- Email: [email protected]
Cost-effectiveness of pegaptanib compared to photodynamic therapy with verteporfin and to standard care in the treatment of subfoveal wet age-related macular degeneration in Canada
Cost-Effectiveness of Pegaptanib Compared to Photodynamic Therapy with Verteporfin and to Standard Care in the Treatment of Subfoveal Wet Age-Related Macular Degeneration in Canada Stephanie R. Earnshaw, PhD1; Yola Moride, PhD2; and Sophie Rochon, MA 3 IRTI Health Solutions, Research Triangle Park, North Carolina; 2Universit~de Montreal, Montreal, Quebec, Canada; and sPfizer Canada Inc., Kirkland, Quebec, Canada ABSTRACT Background: Age-related macular degeneration (AMD) is characterized by loss of central vision and is the leading cause of blindness among persons over the age of 50 years in Canada. The wet form of AMD has 3 subtypes~occult, minimally classic, and predominantly classic. Photodynamic therapy (PDT) with verteporfin is indicated only for the category of predominantly classic wet AMD. Currently, there are no treatments available for the other AMD subtypes. Pegaptanib sodium was the first pharmacologic therapy approved in Canada for the treatment of subfoveal wet AMD regardless of subtype. Objective: The aim of this study was to examine the cost-effectiveness of pegaptanib versus PDT with verteporfin and versus standard care for the treatment of subfoveal wet AMD in patients aged _>65 years in Canada. Methods: A Markov model based on visual acuity in the better-seeing eye was developed. Clinical efficacy was taken from the clinical trials. Costs of treatment, comorbidities (eg, depression, fractures, need for assisted living), vision rehabilitation, visual aids, and adverse events were considered. Costs, utilities, and mortality were estimated from data from the available published literature. Costs were reported in 2004 Canadian dollars, and costs and outcomes were discounted at 3% per annum. Lifetime costs, qualityadjusted life-years (QALYs), and vision years gained (VYGs) were estimated. Sensitivity analyses were performed to determine model robustness. Results: Patients who received pegaptanib experienced more QALYs gained (4.17) and VYGs (3.83) compared with patients who received PDT (3.87 and 3.01, respectively) or standard care (3.96 and 3.26). Mean total costs per patient were greater in patients
who received pegaptanib compared to those who received PDT or standard care ($20,016 vs $15,345 or $7669, respectively). The incremental cost per QALY in patients receiving pegaptanib compared to those receiving PDT was $49,052 and $59,039 for patients receiving pegaptanib versus standard care. The incremental cost per VYG was $20,401 and $21,559 with pegaptanib versus PDT and standard care, respectively. Sensitivity analyses found that the model was relatively robust to changes in various model parameters. Conclusion: The results of this analysis suggest that in Canada, pegaptanib is a cost-effective treatment for subfoveal wet AMD in elderly patients, regardless of lesion subtype, compared to PDT with verteporfin and to standard care. (Clin Ther. 2007;29:20962106) Copyright © 2007 Excerpta Medica, Inc. Key words: age-related macular degeneration, pegaptanib, photodynamic therapy with verteporfin, cost-effectiveness, quality-adjusted life-year.
INTRODUCTION Age-related macular degeneration (AMD) is the leading cause of blindness in developed countries) AMD is a retinal disease that results from deterioration of the macula and is characterized by blurring, distorThis work was presented in abstract form at the Third Annual Canadian Joint Therapeutics Congress, May 10-13, 2006, Toronto, Canada; and the Canadian Ophthalmology Society Annual Meeting and Exhibition, June 21-24, 2006, Toronto, Canada.
Accepted for publication March 8, 2007. doi:l 0.1016/j.clinthera.2007.09.001 0149-2918/$32.00 Printed in the USA. Reproduction in whole or part is not permitted. Copyright © 2007 ExcerptaMedica, Inc.
tion, or loss of central vision. 2 AMD can be categorized into atrophic (dry) or exudative (wet) disease. In wet AMD, abnormal blood vessels form behind the macula and their contents leak into the retina or subretinal space. Three subtypes of wet AMD exist and are classified according to the pattern of lesion as visualized by fluorescein angiography: occult (35 %-73 % of patients), minimally classic (35%), and predominantly classic (20%-44%). 3-s Wet AMD is subclassified according to the location of lesions relative to the fovea3 - s extrafoveal, juxtafoveal, and subfoveal--with the majority of cases (76%-83%) being subfoveal.4 In Canada, >2 million people over the age of 50 years have some form of AMD. 6 Because the incidence of AMD increases with age, 1 this figure is likely to triple over the next 25 years, coincident with the aging population of Canada. 7 Central vision impairment results in diminished ability to perform routine daily activities. 8-11 Patients with AMD have an increased risk for fallsl2,13; fractures14; depressionlS; and deterioration of their independence and emotional and social health. 16-18 The quality of life of patients with AMD is compromised as seriously as that of patients with arthritis, asthma, diabetes, and stroke. 19 Costs incurred from blindness are associated directly with treatment, loss of personal income due to diminished work capacity, and social support services (eg, visual rehabilitation services, social services). 2° Conventional treatments for wet AMD include thermal laser photocoagulation and photodynamic therapy (PDT) with verteporfin21; however, these treatments are not indicated for all subtypes of wet AMD. These treatments slow the progression of vision loss but do not restore a patient's vision to the pre-AMD state. 2,22 Thermal laser photocoagulation 23 is indicated solely for extrafoveal and juxtafoveal wet AMD (15% of wet AMD cases), whereas PDT is indicated only for predominantly classic subfoveal wet AMD. 24 Because most cases of AMD present as the occult form of the disease, 2-s the majority of patients are not eligible for PDT treatment. Thus, an AMD treatment that successfully treats all categories would be invaluable in preventing blindness. Pegaptanib sodium was the first pharmacologic agent approved in Canada for the treatment of subfoveal wet AMD regardless of subtype. 25 Being a vascular endothelial growth factor (VEGF) antagonist, pegaptanib inhibits angiogenesis, vascular permeability, and inflammation of the retina, all of which may
contribute to the progression of the disease. 26 The efficacy and tolerability of pegaptanib in 1186 patients with subfoveal wet AMD were studied in 2 concurrent randomized clinical trials, 27 with a significant treatment benefit compared to standard therapy observed. The economic impact of AMD is not well understood, either in terms of the overall burden of illness or the costs and cost-effectiveness of specific interventions. Using a Markov model projecting lifetime progression and 2-year treatment of wet AMD, populated with data obtained from published clinical trials, Canadian provincial and federal government sources, and clinical experts, the present study examined the cost-effectiveness of pegaptanib versus PDT or standard care for the treatment of subfoveal wet AMD in Canada. The perspective of a Canadian provincial Ministry of Health was used in this analysis, and only direct costs were considered. The analysis was targeted to decision makers involved in public reimbursement of costs of new medicines.
MATERIALS AND METHODS Study Design A Markov model was developed to simulate a cohort of patients presenting with subfoveal wet AMD. The model structure is presented in the figure. In the model, patients transitioned between health states based on visual acuity (VA), as follows: better than 20/40, VA in the better-seeing eye was better than 20/40; 20/40 to >20/80, VA in the better-seeing eye was better than 20/80 but worse than 20/40; 20/80 to >20/200, VA in the better-seeing eye was better than 20/200 but worse than 20/80; 20/200 to >20/400, VA in the better-seeing eye was better than 20/400 but worse than 20/200 (legally blind); and <_20/400, VA in the better-seeing eye was worse than 20/400 (light perception only). Patients transitioned between the health states every 3 months for the remainder of their lifetime. The patient population was stratified by age (group 1, 65-74 years; group 2, >74 years), sex, and lesion subtype (occult, minimally classic, or predominantly classic). Patients received 2 years of treatment with pegaptanib, PDT, and standard care. Treatment of subfoveal wet AMD was considered for a maximum of 2 years in the model. During each model cycle, patients could remain in their current state, or move up by 1 line, indicating a gain in VA, or move down by 1 or 2 lines, indicating
among the VA health states before treatment within the model was obtained from the VEGF Inhibition Study in Ocular Neovascularization (VISION) trials 27 (Table I). VISION consisted of randomized trials of intravitreous injections of pegaptanib compared with standard care in patients _>50 years of age diagnosed with AMD and had a range of best-corrected VA of 20/40 to 20/320 in the study eye and of 20/800 or better in the other eye.
>20/40 in treated eye
20/40 to >20/80 in treated eye
20/80 to >20/200 in treated eye
20/200 to
>20/400 in treated eye
Patients may transition to the death state from any state in the model
_<20/400 in treated eye
Figure. Markov model structure.
a loss in VA. Transitions between VA levels are based on _>3-line gain, 3- to 6-line loss, and >6-line loss on the VA scale. Patients could transition to the death (absorbing) state from any health state in the model. The model is based on bilateral disease in which only the better-seeing eye was treated.
