Economic Modeling of the Potential Impact of Chronic Myeloid Leukemia (CML) Monitoring on Healthcare Costs

Abstract

Introduction: Molecular monitoring (qPCR testing) of patients with chronic myeloid leukemia (CML) is necessary, as it allows physicians to optimize treatment and make timely changes to therapy. National Comprehensive Cancer Network (NCCN) guidelines recommend molecular testing as often as every 3 months, but previous studies indicate that only approximately 30% of patients receive testing in accordance with these guidelines and up to 50-79% of patients do not receive testing within 3-6 months of starting treatment (Latremouille-Viau D, et al. J Manag Care Spec Pharm 2017, Guérin A, et al. J Med Econ 2014, Di Bella NJ, et al. Leuk 2015, Saleh MN, et al. Curr Med Res Opin 2014). Appropriate molecular testing has been shown to be associated with a lower risk of disease progression (Haque R, et al. J Am Pharm Assoc 2017).Studies have also shown that testing according to guidelines is expected to lead to cost savings (Latremouille-Viau D, et al. J Manag Care Spec Pharm 2017).Finally, previous research has concluded that the total cost of care for patients with CML progression is significantly greater than for patients with no evidence of CML progression (Jabbour EJ, et al. J Med Econ 2017).Currently data that can directly link these elements are lacking, thus an economic model was developed to estimate the potential impact of poor monitoring on patient healthcare costs.Methods: An Excel based decision analytic economic model was developed to evaluate the potential impact of CML monitoring on healthcare costs from a U.S. payer perspective. This model calculates the expected cost savings from molecular monitoring of CML patients as compared to not monitoring. The following model inputs were estimated or obtained based on published literature (Haque R, et al. J Am Pharm Assoc 2017, Jabbour EJ, et al. J Med Econ 2017): CML progression rates among patients with vs. without monitoring, the annual cost of CML progression, the average number of monitoring tests per year, and the average cost per CML monitoring test. Such information was used to assess the total healthcare cost and the cost differences per patient among patients with vs. without monitoring. The total healthcare costs were defined as the combined costs of CML monitoring tests and CML progression. Additionally, univariate and multivariate sensitivity analyses were conducted to test the robustness of the results by varying the model inputs.Results: Estimated average CML progression rates of 0.35% per patient year for patients with monitoring and 5.12% per patient year among patients without monitoring were used in the calculations (Table). The annual incremental cost of CML progression per patient of $136,308 was obtained from the published literature. The average number of monitoring tests per year was estimated at 3 and 0 among patients with and without molecular monitoring, respectively, with an average cost of $223 per monitoring test. The total healthcare costs including costs associated with CML progression and molecular monitoring tests were estimated to be $1,142 for patients with monitoring and $6,982 for patients without monitoring (difference = $5,840). In a hypothetical cohort of 100 CML patients, achieving a 90% monitoring rate (i.e., 90 out of 100 patients were monitored) was associated with a total cost reduction of $525,604 as compared to a 0% monitoring rate. This cost reduction remained consistent under both univariate and multivariate sensitivity analyses, demonstrating the validity of the model calculations.Conclusion: Improving CML monitoring rates is associated with improved patient outcomes in CML and therefore reduced healthcare costs. [Display omitted] DisclosuresJabbour:Bristol-Myers Squibb: Consultancy. Siegartel:Bristol-Myers Squibb: Employment. Lin:Novosys Health: Employment; Bristol-Myers Squibb: Consultancy. Lingohr-Smith:Novosys Health: Employment. Menges:Novosys Health: Employment. Makenbaeva:Bristol-Myers Squibb: Employment.