Philadelphia chromosome-like B-cell acute lymphoblastic leukemia (Ph-like ALL) is driven by genetic alterations that induce constitutive kinase signaling and is associated with chemoresistance and high relapse risk in children and adults. Preclinical studies in the most common CRLF2-rearranged/JAK pathway-activated Ph-like ALL subtype have revealed variable responses to JAK inhibitor-based therapies, suggesting incomplete oncogene addiction and highlighting a need to elucidate alternative biologic dependencies and therapeutic vulnerabilities, whereas the ABL-class Ph-like ALL subtype seems preferentially sensitive to SRC/ABL- or PDGFRB-targeting inhibitors. Which patients may be responsive vs resistant to tyrosine kinase inhibitor (TKI)-based precision medicine approaches remains a critical knowledge gap. Using bulk and single-cell multiomics analyses, we profiled residual cells from CRLF2-rearranged or ABL1-rearranged Ph-like ALL patient-derived xenograft models treated in vivo with targeted inhibitors to identify TKI-resistant subpopulations and potential mechanisms of therapeutic escape. We detected a specific MYC dependency in Ph-like ALL cells and defined a new leukemia cell subpopulation with senescence-associated stem cell-like features regulated by AP-1 transcription factors. This dormant ALL subpopulation was effectively eradicated by dual pharmacologic inhibition of BCL-2 and JAK/STAT or SRC/ABL pathways, a clinically relevant therapeutic strategy. Single cell-derived molecular signatures of this senescence and stem/progenitor-like subpopulation further predicted poor clinical outcomes associated with other high-risk genetic subtypes of childhood B-ALL and thus may have broader prognostic applicability beyond Ph-like ALL.