater occurrence of hyperemia with bimatoprost compared with both travoprost and latanoprost, one SR reported a greater occurrence of hyperemia with bimatoprost compared with travoporst, and two SRs reported a greater occurrence of hyperemia with bimatoprost compared with latanoprost. In this overview, one RCT showed that bimatoprost reduced IOP to a greater extent compared with either travoprost or latanoprost, one RCT showed there was no difference between the three drugs and one RCT showed there was no difference in IOP reduction between bimatoprost and latanoprost. Details are available in Appendix 5. In this overview, one RCT showed that eye irritation was higher with bimatoprost compared with latanoprost (20% versus 15%), ocular surface disease was similar in both groups (7.5%), and one observational study showed that periocular pigmentation was higher with bimatoprost compared with latanoprost (5.8% versus 0.8%). Randomized controlled trials Twelve relevant RCTs, were identified. Comparison of Bimatoprost, Travoprost, and Latanoprost Monotherapies: Six RCTs 15,16,19,21-23 comparing bimatoprost, travoprost, and latanoprost as monotherapies, reported on IOP (Table 2). Two RCTs 16,22 reported that IOP was statistically significantly reduced from baseline in all three group. One RCT 19 reported that in patients previously treated for at least one month with latanoprost, IOP was statistically significantly reduced from baseline in the bimatoprost and travoprost groups but not in the group that continued with latanoprost treatment. One RCT 15 reported statistically significantly reduction in IOP with bimatoprost compared with both travoprost and latanoprost. Three RCTs 19,21,23 reported that reductions in IOP were not statistically significant different between the three treatment groups. Prostaglandin Analogues for Ophthalmic Use 7

8 Table 2: Intra-ocular Pressure (IOP) with Bimatoprost, Travoprost and Latanoprost Monotherapies Study Outcome Bimatoprost (B) Travoprost (Tr) Latanoprost (L) P-values Aydin Kurna, IOP IOP was significantly reduced from baseline in all the groups < 0.05 Mishra, Crichton, Birt, , Faridi, Whitson, Reduction in IOP ( mean ± SD) (mm Hg), at 12 weeks IOP (mm Hg) change from baseline, at 12 weeks a IOP (mm Hg) (mean ± SD),at 24 weeks IOP (mm Hg) (mean ± SD), % reduction at 6 months IOP (mm Hg) (mean ± SD) at 3 months 8.8 ± ± ± 1.1 For B vs Tr, < 0.001; for B vs L, < 0.001, and for Tr vs L, (P = 0.02) b -1.0 (P = 0.001) b -0.1 (P = 0.78) b 0.24 for between groups at 12 weeks 17.9 ± 5.2 c 18.2 ± 3.8 c 17.7 ± 3.4 c NR (P-values for between group ) ± 3.80, 35% ± 3.71, 30% ± 4.14, 29% 0.15, for between groups 15.6 ± ± ± for between groups IOP = intraocular pressure, SD = standard deviation a Note: baseline IOP is after treatment with latanoprost for at least 1 month b P values for comparison with baseline IOP. c P < compared with baseline IOP Five RCTs 15,19,21-23 comparing bimatoprost, travoprost, and latanoprost as monotherapies, reported on hyperemia (Table 3). Three RCTs 19,22,23 reported that incidence of hyperemia was not statistically significantly different between the three treatment groups. Two RCTs reported numerically higher incidences of hyperemia in the bimatoprost group compared with the travoprost and latanoprost groups, however it was unclear if the differences were statistically significant Prostaglandin Analogues for Ophthalmic Use 8

