Oral therapies for the treatment of pulmonary arterial hypertension: a population-based cost-minimization analysis

Abstract

Background and objective: Pulmonary arterial hypertension (PAH) is a rare but life-threatening condition that is characterized by progressive elevation of pulmonary artery pressure and pulmonary vascular resistance, leading to right-sided heart failure and frequently death. Orally administered agents used for the treatment of symptomatic, moderate-to-severe PAH include sildenafil and the endothelin (ET) receptor antagonists (ERAs), bosentan and sitaxentan (sitaxsentan). Ambrisentan is a new oral ET(A) receptor-selective ERA, with higher ET receptor affinity than bosentan. Placebo-controlled, randomized clinical trials (RCTs) have demonstrated that ambrisentan (5 or 10 mg/day) is safe and effective. To provide health economic data on the multiple oral PAH therapies currently available, a population-based cost-minimization analysis (CMA) was conducted for Canada.

Methods: The primary requirement for a CMA is that all clinical outcomes be equivalent between comparator treatments. To provide such supporting data, a literature search was conducted for RCTs of oral agents used to treat symptomatic PAH. This was followed by application of direct and indirect statistical methods to support the hypothesis of clinical equivalence between the oral agents. Estimates for PAH prevalence, incidence and death rates were then used to build a population-based CMA model. The base-case analysis considered costs for drug therapy, outpatient pharmacy costs, medical consultations and visits, laboratory and diagnostic procedures and other healthcare-related resources. In addition, costs for secondary pharmacotherapy in cases where the primary agent had to be discontinued because of adverse effects were also included. The time horizon for evaluating pharmacotherapy was 3 years, all costs were in 2008 Canadian dollars ($Can) and the costs were discounted at a rate of 3% annually. The study perspective was the Canadian healthcare system.

Results: There were no double-blind RCTs comparing ambrisentan with any of the other oral agents. Therefore, an indirect comparison of placebo-controlled trials of PAH drugs had to be used to support the clinical equivalence. This included a calculation of standardized mean differences (SMD) between agents (vs placebo) and a meta-regression analysis on the primary and secondary trial endpoints. Keeping in mind the caveats associated with indirect trial comparisons, the data suggested similar clinical efficacy over 12-16 weeks between agents, as indicated by the identical magnitude of the SMD between the active agent and placebo and the non-significant differences between drugs as determined by the meta-regression analysis. The population-based model projected that the number of PAH patients clinically suitable for these drugs in Canada would be 931 in the first full-budget year (i.e. 2009) with an increase to 1114 by the third full year. The CMA revealed the following rank order of the least to most costly agent; sildenafil, ambrisentan, sitaxentan and bosentan. Sildenafil was the least costly, primarily because of the lower daily drug-acquisition cost. Of the three ERAs, ambrisentan would be associated with annual cost savings of $Can3.4 and $Can5.6 million when used as an alternative to sitaxentan or bosentan, respectively.

Conclusions: Ambrisentan is less costly than other available ERAs, including bosentan and sitaxentan, but is more costly than sildenafil. In PAH patients in whom an ERA is the preferred agent, ambrisentan may be the drug of choice because of its economic advantages and improved safety profile.