Effectiveness and Costs of Bevacizumab vs Ranibizumab in DME
Effectiveness and Costs of Bevacizumab vs Ranibizumab in DME
The BRDME study is a multicenter, randomized, double-masked comparative clinical trial.
All patients with vision loss due to DME and foveal thickening (as determined on OCT) that may benefit from anti-VEGF treatment are potentially eligible for the study (Table 1 and Table 2).
Included patients are randomized to receive either bevacizumab or ranibizumab.
Patient, treating physician, and evaluating investigator staff are blinded for treatment allocation. Study medication is repackaged in masked syringes at the local pharmacy.
The randomization procedure is computer- and web based, using permuted blocks and stratified by centre, BCVA of the study eye (52 letters or less versus 53 letters or more), and BCVA of the non-study eye (52 letters or less versus 53 letters or more). Randomization is made available by the AMC Clinical Research Unit. Randomization code breaking information and sheets for emergency use only is kept in the local hospital safe. Indications to break the randomization code are not predefined.
Study Procedures. At the baseline visit, the patient signs the written informed consent form, and the medical and ophthalmic history is taken. Within 14 days after randomization the patient receives the first intravitreal injection of the study drug. Investigations and measurements of the BRDME trial are carried out according to the following diagram.
During each visit, vital signs (pulse and blood pressure), concomitant medication and adverse events are recorded. BCVA is assessed and an OCT examination is performed by certified personnel prior to the intravitreal injection. The interval between visits is 30 days, ±7 days to allow for flexibility in scheduling. At baseline and 6 months, an ophthalmic exam and fluorescein angiography are performed. At baseline, 3 and 6 months patients are asked to complete a short, 16-item questionnaire on health status (EQ-5D), health care resource utilization, and out-of-pocket expenses (shortened Health and Labour questionnaire) (Table 3).
The primary outcome is the mean change in best-corrected visual acuity (BCVA) in the study eye from baseline to month 6.
The study adheres to the tenets of the Declaration of Helsinki. The trial protocol has been approved by the Medical Ethical Committee of the Academic Medical Center (Amsterdam). The participation of the other centers is reviewed at each center according to Dutch law. Registration of the trial was requested June 28, 2012 at clinicaltrials.gov. NCT01635790. The study is ongoing. At the time of submission there were no publications of this trial.
The sponsor of this trial is the Netherlands Organisation for Health Research and Development, ZonMW. The study design has been peer reviewed. There is no contribution of commercial organisations.
A Data Safety Monitoring Board is installed and can recommend the steering committee of the BRDME trial to terminate the study before completion depending on the level of discrepancy between both study arms in incidences of (serious) adverse events and numbers of patients who terminate the initially assigned drug prematurely.
The DSMB members are independent and have no competing interests related to any intervention in the BRDME trial. The DSMB is composed of two members with relevant clinical expertise and one member with a background in statistics.
The difference in the BCVA change scores from baseline at month 6 is tested statistically for non-inferiority. Starting from a common standard deviation - based on observations in previous trials with ranibizumab in DME - of the change in BCVA score of 11 letters in both groups, and assuming an improvement from baseline of 6–8 letters in both the ranibizumab group and the bevacizumab group, a sample size of 246 patients (123 in each group) has an 80 % power of demonstrating non-inferiority by excluding a difference of 3.5 letters or more, using a one-sided Student's T-test and a significance level of 0.05. The mean improvement of 7 letters is the average of the changes observed in the placebo-controlled trials discussed earlier. The margin of non-inferiority is equivalent to less than half this improvement.
According to the intention-to-treat principle all randomized patients are included in the final analyses.
The primary outcome measure is the change in best-corrected visual acuity (BCVA) from baseline to month 6 as assessed with ETDRS-like VA charts. When visual acuity is measured in this manner, a 15 letter gain means a doubling of the visual acuity, and a 15 letter loss means that acuity is halved. As the measure of effectiveness, the mean BCVA score at month 6 in the ranibizumab group is compared with the mean score in the bevacizumab group, corrected for baseline differences. The confidence interval for the difference in mean BCVA change scores are reported. We also test statistically for non-inferiority of bevacizumab to ranibizumab using the T-test for independent samples.