9 Table 3: Hyperemia with Bimatoprost, Travoprost and Latanoprost Monotherapies Study Outcome Bimatoprost Travoprost Latanoprost P-values (B) (Tr) (L) Mishra, Number (%) of patients with hyperemia, 12 7 (24.1) 4 (12.9) 2 (6.4) NR Crichton, Birt, , Faridi, Whitson, weeks Hyperemia score, 12 weeks Hyperemia, 24 weeks Number of patients with hyperemia, 6 months Hyperemia score (mean ± SD) NR = not reported, SD = standard deviation 0.42 ± ± ± The incidences of hyperemia, were not 0.40 statistically significantly different between the different treatments NR 0.80 ± ± ± Three RCTs 15,19,21-23 comparing bimatoprost, travoprost, and latanoprost as monotherapies, reported on tear break up time (TBUT) (Table 4).Two RCTs 19,23 reported that there were no statistically significantly differences in TBUT between the three treatment groups and one RCT 16 did not report on statistical significance. Table 4: Tear Break-up Time (TBUT) with Bimatoprost, Travoprost and Latanoprost Monotherapies Study Outcome Bimatoprost Travoprost Latanoprost P-values (B) (Tr) (L) Aydin TBUT (mean 16.5 ± ± ± 4.15 NR Kurna, ± SD) (sec) at beginning TBUT (mean ± SD) (sec) at ± ± ± 2.87 NR Crichton, Whitson, months TBUT (mean ± SD) (sec) at baseline TBUT (mean ± SD) (sec) at 12 weeks TBUT (mean ± SD) (sec) at baseline TBUT (mean ± SD) (sec) at 3 months NR = not reported, SD = standard deviation 9.7 ± ± ± (between groups) 9.7 ± ± ± (between groups) 9.1 ± ± ± (between groups) 9.7 ± ± ± (between groups) Prostaglandin Analogues for Ophthalmic Use 9

10 Comparison of Bimatoprost and Travoprost, Monotherapies: Four RCTs 14,18,20,24 comparing bimatoprost and travoprost as monotherapies reported on IOP (Table 5). Two RCTs 18,24 reported that reductions in IOP were statistically significantly greater with bimatoprost compared with travoprost treatment. One RCT 14 reported that travoprost was non inferior to bimatoprost in reducing IOP. One RCT 20 reported that IOP after treatment with bimatoprost or travoprost were not statistically significantly different. Table 5: IOP with Bimatoprost and Travoprost Monotherapies Study Outcome Bimatoprost (B) Travoprost (Tr) P-values Dubiner, , IOP (mm Hg) 17.2 ± ± 2.7 NR USA (mean ± SD), 6 Chander, , India Kammer, Macky, weeks Mean difference in IOP (mm Hg), mean and 95% CI, Reduction in IOP from baseline (%), 12 weeks Diurnal IOP reduction from baseline (mean ± SD) (mm Hg), 3 month IOP (mm Hg) (mean ± SD), 6 months NR = not reported, SD = standard deviation The between group difference in IOP : NR 0.22 and 95% CI: to 0.67 and as the upper limit of 95% CI was 1.5, the effect of Tr was considered to be noninferior to B (between groups) 2.1 ± ± (between group) ± ± (between group) Four RCTs 14,18,20,24 comparing bimatoprost and travoprost as monotherapies reported on hyperemia or ocular redness (Table 6). The incidence of hyperemia was numerically higher with bimatoprost compared with travoprost, however it was unclear if the difference was statistically significant. Ocular redness was numerically higher with bimatoprost compared to travoprost in one RCT, 20 however it was unclear if the difference was statistically significant. Ocular redness was numerically lower with bimatoprost compared to travoprost in one RCT 18 however the difference was not statistically significant. Table 6: Hyperemia with Bimatoprost, and Travoprost monotherapies Study Outcome Bimatoprost (B) Travoprost (Tr) P values Dubiner, Patients with Mild: 39 (16/41) Mild: 31 (13/42) NR hyperemia % (ratio) Moderate: 2 (1/41) Severe: 0 Moderate: 0 Severe: 0 Chander, Ocular redness NS (%) Kammer, Patients with 4 (3.1) 2 (1.5) NR hyperemia, n (%) Macky, Ocular redness, n 7 6 NR n = number of patients, NR = not reported, NS = not significant Prostaglandin Analogues for Ophthalmic Use 10

