Open Collections

Abstract

B cell acute lymphoblastic leukemia (B-ALL) is the most common malignancy in children, accounting for more than 10% of childhood cancer deaths. In most cases, B-ALL initiates in utero, with a clinically silent pre-leukemic phase preceding diagnosis. Prenatal genetic events give rise to pre-leukemic clones in more than 1% of newborns, yet fewer than 1 in 100 progresses to overt leukemia. This suggests the presence of both protective and/or promoting factors may influence the fate of pre-leukemic clones. My thesis investigates how immune modulation during the pre-leukemic phase affects B-ALL progression. Infections have been implicated in B-ALL development, yet epidemiological studies show conflicting outcomes, and no specific pathogen has been definitively linked to the disease. Using murine cytomegalovirus as a model pathogen, I show that infection induces depletion of pre-leukemia in B-ALL predisposed mice. Conversely, Poly(I:C), which mimics pathogenic double-stranded RNA, increases pre-leukemia cell numbers. This divergent response is mediated by engagement of distinct pattern recognition receptors (PRRs): signaling via TLR3 promotes pre-leukemia through TNF-α, while other pathways can restrict it. This mechanism is age-dependent and conserved between mouse and human systems. These findings help explain contradictory epidemiological links by identifying both pro- and anti-leukemic immune pathways. While most children with pre-leukemia do not progress to leukemia, the role of T cells in protection from leukemia is unclear. At diagnosis, children with B-ALL frequently exhibit signatures of T cell exhaustion, suggesting prior anti-leukemic activity. Using B-ALL mouse models, I demonstrate that T cell exhaustion—marked by phenotypic, functional, and transcriptional changes— arises specifically during overt leukemia, but is absent during the pre-leukemic stage. Additionally, PRR agonists such as CpG-ODN and 5′pppdsRNA modulate T cell responses, and early immune checkpoint blockade prolongs leukemia-free survival. Further, CAR-T cells derived from pre-leukemic mice are more effective at clearing leukemia compared to those from leukemia-bearing mice. Overall, my findings reveal how distinct immune pathways modulate B-ALL, offering a mechanistic explanation for conflicting epidemiological links between infection and leukemia. Additionally, by identifying T cell exhaustion as a feature of overt leukemia—but not pre-leukemia—this work identifies key windows for immune-based prevention and therapeutic intervention.