Vitamin K antagonists (VKAs) are used off-label in pediatric patients to treat or prevent thromboembolic events. Dosing of VKAs in pediatric patients is complex, partly because of the developing hemostatic system in these patients. Furthermore, there is a large inter- and intra-individual variability in VKA dose requirement. Little is known about the use of acenocoumarol and phenprocoumon in pediatric patients.
Therefore, this thesis aimed to study important knowledge gaps in the use of VKAs in pediatric patients and to develop dosing algorithms to improve their application.
To achieve this, we conducted a multicenter retrospective follow-up study, The Children Anticoagulation and Pharmacogenetics Study (CAPS). Patients were selected and invited for participation in four pediatric hospitals (Amsterdam, Rotterdam, Utrecht and Groningen) and one anticoagulation clinic in Leiden. In total, 213 patients participated in CAPS and all their treatment related information was collected in the hospitals and anticoagulation clinics involved in the management of their treatment. Furthermore, the patients were asked to collect a saliva sample to genotype them for 7 genetic variations in 5 genes (VKORC1, CYP2C9, CYP4F2, CYP2C18, and CYP3A4).
We have shown that pediatric patients are approximately 50% of their time within therapeutic International Normalized Ratio (INR) range (TTR) during the first month of VKA treatment. Over the course of the first year, this percentage increased to more than 64%. 14.6% of the patients using acenocoumarol and 31.3% of the patients using phenprocoumon developed a bleeding during the first year of VKA treatment. A large proportion (20-22%) of these patients developed their first bleeding within the first few weeks of treatment. Furthermore, we have shown that younger patients were more prone to have a lower TTR, lower chance of reaching stable INR values, and a higher risk for developing a bleeding event within the first year of treatment.
We have shown that for acenocoumarol 45.0% of the variability in dose requirement was explained by body surface area and the indication for its use. By adding the genotypes of VKORC1, CYP2C9 and CYP2C18 to the model this increased to 61.8%. For phenprocoumon age explained 56.2% of the variability in dose requirement. By adding the genotype of VKORC1, CYP2C9 and CYP3A4 to the model 80.4% of the variability was explained.
In conclusion, we have shown that the quality of anticoagulation in the first month of VKA use was relatively low and should be improved. Bleeding events were common in the first year of treatment and often occurred in the first few weeks of treatment. Younger pediatric patients were more likely to have a low TTR in the first 3 months of VKA use and to develop a bleeding event. The required VKA dose is largely depending on clinical factors, but also genetic variations have an important role in explaining dose variations. The developed dosing algorithms can be used to predict the required dose and it is hypothesized that this will improve anticoagulation control especially in the first 3 months of treatment by improving the time to reach stable INRs within TR.