A clinical case of hemorrhagic complication during anticoagulant therapy: the role of pharmacogenetic testing

Introduction

There are many different diseases for which anticoagulant therapy is used. These diseases include atrial fibrillation (AF), deep vein thrombosis, pulmonary embolism, and others. In patients with AF, direct oral anticoagulants (DOACs) are currently the first-line drugs for the prevention of thromboembolic complications and are used lifelong. DOACs have high bioavailability, ease of use, a minimal spectrum of interactions with other drugs, a wide therapeutic window, and no need for laboratory monitoring [1-3]. Despite proven efficacy and safety, the use of DOACs, in the presence of certain factors, may be accompanied by an increased risk of hemorrhagic complications, both life-threatening and minor but clinically significant [4]. Factors that increase the risk of hemorrhagic complications are usually divided into modifiable, partially modifiable, and non-modifiable [1, 2]. One of the non-modifiable risk factors is genetic, which includes mutations in the genes responsible for DOAC biotransformation [1, 2]. Currently, the role of pharmacogenetic testing for the prevention of hemorrhagic complications during DOAC therapy is widely discussed.

Clinical case

Patient D., 83 years old, was hospitalized in a multidisciplinary hospital in Moscow on a planned basis for examination. Upon admission, the patient complained of gingival bleeding, subcutaneous hemorrhages, and epigastric pain.

History of present illness. From the provided medical documentation, it was found that the patient had long suffered from arterial hypertension (AH), permanent atrial fibrillation, chronic heart failure, and in 2017 had suffered an acute myocardial infarction (AMI) of the inferior wall of the left ventricle. For the past several years, the patient had been taking the following drug therapy: rivaroxaban tablets 15 mg once daily in the morning, metoprolol tablets 25 mg twice daily, enalapril tablets 5 mg twice daily, torasemide tablets 5 mg once daily in the morning, and spironolactone tablets 25 mg in the morning.

Past medical history. Family history for cardiovascular disease was unremarkable. The patient has two children; the eldest daughter also suffers from AH. In 1986, the patient underwent appendectomy; in 2000 — cholecystectomy.

Physical examination. Height 161 cm, weight 85 kg, body mass index (BMI) 32.79 kg/m². General condition satisfactory. Body temperature 36.7°C. Petechial hemorrhages were noted on the skin of the trunk and upper extremities; gingivitis was noted upon examination of the oral cavity; visible mucous membranes were of physiological color and moisture, without pathological changes. Edema of the lower extremities on the feet and up to the middle third of the legs bilaterally. Respiratory rate 17 per minute. SpO₂ 98%. On auscultation of the lungs, vesicular breathing was audible in all areas, no wheezing. On cardiac auscultation — irregular rhythm due to atrial fibrillation, no pathological murmurs. Heart rate 64-86/min. Blood pressure 134/82 mmHg. Abdomen enlarged due to subcutaneous fat, soft, painless on palpation. Physiological functions within normal limits: urination regular and painless; stool regular, without pathological inclusions. Percussion sign negative bilaterally.

Examination findings. Complete blood count: erythrocytes 4.3×10¹²/L, hemoglobin 115 g/L (N 120-140 g/L), leukocytes 6.2×10⁹/L, platelets 280×10⁹/L (N 150-400×10⁹/L). Biochemical blood test: Urea 5.2 mmol/L, Creatinine 115 μmol/L (N 53-115 μmol/L). Calculated creatinine clearance by Cockcroft-Gault formula = 44 mL/min. Glomerular filtration rate (GFR) (by CKD-EPI formula): 38 mL/min/1.73 m². Coagulogram: Activated partial thromboplastin time 32 seconds, INR 1.5. Prothrombin time (PT) 15 seconds (N 10.0-13.2 s), Fibrinogen 2.2 g/L. Urinalysis — within reference values. Electrocardiography (ECG): AF with ventricular rate 60-84/min. ECG signs of left ventricular myocardial hypertrophy. Signs of previous focal lesion of the inferior wall of the left ventricle. Echocardiography: study performed during atrial fibrillation, normosystolic form. Left ventricular ejection fraction (LVEF) 55%. Left ventricular hypertrophy (LVH). Scar changes in the myocardium of the inferior wall of the left ventricle. Esophagogastroduodenoscopy: Cardia insufficiency. Reflux esophagitis. Superficial gastritis with pinpoint hemorrhages. Bulbitis.

