To evaluate the cost-effectiveness of nilotinib 300 mg bid and dasatinib 100 mg qd, each compared to imatinib 400 mg qd, as first line therapy in chronic myeloid leukemia in Colombia, from third party payer perspective.
We used a markov model to evaluate a hypothetical cohort of 100 patients, aged 55 years, with newly diagnosed CML in chronic phase, in a 10 year time horizon. Progression free life years saved (PF-LYS) were considered the primary efficacy outcome. Transition probabilities for major molecular response (MMR), disease progression (AP/BC) and CML related death, were analyzed in the model for each arm (nilotinib, imatinib and dasatinib) according to literature review and clinical experts validation. A 3% discount rate was applied to all costs and patient outcomes. In the absence of any head to head trials to compare nilotinib and dasatinib, comparisons were made independently for each one vs imatinib. Final comparison between both nilotinib and dasatinib was made indirectly. Costs analysis included direct medical costs for patient care at different stages of the disease as well as progression and hematology adverse events management, obtained from local health care providers databases at prices for year 2010. Transplantation costs were excluded, due to the age of the base case (55 years old). Prices for medicines were estimated from official government top reimbursement prices. Model robustness was evaluated through univariate and multivariate Montecarlo sensibility analysis. In the abscense of an official reference cost-effectiveness threshold in Colombia, all comparisons in the sensitivity analysis were made taking as reference the estimated average costs to treat patient disease progression during three months (cycle).
For the model cohort of 100 patients, nilotinib was projected to result in greater expected PF-LYS (15,376 for nilotinib vs 14,643 for imatinib), followed by dasatinib (15,108 vs 14,789 for imatinib). Imatinib had lower total lifetime costs, due in part to lower efficacy rates and shorter survival. The incremental cost-effectiveness ratio (ICER) was USD $12,640 per PF-LYS in the nilotinib arm and USD $46,348 per PF-LYS for dasatinib arm, each compared to imatinib. When analyzing indirectly nilotinib vs dasatinib, nilotinib was found to be dominant due to higher efficacy (267 PF-LYS) and less costs (USD $5,225) in the base case. The multivariate sensitivity analysis after 1000 iterations showed that nilotinib maintained its cost-effectiveness against imatinib in more than 85% of scenarios and proved dominant against dasatinib in 82% of scenarios. The average estimated cost to manage disease progression per cycle (three months) was USD $16,109 which was considered as the cost effectiveness threshold.
From a third party payer perspective in Colombia, using PF-LYS as the efficacy outcome, nilotinib is highly cost-effective when compared to imatinib and dominant vs dasatinib in first line therapy for CML in chronic phase.
de Acevedo:Novartis Colombia: Consultancy; Pfizer Colombia: Consultancy.
Asterisk with author names denotes non-ASH members.