11 Comparison of Combination Therapies with Bimatoprost, Travoprost, or Latanoprost, each in combination with Timolol: One RCT 17 comparing bimatoprost, travoprost, and latanoprost as combination therapies, each in combination with timolol, and one RCT 13 comparing bimatoprost and timolol with travoprost and timolol, reported on IOP (Table 7). It was unclear if there were any statistically significant differences between the treatment groups with respect to IOP reduction. Table 7 : IOP with Bimatoprost, Travoprost or Latanoprost in Combination with Timolol Study Outcome Bimatoprost Travoprost+ Latanoprost+ P values +Timolol (BT) Timolol (TT) Timolol (LT) Russ, a IOP (mean ± SD) ( mm Hg), 3 months ± 2.96 P = ± 0.44 P = ± 3.11 P = 0 P values compared to baseline Lee, Diurnal IOP (mean ± SD) ( mm Hg), 8 weeks ± 3.39 P = ± 3.48 P = NA = not applicable, SD = standard deviation a Fixed combinations of bimatoprost with timolol and travoprost with timolol NA P values compared to baseline One RCT 13 comparing bimatoprost and travoprost as combination therapies with timolol, reported on hyperemia (Table 8). The incidence of worsening hyperemia was numerically higher with the bimatoprost plus timolol combination therapy compared with travoprost plus timolol combination therapy, however it was unclear in the difference was statistically significant. Table 8: Hyperemia with Bimatoprost, and Travoprost combination therapies with Timolol Study Outcome Bimatoprost Travoprost+Timolol P values +Timolol (BT) (TT) Lee, Hyperemia worse, n (%) 5 (23.8) 2 (10.0) NR n = number of patients, NR = not reported, a Fixed combination of bimatoprost with timolol and travoprost with timolol Details of less frequently reported adverse events are available in Appendix 6. What is the cost effectiveness of bimatoprost versus other prostaglandin analogues for ophthalmic use? Three relevant economic studies were identified. One economic study 26 was a cost-effectiveness study. It used a Markov model and reported the costs per patient by treatment strategy, considering the time duration when the patient was stable on therapy, the time duration on first line, second line and third line therapy, vision status, follow-up visits, other procedural costs and cost of low vision (Table 9). Cost details are available in Appendix 7. Drug acquisition costs were not a key driver of the total cost of glaucoma management. The total long term costs with all the three prostaglandins were similar. Increased drug costs were offset by fewer clinic visits to initiate treatment switches and by avoiding surgery or costs for managing low vision. Treatment switches were triggered by Prostaglandin Analogues for Ophthalmic Use 11

12 adverse events and lack of tolerance, or failure to achieve target IOPs or progression in visual field defect. Considering reduced hyperemia as a proxy for tolerance, the latanoprost strategy would have longer intervals between the need to switch therapy. For every 1,000 clinic appointments, 719 patients can be managed on latanoprost-based therapy per year compared with 586 and 568 patients on bimatoprost- and travoprost-based therapies respectively. Table 9: Average cumulative results per patient by treatment strategy Analysis Item Bimatoprost (B) First line Travoprost (Tr) First line Latanoprost (L) First line Base case Total cost ( ) 6, , , QALY Cost/ QALY a Sensitivity analyses Low risk patients Total cost ( ) 3, , , QALY Cost/ QALY a High risk patients Total cost ( ) 8, , , QALY Cost/ QALY a QALY = quality adjusted life years a Calculated by CADTH reviewer One economic study 27 conducted a cost-offset analysis using a hypothetical health plan model. Considering the 1 mm Hg IOP reduction advantage of bimatoprost, the estimated number of prostaglandin treated glaucoma patients who would develop visual field defect or glaucomatous progression over 7 years was 630 if all the patients were treated with bimatoprost and 766 if all the patients were treated with either travoprost or latanoprost. The cost savings per delayed or avoided progression were estimated to be US$4,009 in case of patients with early glaucoma and US$4,543 in case of patients with advanced glaucoma. The total cost savings from delayed or avoided progression with exclusive treatment with bimatoprost instead of travoprost or latanoprost were US$545,224 and US$617,848 in the case of early glaucoma and advanced glaucoma respectively. Details are available in Appendix 7. One economic study 25 was a cost consequence analysis. The analysis considered resources consumed during treatment and following failure. Treatment failure was experienced by 30.0% of patients on BBT, 25.4% of patients on TTT and 51.6% of patients on LLT. The average monthly total costs were 20.64, 17.21, and for the treatment sequences BTT, TTT, and LTT respectively. Costs were statistically significantly lower with TTT than with LLT or BBT (P = for TTT vs BBT and P < for TTT vs LLT) Limitations The two overviews of SRs had an overlap of four SRs, hence results were not completely exclusive. All the RCTs had restrictive inclusion and exclusion criteria hence generalizability was limited to the study population. Exclusion criteria also varied across studies. Examples of exclusion Prostaglandin Analogues for Ophthalmic Use 12

