In This Issue

Abstract

See article, p. 16.B-precursor acute lymphoblastic leukemia (B-ALL) relapse often involves the central nervous system (CNS). In this article, Vanner, Dobson, et al. characterize B-ALL subclones with differential CNS involvement in patient-derived xenografts. By comparing transcriptional profiles, the authors identify complement component C3 and stem cell, metabolic, and mRNA translation pathways enriched in subclones with preferential CNS residence regardless of mutational composition. These findings are validated by proteomics and by genetic or pharmacologic targeting of the candidate pathways. The protein translation inhibitor omacetaxine mepesuccinate strongly reduces CNS infiltration.See article, p. 32.In this study, Petti et al. evaluated mechanisms that contribute to the relapse of patients with acute myeloid leukemia (AML) after therapy. Bone marrow samples from presentation and relapse from six patients with normal karyotype AML were subjected to enhanced whole-genome sequencing to identify all somatic mutations and their evolution. Single-cell RNA sequencing defined transcriptional and cellular changes at relapse. Each case had a unique set of genetic and transcriptional adaptations associated with relapse, but most acquired a conserved relapse-enriched leukemia cell (RELC) signature that had features of stemness, quiescence, and adhesion.See article, p. 50.Diffuse large B-cell lymphomas (DLBCL) are dependent on anaplerotic glutaminolysis driven by SIRT3. Li et al. show that the metabolic collapse induced by SIRT3 depletion leads to marked repression of ATF4 target genes that are highly expressed in DLBCL. ATF4 is required for DLBCL cell proliferation and is translationally activated when sensing depletion of nonessential amino acids (NEAA). Reciprocally, loss of SIRT3 induces accumulation of NEAAs due to induction of autophagy, which in turn suppresses ATF4 translation, further aggravates nutrient starvation, and ultimately leads to cell death, which is further enhanced by exposure to GCN2 inhibitors.See article, p. 66.Although 80% to 90% of patients treated with CD19 CAR T-cell therapy for ALL achieve remission, half will relapse. Biomarkers predicting relapse have so far been unknown. In this study, Pulsipher et al. hypothesize that the presence of minimal residual disease detected by next-generation sequencing (NGS-MRD) after tisagenlecleucel therapy would predict relapse of both CD19+ and CD19− blasts. They demonstrate that any level of NGS-MRD positivity was highly associated with relapse risk. A combination of early loss of B-cell aplasia with NGS-MRD >0 predicted relapse with sufficient time to intervene with various approaches aimed at preventing relapse.