Introduction: Cancer associated venous thromboembolism (CAT) is a major complication of cancer; results in more than 2-fold increased mortality in cancer patients compared to those without CAT. Approximately 20% of all venous thromboembolism is contributed to by patients with cancer, indicating a huge economic burden. While low molecular weight heparin (LMWH) has been the standard treatment for CAT, more recently several direct oral anticoagulants (DOACs) have been tested in randomized controlled trials and are now available for use (Table 1). However, these agents have not all been compared head-to-head and the decision to choose is often based on arbitrary factors such as patient comorbidities, scheduling ease, availability of agents, patient and physician preference, insurance coverage etc. The objective of our study was to examine the cost-effectiveness of DOACs when compared to the traditional LMWH.
Methods: We constructed a 9-state Markov transition decision model to examine strategies of 1) LMWH, 2) Apixaban, 3) Edoxaban, and 4) Rivaroxaban for the treatment of CAT (Figure 1). Our hypothetical cohort consisted of cancer patients who entered the Markov model at almost 50 years of age. These patients were assumed to have a base case utility of 0.609 (adapted from the literature, specific for cancer patients with VTE with no distant metastasis residing in the US). Survival rates were based on 2011 life tables of the US population extracted from the Centers for Disease Control and Prevention database. Data sources included MEDLINE searches, bibliographies from relevant articles in English language. The Center for Medicare/Medicaid Services 2018 data was used to extract drug costs. Cycle length was one month and we used a lifelong time horizon with a healthcare system perspective. Outcomes measure included effectiveness measured in quality-adjusted life years (QALYs) and costs measured in 2019 U.S. dollars.
We made some "simplifying" assumptions to run our model: 1) Effect of treatment and side effects of each medication were maintained through the life cycle and no crossover was allowed after failure 2) Cost and morbidity of major bleeding was considered temporary; (one cycle length i.e. one month), after which the patient would recover and his utility would be equivalent to that of base case. Anticoagulants would be discontinued for this patient population for the rest of their life time. 3) Patients whose anticoagulants were discontinued were assumed to not experience another bleeding event but they were still at risk for developing a recurrent VTE. 5) Due to the paucity of literature evaluating the utility value of anticoagulants in cancer population, we assumed that the all DOACs had the same utility (=1) as they are taken orally. Meanwhile, LMWH was assumed to have a lower utility of about 0.98 as it is administered as a subcutaneous injection.
Results: Apixaban dominated all other strategies by being less costly and more effective than LMWH or the other DOACs. All relative risks and hazard ratios were compared with LMWH. One-way sensitivity analyses were conducted to examine the impact of uncertainty in selected parameter values (relative risks, hazard ratios, costs of different agents, age and utilities). 1) Apixaban remains cost-saving, unless the hazard ratio for recurrent VTE on Apixaban exceeds 3.8. It remains cost-effective with an incremental cost-effectiveness ratio (ICER) < $50,000/QALY up to a hazard ratio 4.08. 2) Apixaban remains cost saving till the relative risk for major bleeding exceeds 1.58 and it remains cost-effective (ICER < $50,000/ QALY) up to a relative risk of 1.61. However when the relative risk for major bleeding for Apixaban exceeds 1.7, the ICER exceeds $150,000/QALY. 3) Apixaban remains cost saving as long as its monthly cost is under $270. It is no longer cost-effective (ICER > $50,000/ QALY) when the monthly cost of Apixaban exceeds $299.
Conclusion: Treatment of CAT with Apixaban is both more effective and less costly when compared to other modalities, including LMWH, Edoxaban and Rivaroxaban.
No relevant conflicts of interest to declare.
This icon denotes a clinically relevant abstract
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal