Cost effectiveness of darunavir/ritonavir combination antiretroviral therapy for treatment-naive adults with HIV-1 infection in Canada

Abstract

Objective
The AntiRetroviral Therapy with TMC114 ExaMined In naive Subjects (ARTEMIS) clinical trial examined the efficacy and safety of two ritonavir-boosted protease inhibitors (PI/r), darunavir/r 800/100 mg once daily (QD) and lopinavir/r 800/200 mg daily, both used in combination with tenofovir disoproxil fumarate/emtricitabine. This study aimed to assess the cost effectiveness of the darunavir/r regimen compared with the lopinavir/r regimen in treatment-naive adults with HIV-1 infection in Canada.

Methods
A Markov model with a 3-month cycle time and six CD4 cell-count-based health states (>500, 351–500, 201–500, 101–200, 51–100, and 0–50 cells/mm3) followed a cohort of treatment-naive adults with HIV-1 infection through initial darunavir/r or lopinavir/r combination therapy and a common set of subsequent regimens over the course of their remaining lifetimes. Population characteristics and transition probabilities were estimated from the ARTEMIS clinical trial and other trials. Costs (in 2014 Canadian dollars), utilities, and mortality were estimated from Canadian sources and published literature. Costs and health outcomes were discounted at 5 % per year. One-way and probabilistic sensitivity analyses were performed, including a simple indirect comparison of the darunavir/r initial regimen with an atazanavir/r-based regimen.

Results
In the base-case lifetime analysis, individuals receiving initial therapy with the darunavir/r regimen experienced 0.25 more quality-adjusted life-years (QALYs) with lower antiretroviral drug costs (?$14,246) and total costs (?$18,402) than individuals receiving the lopinavir/r regimen, indicating that darunavir/r dominated lopinavir/r. In an indirect comparison with an atazanavir/r-based regimen, the darunavir/r regimen remained the dominant choice, but with lower cost savings (?$2,303) and QALY gains (0.02). Results were robust to a wide range of other changes in input parameter values, population characteristics, and modeling assumptions. The probabilistic sensitivity analysis demonstrated that the darunavir/r regimen was cost effective compared with the lopinavir/r regimen in over 86 % of simulations for willingness-to-pay thresholds between $0 and $100,000 per QALY gained.

Conclusions
Darunavir/r 800/100 mg QD may be a cost-effective PI/r component of initial antiretroviral therapy for treatment-naive adults with HIV-1 infection in Canada.