13 criteria include history of intraocular surgery, patients currently using contact lenses, pregnant women, eyelid deformity, and progressive retinal or optic nerve disease. Some studies presented IOP measurements over the circadian cycle whereas other studies presented results at one or two time points during the day. The IOP varies during the day hence IOP determinations at one or two time points may not capture the IOP changes over the entire day. In addition IOP results were presented in different formats such as final value after treatment, change from baseline and percentage change. Adverse event reporting varied across studies. Not all RCTs reported all outcomes. Hence comparison across studies was difficult. The extent of compliance with study medication was not reported hence its impact on the results was unclear. Active drug as well as preservatives that are used with some of the drug formulations may affect ocular surface. Some studies included preservatives in one treatment arm but not in the other treatment arm and in some studies different preservatives were used in the different treatment arms. Hence it was unclear to what extent preservatives impacted the results. The majority of studies were of duration six months or less hence long term effects are not known. No studies on the comparison of bimatoprost with tafluprost were identified. The economic studies used different methods of analyses, hence comparison of results was not possible. The studies were based on several assumptions hence the results need to be interpreted with caution. There were inconsistencies in the findings of the economic studies and definitive conclusions were not possible. None of the economic studies pertained to a Canadian setting, hence generalizability of the findings in the Canadian context was unclear. CONCLUSIONS AND IMPLICATIONS FOR DECISION OR POLICY MAKING Eighteen relevant studies, comprising of one SR 11 and two SR of SRs, 2,12 twelve RCTs, and three economic studies, were identified. One SR on combination therapies reported statistically significant greater IOP reduction with BTFC compared with TTFC and LTFC and higher incidences of hyperemia with BTFC compared with TTFC and LTFC, though the differences were not statistically significant. Two SR of SRs reported greater reduction in IOP with bimatoprost compared with either travoprost or latanoprost in some instances and equivalence among the three treatments in some instances. One SR of SRs reported a numerically higher incidence of hyperemia with bimatoprost compared with travoprost and latanoprost, however statistical significance of the difference was unclear. Two RCTs reported statistically significant reduction in IOP from baseline with bimatoprost, travoprost and latanoprost, however it was unclear if the difference between treatments were statistically significant. One RCT reported a statistically significant greater reduction in IOP with bimatoprost compared with travoprost or latanoprost. Two RCTs reported greater reduction in IOP with bimatoprost compared with travoprost or latanoprost, but the differences were not statistically significant. Two RCTs on combination therapies of prostaglandins with timolol reported a statistically significant reduction in IOP from baseline but the statistical significance of the differences in IOP reductions between the three treatments were unclear. Generally, in the RCTs, the incidence of hyperemia was numerically higher with Prostaglandin Analogues for Ophthalmic Use 13