If the non-inferiority of bevacizumab to ranibizumab can be demonstrated, then the economic evaluation will be performed as a cost-minimization analysis from a societal perspective. If bevacizumab turns out to be inferior, then the question arises whether the health losses are in reasonable balance with the expected cost savings. In that situation a cost-utility analysis will be performed with the cost per quality-adjusted life-year as outcome parameter. The analysis will be based on (i) the observed cost and visual acuity data, and (ii) available and upcoming literature on health utility associated with different levels of visual acuity. If a cost-utility analysis seems opportune, sensitivity analyses will be done to study the robustness of using patient-based preferences instead of general population based preferences in order to derive health utilities. The latter ones will be reported as the main outcome.
Costs will include the direct medical costs of diagnosis and treatment restricted to (potential) vision loss, including the use of visual aids. Only the medical costs attributable to loss of vision or the prevention thereof will be included in this population with a high risk of co-morbid conditions. Costs will be estimated as the product-sum of the volumes of resources used and their respective unit costs. The cost items will include visits to the health care providers (e.g., ophthalmologist, optometrists, and general practitioners), medication use, and ophthalmic equipment for imaging (Digital Imaging Systems, Fundus Camera's, Optical Coherence Tomography) and operating theatres. Patient-related costs will include the costs of health-related travel and over-the-counter medication. In this population data on loss of productivity will also be collected. The use of resources will be documented in the case record forms and by an additional questionnaire to be completed at baseline and at 3-monthly intervals by the study participants. Unit costing will be based on the national guideline on costing in health care research, supplemented by mean local unit costing data from participating reference centres and practices. Unit costs will be based on relevant national guide prices (Farmacotherapeutisch kompas, G-standaard van de Z-index).
The base year for costing will be 2013. Unit costs will be price-indexed when originating from other calendar years using general yearly price-indices.
The time horizon of the economic evaluation will be 6 months. No discounting of costs (and effects) will be performed.
Methods
Design
The BRDME study is a multicenter, randomized, double-masked comparative clinical trial.
Patient Population
All patients with vision loss due to DME and foveal thickening (as determined on OCT) that may benefit from anti-VEGF treatment are potentially eligible for the study (Table 1 and Table 2).
Randomization, Blinding and Treatment Allocation
Included patients are randomized to receive either bevacizumab or ranibizumab.
Patient, treating physician, and evaluating investigator staff are blinded for treatment allocation. Study medication is repackaged in masked syringes at the local pharmacy.
The randomization procedure is computer- and web based, using permuted blocks and stratified by centre, BCVA of the study eye (52 letters or less versus 53 letters or more), and BCVA of the non-study eye (52 letters or less versus 53 letters or more). Randomization is made available by the AMC Clinical Research Unit. Randomization code breaking information and sheets for emergency use only is kept in the local hospital safe. Indications to break the randomization code are not predefined.
Intervention
Study Procedures. At the baseline visit, the patient signs the written informed consent form, and the medical and ophthalmic history is taken. Within 14 days after randomization the patient receives the first intravitreal injection of the study drug. Investigations and measurements of the BRDME trial are carried out according to the following diagram.
During each visit, vital signs (pulse and blood pressure), concomitant medication and adverse events are recorded. BCVA is assessed and an OCT examination is performed by certified personnel prior to the intravitreal injection. The interval between visits is 30 days, ±7 days to allow for flexibility in scheduling. At baseline and 6 months, an ophthalmic exam and fluorescein angiography are performed. At baseline, 3 and 6 months patients are asked to complete a short, 16-item questionnaire on health status (EQ-5D), health care resource utilization, and out-of-pocket expenses (shortened Health and Labour questionnaire) (Table 3).
Primary Outcome
The primary outcome is the mean change in best-corrected visual acuity (BCVA) in the study eye from baseline to month 6.
Secondary Outcomes
the difference between ranibizumab and bevacizumab in their effect on retinal vascular leakage and edema as determined by FA and OCT at 6 months
the proportions of dropouts in the two treatment arms before the final 6 months assessment
the proportions of BCVA gainers, BCVA losers and OCT/FA non-responders in the two treatment arms at the 6 month assessment
the difference in the occurrence of (serious) adverse events in the 6 months study period
the difference in costs and costs per quality adjusted life-year between the two treatment strategies over the 6 months treatment period
Ethics and Funding
The study adheres to the tenets of the Declaration of Helsinki. The trial protocol has been approved by the Medical Ethical Committee of the Academic Medical Center (Amsterdam). The participation of the other centers is reviewed at each center according to Dutch law. Registration of the trial was requested June 28, 2012 at clinicaltrials.gov. NCT01635790. The study is ongoing. At the time of submission there were no publications of this trial.
