Projecting the Long-Term Quality-Adjusted Survival of Patients with Imatinib-Resistant Chronic Myelogenous Leukemia in Chronic Phase Receiving Nilotinib, Dasatinib, Dose Escalation Imatinib, or Palliative Care: An Essential Element for Cost-Effectiveness Analysis.

Background: Imatinib is the mainstay in the treatment of chronic phase CML (CML-CP). However, in a minority of patients, resistance to imatinib is observed. Until recently, these patients underwent dose-escalation with imatinib (DEI) or stopped therapy and received no further tyrosine kinase inhibitors (TKIs). Recent evidence suggests that the new-generation TKIs nilotinib and dasatinib may provide an alternative for these patients. This study modeled the quality-adjusted life years (QALYs) of nilotinib, dasatinib, DEI, or no additional TKI (i.e., palliative care) in imatinib-resistant CML-CP patients, with the aim of using these estimates in cost effectiveness analyses.

Methods: A Markov model linking major cytogenetic response (MCyR) at 6 months to long-term survival was developed to project disease progression and estimate QALYs associated with each treatment option for a CML-CP patient (male, age 56 years old). The 6-month MCyR was used as it was the furthest assessment time common across all treatment options. The fundamental tenet of the model was that imatinib-resistant patients achieving a MCyR at 6 months on any option live longer than those who do not. The proportions of patients achieving a MCyR at 6 months were derived from separate single-arm registration trials and epidemiologic studies of each option. The duration of survival for patients achieving/not achieving a MCyR at 6 months was determined by estimating the areas under the Kaplan-Meier survival curves of CML-CP patients with documented resistance to regular-dose imatinib who were receiving DEI (based on a subgroup analysis of patients enrolled in the IRIS study who underwent dose escalation) or no additional TKI. Duration of survival for responders/non responders was further divided into time spent in chronic phase (CP), accelerated phase (AP), and blast crisis (BC). The time spent in AP and BC was assumed independent of the treatment option received and was estimated using the literature. Thus, patients with MCyR were projected to live longer, spend more time in CP, and progressed more slowly compared to AP, BC, and ultimately CML-related death than patients without MCyR. Assumptions regarding the quality of life impacts of CML and the adverse events of the treatments were obtained from the literature.

Results: Patients achieving and not achieving a MCyR at 6 months were projected to live 12.78 and 6.14 years, respectively. The proportion of patients achieving a MCyR at 6 months was 51% for nilotinib, 31% for dasatinib, 25% for DEI, and 2.5% for no additional TKI. Total survival and QALY were projected at 6.31 years and 4.62 QALYs for patients not receiving additional TKI therapy. Compared to these patients, those treated with nilotinib, dasatinib, DEI lived an additional 3.22, 1.89, and 1.49 years, respectively. Corresponding QALY gains were estimated to be 2.80, 1.62, and 1.30, respectively.

Conclusion: This analysis suggests that nilotinib may be superior to dasatinib, followed by DEI, and no additional TKI in terms of projected survival and QALYs. These findings reflect the MCyR for each option observed at 6 months and are based on modeling and indirect comparisons. As such, they require confirmation through comparative studies but can nevertheless serve as a starting point for cost effectiveness analyses.