14 bimatoprost compared with travoprost and latanoprost, however the statistical significance of the difference was unclear. Overall, there is a suggestion that bimatoprost is better or equivalent to travoprost and latanoprost in reducing IOP, and is associated with a numerically higher incidence of hyperemia hence definitive conclusions are difficult. There were inconsistencies in the findings of the economic studies. A cost-effectiveness study reported that total long term costs with bimatoprost, travoprost and latanoprost were similar. One cost offset study reported cost savings from delayed or avoided glaucoma progression with bimatoprost treatment compared with travoprost and latanoprost treatments. One cost consequence study on monotherapy followed by combination therapy, reported that considering resources consumed due to treatment and following treatment failure, the costs were lower with TTT compared with BBT and LLT. These differences among economic studies, may be due to variation in the population and treatment strategies studied, and the methodologies used. Hence definitive conclusions were not possible. PREPARED BY: Canadian Agency for Drugs and Technologies in Health Tel: Prostaglandin Analogues for Ophthalmic Use 14

15 REFERENCES 1. NIHR Horizon Scanning Centre. Glaucoma disease-specific review [Internet]. Birmingham, United Kingdom: National Institute for Health Research; 2013 Apr. [cited 2015 Jul 15]. Available from: 2. Treatment for glaucoma: comparative effectiveness. Executive summary [Internet]. Rockville (MD): Agency for Healthcare Research and Quality; 2012 Apr. [cited 2015 Jul 15]. (Comparative effectiveness review no. 60). Available from: execsumm.pdf 3. Rock WJ, Buhrmann R, Miller G. Glaucoma [Internet]. Ottawa: The Ottawa Hospital; [cited 2015 Jul 15]. Available from: mcs/programs/eyeinstitute/clinicaldiagnosticservices/glaucoma 4. Tortorice K, Heron B, VHA Pharmacy Benefits Management Strategic Healthcare Group. Drug class review: ophthalmic prostaglandin analogs [Internet]. Washington (DC): U.S. Department of Veterans Affairs; 2011 Jun. [cited 2015 Jul 15]. Available from: ogdrugclassreviewupdate.doc 5. Rigollet JP, Ondategui JA, Pasto A, Lop L. Randomized trial comparing three fixed combinations of prostaglandins/prostamide with timolol maleate. Clin Ophthalmol [Internet] [cited 2015 Jul 6];5: Available from: 6. University of Massachusetts Medical School. Therapeutic class overview ophthalmic prostaglandin analogues [Internet]. Carson City (NV): Nevada Department of Health and Human Services; 2013 Jun 3. [cited 2015 Jul 15]. Available from: _ pdf 7. Hodge WG, Lachaine J, Steffensen L, Murray C, Barnes D, Foerster V, et al. Prostaglandin analogues for ophthalmic use: analysis of clinical and cost-effectiveness [Internet]. Ottawa: CADTH; 2007 Jun. [cited 2015 Jul 28]. (Technology report no 89). Available from: 8. Shea BJ, Grimshaw JM, Wells GA, Boers M, Andersson N, Hamel C, et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Med Res Methodol [Internet] [cited 2015 Jul 13];7:10. Available from: 9. Downs SH, Black N. The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health [Internet] Jun [cited 2015 Jul 13];52(6): Available from: Prostaglandin Analogues for Ophthalmic Use 15

