Verhoef, TI; (2013) Personalised treatment with oral anticoagulant drugs. Clinical and economic issues. Doctoral thesis , Utrecht University.

Preview

PDF Verhoef - Thesis.pdf Available under License : See the attached licence file. Download (3MB)

Abstract

Coumarin derivatives such as acenocoumarol, phenprocoumon and warfarin are frequently used for the prevention of stroke and systemic embolism in patients with atrial fibrillation or for the treatment of venous thromboembolism. These oral anticoagulants have a narrow therapeutic range and a large variability in dose requirement among patients. The anticoagulant effect is monitored by regular measurement of the International Normalised Ratio (INR). An INR below 2 is related to therapy failure and thereby leads to an increased risk of stroke or systemic embolism, while an INR above 3.5 leads to an increased risk of bleeding. Personalised dosing using genetic information is expected to help physicians to prescribe the required dose from the start of the therapy and thereby increase the efficacy and safety of the treatment. In the European Pharmacogenetics of Anticoagulant Therapy (EU-PACT) trial, we tested a dosing algorithm based on age, sex, height, weight, concomitant amiodarone use and CYP2C9 and VKORC1 genotypes. It is important to assess the cost-effectiveness of pharmacogenetic-guided coumarin dosing, because this information is used by payers such as health insurers to make decisions about reimbursement. In this thesis we describe the clinical and economic issues of personalised treatment with oral anticoagulant drugs. In Chapter 1 we provide a general introduction and describe the aims of this thesis. Our aim was to study genetic and other determinants that explain the variability in response to coumarin derivatives. We also aimed to study the economic consequences of different options (personalised medicine or using new oral anticoagulant drugs) to improve anticoagulant therapy. More background information is provided in Chapter 2. Currently, patients initiating coumarin anticoagulant therapy receive a standard loading dose for the first few days. The dose is adjusted when the INR is measured after a few days. When patients have reached a stable INR in the therapeutic range and a stable dose, INR measurements will be repeated every 4-6 weeks. The stable dose can vary up to 10-fold between patients. Polymorphisms in the CYP2C9 gene, coding for the main metabolising enzyme of coumarin anticoagulants, and the VKORC1 gene, coding for the target enzyme of coumarin anticoagulants, together explain approximately one-third of the variation in dose requirement. Many genotype-guided dosing algorithms for warfarin have been developed in different populations. In contrast, fewer algorithms have been published for acenocoumarol and phenprocoumon. With these algorithms, more than 50% of dose variation can be explained. Several clinical studies have been published describing the effect of pharmacogenetic-guided dosing on anticoagulation control, but no convincing evidence about the clinical significance exists yet. Currently ongoing large randomised trials such as EU-PACT are expected to provide this evidence and this information can then also be used to assess the cost-effectiveness of pharmacogenetic-guided dosing. Different determinants of variation in response to coumarin anticoagulants are studied in Part I of this thesis. We described the association between polymorphisms in CYP2C9 and VKORC1 and over- and under-anticoagulation in different time periods after treatment initiation in Chapter 3. VKORC1 genotype had the largest influence on over- or under-anticoagulation. Among wild-type patients, 73% of the patients had a subtherapeutic INR and 30% a supratherapeutic INR in the first month, compared to 45% with a subtherapeutic INR and 74% with a supratherapeutic INR among patients carrying two variant alleles. After the first month, the differences were much smaller and no difference was seen after the sixth month. In Chapter 4 we performed similar analysis for phenprocoumon and found no differences in risk of over- or under-anticoagulation between the genotypes after the first month. Pharmacogenetic information could therefore be used to prevent subtherapeutic or supratherapeutic INRs in the first month, but not after the first month for phenprocoumon or after 3-6 months for acenocoumarol. Concomitant medication use can also influence the required coumarin dose. In Chapter 5 we investigated the interaction between the proton pump inhibitors omeprazole and esomeprazole and phenprocoumon. On average, patients using omeprazole or esomeprazole required a dose that was 0.49 or 0.39 mg per day lower than non-users respectively. When these proton pump inhibitors were included in a genotype-guided dosing algorithm, 56.7% of the dose variation could be explained, which is only marginally higher than an algorithm not including proton pump inhibitors. Next to prescribing the correct dose to a patient, it is important that a patient is compliant with this dose. Therapy adherence is associated with the beliefs that patients have about their medicines. We described the beliefs that patients included in the EU-PACT trial have about acenocoumarol and phenprocoumon in Chapter 6. On average, coumarin users had a positive attitude towards their treatment. The beliefs score about the necessity of the treatment was higher than the beliefs score about the concerns (for example about side effects). This was not different from users of other cardiovascular drugs. However, patients with atrial fibrillation did have a lower score on the necessity beliefs