Study Population The Markov model was based on a hypothetical cohort of patients aged _>65 years, classified by the 3 AMD subtypes. Because the perspective of the Canadian provincial Ministry of Health was used for this analysis, the age group used for analysis was aged _>65 years to be in accordance with the majority of beneficiaries of a Canadian provincial Ministry of Health. The proportion of patients in age and sex strata within each AMD subgroup was determined by multiplying the age and sex distribution in the general population 28 by the 1-year incidence rate derived from Klein et al. 29 The distribution of patients
Treatments Study comparators examined included pegaptanib sodium,* PDT with verteporfin,t and standard care, as seen in the VISION, 27 Verteporfin in Photodynamic Therapy (VIP), 21 and Treatment of Age-Related Macular Degeneration with Photodynamic Therapy (TAP) 24,3°,31 clinical trials. Patients receiving pegaptanib or PDT were assumed to have been receiving this treatment regardless of lesion subtype. PDT is indicated in patients with predominantly classic lesions. 24 However, for ethical reasons, patients in all arms of the VISION clinical trials were allowed treatment with PDT at the discretion of the ophthalmologist. 27 Therefore, standard care (ie, the sham or placebo arm in the VISION trials) was defined as treatment with PDT if lesion subtype and/or size qualified for treatment as indicated in the product monograph. The cost of PDT was added in both the pegaptanib and sham arm for the purpose of this economic analysis.
Clinical Efficacy To compare pegaptanib with PDT, clinical efficacy was taken from the published clinical trials. 27 Since there are no direct comparative studies (based on a literature search of MEDLINE [search terms: pegaptanib AND photodynamic therapy, LIMIT clinical trial]) on clinical efficacy for pegaptanib versus PDT treatment, the efficacy data used to estimate the costeffectiveness of pegaptanib versus PDT were derived from the VISION, 27 VIP,21 and TAP 24,30,31 trials. In the VISION trials, the efficacy of pegaptanib was compared with that of standard care (sham injection for patients with occult and minimally classic AMD subtypes or, for patients with predominantly classic *Trademark: Macugen ® ([OSI], Eyetech Inc., New York, New York; Pfizer Canada Inc., licensee, Kirkland, Quebec, Canada). tTrademark: Visudyne ® (Novartis AG, BiJlach, Switzerland).
Table I. Parameters and costs used in the development of the Markov study model. Distribution of patient VA state used in model*
Better than 20/40 20/40 to >20/80 20/80 to >20/200 20/200 to >20/400 _<20/400 Parameter Probability of patients with normal vision needing service,f % Hazard ratio of resource use by vision health state~ Better than 20/40 20/40 to >20/80 20/80 to >20/200, 20/200 to >20/400, ___20/400 Annual cost ofcomorbidity, t Can $ Mean number of PDT treatments per patient
Patients, %
0.34 36.05 46.05 17.06 0.50 Depression/Anxiety
Assisted Living
Fractures
11.10
2.10
3.10
1.0
1.0
1.0
1.20
2.67
1.44
1.29
2.34
1.16
1060
15,701
139
Pegaptanib
Standard Care
Year 1
1.71
2.05
Year 2
1.00
1.54
18.11 6.77
20.59 8.82
Treated, %
Mean Cost, Can $ (2004)
100 100 33 90 100 3 15
343 33 462 818 1801 33 2071
Patients who received PDT, % Year 1 Year 2
AE 21'24'27'30'31 Endophthalmitis Injection site of AE Retinal capillary nonperfusion Retinal detachment Traumatic injury to lens Visual disturbance Vitreous hemorrhage
VA = visual acuity; PDT = photodynamic therapy with verteporfin; AE = adverse event. *Data from Gragoudas el: al. 27 t Source: R~gie de I'Assurance Maladie du Quebec database. Sources: Comorbidil:y costs were obtained from Center for Addiction and Mental Health, Canadian Psychiatric Association and the Canadian Psychological Association, Community Care Access, and Ministry of Health and Long Term Care-Ontario databases.
AMD, sham plus PDT with verteporfin was administered), defined as the sham arm in the VISION trials. In the VIP and TAP trials, PDT with verteporfin was compared with PDT with placebo (injection of 5% dextrose in water). Specifically, transitions between VA health states during years 1 and 2 in patients who received pegaptanib, standard care, PDT, or placebo were based on the percentages of patients experiencing a 3-line gain, 3- to 6-line loss, and >6-line loss on the VA scale as estimated from the clinical end points from the trials for those years, respectively. Threemonth probabilities for the first year were estimated from the first-year clinical probabilities for pegaptanib or PDT assuming an exponential distribution. As such, the 3-month transition probabilities were estimated as: Pn = 1 - (1 - Pro) n/m, where n = 1/4 (3 months in a 12-month period), m = 1 (1 year), Pn = the 3-month transition probability, and P m - the 1-year probability. For the second year, the model estimates the difference between the firstand second-year clinical probabilities (0.2041 and 0.2857, respectively) and then calculates 3-month probabilities using the equation described previously. Annual cumulative probabilities of change in vision for each treatment and lesion subtype as reported in the clinical trials 27 are presented in Table II. After discontinuation, disease progression was assumed to occur similarly to the year-2 disease progression observed in patients receiving standard care/placebo for the respective trials. Age- and sex-specific mortality rates were obtained from Statistics Canada (2003). 28 Relative risk for death due to blindness was 1.37, 9 and the risk for death was allowed to increase with patients' age.
Utility Values Utility values for ocular diseases have been found to correlate more closely with VA in the better-seeing eye. 9 In a study by Brown et al, 9 visual utility scores in patients with ocular disease were determined by VA health states. The utility values were estimated using the time-tradeoff and standard gamble methods. 32 A greater correlation between utility values and VA was found in the better-seeing eye than between the same utility values and VA in the poorer-seeing eye. 9 For this reason, only treatment in the better eye in bilateral disease was assessed.
Cost Data Costs of pegaptanib and PDT were estimated from annual use as reported in the clinical trialsY, 24,27,3° The numbers of pegaptanib treatments were assumed to be 8.4 and 6.9 for the first and second years, respectively. Drug cost included the cost of the drug ($995) and the cost of the procedure ($280). For PDT, the numbers of treatment were 3.4 and 2.2 in the first and second years. Drug cost included the cost of the drug ($1750) and the cost of the procedure ($300). The number of treatments with pegaptanib in years 1 and 2 was obtained from the VISION 27 trial, where patients were randomly assigned to receive either intravitreous injection of pegaptanib or standard care. Patients received a mean of 8.4 and 6.9 pegaptanib treatments in year 1 and 2, respectively. 27 The mean number of treatments in the PDT group was extracted from TAP reports 1 and 2. 24,30 Patients received either placebo or verteporfin IV. 24,30 Patients in the verteporfin-treated group received a mean of 3.4 treatments per participant in the first year and 2.2 treatments in the second year of the study. 24,3° Patients in the VISION trial were allowed to receive treatment with PDT at the physician's discretion. 27 The proportion of patients who received PDT and the mean number of PDT treatments per patient in the pegaptanib and standard-care therapy groups in both years of treatment are shown in Table I. The cost associated with 1 PDT and 1 pegaptanib treatment, including drug and administration costs, was obtained from the physician payment schedule from the universal health care plan of the province of Quebec, the R4gie de l'Assurance Maladie du Qu6bec (RAMQ). These costs were estimated at $2512 and $1308 for PDT and pegaptanib, respectively. The mean cost per annum associated with PDT and pegaptanib treatment in these groups was estimated by multiplying the average number of treatments per year by the percentage of patients receiving the treatment and by the average cost of treatment. Costs of treating adverse events (AEs) were estimated using treatment algorithms specified by a panel of clinical experts and from the rates of AE occurrence obtained from the clinical trials. 21,27 The treatment patterns for AEs were determined and the percentage of patients treated for each AE was derived from interviews with clinical experts (Table I). Costs of comorbidities (ie, depression/anxiety, fractures, need for assisted living) were estimated from the
Table II. Cumulative probability of change in vision by year, lesion subtype, and treatment. Values are %. Cumulative Efficacy Probabilities 12 Months
Parameter All lesion subtypes 27 Loss of 3-6 lines Loss of_>6 lines Gain of>3 lines
24 Months
Pegaptanib
Standard Care
Pegaptanib
Standard Care
20.41 9.52 6.12
22.64 21.96 2.03
28.57 12.78 9.77
28.97 26.17 3.74
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Predominantly classic 21'24'30'31 Loss of 3-6 lines Loss of_>6 lines Gain of>3 lines
PDT 26.51 13.46 5.66
Placebo 23.10 29.14 2.41
PDT 32.86 16.46 7.41
Placebo 28.62 36.00 3.89
Minimally classic Loss of 3-6 lines Loss of>6 lines Gain of_>3 lines
27.23 16.83 6.44
28.85 16.35 1.92
32.67 19.80 8.42
28.85 26.92 4.81
Occult Loss of 3-6 lines Loss of>6 lines Gain of>_3 lines
28.92 22.29 3.01
22.83 32.61 2.17
25.90 28.92 4.82
21.74 46.74 1.09
PDT = photodynamic therapy with verteporfin.
R A M Q database (Table I). Costs are reported in 2004 Canadian dollars, and both costs and outcomes are discounted at 3% per annum in the base case.
