P-33 EVALUATION OF CIRCULATING METABOLOME IN THE SEARCH OF POTENTIAL DRUG-INDUCED LIVER INJURY BIOMARKERS

Introduction

Idiosyncratic drug-induced liver injury (DILI) is a complex hepatic condition whose diagnosis is challenging due to lack of specific biomarkers.

Objectives

We aimed to evaluate serum metabolomic differences between patients with DILI and with other causes of liver injury in search for specific DILI biomarkers.

Methods

Metabolomic profiles of serum samples from 26 Spanish DILI patients, 34 with non-DILI acute liver injury (ALI) and 48 healthy controls, were analyzed using UHPLC-MS. To assess changes in disease progression, DILI and ALI patients were followed-up from detection (visit 1), one week (visit 2) and >30 days (visit 3). Data were analyzed using Shapiro-Wilk, Student's t and Wilcoxon tests.

Results

Several amino acids, fatty acids (FAs), LPI and bile acids were increased, whereas the glycerophospholipids MEPE and MAPC were decreased (p<0.05) when DILI was compared to control (visits 1&2). More metabolites were altered when ALI was compared to controls, with higher levels of FAs and lower levels of MEPE and MEPC. Both DILI and ALI showed fewer differences at visit 3 compared to controls, although several FAs remained increased. The differences found were more limited when ALI vs DILI were compared. However, at visit 1 ALI showed a significant higher increase in the bile acids and 31 FAs than DILI patients, but with lower levels of MEPE, tryptophan and alanine. Remarkably, the amino acids Phe-Phe, taurine, glutamic acid and lysine were significantly increased in DILI patients as compared to controls (p<0,05) but did not differ between ALI and controls (p>0,05).

Conclusion

Most metabolomic differences are found at times closer to DILI recognition (visits 1&2), although abnormal values of FAs remain during recovery. Some FAs species and the amino acids taurine, Phe-Phe glutamic acid, lysine, tryptophan and alanine seem promising DILI biomarker candidates that should be further explored. Funding: CIBERehd, ISCiii-FEDER PI18/00901, PI19/00883.