The sponsor of this trial is the Netherlands Organisation for Health Research and Development, ZonMW. The study design has been peer reviewed. There is no contribution of commercial organisations.
Data Safety and Monitoring Board
A Data Safety Monitoring Board is installed and can recommend the steering committee of the BRDME trial to terminate the study before completion depending on the level of discrepancy between both study arms in incidences of (serious) adverse events and numbers of patients who terminate the initially assigned drug prematurely.
The DSMB members are independent and have no competing interests related to any intervention in the BRDME trial. The DSMB is composed of two members with relevant clinical expertise and one member with a background in statistics.
Sample Size
The difference in the BCVA change scores from baseline at month 6 is tested statistically for non-inferiority. Starting from a common standard deviation - based on observations in previous trials with ranibizumab in DME - of the change in BCVA score of 11 letters in both groups, and assuming an improvement from baseline of 6–8 letters in both the ranibizumab group and the bevacizumab group, a sample size of 246 patients (123 in each group) has an 80 % power of demonstrating non-inferiority by excluding a difference of 3.5 letters or more, using a one-sided Student's T-test and a significance level of 0.05. The mean improvement of 7 letters is the average of the changes observed in the placebo-controlled trials discussed earlier. The margin of non-inferiority is equivalent to less than half this improvement.
Statistical Analyses
According to the intention-to-treat principle all randomized patients are included in the final analyses.
The primary outcome measure is the change in best-corrected visual acuity (BCVA) from baseline to month 6 as assessed with ETDRS-like VA charts. When visual acuity is measured in this manner, a 15 letter gain means a doubling of the visual acuity, and a 15 letter loss means that acuity is halved. As the measure of effectiveness, the mean BCVA score at month 6 in the ranibizumab group is compared with the mean score in the bevacizumab group, corrected for baseline differences. The confidence interval for the difference in mean BCVA change scores are reported. We also test statistically for non-inferiority of bevacizumab to ranibizumab using the T-test for independent samples.
Ecomomic Evaluation
If the non-inferiority of bevacizumab to ranibizumab can be demonstrated, then the economic evaluation will be performed as a cost-minimization analysis from a societal perspective. If bevacizumab turns out to be inferior, then the question arises whether the health losses are in reasonable balance with the expected cost savings. In that situation a cost-utility analysis will be performed with the cost per quality-adjusted life-year as outcome parameter. The analysis will be based on (i) the observed cost and visual acuity data, and (ii) available and upcoming literature on health utility associated with different levels of visual acuity. If a cost-utility analysis seems opportune, sensitivity analyses will be done to study the robustness of using patient-based preferences instead of general population based preferences in order to derive health utilities. The latter ones will be reported as the main outcome.
Costs will include the direct medical costs of diagnosis and treatment restricted to (potential) vision loss, including the use of visual aids. Only the medical costs attributable to loss of vision or the prevention thereof will be included in this population with a high risk of co-morbid conditions. Costs will be estimated as the product-sum of the volumes of resources used and their respective unit costs. The cost items will include visits to the health care providers (e.g., ophthalmologist, optometrists, and general practitioners), medication use, and ophthalmic equipment for imaging (Digital Imaging Systems, Fundus Camera's, Optical Coherence Tomography) and operating theatres. Patient-related costs will include the costs of health-related travel and over-the-counter medication. In this population data on loss of productivity will also be collected. The use of resources will be documented in the case record forms and by an additional questionnaire to be completed at baseline and at 3-monthly intervals by the study participants. Unit costing will be based on the national guideline on costing in health care research, supplemented by mean local unit costing data from participating reference centres and practices. Unit costs will be based on relevant national guide prices (Farmacotherapeutisch kompas, G-standaard van de Z-index).
The base year for costing will be 2013. Unit costs will be price-indexed when originating from other calendar years using general yearly price-indices.
The time horizon of the economic evaluation will be 6 months. No discounting of costs (and effects) will be performed.
Source...