Use of Pharmacogenetic Factors to Predict Warfarin Dose in Children - a Retrospective Cohort Study

Background: Warfarin is the most commonly used oral anticoagulant, but its use is limited by the narrow therapeutic index. Single nucleotide polymorphisms in cytochrome P450 2C9 (CYP2C9) and vitamin K epoxide reductase (VKORC1) have been shown to impact warfarin dose requirements. Use of such pharmacogenetic data to improve warfarin dosing has been widely studied in adult populations and is being integrated into clinical practice. However, there remains a paucity of data on the appropriate use of this genetic information in the pediatric population. We developed a retrospective review of pediatric patients on warfarin therapy to evaluate associations between CYP2C9 and VKORC1 polymorphisms and warfarin dose requirements in children.

Methods: Patients were recruited from outpatient clinics at Texas Children’s Hospital and Stollery Children’s Hospital. Any patient less than 18 years of age that had taken warfarin and achieved maintenance dose was eligible. Blood sample or buccal swab was collected for genotyping and clinical data was collected from the medical record. Genotyping was performed by melting curve analysis to detect CYP2C9*2, CYP2C9*3, or VKORC1 −1639 mutations. Warfarin genotyping reagents were obtained from Idaho Technology and assays were run using the Roche LightCyler instrument.

Results: This study was initiated in March 2008 and is ongoing. To date 47 children (23M: 24F) have been enrolled. Indications for warfarin therapy include: arrhythmia (n=2), artificial heart valve (n=10), pulmonary hypertension (n=10), deep vein thrombosis (n=9), pulmonary embolism (n=2), Fontan (n=10), stroke (n=2), other (n=2). The mean age is 9.3 years (range 1–18); median target International Normalized Ratio (INR) range is 2–3 (with similar target INRs across genotypes). Twenty-nine patients have fully evaluable data at the time of this report, of whom 21 (72%) are wild type CYP2C9, 6 (21%) carry at least 1 mutant allele, and 2 (7%) are homozygous or compound heterozygote. As compared with wild type patients (mean dose 0.14mg/ kg/day), patients with heterozygous (0.09mg/kg/day; p=0.04) or homozygous (0.05mg/ kg/day; p=0.001) genotype required significantly lower maintenance daily warfarin dose. For VKORC1 −1639 genotype 6 (21%) are AA, 11 (38%) are GA, and 12 (41%) are GG. As compared with GG patients (mean dose 0.16mg/kg/day), patients with 1 (0.11mg/kg/ day; p=0.12) or 2 (0.09mg/kg/day; p=0.02) A allele required lower doses.

Discussion: We present the first pediatric series to evaluate the influence of CYP2C9 and VKORC1 genotype on warfarin dosing. These data demonstrate genotype frequency similar to adults, and suggests that genotype influences the maintenance warfarin dose in children. Enrollment to this study is ongoing, and will allow a more precise evaluation of the impact of genotype and additional demographic data on pediatric dosing. These data will provide for development of a pediatric dosing algorithm to predict maintenance warfarin dose a priori.