Objective: This study compared the cost-effectiveness of ranibizumab with that of photodynamic therapy (PDT) in the treatment of predominantly classic choroidal neovascularization secondary to age-related macular degeneration (AMD) from the perspective of a third-party payer in a Spanish setting.
Methods: We constructed a Markov model with 5 states defined by visual acuity (VA) in the better-seeing eye (Snellen scale), as follows: VA >20/40, <or=20/40 to >20/80, </or20/80 to >20/200, <or=20/200 to >20/400, and <or=20/400. A death state was also included. We took transition probabilities, number of ranibizumab injections, and number of PDT treatments from the ANCHOR (Anti-Vascular Endothelial Growth Factor Antibody for the Treatment of Predominantly Classic Choroidal Neovascularization in Age-Related Macular Degeneration) trial. Utilities were taken from a published study of patients' preferences. We used unit costs from our hospital and drug costs from a national database. Resource utilization was determined by an ophthalmologist according to current clinical practice. We performed univariate, threshold, and probabilistic sensitivity analyses. Incremental costs (2007 euro) and quality-adjusted life-years (QALYs), both discounted at a 3% annual rate, and incremental cost-effectiveness ratios (ICERs; euro/QALY) were determined for the 2-year and life-expectancy time horizons.
Results: Treating patients with varying degrees of visual impairment with ranibizumab instead of PDT, with a 2-year time horizon, was found to be euro18,328 more costly and to confer 0.140 additional QALY (euro131,275/QALY). This ICER was reduced to euro39,398/ QALY for the longer life-expectancy time horizon. According to the probabilistic sensitivity analysis, PDT is the therapy of choice in all cases below the threshold of euro30,000/QALY for the 2-year time horizon. Ranibizumab was the optimal intervention in 26% of cases in the longer lifetime horizon. When the initial VA was <or=20/400, the ICER increased to euro255,477 over 2 years. When ranibizumab was administered on an as-needed basis, as in the PrONTO (Prospective Optical coherence tomography imaging of patients with Neovascular AMD Treated with intra-Ocular ranibizumab) trial, the ICERs were reduced to euro29,566/QALY and euro11,469/QALY in the 2-year and life-expectancy horizons, respectively.
Conclusions: Based on these results, ranibizumab was not cost-effective when administered on a monthly basis. When administered as needed, ranibizumab was cost-effective compared with PDT for the treatment of AMD.