Model Calculations Lifetime costs, quality-adjusted life-years (QALYs), and vision years gained (VYGs) (calculated as the time at VA >20/200) were estimated. The incremental cost per QALY and incremental cost per VYG were obtained. The incremental cost per QALY or VYG for pegaptanib versus standard care was calculated as follows, based on typical incremental cost-effectiveness ratio equations: Incremental cost per QALY or VYG = (CpE G - CSTD)/(EpEG - ESTD)
where CpEc and EpEG were the per-patient cost and efficacy (QALY or VYG) experienced by patients treated with pegaptanib in the pegaptanib clinical trial, 27 and CSTD and ESTD were the per-patient cost and effi-
cacy experienced by patients receiving standard care in the pegaptanib clinical trial. 27 Incremental costs per QALY or VYG for PDT versus PDT placebo were calculated in a similar manner. Since no head-to-head comparison of pegaptanib versus PDT was available and normalization of clinical efficacy data was not performed prior to running the model, the incremental cost per QALY or VYG was calculated as follows, modified from typical incremental cost-effectiveness ratio equations: Incremental cost per QALY or VYG = ([CpE6 - CSTD] [CpDT - CpLAC])/([EPEG -- ESTD] -- [EpDT - EpLAC]), where CpEG, EpEG, CSTD, and ESTD were defined as mentioned previously, and CpDw and EpDw w e r e the per-patient cost and efficacy experienced by patients treated with PDT in the PDT clinical trials, 21 and CpLAC and EpLACw e r e the per-patient cost and efficacy experienced by patients who received placebo in the PDT clinical trials. 21
Sensitivity Analysis To assess the robustness of the base-case model, sensitivity analyses were performed on the time horizon (from lifetime in the base case to a 2- or 5-year model), the costs associated with vision-loss comorbidities (±10%), the impact of vision loss on mortality (ie, patients were or were not at greater risk for death compared with the nonblind general population), AMD treatment costs (±10%); angiographic subtype prevalence (55% occult; 24% minimally classic; 21% predominantly classic), the effect of continuing treatment after reaching legal blindness (treatment extended to a VA of 20/400 or worse in the betterseeing eye), patient age (cohort of patients aged >45 years), discount rate (varied from 0% to 5%), utility score (varied using reported 95% CIs), and health care resource use for vision loss-associated comorbidities data (varied using alternative set of hazard ratios for comorbidities obtained from an analysis of US Medicare claims). TM The model was built, and all analyses conducted, using Microsoft Office Excel 2003 (Microsoft Corporation, Redmond, Washington). RES U LTS Study Population V I S I O N 27 enrolled 1190 patients (58.5% female; 96.1% white; AMD subtype distribution: 27.6% pre-
dominantly classic, 35.8% minimally classic, and 38.5% occult). Of these patients, 1111 were >_65years of age, with 834 in the pegaptanib group and 277 in the standard-care therapy group.
Outcomes and Mean Costs Patients who received pegaptanib treatment experienced more QALYs and VYGs compared with patients who received PDT or standard care (Table III). The mean total cost per patient was greater in patients who received pegaptanib treatment compared with that in those who received PDT or standard care ($20,016 vs $15,345 or $7669). The incremental total cost for patients who received pegaptanib over those who received PDT was $3741 and $12,347 in patients who received pegaptanib over standard care. Drug costs were the key cost driver in the pegaptanib group, whereas in the PDT group, average total costs of comorbidities were the cost drivers. Cost-Effectiveness
The incremental cost per QALY gained in patients receiving pegaptanib versus standard care was $59,039 (ie, the difference between the mean costs of the 2 therapies divided by the difference in QALY gained per patient between the 2 therapies: [$20,016 $7669]/[4.17- 3.96]). The incremental cost per QALY
Table III. Mean expected clinical outcomes and costs per patient over a lifetime. Parameter
Pegaptanib
Standard Care*
PDT
Placebot
Expected outcomes Mean vision years per patient
3.83
3.26
3.01
2.62
Mean QALYs per patient
4.17
3.96
3.87
3.73
9932 3208 24 6852 20,016
568 292 1 6808 7669
5569 2996 8 6772 15,345
0 0 0 6737 6737
Expected costs per patient, CanS Mean drug costs Mean treatment costs (not including drug costs) Mean adverse-event costs Mean total costs ofcomorbidities* Mean total costs
PDT = photodynamic therapy; QALYs = quality-adjusted life-years. *Standard care occurred when PDT with verteporfin was allowed For all patients with predominantly classic wet age-related degeneration at the discretion of'the ophthalmologist. CPlacebo was 5% dextrose in water. CData from Center For Addiction and Mental Health, Canadian Psychiatric Association and the Canadian Psychological Association, Community Care Access, and Ministry of Health and Long Term Care-Ontario Databases.
in patients who received pegaptanib versus PDT was $49,052 (ie, the difference between the average costs of the 2 therapies compared with their respective comparison groups [standard care or placebo] divided by the difference in QALYs gained between the 2 therapies compared with their respective comparison groups [ie, standard care or placebo]: [($20,016 - $7669) ($15,345 - $6737)]/[(4.17 - 3.96) - (3.87 - 3.73)]). The incremental costs per VYG per patient were $21,559 and $20,401 for pegaptanib versus standard care and pegaptanib versus PDT, respectively (Table III).
Sensitivity Analysis Results of the sensitivity analysis are presented as incremental cost-effectiveness ratios (cost per VYG and cost per QALY gained) (Table IV). The results of
Table IV.
the majority of the sensitivity analyses varied only slightly from the base-case results. The model was not sensitive (ie, results were similar to that for the base case) to variations in vision loss, comorbidity costs, AMD treatment costs (excluding drug costs), utility scores, or changes in the hazard ratio of resource use related to vision loss-associated comorbidities by VA health state. The model results were most sensitive to variations in the time horizon considered; when the lifetime time horizon was decreased to 5 years, the incremental cost per VYG and per QALY for pegaptanib versus standard care increased 1.67- and 1.90-fold, respectively, and for pegaptanib versus PDT, increased by 1.98and 2.15-fold, respectively. When the lifetime time horizon was decreased further to 2 years, the incre-
Sensitivity analysis of
the cost-effectiveness of therapies for wet age-related macular degeneration (AMD). Values are CanS. Incremental Cost per VY
Incremental Cost per QALY
Pegaptanib vs Standard Care
Pegaptanib vs PDT
Pegaptanib vs Standard Care
Pegaptanib vs PDT
Base case*
21,559
20,401
59,039
49,052
Model time horizon 2 Years 5 Years
80,863 35,990
74,406 40,459
274,704 112,466
225,659 105,303
No greater risk for death due to vision loss
21,458
20,335
69,379
57,259
AMD subtype distribution 55% occ; 24% mc; 21% pc
21,559
17,021
59,039
42,300
Sensitivity Analysis Variable
Treatment costs (excluding drug costs) Decreasing cost by 10% Increasing cost by 10% Treatment after legally blind No treatment at VA _<20/400
21,050 22,068
20,445 20,357
57,645 60,433
49,157 48,946
25,870
20,958
70,515
53,126
Cost for patients aged _>45 years
20,384
18,828
56,385
45,950
Discount rates 0% 5%
18,894 23,336
17,412 22,401
50,428 64,881
41,120 54,450
VY = vision year; QALY= quality-adJusted life-year; PDT = photodynamic therapy; occ = occult; mc = minimally classic; pc = predominantly classic; VA = visual acuity. *Base-case parameters: lifetime horizon, 1.37 greater risk of death, AMD subtype distribution occ: 35%; mc: 21%; pc: 44%; no treatment at VA <20/200, age of patients _>65years, discount rate 3%.
mental cost per VYG and per QALY for pegaptanib versus standard care increased by 3.75- and 4.65-fold, respectively, and for pegaptanib versus PDT, increased by 3.65- and 4.6-fold, respectively. When testing no change in risk for death due to vision loss, there was no effect on incremental cost per VYG for pegaptanib compared with either treatment; however, incremental cost per QALY was increased by 18 % for pegaptanib versus standard care and by 17% versus PDT. The resuits were slightly sensitive to continuing treatment after legal blindness was reached, with the incremental cost per VYG and QALY increasing by 20% and 19% for pegaptanib versus standard care, respectively, and the incremental cost per VYG and QALY increasing by 3% and 8% for pegaptanib versus PDT, respectively. Varying the lower age limit of the patients to 45 years of age from 65 years of age reduced incremental cost-effectiveness ratios by 4% to 6%. The model was sensitive to variations in the discount rate; increasing the discount rate to 5% increased incremental cost-effectiveness ratios by 8% to 11%, while setting the discount rate to 0% resulted in decreasing the incremental cost-effectiveness ratios by 12% to 16%. The model results were minimally affected by varying the angiographic subtype prevalence of wet AMD; decreases in cost-effectiveness were seen for comparisons with PDT only (17% for vision years and 14% QALYs). DISCUSSION The results of this economic assessment suggest that pegaptanib is a cost-effective treatment alternative to PDT for subfoveal wet AMD. Incremental costs per QALY of $49,052 and $59,039 for pegaptanib treatment versus PDT and standard care were observed, respectively. The TAP and VIP studies reported efficacy in terms of improved transition between health states. However, PDT has not been approved as treatment for minimally classic and occult AMD lesion subtypes; thus, when using this efficacy and adding drug costs, it is expected that the incremental cost-effectiveness ratio would be higher than when only treating (ie, giving drug cost and efficacy) those patients who would benefit from the treatment. Furthermore, results of the model show incremental costs per VYG of $20,401 and $21,559 for pegaptanib treatment versus PDT and standard care, respectively. Based on the literature search, there have been few published economic analyses of the treatment of AMD.
