ASPO Abstracts
Metabolite Set Enrichment Analysis Reveals Pathways Associated with Minimal Residual Disease and Relapse in Acute Lymphoblastic Leukemia
Category: Molecular Epidemiology & Environment, Early Detection & Risk Prediction
Conference Year: 2018
Abstract Body:
Purpose: The purpose of this study was to identify alterations in metabolites and metabolic pathways in diagnostic bone marrow plasma which were associated with pediatric acute lymphoblastic leukemia (ALL) minimal residual disease (MRD) and relapse. Methods: Diagnostic bone marrow plasma from N=100 ALL cases treated at Texas Children’s Hospital from 2007- 2015 were sent to Metabolon for global metabolomic profiling by UPLC-MS/MS. Metabolites (N=416) were identified and mapped to the Kyoto Encyclopedia of Genes and Genomes Compound database. Concentrations of each metabolite were re-scaled such that the median was 1. We conducted metabolite set enrichment analysis using IMPaLA to identify metabolic pathways with differential activity in MRD+ vs. MRD- and relapsed vs. non-relapsed patients. The false discovery rate (FDR) was used to account for multiple comparisons. Results: The sample included 39 MRD+ and 23 relapsed patients. At FDR<0.05, MRD+ patients were distinguished by altered tricarboxylic acid cycle (TCA, p = 1.9 x 10-6) activity as well as increased concentrations of glutathione, spermine, and spermidine at the time of diagnosis (“glutathione metabolism” pathway, p < 0.005). Mean concentrations of long-chain and unsaturated fatty acids were decreased among relapsed as compared to non- relapsed patients (p = 7.6 x 10-6). Relapsed patients also demonstrated a trend towards increased concentrations of early, but not late, TCA cycle intermediates (citrate, isocitrate and cis-aconitate, p = 0.008), adenosine monophosphate (AMP) and cyclic AMP (p = 0.001). Conclusions: In this assessment, we identified differential activity of key metabolic pathways associated with poor treatment response in ALL. Increased abundances of TCA cycle intermediates were associated with both MRD and relapse. Additionally, increased concentrations of polyamines were associated with MRD whereas decreased concentrations of long-chain fatty acids were associated with relapse. These changes may drive chemoresistance by affecting proliferation and attenuating chemotherapy-induced apoptotic signaling. Metabolomics demonstrates potential utility for early identification of high-risk ALL patients, and for identifying mechanisms of chemoresistance. This work was supported by CPRIT RP160097
Keywords: acute lymphoblastic leukemia; pediatric cancer; metabolomics; minimal residual disease; relapse