Based on complaints, medical history, and examination findings, the following diagnosis was established: Ischemic heart disease: Post-infarction cardiosclerosis (AMI 2017). Stage 3 hypertension, Controlled hypertension, cardiovascular complication risk 4 (very high). LVH. Cardiac arrhythmia: Permanent atrial fibrillation, normosystolic form (CHA₂DS₂-VASc 4 points, HAS-BLED 2 points). Chronic heart failure with preserved left ventricular ejection fraction (LVEF 55%) stage IIA, NYHA class II. Class I obesity (BMI 32.79 kg/m²). Superficial gastritis with pinpoint hemorrhages. Reflux esophagitis. Bulbitis.

Additional examination findings. Therapeutic drug monitoring (TDM) of DOAC: Minimal steady-state concentration of rivaroxaban — 98 ng/mL (N 6-87 ng/mL). Pharmacogenetic testing: homozygous carrier of the ABCB1 (rs1045642) gene polymorphism — mutant TT genotype.

Discussion

Hemorrhagic complications in the form of petechial skin rashes, gingival bleeding, and pinpoint hemorrhages on the gastric mucosa in this patient could be due to DOAC (rivaroxaban) intake. Although the rivaroxaban dose was correctly adjusted considering GFR, the serum drug concentration was elevated. This may have been caused by a mutation in the ABCB1 (MDR1) gene. This gene is responsible for the function of the P-glycoprotein transporter protein, whose role is to prevent the absorption of xenobiotics and, once they enter the body, to ensure their rapid elimination [5]. It is known that rivaroxaban is metabolized ²/₃ via cytochrome P450 enzymes (CYP3A4/5 and CYP2J2) and ¹/₃ is excreted unchanged in the urine by P-glycoprotein (P-gp) and Breast Cancer Resistance Protein (BCRP) transporters [6]. Thus, carriage of the mutant TT allele of the ABCB1 gene could have slowed rivaroxaban elimination, which, according to TDM data, led to increased concentration and prolonged PT, and ultimately to hemorrhagic complications. The absence of anti-Xa activity data in the presented clinical case is due to a fundamental methodological choice: the aim of the work was to demonstrate the possibility of using an alternative monitoring method (measuring rivaroxaban concentration).

A clinical case of hemorrhagic complication during rivaroxaban therapy was also previously described in the work of Lorenzini K. et al. [7]. The authors reported gastrointestinal bleeding in a 79-year-old patient while taking rivaroxaban 20 mg daily (for 3 months). The patient suffered from heart failure and type 2 diabetes mellitus. Concomitant therapy included: insulin, simvastatin 40 mg once daily, levothyroxine 75 mcg once daily, extended-release metoprolol 25 mg once daily, and enalapril 10 mg once daily. To assess the reasons for the potential enhancement of rivaroxaban's effect at therapeutic doses, anti-Xa activity measurement, plasma rivaroxaban concentration measurement, ABCB1 genotyping, and CYP3A4/5 phenotyping were performed. Laboratory tests showed high levels of anti-Xa activity and plasma rivaroxaban concentration (24 hours after the last dose) and an unexpected delay in rivaroxaban clearance, indicating impaired rivaroxaban elimination. The patient was a homozygous carrier of both tested ABCB1 variant alleles: his genotype was TT for both polymorphisms (rs1045642 and rs2032582). CYP3A4/5 phenotyping showed moderately reduced enzymatic activity with an OH-midazolam/midazolam metabolic ratio of 0.31. The authors suggested that both genetic factors and environmental factors could have contributed to increased susceptibility to rivaroxaban in the presented patient, for example, homozygous presence of ABCB1 variant alleles and reduced CYP3A4/5 activity due to drug interaction with simvastatin in addition to moderately reduced renal function (calculated creatinine clearance by Cockcroft-Gault formula was 39 mL/min).

To reduce the risk of hemorrhagic complications in the patient from our clinical example, switching to another DOAC with lower renal clearance compared to rivaroxaban may be considered. Apixaban could be such a drug.

Conclusion

This clinical case demonstrates the significant role of pharmacogenetic testing in diagnosing the cause of hemorrhagic complications in a patient with AF while taking DOACs (rivaroxaban). Although this method is not part of routine practice, its use may be appropriate in cases of bleeding during long-term DOAC therapy in patients with a low risk of bleeding according to the HAS-BLED scale.