In a 2-year model of patients receiving PDT versus placebo, an incremental cost per QALY of $86,721 (1999 US $) (-Can $96,961) was observed in patients whose better-seeing eye had only minor visual loss at baseline (20/40). For patients whose better-seeing eye had more severe vision loss at baseline (20/200), an incremental cost of $173,984/QALY (1999 US $) was observed. 33 An analysis conducted in 2003 in the United Kingdom examined costs associated with diminishing VA and blindness. The incremental cost per QALY gained with PDT compared to placebo was £120,000 (-Can $253,000 at an exchange rate of 2.0981) over 1 year and between £151,000 and £182,000 (-Can $318,000 and $383,000, respectively) for 2 years of treatment. 34 These studies of AMD treatment used much shorter time horizons than the one used in our study, making comparison of results difficult. Multiple 1-way sensitivity analyses were performed to test the robustness of the model. Since the outcomes of the majority of these analyses did not result in great variation from the base-case results, the robustness of the main analysis is supported. Changes in the time horizon had the largest impact on the model. Differences in cost-effectiveness were likely due to changes in time horizon because, for the most part, full health benefits are not considered in short time horizons. Thus, it is important to consider a long time horizon as for the base case, one that is long enough for full benefit to be realized. Some sensitivity was seen following changes in age, mortality rates, treatment costs (excluding drug costs), discounting rates, and treatment duration after legal blindness was diagnosed. This study is one of the few investigations of AMD that considers the costs associated with treatment of AMD-related comorbidities. The base-case results were robust to the key input variables surrounding comorbidities, confirmed by sensitivity analyses. Although the standard care treatment group to which pegaptanib has been compared included PDT treatment in some patients, the pegaptanib group also contained some PDT use (refer to Table I), and pegaptanib was nevertheless found to be clinically better than standard care (refer to Table 11). 27
Study Limitations Our study comprised a few limitations. First, a head-to-head trial with pegaptanib versus PDT was not available for analysis. For this reason, information
from previously published clinical trials was used for comparing treatment groups by inference. Second, although appropriate for the perspective of this article (ie, the public payer perspective), this study did not address indirect costs. The loss of productivity in performing household chores or paid work may be substantial from a patient's perspective. One may conclude that there would be a bigger impact on costs and perhaps a smaller cost-effectiveness ratio; thus, this analysis underestimates the cost-effectiveness of the treatment. Third, the ages of the patient populations from the trials used in the study were different. However, the sensitivity analyses performed in this study found that the Markov model was slightly sensitive to alterations in the age of the study population. Finally, the model examined only pegaptanib treatment for 2 years, representing a conservative estimate of treatment cost-effectiveness. As data from longer trials become available, cost-effectiveness will need to be reassessed. CONCLUSION The results of this analysis suggest that in Canada, pegaptanib is a cost-effective treatment for subfoveal wet AMD in elderly patients, regardless of lesion subtype, compared to PDT with verteporfin and to standard care. ACKNOWLEDGM ENTS The authors would like to thank the panel of clinical experts, including Drs. S. Sharma, S. Fekrat, D. Marcus, D. Fletcher, S. Bressler, E. Kim, and R. Buggage. This article was prepared with the assistance of BioMedCom Consultants inc, Montreal, Canada. Dr. Earnshaw is an employee of RTI Health Solutions, an independent contract research organization that has received research funding from Pfizer Inc. for pegaptanib-related studies and from other pharmaceutical companies that market drugs for the treatment of AMD and other medical conditions. Dr. Moride is a consultant for the pharmaceutical industry and regulatory authorities. REFERENCES 1. World Health Organization. Magnitude and causes of visual impairment. Geneva, Switzerland: WHO; 2004. 2. Ambati J, Ambati BK, Yoo SH, et al. Age-related macular degeneration: Etiology, pathogenesis, and therapeutic strategies. Surv Opktkalmol. 2003;48:257-293.
3. Margherio RR, Margherio AR, DeSantis ME. Laser treatments with verteporfin therapy and its potential impact on retinal practices. Retina. 2000;20:325-330. 4. Olsen TW, Feng X, Kasper TJ, et al. Fluorescein anglographic lesion type Frequency in neovascular age-related macular degeneration. Ophtkalmolo~,. 2004;111:250-
255. 5. Zawinka C, Ergun E, Stur M. Prevalence of patients presenting with neovascular age-related macular degeneration in an urban population. Retina. 2005;25:324-331. 6. Patterson C. Screening For visual impairment in the elderly. Hamilton, Ontario, Canada: Public Health Agency of Canada; 1994. Available at: http://www.phac-aspc.gc.ca/ publicat/clinicclinique/pdf/sl 1c78e.pdf#search=%22 screening%20For%20visual%20im pairment%20in%20the% 20elderly%20and%20patterson%22. Accessed August 21, 2006. 7. Klein R, Klein BE, Linton KL. Prevalence of age-related maculopathy. The Beaver Dam Eye Study. Ophthalmalo~. 1992;99:933-943. 8. Hart PM, Chakravarthy U, Stevenson MR, Jamison JQ. A vision specific Functional index for use in patients with age related macular degeneration. Br J Opktkalmol. 1999;83:1115-I 120. 9. Brown GC, Sharma S, Brown MM, KistlerJ. Utility values and age-related macular degeneration. Arck Opktkalmol. 2000;118:47-51. 10. Brown MM, Brown GC, Sharma S, LandyJ. Health care economic analyses and value-based medicine. Surv Ophtkalmol. 2003;48:204-223. 1 I. SteinJD, Brown MM, Brown GC, et al. Quality of life with macular degeneration: Perceptions of patients, clinicians, and community members. BrJ Opktkalmol. 2003;87:812. 12. Dargent-Molina P, Favier F, Grandjean H, et al. Fall-related Factors and risk of hip Fracture: The EPIDOS prospective study [published correction appears in Lancet. 1996;348: 416]. Lancet. 1996;348:145-149. 13. Harwood RH, Ebrahim S. A comparison of the responsiveness of the Nottingham extended activities of'daily living scale, London handicap scale and SF-36. Disafiil Rekabil. 2000;22:786-793. 14. Ivers RQ, Norton R, Cumming RG, et al. Visual impairment and risk of hip Fracture. AmJ Epidemiol. 2000;152: 633-639. I S. Brody BL, Gamst AC, Williams IRA, et al. Depression, visual acuity, comorbidity, and disability associated with agerelated macular degeneration. Opktkalmolo~. 2001;108: 1893-1900, discussion 1900-1901. 16. Mangione CM, Gutierrez PR, Lowe G, et al. Influence of age-related maculopathy on visual Functioning and health-related quality of life. AmJ Ophthalmol. 1999;128:
45-53.
17. Williams IRA, Brody BL, Thomas RG, et al. The psychosocial impact ofmacular degeneration. Arch Ophthalmol. 1998;116:514-520. 18. Javitt JC, Zhou Z, Willke RJ. Association between vision loss and higher medical care costs in Medicare beneficiaries costs are greater for those with progressive vision loss. Ophthalmolo~. 2007;114: 238-245. 19. Chia EM, WangJJ, Rochtchina E, et al. Impact of" bilateral visual impairment on health-related quality of'life: The Blue Mountains Eye Study. Invest Ophthalmol VisSci. 2004;45:71-76. 20. Brown MM, Brown GC, Stein JD, et al. Age-related macular degeneration: Economic burden and value-based medicine analysis. Can J Ophthalmol. 2005;40:277-287. 21. Verteporfin in PhotodynamicTherapy Study Group. Verteporfin therapy ogsubfoveal choroidal neovascularization in age-related macular degeneration: Two-year results of" a randomized clinical trial including lesions with occult with no classic choroidal neovascu larizatio n-verteporfin in photodynamic therapy report 2. AmJ Ophthalmol. 2001;131: 541-560. 22. Weir E. Age-related macular degeneration: Armed against ARMD. CMAJ. 2004;170:463-464. 23. Macular Photocoagulation Study Group. Visual outcome after laser photocoagulation for subfoveal choroidal neovascularization secondary to age-related macular degeneration. The influence of initial lesion size and initial visual acuity. Arch Ophthalrnol. 1994;112:480-488. 24. Bressler NM, for the Treatment of" Age-Related Macular Degeneration with Photodynamic Therapy (TAP) Study Group. Photodynamic therapy of" subfoveal choroidal neovascularization in age-related macular degeneration with verteporfin: Two-year results of" 2 randomized clinical trials--TAP report 2. Arch Ophthalmol. 2001 ;119:198-207.
25. Doggrell SA. Pegaptanib: The first antiangiogenic agent approved For neovascular macular degeneration. Expert Opin Pharmacother. 2005;6: 1421-1423. 26. Ng EW, Adamis AP. Targeting angiogenesis, the underlying disorder in neovascular age-related macular degeneration. CanJOphthalmo/. 2005; 40:352-368. 27. Gragoudas ES, Adamis AP, Cunningham ET Jr, et al, For the VEGF Inhibition Study in Ocular Neovascularization Clinical Trial Group. Pegaptanib for neovascular agerelated macular degeneration. N EnglJ Med. 2004;351:2805-2816. 28. Annual demographic statistics 2003. Ottawa, Ontario, Canada: Statistics Canada; 2004. 29. Klein R, Klein BE, Tomany SC, et al. Ten-year incidence and progression of" age-related maculopathy: The Beaver Dam eye study. Ophthalmology. 2002;109:1767-1779. 30. Treatment of" age-related macular degeneration with photodynamic therapy (TAP) Study Group. Photodynamic therapy of subfoveal choroidal neovascularization in agerelated macular degeneration with verteporfin: One-year results of" 2 randomized clinical trials--TAP report [published correction appears
in Arch Ophthalmol. 2000;118:488]. Arch Ophthalmol. 1999;117:13291345. 31. BresslerNM, ArnoldJ, Benchaboune M, et al. Treatment of'Age-Related Macular Degeneration with Photodynamic Therapy (TAP) Study Group. Verteporfin therapy of subfoveal choroidal neovascularization in patients with age-related macular degeneration: Additional information regarding baseline lesion composition's impact on vision outcomes-TAP report No. 3. Arch Ophthalmol. 2002; 120:1443-I 454. 32. Green C, Brazier J, Deverill M. Valuing health-related quality of'life. A review of health state valuation techniques. PharmacoEconomics.2000; 17:151-165.