16 10. Drummond MF. Figure 15.5.a: Drummond checklist [Internet]. In: Higgins JPT, Green S, editors. Cochrane handbook for systematic reviews of interventions. Version Oxford, UK: The Cochrane Collaboration; 2011 Mar [cited 2015 Jul 13]. Available from: nd_1996.htm. 11. Lou H, Wang H, Zong Y, Cheng JW, Wei RL. Efficacy and tolerability of prostaglandintimolol fixed combinations: an updated systematic review and meta-analysis. Curr Med Res Opin Jun;31(6): Daka Q, Trkulja V. Efficacy and tolerability of mono-compound topical treatments for reduction of intraocular pressure in patients with primary open angle glaucoma or ocular hypertension: an overview of reviews. Croat Med J [Internet] Oct [cited 2015 Jul 6];55(5): Available from: Lee MY, Teh NC, Nur ZM, Thayanithi S, Jelinar MN, Rizal AM, et al. The effects of fixed combination of bimatoprost-timolol and travoprost-timolol on intraocular pressure in patients with primary open-angle glaucoma or ocular hypertension, previously on nonfixed combination of latanoprost and timolol. Asia Pac J Ophthalmol (Phila) Jul;1(4): DuBiner HB, Hubatsch DA. Late-day intraocular pressure-lowering efficacy and tolerability of travoprost 0.004% versus bimatoprost 0.01% in patients with open-angle glaucoma or ocular hypertension: a randomized trial. BMC Ophthalmol [Internet] [cited 2015 Jul 6];14:151. Available from: Mishra D, Sinha BP, Kumar MS. Comparing the efficacy of latanoprost (0.005%), bimatoprost (0.03%), travoprost (0.004%), and timolol (0.5%) in the treatment of primary open angle glaucoma. Korean J Ophthalmol [Internet] Oct [cited 2015 Jul 6];28(5): Available from: Aydin Kurna S, Acikgoz S, Altun A, Ozbay N, Sengor T, Olcaysu OO. The effects of topical antiglaucoma drugs as monotherapy on the ocular surface: a prospective study. J Ophthalmol [Internet] [cited 2015 Jul 6];2014: Available from: Russ HH, Nogueira-Filho PA, Barros JN, Faria NV, Montiani-Ferreira F, Gomes JA, et al. Ocular surface evaluation in patients treated with a fixed combination of prostaglandin analogues with 0.5% timolol maleate topical monotherapy: a randomized clinical trial. Clinics [Internet] Oct [cited 2015 Jul 6];68(10): Available from: Chander A, Kapoor H, Thomas S. Comparison of the efficacy and safety of bimatoprost (0.03 %) and travoprost (0.004 %) in patients with primary open angle glaucoma. Nepal J Ophthalmol Jan;5(9): Crichton AC, Vold S, Williams JM, Hollander DA. Ocular surface tolerability of prostaglandin analogs and prostamides in patients with glaucoma or ocular hypertension. Adv Ther Mar;30(3): Prostaglandin Analogues for Ophthalmic Use 16

17 20. Macky TA. Bimatoprost versus travoprost in an Egyptian population: a hospital-based prospective, randomized study. J Ocul Pharmacol Ther Dec;26(6): Faridi UA, Saleh TA, Ewings P, Venkateswaran M, Cadman DH, Samarasinghe RA, et al. Comparative study of three prostaglandin analogues in the treatment of newly diagnosed cases of ocular hypertension, open-angle and normal tension glaucoma. Clin Experiment Ophthalmol Oct;38(7): Birt CM, Buys YM, Ahmed II, Trope GE, Toronto Area Glaucoma Society. Prostaglandin efficacy and safety study undertaken by race (the PRESSURE study). J Glaucoma Sep;19(7): Whitson JT, Trattler WB, Matossian C, Williams J, Hollander DA. Ocular surface tolerability of prostaglandin analogs in patients with glaucoma or ocular hypertension. J Ocul Pharmacol Ther Jun;26(3): Kammer JA, Katzman B, Ackerman SL, Hollander DA. Efficacy and tolerability of bimatoprost versus travoprost in patients previously on latanoprost: a 3-month, randomised, masked-evaluator, multicentre study. Br J Ophthalmol [Internet] Jan [cited 2015 Jul 6];94(1):74-9. Available from: Lafuma A, Salmon JF, Robert J, Berdeaux G. Treatment persistence and costeffectiveness of latanoprost/latanoprost-timolol, bimatoprost/bimatoprost-timolol, and travoprost/travoprost-timolol in glaucoma: an analysis based on the United Kingdom general practitioner research database. Clin Ophthalmol [Internet] [cited 2015 Jul 6];5: Available from: Orme M, Collins S, Loftus J. Long-term medical management of primary open-angle glaucoma and ocular hypertension in the UK: optimizing cost-effectiveness and clinic resources by minimizing therapy switches. J Glaucoma Sep;21(7): Berenson KL, Kymes S, Hollander DA, Fiscella R, Burk C, Patel VD. Cost-offset analysis: bimatoprost versus other prostaglandin analogues in open-angle glaucoma. Am J Manag Care Sep;17(9):e365-e374. Prostaglandin Analogues for Ophthalmic Use 17

