Drug-Induced Liver Injury During Consolidation Therapy in Childhood Acute Lymphoblastic Leukemia As Assessed for Causality Using the Updated RUCAM

Purpose: The presence of serious toxicities is a major problem in the treatment of childhood acute lymphoblastic leukemia (ALL). The objective of this research is to evaluate drug-induced liver injury (DILI) during consolidation therapy in childhood ALL.

Methods: Clinical data of pediatric patients who received consolidation therapy between August 2012 and July 2018 were collected. Characteristics (incidences and patterns) of DILI at different stratifications were determined. Risks of DILI were evaluated using binary logistic regression analysis. Drug causality assessment was carried out by the updated Roussel Uclaf Causality Assessment Method (RUCAM).

Results: Patients with high risk (HR) and standard risk (SR)/intermediate risk (IR) received 270 and 1539 courses of consolidation therapy, respectively; among these courses, 15 (5.6%) and 38 (2.5%) developed DILI. The occurrences of DILI in SR/IR patients were primarily associated with age (≤5.2 years), treatment course (≥5), and baseline serum parameters before treatment (cystatin C > 0.79 mg/L, albumin ≤45 g/L, and gamma-glutamyl transpeptidase (GGT) > 17 U/L). The ROC curve generated using the parameters assigned to specific values achieved an area under the curve (AUC) of 0.846 (95% CI 0.827-0.863) with a cutoff value of 3, and the sensitivity and specificity were 94.7% and 62.3%, respectively. For HR patients, a decrease in baseline albumin and elevation of baseline liver enzymes (GGT and aspartate aminotransferase) were observed in DILI cases compared with the non-DILI subjects. In the SR/IR group with DILI, the causality gradings for high-dose methotrexate (HD-MTX) were highly probable in 5 (13.2%) cases, probable in 31 (81.6%) cases, and possible in 2 (5.3%) cases. Among the DILI cases in HR-1, HR-2, and HR-3 groups, high causality gradings (probable + highly probable) were detected in "100% of HD-MTX + 57% of high-dose cytarabine (HD-Ara-C)," "100% of HD-MTX + 20% of pegylated asparaginase (PEG-ASP)," and "100% of HD-Ara-C + 33.3% of PEG-ASP," respectively.

Conclusion: Incidence of DILI in HR patients was significantly higher than that in SR/IR patients. A number of potential risk factors were identified, among which the preexisting liver conditions were suggested as shared risk factors in all stratification groups. HD-MTX, HD-Ara-C, and PEG-ASP were the main causative agents of DILI. The knowledge generated from this study will be helpful for understanding characteristics of DILI during consolidation treatment in childhood ALL.