33. Sharma S, Brown GC, Brown MM, et al. The cost-effectiveness of" photodynamic therapy for Fellow eyes with subfoveal choroidal neovascularization secondary to age-related macular degeneration. Ophthalmology. 2001 ;108:2051-2059. 34. Meads C, Salas C, Roberts T, et al. Clinical effectiveness and cost-utility of" photodynamic therapy For wet age-related macular degeneration: A systematic review and economic evaluation. Health Technol Assess. 2003;7:v-vi, 1-98.
Address correspondence to: Sophie R o c h o n , M A , Pfizer C a n a d a Inc., 17300 T r a n s C a n a d a Highway, Kirkland, Q u e b e c H 9 J 2 M 5 , C a n a d a . E-maih [email protected]
who received pegaptanib compared to those who received PDT or standard care ($20,016 vs $15,345 or $7669, respectively). The incremental cost per QALY in patients receiving pegaptanib compared to those receiving PDT was $49,052 and $59,039 for patients receiving pegaptanib versus standard care. The incremental cost per VYG was $20,401 and $21,559 with pegaptanib versus PDT and standard care, respectively. Sensitivity analyses found that the model was relatively robust to changes in various model parameters. Conclusion: The results of this analysis suggest that in Canada, pegaptanib is a cost-effective treatment for subfoveal wet AMD in elderly patients, regardless of lesion subtype, compared to PDT with verteporfin and to standard care. (Clin Ther. 2007;29:20962106) Copyright © 2007 Excerpta Medica, Inc. Key words: age-related macular degeneration, pegaptanib, photodynamic therapy with verteporfin, cost-effectiveness, quality-adjusted life-year.
INTRODUCTION Age-related macular degeneration (AMD) is the leading cause of blindness in developed countries) AMD is a retinal disease that results from deterioration of the macula and is characterized by blurring, distorThis work was presented in abstract form at the Third Annual Canadian Joint Therapeutics Congress, May 10-13, 2006, Toronto, Canada; and the Canadian Ophthalmology Society Annual Meeting and Exhibition, June 21-24, 2006, Toronto, Canada.
Accepted for publication March 8, 2007. doi:l 0.1016/j.clinthera.2007.09.001 0149-2918/$32.00 Printed in the USA. Reproduction in whole or part is not permitted. Copyright © 2007 ExcerptaMedica, Inc.
tion, or loss of central vision. 2 AMD can be categorized into atrophic (dry) or exudative (wet) disease. In wet AMD, abnormal blood vessels form behind the macula and their contents leak into the retina or subretinal space. Three subtypes of wet AMD exist and are classified according to the pattern of lesion as visualized by fluorescein angiography: occult (35 %-73 % of patients), minimally classic (35%), and predominantly classic (20%-44%). 3-s Wet AMD is subclassified according to the location of lesions relative to the fovea3 - s extrafoveal, juxtafoveal, and subfoveal--with the majority of cases (76%-83%) being subfoveal.4 In Canada, >2 million people over the age of 50 years have some form of AMD. 6 Because the incidence of AMD increases with age, 1 this figure is likely to triple over the next 25 years, coincident with the aging population of Canada. 7 Central vision impairment results in diminished ability to perform routine daily activities. 8-11 Patients with AMD have an increased risk for fallsl2,13; fractures14; depressionlS; and deterioration of their independence and emotional and social health. 16-18 The quality of life of patients with AMD is compromised as seriously as that of patients with arthritis, asthma, diabetes, and stroke. 19 Costs incurred from blindness are associated directly with treatment, loss of personal income due to diminished work capacity, and social support services (eg, visual rehabilitation services, social services). 2° Conventional treatments for wet AMD include thermal laser photocoagulation and photodynamic therapy (PDT) with verteporfin21; however, these treatments are not indicated for all subtypes of wet AMD. These treatments slow the progression of vision loss but do not restore a patient's vision to the pre-AMD state. 2,22 Thermal laser photocoagulation 23 is indicated solely for extrafoveal and juxtafoveal wet AMD (15% of wet AMD cases), whereas PDT is indicated only for predominantly classic subfoveal wet AMD. 24 Because most cases of AMD present as the occult form of the disease, 2-s the majority of patients are not eligible for PDT treatment. Thus, an AMD treatment that successfully treats all categories would be invaluable in preventing blindness. Pegaptanib sodium was the first pharmacologic agent approved in Canada for the treatment of subfoveal wet AMD regardless of subtype. 25 Being a vascular endothelial growth factor (VEGF) antagonist, pegaptanib inhibits angiogenesis, vascular permeability, and inflammation of the retina, all of which may
contribute to the progression of the disease. 26 The efficacy and tolerability of pegaptanib in 1186 patients with subfoveal wet AMD were studied in 2 concurrent randomized clinical trials, 27 with a significant treatment benefit compared to standard therapy observed. The economic impact of AMD is not well understood, either in terms of the overall burden of illness or the costs and cost-effectiveness of specific interventions. Using a Markov model projecting lifetime progression and 2-year treatment of wet AMD, populated with data obtained from published clinical trials, Canadian provincial and federal government sources, and clinical experts, the present study examined the cost-effectiveness of pegaptanib versus PDT or standard care for the treatment of subfoveal wet AMD in Canada. The perspective of a Canadian provincial Ministry of Health was used in this analysis, and only direct costs were considered. The analysis was targeted to decision makers involved in public reimbursement of costs of new medicines.
MATERIALS AND METHODS Study Design A Markov model was developed to simulate a cohort of patients presenting with subfoveal wet AMD. The model structure is presented in the figure. In the model, patients transitioned between health states based on visual acuity (VA), as follows: better than 20/40, VA in the better-seeing eye was better than 20/40; 20/40 to >20/80, VA in the better-seeing eye was better than 20/80 but worse than 20/40; 20/80 to >20/200, VA in the better-seeing eye was better than 20/200 but worse than 20/80; 20/200 to >20/400, VA in the better-seeing eye was better than 20/400 but worse than 20/200 (legally blind); and <_20/400, VA in the better-seeing eye was worse than 20/400 (light perception only). Patients transitioned between the health states every 3 months for the remainder of their lifetime. The patient population was stratified by age (group 1, 65-74 years; group 2, >74 years), sex, and lesion subtype (occult, minimally classic, or predominantly classic). Patients received 2 years of treatment with pegaptanib, PDT, and standard care. Treatment of subfoveal wet AMD was considered for a maximum of 2 years in the model. During each model cycle, patients could remain in their current state, or move up by 1 line, indicating a gain in VA, or move down by 1 or 2 lines, indicating
among the VA health states before treatment within the model was obtained from the VEGF Inhibition Study in Ocular Neovascularization (VISION) trials 27 (Table I). VISION consisted of randomized trials of intravitreous injections of pegaptanib compared with standard care in patients _>50 years of age diagnosed with AMD and had a range of best-corrected VA of 20/40 to 20/320 in the study eye and of 20/800 or better in the other eye.
>20/40 in treated eye
20/40 to >20/80 in treated eye
20/80 to >20/200 in treated eye
20/200 to
>20/400 in treated eye
Patients may transition to the death state from any state in the model
_<20/400 in treated eye
Figure. Markov model structure.
a loss in VA. Transitions between VA levels are based on _>3-line gain, 3- to 6-line loss, and >6-line loss on the VA scale. Patients could transition to the death (absorbing) state from any health state in the model. The model is based on bilateral disease in which only the better-seeing eye was treated.
Study Population The Markov model was based on a hypothetical cohort of patients aged _>65 years, classified by the 3 AMD subtypes. Because the perspective of the Canadian provincial Ministry of Health was used for this analysis, the age group used for analysis was aged _>65 years to be in accordance with the majority of beneficiaries of a Canadian provincial Ministry of Health. The proportion of patients in age and sex strata within each AMD subgroup was determined by multiplying the age and sex distribution in the general population 28 by the 1-year incidence rate derived from Klein et al. 29 The distribution of patients
Treatments Study comparators examined included pegaptanib sodium,* PDT with verteporfin,t and standard care, as seen in the VISION, 27 Verteporfin in Photodynamic Therapy (VIP), 21 and Treatment of Age-Related Macular Degeneration with Photodynamic Therapy (TAP) 24,3°,31 clinical trials. Patients receiving pegaptanib or PDT were assumed to have been receiving this treatment regardless of lesion subtype. PDT is indicated in patients with predominantly classic lesions. 24 However, for ethical reasons, patients in all arms of the VISION clinical trials were allowed treatment with PDT at the discretion of the ophthalmologist. 27 Therefore, standard care (ie, the sham or placebo arm in the VISION trials) was defined as treatment with PDT if lesion subtype and/or size qualified for treatment as indicated in the product monograph. The cost of PDT was added in both the pegaptanib and sham arm for the purpose of this economic analysis.