than patients with venous thromboembolisms. Because of this less positive attitude, the risk of non-adherence is higher in patients with atrial fibrillation than in patients with venous thromboembolism. In Part II of this thesis we described studies on the cost-effectiveness of pharmacogenetic-guided dosing of coumarin derivatives, as well as the cost-effectiveness of the new oral anticoagulants dabigatran, apixaban and rivaroxaban. We provide an overview of cost-effectiveness studies on pharmacogenetic-guided dosing of coumarin derivatives published up to the end of 2009 in Chapter 7. All studies focused on genotype-guided dosing of warfarin in the United States, except one study on acenocoumarol in The Netherlands. The results of these studies varied considerably. In most studies genotyping led to improved health outcomes, but the costs were also higher than for standard care, with an incremental cost-effectiveness ratio up to almost 350,000 US$ per Quality-Adjusted Life-Year (QALY) gained. The wide variation in results made it impossible to conclude whether or not pharmacogenetic-guided dosing is a cost-effective strategy. More evidence on the effectiveness of genotyping on anticoagulation control was required and the costs of the genetic test needed to be defined more precisely. In Chapter 8 we evaluated a study by Meckley et al. on the benefits, risks and costs of pharmacogenetic-guided dosing of warfarin, published in 2010. In this study, the probabilities to experience a haemorrhagic or thromboembolic event were based on the time spent in different INR ranges. The percentage time spent in these ranges was derived from a clinical trial (CoumaGen) published in 2007. The incremental cost-effectiveness ratio of genotype-guided dosing versus standard care was US$60,000 per QALY gained, but the sensitivity analyses showed large uncertainty because the effectiveness of genotyping in clinical practice is still unclear. The model was adapted to study the cost-effectiveness of pharmacogenetic-guided phenprocoumon dosing in The Netherlands in Chapter 9. In 2012, the results of the CoumaGenII trial were published, which showed a larger effect of genotyping than the CoumaGen trial published in 2007. We used data on the percentage time spent in the different INR ranges from this trial and assumed that genotyping one patient using a point-of-care genetic test would cost €40. Pharmacogenetic-guided dosing increased the QALYs only by 0.0057 (2 days in full health), but the incremental costs were also low (€15). The incremental cost-effectiveness ratio was €2658. Because there was still a large uncertainty regarding the effectiveness, this study could not provide enough evidence to conclude whether or not pharmacogenetic-guided dosing should be implemented. The management and costs of anticoagulant care can influence the cost-effectiveness of pharmacogenetic-guided coumarin dosing. We described the organization and costs of anticoagulant care for the treatment of atrial fibrillation in 6 European countries in Chapter 10. The setting in which the management of the treatment took place varied from a specialised anticoagulation clinic to the general practitioner and hospital settings. The percentage time spent in the target range, which is a measure for quality of anticoagulation, also varied considerably between the countries. The highest percentage time within range was seen in The Netherlands, which uses a system of specialised anticoagulation clinics. Because of these differences and the differences in costs associated with coumarin therapy and management of complications, it is likely that the cost-effectiveness of pharmacogenetic-guided dosing will also vary appreciably among countries. It is therefore important to perform country specific cost-effectiveness analyses. In Chapter 11 we conducted country specific cost-effectiveness analyses on the new oral anticoagulants (NOACs) versus coumarin derivatives in The Netherlands and in the United Kingdom. In The Netherlands, acenocoumarol is most frequently used and treatment is monitored and guided by specialised anticoagulation clinics with a percentage time spent in the therapeutic range of 76-79%. In The United Kingdom, warfarin is most frequently used and many patients are treated by a general practitioner with a percentage time spent in the therapeutic range of approximately 63%. Because of these differences, the incremental cost-effectiveness ratios of the NOACs were higher in The Netherlands than in the United Kingdom, although dabigatran and apixaban could be considered cost-effective in both countries. The incremental cost-effectiveness ratio of rivaroxaban versus acenocoumarol in The Netherlands was almost €33,000 per QALY gained and this drug was therefore not considered as cost-effective in this country. In Chapter 12 we discuss the findings described in this thesis and their relevance, including the strengths and limitations of the studies, implications for clinical practise and future research. Pharmacogenetic-guided dosing of coumarin derivatives could be used to improve the therapy for patients with atrial fibrillation or venous thromboembolism. NOACs were also shown to be promising alternatives to coumarin derivatives. The results of the cost-effectiveness studies in this thesis underline the importance of country specific cost-effectiveness analysis when looking at the economic consequences of improving oral anticoagulant therapy.

Download activity - last month

Export as CSVJSONXML

Download activity - last 12 months

Export as XMLJSONCSV

Downloads by country - last 12 months

Export as CSVJSONXML

Archive Staff Only

View Item