Evaluating miR-122-5p as a biomarker for anti-tuberculosis drug-induced liver injury: diagnostic efficacy and correlation with liver function indicators

Background Drug-induced liver injury (DILI) is characterized by hepatic dysfunction due to exogenous agents, potentially progressing to acute liver failure. MicroRNA-122-5p (miR-122-5p) constitutes about 70% of total liver microRNAs, offering high liver specificity and superior sensitivity for detecting injury compared to traditional transaminases, without interference from renal or muscle injury. This cross-sectional study aims to evaluate the diagnostic utility of miR-122-5p as a biomarker for anti-tuberculosis DILI. Methods miR-122-5p expression was measured by real-time quantitative polymerase chain reaction (PCR). Descriptive statistics summarized sample characteristics. For normally distributed continuous variables with equal variances, data were presented as mean ± standard deviation, with comparisons among three groups using one-way analysis of variance (ANOVA); otherwise, data were expressed as median (interquartile range) with Kruskal–Wallis H test, followed by Mann–Whitney U pairwise comparisons. Categorical variables were analyzed by Chi-square test. Receiver operating characteristic (ROC) analysis assessed diagnostic efficacy (area under the curve (AUC), optimal cutoff, sensitivity, specificity). Correlations with alanine aminotransferase (ALT), aspartate aminotransferase (AST), and alkaline phosphatase (ALP) were evaluated by Spearman’s rank correlation, and scatter plots with fitted lines were generated using Python. Results The DILI group showed significantly higher miR-122-5p expression (2.48 ± 0.81) than the Liver biochemical abnormality and Healthy control groups ( P < 0.001). ROC analysis yielded an AUC of 0.829 (95% confidence interval (CI) [0.732–0.926]), with sensitivity 0.833 and specificity 0.760 for distinguishing DILI from non-DILI. miR-122-5p positively correlated with ALT ( ρ = 0.626, P < 0.001) and AST ( ρ = 0.522, P < 0.001), but not with ALP ( ρ = 0.193, P = 0.114). No significant intergroup differences in age or gender were found ( P > 0.05). Conclusions Our research indicates that miR-122-5p has potential utility as a biomarker for anti-tuberculosis DILI, particularly in distinguishing it from healthy controls. However, these findings are exploratory, based on a small cohort with no independent validation, and its incremental value beyond conventional liver function tests remains to be determined.