Clinical Efficacy To compare pegaptanib with PDT, clinical efficacy was taken from the published clinical trials. 27 Since there are no direct comparative studies (based on a literature search of MEDLINE [search terms: pegaptanib AND photodynamic therapy, LIMIT clinical trial]) on clinical efficacy for pegaptanib versus PDT treatment, the efficacy data used to estimate the costeffectiveness of pegaptanib versus PDT were derived from the VISION, 27 VIP,21 and TAP 24,30,31 trials. In the VISION trials, the efficacy of pegaptanib was compared with that of standard care (sham injection for patients with occult and minimally classic AMD subtypes or, for patients with predominantly classic *Trademark: Macugen ® ([OSI], Eyetech Inc., New York, New York; Pfizer Canada Inc., licensee, Kirkland, Quebec, Canada). tTrademark: Visudyne ® (Novartis AG, BiJlach, Switzerland).
Table I. Parameters and costs used in the development of the Markov study model. Distribution of patient VA state used in model*
Better than 20/40 20/40 to >20/80 20/80 to >20/200 20/200 to >20/400 _<20/400 Parameter Probability of patients with normal vision needing service,f % Hazard ratio of resource use by vision health state~ Better than 20/40 20/40 to >20/80 20/80 to >20/200, 20/200 to >20/400, ___20/400 Annual cost ofcomorbidity, t Can $ Mean number of PDT treatments per patient
Patients, %
0.34 36.05 46.05 17.06 0.50 Depression/Anxiety
Assisted Living
Fractures
11.10
2.10
3.10
1.0
1.0
1.0
1.20
2.67
1.44
1.29
2.34
1.16
1060
15,701
139
Pegaptanib
Standard Care
Year 1
1.71
2.05
Year 2
1.00
1.54
18.11 6.77
20.59 8.82
Treated, %
Mean Cost, Can $ (2004)
100 100 33 90 100 3 15
343 33 462 818 1801 33 2071
Patients who received PDT, % Year 1 Year 2
AE 21'24'27'30'31 Endophthalmitis Injection site of AE Retinal capillary nonperfusion Retinal detachment Traumatic injury to lens Visual disturbance Vitreous hemorrhage
VA = visual acuity; PDT = photodynamic therapy with verteporfin; AE = adverse event. *Data from Gragoudas el: al. 27 t Source: R~gie de I'Assurance Maladie du Quebec database. Sources: Comorbidil:y costs were obtained from Center for Addiction and Mental Health, Canadian Psychiatric Association and the Canadian Psychological Association, Community Care Access, and Ministry of Health and Long Term Care-Ontario databases.
AMD, sham plus PDT with verteporfin was administered), defined as the sham arm in the VISION trials. In the VIP and TAP trials, PDT with verteporfin was compared with PDT with placebo (injection of 5% dextrose in water). Specifically, transitions between VA health states during years 1 and 2 in patients who received pegaptanib, standard care, PDT, or placebo were based on the percentages of patients experiencing a 3-line gain, 3- to 6-line loss, and >6-line loss on the VA scale as estimated from the clinical end points from the trials for those years, respectively. Threemonth probabilities for the first year were estimated from the first-year clinical probabilities for pegaptanib or PDT assuming an exponential distribution. As such, the 3-month transition probabilities were estimated as: Pn = 1 - (1 - Pro) n/m, where n = 1/4 (3 months in a 12-month period), m = 1 (1 year), Pn = the 3-month transition probability, and P m - the 1-year probability. For the second year, the model estimates the difference between the firstand second-year clinical probabilities (0.2041 and 0.2857, respectively) and then calculates 3-month probabilities using the equation described previously. Annual cumulative probabilities of change in vision for each treatment and lesion subtype as reported in the clinical trials 27 are presented in Table II. After discontinuation, disease progression was assumed to occur similarly to the year-2 disease progression observed in patients receiving standard care/placebo for the respective trials. Age- and sex-specific mortality rates were obtained from Statistics Canada (2003). 28 Relative risk for death due to blindness was 1.37, 9 and the risk for death was allowed to increase with patients' age.
Utility Values Utility values for ocular diseases have been found to correlate more closely with VA in the better-seeing eye. 9 In a study by Brown et al, 9 visual utility scores in patients with ocular disease were determined by VA health states. The utility values were estimated using the time-tradeoff and standard gamble methods. 32 A greater correlation between utility values and VA was found in the better-seeing eye than between the same utility values and VA in the poorer-seeing eye. 9 For this reason, only treatment in the better eye in bilateral disease was assessed.
Cost Data Costs of pegaptanib and PDT were estimated from annual use as reported in the clinical trialsY, 24,27,3° The numbers of pegaptanib treatments were assumed to be 8.4 and 6.9 for the first and second years, respectively. Drug cost included the cost of the drug ($995) and the cost of the procedure ($280). For PDT, the numbers of treatment were 3.4 and 2.2 in the first and second years. Drug cost included the cost of the drug ($1750) and the cost of the procedure ($300). The number of treatments with pegaptanib in years 1 and 2 was obtained from the VISION 27 trial, where patients were randomly assigned to receive either intravitreous injection of pegaptanib or standard care. Patients received a mean of 8.4 and 6.9 pegaptanib treatments in year 1 and 2, respectively. 27 The mean number of treatments in the PDT group was extracted from TAP reports 1 and 2. 24,30 Patients received either placebo or verteporfin IV. 24,30 Patients in the verteporfin-treated group received a mean of 3.4 treatments per participant in the first year and 2.2 treatments in the second year of the study. 24,3° Patients in the VISION trial were allowed to receive treatment with PDT at the physician's discretion. 27 The proportion of patients who received PDT and the mean number of PDT treatments per patient in the pegaptanib and standard-care therapy groups in both years of treatment are shown in Table I. The cost associated with 1 PDT and 1 pegaptanib treatment, including drug and administration costs, was obtained from the physician payment schedule from the universal health care plan of the province of Quebec, the R4gie de l'Assurance Maladie du Qu6bec (RAMQ). These costs were estimated at $2512 and $1308 for PDT and pegaptanib, respectively. The mean cost per annum associated with PDT and pegaptanib treatment in these groups was estimated by multiplying the average number of treatments per year by the percentage of patients receiving the treatment and by the average cost of treatment. Costs of treating adverse events (AEs) were estimated using treatment algorithms specified by a panel of clinical experts and from the rates of AE occurrence obtained from the clinical trials. 21,27 The treatment patterns for AEs were determined and the percentage of patients treated for each AE was derived from interviews with clinical experts (Table I). Costs of comorbidities (ie, depression/anxiety, fractures, need for assisted living) were estimated from the
Table II. Cumulative probability of change in vision by year, lesion subtype, and treatment. Values are %. Cumulative Efficacy Probabilities 12 Months
Parameter All lesion subtypes 27 Loss of 3-6 lines Loss of_>6 lines Gain of>3 lines
24 Months
Pegaptanib
Standard Care
Pegaptanib
Standard Care
20.41 9.52 6.12
22.64 21.96 2.03
28.57 12.78 9.77
28.97 26.17 3.74
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Predominantly classic 21'24'30'31 Loss of 3-6 lines Loss of_>6 lines Gain of>3 lines
PDT 26.51 13.46 5.66
Placebo 23.10 29.14 2.41
PDT 32.86 16.46 7.41
Placebo 28.62 36.00 3.89
Minimally classic Loss of 3-6 lines Loss of>6 lines Gain of_>3 lines
27.23 16.83 6.44
28.85 16.35 1.92
32.67 19.80 8.42
28.85 26.92 4.81
Occult Loss of 3-6 lines Loss of>6 lines Gain of>_3 lines
28.92 22.29 3.01
22.83 32.61 2.17
25.90 28.92 4.82
21.74 46.74 1.09
PDT = photodynamic therapy with verteporfin.
R A M Q database (Table I). Costs are reported in 2004 Canadian dollars, and both costs and outcomes are discounted at 3% per annum in the base case.
Model Calculations Lifetime costs, quality-adjusted life-years (QALYs), and vision years gained (VYGs) (calculated as the time at VA >20/200) were estimated. The incremental cost per QALY and incremental cost per VYG were obtained. The incremental cost per QALY or VYG for pegaptanib versus standard care was calculated as follows, based on typical incremental cost-effectiveness ratio equations: Incremental cost per QALY or VYG = (CpE G - CSTD)/(EpEG - ESTD)
where CpEc and EpEG were the per-patient cost and efficacy (QALY or VYG) experienced by patients treated with pegaptanib in the pegaptanib clinical trial, 27 and CSTD and ESTD were the per-patient cost and effi-
cacy experienced by patients receiving standard care in the pegaptanib clinical trial. 27 Incremental costs per QALY or VYG for PDT versus PDT placebo were calculated in a similar manner. Since no head-to-head comparison of pegaptanib versus PDT was available and normalization of clinical efficacy data was not performed prior to running the model, the incremental cost per QALY or VYG was calculated as follows, modified from typical incremental cost-effectiveness ratio equations: Incremental cost per QALY or VYG = ([CpE6 - CSTD] [CpDT - CpLAC])/([EPEG -- ESTD] -- [EpDT - EpLAC]), where CpEG, EpEG, CSTD, and ESTD were defined as mentioned previously, and CpDw and EpDw w e r e the per-patient cost and efficacy experienced by patients treated with PDT in the PDT clinical trials, 21 and CpLAC and EpLACw e r e the per-patient cost and efficacy experienced by patients who received placebo in the PDT clinical trials. 21
Sensitivity Analysis To assess the robustness of the base-case model, sensitivity analyses were performed on the time horizon (from lifetime in the base case to a 2- or 5-year model), the costs associated with vision-loss comorbidities (±10%), the impact of vision loss on mortality (ie, patients were or were not at greater risk for death compared with the nonblind general population), AMD treatment costs (±10%); angiographic subtype prevalence (55% occult; 24% minimally classic; 21% predominantly classic), the effect of continuing treatment after reaching legal blindness (treatment extended to a VA of 20/400 or worse in the betterseeing eye), patient age (cohort of patients aged >45 years), discount rate (varied from 0% to 5%), utility score (varied using reported 95% CIs), and health care resource use for vision loss-associated comorbidities data (varied using alternative set of hazard ratios for comorbidities obtained from an analysis of US Medicare claims). TM The model was built, and all analyses conducted, using Microsoft Office Excel 2003 (Microsoft Corporation, Redmond, Washington). RES U LTS Study Population V I S I O N 27 enrolled 1190 patients (58.5% female; 96.1% white; AMD subtype distribution: 27.6% pre-
dominantly classic, 35.8% minimally classic, and 38.5% occult). Of these patients, 1111 were >_65years of age, with 834 in the pegaptanib group and 277 in the standard-care therapy group.
