Abstract
Arsenic trioxide (ATO)-induced hepatotoxicity is often observed in acute promyelocytic leukemia (APL) patients and decreases therapeutic effect of ATO. Thus, concerns over hepatotoxicity have been raised. The aim of this study was to explore some noninvasive clinical indicators that can be used to guide the individualized application of ATO in the future. APL patients treated with ATO were identified retrospectively via electronic health records at our hospital from August 2014 through August 2019. APL patients without hepatotoxicity were selected as controls. The association between putative risk factors and ATO-induced hepatotoxicity was estimated with ORs and 95% CIs, which were calculated using the chi-square test. The subsequent multivariate analysis was performed using logistic regression analysis. In total, 58.04% of patients experienced ATO-induced hepatotoxicity during the first week. Elevated hemoglobin (OR 8.653, 95% CI, 1.339-55.921), administration of nonprophylactic hepatoprotective agents (OR 36.455, 95% CI, 7.409-179.364), non-single-agent ATO to combat leukocytosis (OR 20.108, 95% CI, 1.357-297.893) and decreased fibrinogen (OR 3.496, 95% CI, 1.127-10.846) were found to be statistically significant risk factors for ATO-induced hepatotoxicity. The area under the ROC curve values were 0.846 for "overall ATO-induced hepatotoxicity" and 0.819 for "early ATO-induced hepatotoxicity." The results revealed that hemoglobin ≥ 80 g/L, nonprophylactic hepatoprotective agents, and non-single-agent ATO and fibrinogen < 1 g/L are risk factors for ATO-induced hepatotoxicity in newly diagnosed APL patients. These findings can enhance the clinical diagnosis of hepatotoxicity. Prospective studies should be performed in the future to validate these findings.