Predictive model of bosentan-induced liver toxicity in Japanese patients with pulmonary arterial hypertension.

Bosentan, an endothelin receptor antagonist, has been widely used as a first-line medication for the treatment of pulmonary arterial hypertension (PAH). It has been shown to improve symptoms of hypertension, exercise capacity, and hemodynamics and prolong time to clinical worsening. However, liver dysfunction is a major side effect of bosentan treatment that could hamper the optimal management of patients with PAH. Previously, we demonstrated, using drug metabolism enzymes and transporters analysis, that the carbohydrate sulfotransferase 3 (CHST3) and CHST13 alleles are significantly more frequent in patients with elevated aminotransferases during therapy with bosentan than they are in patients without liver toxicity. In addition, we constructed a pharmacogenomics model to predict bosentan-induced liver injury in patients with PAH using two single-nucleotide polymorphisms and two nongenetic factors. The purpose of the present study was to externally validate the predictive model of bosentan-induced liver toxicity in Japanese patients. We evaluated five cases of patients treated with bosentan, and one presented with liver dysfunction. We applied mutation alleles of CHST3 and CHST13, serum creatinine, and age to our model to predict liver dysfunction. The sensitivity and specificity were calculated as 100% and 50%, respectively. Considering that PAH is a rare disease, multicenter collaboration would be necessary to validate our model.

CHST3 and CHST13 polymorphisms as predictors of bosentan-induced liver toxicity in Japanese patients with pulmonary arterial hypertension.

Bosentan, an endothelin receptor antagonist, has been widely used as a first-line drug for the treatment of pulmonary arterial hypertension (PAH). In addition, bosentan is approved for patients with digital ulcers related to systemic sclerosis. Liver dysfunction is a major adverse effect of bosentan and may lead to discontinuation of therapy. The purpose of this study was to identify genomic biomarkers to predict bosentan-induced liver injury. A total of 69 PAH patients were recruited into the study. An exploratory analysis of 1936 single-nucleotide polymorphisms (SNPs) in 231 genes involved in absorption, distribution, metabolism, and elimination of multiple medications using Affimetrix DMET™ (Drug Metabolism Enzymes and Transporters) chips was performed. We extracted 16 SNPs (P < 0.05) using the Jonckheere-Terpstra trend test and multiplex logistic analysis; we identified two SNPs in two genes, CHST3 and CHST13, which are responsible for proteoglycan sulfation and were significantly associated with bosentan-induced liver injury. We constructed a predictive model for bosentan-induced liver injury (area under the curve [AUC]: 0.89, sensitivity: 82.61%, specificity: 86.05%) via receiver operating curve (ROC) analysis using 2 SNPs and 2 non-genetic factors. Two SNPs were identified as potential predictive markers for bosentan-induced liver injury in Japanese patients with pulmonary arterial hypertension. This is the first pharmacogenomics study linking proteoglycan sulfating genes to drug-induced liver dysfunction, a frequently observed clinical adverse effect of bosentan therapy. These results may provide a way to personalize PAH medicine as well as provide novel mechanistic insights to drug-induced liver dysfunction.