Outcomes and Mean Costs Patients who received pegaptanib treatment experienced more QALYs and VYGs compared with patients who received PDT or standard care (Table III). The mean total cost per patient was greater in patients who received pegaptanib treatment compared with that in those who received PDT or standard care ($20,016 vs $15,345 or $7669). The incremental total cost for patients who received pegaptanib over those who received PDT was $3741 and $12,347 in patients who received pegaptanib over standard care. Drug costs were the key cost driver in the pegaptanib group, whereas in the PDT group, average total costs of comorbidities were the cost drivers. Cost-Effectiveness
The incremental cost per QALY gained in patients receiving pegaptanib versus standard care was $59,039 (ie, the difference between the mean costs of the 2 therapies divided by the difference in QALY gained per patient between the 2 therapies: [$20,016 $7669]/[4.17- 3.96]). The incremental cost per QALY
Table III. Mean expected clinical outcomes and costs per patient over a lifetime. Parameter
Pegaptanib
Standard Care*
PDT
Placebot
Expected outcomes Mean vision years per patient
3.83
3.26
3.01
2.62
Mean QALYs per patient
4.17
3.96
3.87
3.73
9932 3208 24 6852 20,016
568 292 1 6808 7669
5569 2996 8 6772 15,345
0 0 0 6737 6737
Expected costs per patient, CanS Mean drug costs Mean treatment costs (not including drug costs) Mean adverse-event costs Mean total costs ofcomorbidities* Mean total costs
PDT = photodynamic therapy; QALYs = quality-adjusted life-years. *Standard care occurred when PDT with verteporfin was allowed For all patients with predominantly classic wet age-related degeneration at the discretion of'the ophthalmologist. CPlacebo was 5% dextrose in water. CData from Center For Addiction and Mental Health, Canadian Psychiatric Association and the Canadian Psychological Association, Community Care Access, and Ministry of Health and Long Term Care-Ontario Databases.
in patients who received pegaptanib versus PDT was $49,052 (ie, the difference between the average costs of the 2 therapies compared with their respective comparison groups [standard care or placebo] divided by the difference in QALYs gained between the 2 therapies compared with their respective comparison groups [ie, standard care or placebo]: [($20,016 - $7669) ($15,345 - $6737)]/[(4.17 - 3.96) - (3.87 - 3.73)]). The incremental costs per VYG per patient were $21,559 and $20,401 for pegaptanib versus standard care and pegaptanib versus PDT, respectively (Table III).
Sensitivity Analysis Results of the sensitivity analysis are presented as incremental cost-effectiveness ratios (cost per VYG and cost per QALY gained) (Table IV). The results of
Table IV.
the majority of the sensitivity analyses varied only slightly from the base-case results. The model was not sensitive (ie, results were similar to that for the base case) to variations in vision loss, comorbidity costs, AMD treatment costs (excluding drug costs), utility scores, or changes in the hazard ratio of resource use related to vision loss-associated comorbidities by VA health state. The model results were most sensitive to variations in the time horizon considered; when the lifetime time horizon was decreased to 5 years, the incremental cost per VYG and per QALY for pegaptanib versus standard care increased 1.67- and 1.90-fold, respectively, and for pegaptanib versus PDT, increased by 1.98and 2.15-fold, respectively. When the lifetime time horizon was decreased further to 2 years, the incre-
Sensitivity analysis of
the cost-effectiveness of therapies for wet age-related macular degeneration (AMD). Values are CanS. Incremental Cost per VY
Incremental Cost per QALY
Pegaptanib vs Standard Care
Pegaptanib vs PDT
Pegaptanib vs Standard Care
Pegaptanib vs PDT
Base case*
21,559
20,401
59,039
49,052
Model time horizon 2 Years 5 Years
80,863 35,990
74,406 40,459
274,704 112,466
225,659 105,303
No greater risk for death due to vision loss
21,458
20,335
69,379
57,259
AMD subtype distribution 55% occ; 24% mc; 21% pc
21,559
17,021
59,039
42,300
Sensitivity Analysis Variable
Treatment costs (excluding drug costs) Decreasing cost by 10% Increasing cost by 10% Treatment after legally blind No treatment at VA _<20/400
21,050 22,068
20,445 20,357
57,645 60,433
49,157 48,946
25,870
20,958
70,515
53,126
Cost for patients aged _>45 years
20,384
18,828
56,385
45,950
Discount rates 0% 5%
18,894 23,336
17,412 22,401
50,428 64,881
41,120 54,450
VY = vision year; QALY= quality-adJusted life-year; PDT = photodynamic therapy; occ = occult; mc = minimally classic; pc = predominantly classic; VA = visual acuity. *Base-case parameters: lifetime horizon, 1.37 greater risk of death, AMD subtype distribution occ: 35%; mc: 21%; pc: 44%; no treatment at VA <20/200, age of patients _>65years, discount rate 3%.
mental cost per VYG and per QALY for pegaptanib versus standard care increased by 3.75- and 4.65-fold, respectively, and for pegaptanib versus PDT, increased by 3.65- and 4.6-fold, respectively. When testing no change in risk for death due to vision loss, there was no effect on incremental cost per VYG for pegaptanib compared with either treatment; however, incremental cost per QALY was increased by 18 % for pegaptanib versus standard care and by 17% versus PDT. The resuits were slightly sensitive to continuing treatment after legal blindness was reached, with the incremental cost per VYG and QALY increasing by 20% and 19% for pegaptanib versus standard care, respectively, and the incremental cost per VYG and QALY increasing by 3% and 8% for pegaptanib versus PDT, respectively. Varying the lower age limit of the patients to 45 years of age from 65 years of age reduced incremental cost-effectiveness ratios by 4% to 6%. The model was sensitive to variations in the discount rate; increasing the discount rate to 5% increased incremental cost-effectiveness ratios by 8% to 11%, while setting the discount rate to 0% resulted in decreasing the incremental cost-effectiveness ratios by 12% to 16%. The model results were minimally affected by varying the angiographic subtype prevalence of wet AMD; decreases in cost-effectiveness were seen for comparisons with PDT only (17% for vision years and 14% QALYs). DISCUSSION The results of this economic assessment suggest that pegaptanib is a cost-effective treatment alternative to PDT for subfoveal wet AMD. Incremental costs per QALY of $49,052 and $59,039 for pegaptanib treatment versus PDT and standard care were observed, respectively. The TAP and VIP studies reported efficacy in terms of improved transition between health states. However, PDT has not been approved as treatment for minimally classic and occult AMD lesion subtypes; thus, when using this efficacy and adding drug costs, it is expected that the incremental cost-effectiveness ratio would be higher than when only treating (ie, giving drug cost and efficacy) those patients who would benefit from the treatment. Furthermore, results of the model show incremental costs per VYG of $20,401 and $21,559 for pegaptanib treatment versus PDT and standard care, respectively. Based on the literature search, there have been few published economic analyses of the treatment of AMD.
