Pharmacogenetics of anticoagulation and clinical events in warfarin-treated patients: A retrospective cohort study with Finnish biobanks and national health registries

Introduction: Warfarin treatment is challenged by a narrow therapeutic range and high inter-individual variation in treatment response. Individuals carrying decreased function alleles of VKORC1 and CYP2C9 are more sensitive to warfarin, and may experience over-anticoagulation which may increase risk of bleeding events. Several studies have shown increased anticoagulation in association with these alleles; however, fewer have investigated clinical events, and the results are contradictory. The aim of the current study was to assess the association between VKORC1 and CYP2C9 variants and the incidence of treatment complications in warfarin-treated patients in a real-world setting by using a register linkage design with Finnish biobanks and national health registries. Methods. Data from multiple Finnish biobanks, national health registries and patient records were linked using personal identity code to investigate pharmacogenetics in warfarin-treated patients. The inclusion criteria were as follows: 1) ≥ 18 years of age, 2) CYP2C9 and VKORC1 genotype information available in the biobank, 3) a diagnosis of a disease of cardiovascular system (I20-26, I48, I63-I66, I67.2, I69.3-I69.8, I70, I80-I82, and 4) at least one warfarin purchase between 1.1.2007 - 30.06.2018. The three biobanks (THL Biobank, Auria Biobank, Helsinki Biobank) first identified individuals genotyped in biobank studies for CYP2C9*2 (rs1799853) and *3 (rs1057910) and VKORC1 g.3588G>A (rs9923231). These data were further linked with medication purchases, patient health records, laboratory data, and healthcare encounters. The primary outcome was the occurrence of bleeding events during warfarin use. INR tests were extracted from the laboratory data and analyzed using the following parameters: time in therapeutic range, time to reach therapeutic range, INR<2, and INR>3. Average daily dosage was calculated from the medication purchases. The patients were divided into two warfarin sensitivity groups normal responders, and sensitive & highly sensitive responders on the basis of their VKORC1 and CYP2C9 genotypes. The study was approved by the ethics committee of the Hospital District of Helsinki and Uusimaa (HUS/513/2019). Results. Of the 2508 participants, 45% were women with a mean age of 69 years, 65% were categorized as normal responders and 35% sensitive or highly sensitive responders. The most common inclusion diagnoses were atrial fibrillation and a vascular disease. The background characteristics did not differ between the genotype groups. Compared to normal responders, sensitive & highly sensitive responders had less INR test results below 2 (33.3% vs. 44.4%, p<0.001) and more above 3 (20.0% vs. 7.1%, p< 0.001). Other analyzed INR parameters did not differ between the groups. Mean daily warfarin dose was higher among normal responders (5.5mg vs. 3.9mg, p<0.001). Incidence (per 100 patient-years) of bleeding outcomes was 5.4 in normal responders and 5.6 in sensitive & highly sensitive responder group (HR=1.03, 95% Cl: 0.74-1.44, p=0.864). Discussion. In a real-world setting, the genetically sensitive & highly sensitive responders to warfarin had more out-of-range INR tests and required a lower daily dose of warfarin than normal responders. However, the risk for bleeding events was not significantly increased in the sensitive & highly sensitive responders, suggesting that the potential clinical benefit of genotype-guided warfarin dosing is limited.