18 ABBREVIATIONS B bak BBT BT BTFC CI IOP ITT L LLT LT m NA NHS NR OAG OH OR OSD OSDI PGA POAG PP QALY RCT RD RR SD SCH SR tim TBUT Tr TT TTT TTFC UK USA vs wk WMD bimatoprost benzylkonium chloride bimatoprost followed by bimatoprost plus timolol bimatoprost plus timolol bromatoprost plus timolol fixed combination confidence interval intra ocular pressure intention-to-treat latanoprost latanoprost followed by latanaprost plus timolol latanaprost plus timolol month not applicable National Health Services not reported open angle glaucoma ocular hypertension odds ratio ocular surface disease ocular surface disease index prostaglandin analogue primary open angle glaucoma per protocol quality adjusted life year randomized controlled trial risk difference relative risk or risk ratio standard deviation Schimer systematic review timolol tear break up time travoprost travoprost plus timolol travoprost followed by travoprost plus timolol Travoprost plus timolol fixed combination United Kingdom United States of America versus week weight mean difference Prostaglandin Analogues for Ophthalmic Use 18

19 APPENDIX 1: Selection of Included Studies 331 citations identified from electronic literature search and screened 287 citations excluded 44 potentially relevant articles retrieved for scrutiny (full text, if available) 1 potentially relevant report retrieved from other sources (grey literature, hand search) 45 potentially relevant reports 27 reports excluded: -irrelevant comparison (9) -irrelevant outcomes (3) -irrelevant design (2) -study already included in at least one of the selected SRs (4) - all studies in SR already included in at least one of the selected SRs (2) -non-english (1) - SR with irrelevant objective (2) -SR included one relevant RCT so RCT included instead (1) -other (review articles, editorials) (3) 18 reports included in review Prostaglandin Analogues for Ophthalmic Use 19

20 APPENDIX 2: References of potential interest RCTs (Comparison not relevant: monotherapy versus combination therapy): 1. Gutierrez-Diaz E, Silva CJ, Munoz-Negrete FJ, Gutierrez-Ortiz C, Morgan-Warren RJ, Maltman J. Bimatoprost/timolol fixed combination versus latanoprost in treatment-naive glaucoma patients at high risk of progression: a pilot study. Clin Ophthalmol [Internet] [cited 2015 Jul 6];8: Available from: 2. Konstas AG, Hollo G, Mikropoulos DG, Haidich AB, Dimopoulos AT, Empeslidis T, et al. 24- hour efficacy of the bimatoprost-timolol fixed combination versus latanoprost as first choice therapy in subjects with high-pressure exfoliation syndrome and glaucoma. Br J Ophthalmol Jul;97(7): Mesci C, Aydin N, Erbil HH. Twenty-four-hour intraocular pressure control with latanoprosttimolol-fixed combination versus bimatoprost in patients who switched from timolol. J Glaucoma Oct;20(8): Nixon DR. A randomized, prospective study of bimatoprost 0.01% or travoprost/timolol in patients previously treated with latanoprost and timolol to reduce intraocular pressure. J Ocul Pharmacol Ther Dec;29(10): Rossetti L, Sacchi M, Karabatsas CH, Topouzis F, Vetrugno M, Centofanti M, et al. Comparison of the effects of bimatoprost and a fixed combination of latanoprost and timolol on 24-hour blood and ocular perfusion pressures: the results of a randomized trial. BMC Ophthalmol [Internet] [cited 2015 Jul 6];15:7. Available from: Prostaglandin Analogues for Ophthalmic Use 20