In a 2-year model of patients receiving PDT versus placebo, an incremental cost per QALY of $86,721 (1999 US $) (-Can $96,961) was observed in patients whose better-seeing eye had only minor visual loss at baseline (20/40). For patients whose better-seeing eye had more severe vision loss at baseline (20/200), an incremental cost of $173,984/QALY (1999 US $) was observed. 33 An analysis conducted in 2003 in the United Kingdom examined costs associated with diminishing VA and blindness. The incremental cost per QALY gained with PDT compared to placebo was £120,000 (-Can $253,000 at an exchange rate of 2.0981) over 1 year and between £151,000 and £182,000 (-Can $318,000 and $383,000, respectively) for 2 years of treatment. 34 These studies of AMD treatment used much shorter time horizons than the one used in our study, making comparison of results difficult. Multiple 1-way sensitivity analyses were performed to test the robustness of the model. Since the outcomes of the majority of these analyses did not result in great variation from the base-case results, the robustness of the main analysis is supported. Changes in the time horizon had the largest impact on the model. Differences in cost-effectiveness were likely due to changes in time horizon because, for the most part, full health benefits are not considered in short time horizons. Thus, it is important to consider a long time horizon as for the base case, one that is long enough for full benefit to be realized. Some sensitivity was seen following changes in age, mortality rates, treatment costs (excluding drug costs), discounting rates, and treatment duration after legal blindness was diagnosed. This study is one of the few investigations of AMD that considers the costs associated with treatment of AMD-related comorbidities. The base-case results were robust to the key input variables surrounding comorbidities, confirmed by sensitivity analyses. Although the standard care treatment group to which pegaptanib has been compared included PDT treatment in some patients, the pegaptanib group also contained some PDT use (refer to Table I), and pegaptanib was nevertheless found to be clinically better than standard care (refer to Table 11). 27
Study Limitations Our study comprised a few limitations. First, a head-to-head trial with pegaptanib versus PDT was not available for analysis. For this reason, information
from previously published clinical trials was used for comparing treatment groups by inference. Second, although appropriate for the perspective of this article (ie, the public payer perspective), this study did not address indirect costs. The loss of productivity in performing household chores or paid work may be substantial from a patient's perspective. One may conclude that there would be a bigger impact on costs and perhaps a smaller cost-effectiveness ratio; thus, this analysis underestimates the cost-effectiveness of the treatment. Third, the ages of the patient populations from the trials used in the study were different. However, the sensitivity analyses performed in this study found that the Markov model was slightly sensitive to alterations in the age of the study population. Finally, the model examined only pegaptanib treatment for 2 years, representing a conservative estimate of treatment cost-effectiveness. As data from longer trials become available, cost-effectiveness will need to be reassessed. CONCLUSION The results of this analysis suggest that in Canada, pegaptanib is a cost-effective treatment for subfoveal wet AMD in elderly patients, regardless of lesion subtype, compared to PDT with verteporfin and to standard care. ACKNOWLEDGM ENTS The authors would like to thank the panel of clinical experts, including Drs. S. Sharma, S. Fekrat, D. Marcus, D. Fletcher, S. Bressler, E. Kim, and R. Buggage. This article was prepared with the assistance of BioMedCom Consultants inc, Montreal, Canada. Dr. Earnshaw is an employee of RTI Health Solutions, an independent contract research organization that has received research funding from Pfizer Inc. for pegaptanib-related studies and from other pharmaceutical companies that market drugs for the treatment of AMD and other medical conditions. Dr. Moride is a consultant for the pharmaceutical industry and regulatory authorities. REFERENCES 1. World Health Organization. Magnitude and causes of visual impairment. Geneva, Switzerland: WHO; 2004. 2. Ambati J, Ambati BK, Yoo SH, et al. Age-related macular degeneration: Etiology, pathogenesis, and therapeutic strategies. Surv Opktkalmol. 2003;48:257-293.
3. Margherio RR, Margherio AR, DeSantis ME. Laser treatments with verteporfin therapy and its potential impact on retinal practices. Retina. 2000;20:325-330. 4. Olsen TW, Feng X, Kasper TJ, et al. Fluorescein anglographic lesion type Frequency in neovascular age-related macular degeneration. Ophtkalmolo~,. 2004;111:250-
255. 5. Zawinka C, Ergun E, Stur M. Prevalence of patients presenting with neovascular age-related macular degeneration in an urban population. Retina. 2005;25:324-331. 6. Patterson C. Screening For visual impairment in the elderly. Hamilton, Ontario, Canada: Public Health Agency of Canada; 1994. Available at: http://www.phac-aspc.gc.ca/ publicat/clinicclinique/pdf/sl 1c78e.pdf#search=%22 screening%20For%20visual%20im pairment%20in%20the% 20elderly%20and%20patterson%22. Accessed August 21, 2006. 7. Klein R, Klein BE, Linton KL. Prevalence of age-related maculopathy. The Beaver Dam Eye Study. Ophthalmalo~. 1992;99:933-943. 8. Hart PM, Chakravarthy U, Stevenson MR, Jamison JQ. A vision specific Functional index for use in patients with age related macular degeneration. Br J Opktkalmol. 1999;83:1115-I 120. 9. Brown GC, Sharma S, Brown MM, KistlerJ. Utility values and age-related macular degeneration. Arck Opktkalmol. 2000;118:47-51. 10. Brown MM, Brown GC, Sharma S, LandyJ. Health care economic analyses and value-based medicine. Surv Ophtkalmol. 2003;48:204-223. 1 I. SteinJD, Brown MM, Brown GC, et al. Quality of life with macular degeneration: Perceptions of patients, clinicians, and community members. BrJ Opktkalmol. 2003;87:812. 12. Dargent-Molina P, Favier F, Grandjean H, et al. Fall-related Factors and risk of hip Fracture: The EPIDOS prospective study [published correction appears in Lancet. 1996;348: 416]. Lancet. 1996;348:145-149. 13. Harwood RH, Ebrahim S. A comparison of the responsiveness of the Nottingham extended activities of'daily living scale, London handicap scale and SF-36. Disafiil Rekabil. 2000;22:786-793. 14. Ivers RQ, Norton R, Cumming RG, et al. Visual impairment and risk of hip Fracture. AmJ Epidemiol. 2000;152: 633-639. I S. Brody BL, Gamst AC, Williams IRA, et al. Depression, visual acuity, comorbidity, and disability associated with agerelated macular degeneration. Opktkalmolo~. 2001;108: 1893-1900, discussion 1900-1901. 16. Mangione CM, Gutierrez PR, Lowe G, et al. Influence of age-related maculopathy on visual Functioning and health-related quality of life. AmJ Ophthalmol. 1999;128:
45-53.
17. Williams IRA, Brody BL, Thomas RG, et al. The psychosocial impact ofmacular degeneration. Arch Ophthalmol. 1998;116:514-520. 18. Javitt JC, Zhou Z, Willke RJ. Association between vision loss and higher medical care costs in Medicare beneficiaries costs are greater for those with progressive vision loss. Ophthalmolo~. 2007;114: 238-245. 19. Chia EM, WangJJ, Rochtchina E, et al. Impact of" bilateral visual impairment on health-related quality of'life: The Blue Mountains Eye Study. Invest Ophthalmol VisSci. 2004;45:71-76. 20. Brown MM, Brown GC, Stein JD, et al. Age-related macular degeneration: Economic burden and value-based medicine analysis. Can J Ophthalmol. 2005;40:277-287. 21. Verteporfin in PhotodynamicTherapy Study Group. Verteporfin therapy ogsubfoveal choroidal neovascularization in age-related macular degeneration: Two-year results of" a randomized clinical trial including lesions with occult with no classic choroidal neovascu larizatio n-verteporfin in photodynamic therapy report 2. AmJ Ophthalmol. 2001;131: 541-560. 22. Weir E. Age-related macular degeneration: Armed against ARMD. CMAJ. 2004;170:463-464. 23. Macular Photocoagulation Study Group. Visual outcome after laser photocoagulation for subfoveal choroidal neovascularization secondary to age-related macular degeneration. The influence of initial lesion size and initial visual acuity. Arch Ophthalrnol. 1994;112:480-488. 24. Bressler NM, for the Treatment of" Age-Related Macular Degeneration with Photodynamic Therapy (TAP) Study Group. Photodynamic therapy of" subfoveal choroidal neovascularization in age-related macular degeneration with verteporfin: Two-year results of" 2 randomized clinical trials--TAP report 2. Arch Ophthalmol. 2001 ;119:198-207.
25. Doggrell SA. Pegaptanib: The first antiangiogenic agent approved For neovascular macular degeneration. Expert Opin Pharmacother. 2005;6: 1421-1423. 26. Ng EW, Adamis AP. Targeting angiogenesis, the underlying disorder in neovascular age-related macular degeneration. CanJOphthalmo/. 2005; 40:352-368. 27. Gragoudas ES, Adamis AP, Cunningham ET Jr, et al, For the VEGF Inhibition Study in Ocular Neovascularization Clinical Trial Group. Pegaptanib for neovascular agerelated macular degeneration. N EnglJ Med. 2004;351:2805-2816. 28. Annual demographic statistics 2003. Ottawa, Ontario, Canada: Statistics Canada; 2004. 29. Klein R, Klein BE, Tomany SC, et al. Ten-year incidence and progression of" age-related maculopathy: The Beaver Dam eye study. Ophthalmology. 2002;109:1767-1779. 30. Treatment of" age-related macular degeneration with photodynamic therapy (TAP) Study Group. Photodynamic therapy of subfoveal choroidal neovascularization in agerelated macular degeneration with verteporfin: One-year results of" 2 randomized clinical trials--TAP report [published correction appears
in Arch Ophthalmol. 2000;118:488]. Arch Ophthalmol. 1999;117:13291345. 31. BresslerNM, ArnoldJ, Benchaboune M, et al. Treatment of'Age-Related Macular Degeneration with Photodynamic Therapy (TAP) Study Group. Verteporfin therapy of subfoveal choroidal neovascularization in patients with age-related macular degeneration: Additional information regarding baseline lesion composition's impact on vision outcomes-TAP report No. 3. Arch Ophthalmol. 2002; 120:1443-I 454. 32. Green C, Brazier J, Deverill M. Valuing health-related quality of'life. A review of health state valuation techniques. PharmacoEconomics.2000; 17:151-165.
33. Sharma S, Brown GC, Brown MM, et al. The cost-effectiveness of" photodynamic therapy for Fellow eyes with subfoveal choroidal neovascularization secondary to age-related macular degeneration. Ophthalmology. 2001 ;108:2051-2059. 34. Meads C, Salas C, Roberts T, et al. Clinical effectiveness and cost-utility of" photodynamic therapy For wet age-related macular degeneration: A systematic review and economic evaluation. Health Technol Assess. 2003;7:v-vi, 1-98.
Address correspondence to: Sophie R o c h o n , M A , Pfizer C a n a d a Inc., 17300 T r a n s C a n a d a Highway, Kirkland, Q u e b e c H 9 J 2 M 5 , C a n a d a . E-maih [email protected]