BIOINFORMATIC CHARACTERIZATION OF IMMUNE CELL TYPES IN THE BONE MARROW OF CHILDHOOD ACUTE LYMPHOBLASTIC LEUKEMIA PATIENTS FROM ARGENTINA THROUGH ANALYSIS OF TRANSCRIPTOME DATA

AVEDAÑO, DANIEL; ABBATE, MERCEDES; RICCHERI, CECILIA; COTIGNOLA, JAVIER; RUIZ, MARÍA SOL

Congreso; LXVI REUNIÓN ANUAL DE LA SOCIEDAD ARGENTINA DE INVESTIGACIÓN CLÍNICA; 2021

SOCIEDAD ARGENTINA DE INVESTIGACIÓN CLÍNICA

Acute lymphoblastic leukemia (ALL), the most incident pediatric cancer, is characterized by the overproduction of immature lymphoid blasts in the bone marrow (BM). While considerable progress has been made on treatment efficacy and survival rates, about 15-30% of patients relapse and/or die. Immunotherapies are promising as complementary treatments to chemotherapy, yet the most relevant therapy targets and patient subsets remain unclear. Aims: To characterize immune cell types in the BM microenvironment of ALL samples using predictive bioinformatic tools on transcriptome data. Methods: we performed RNA-seq on BM samples collected at ALL diagnosis (N=32). The proportion of immune cells was estimated using MIXTURE through two gene expression signatures (LM22, TIL10). A ?cytolytic score? reflecting CD8+ cytotoxic T lymphocytes (CTLs) and Natural Killer cells (NK) abundance was calculated as the geometric mean of 5 genes specifically expressed in CTLs/NK. Gene set enrichment analysis using ImmuneSigDB and Reactome was performed with GSVA package in R. Results: Relative proportions of B and T cells were concordant with B- and T-cell ALL subtype, respectively. Cytolytic score was positively correlated with CD8+ T cells and NK proportions (Spearman Rho>0.38, p-val 1, and half of them relapsed or died. Gene set enrichment analysis for >1 vs.