P324: cell plasticity predetermines treatment response in t-all

Background: Acute T cell lymphoblastic leukemia (T-ALL) is an aggressive lymphoid malignancy in children and young adults that frequently becomes treatment-refractory and relapses. Oncogenic NOTCH signaling is a major driver of T-ALL resulting in malignant transformation of T cell progenitors in the bone marrow. With over 50% of the patients harboring gain-of-function NOTCH mutations, inhibiting the oncogenic signaling presents an attractive targeting strategy, yet patient response is often short-lived and current treatment options for relapsed/refractory disease are of limited success. Existence of epigenetically and transcriptionally distinct cell subpopulations has been recently suggested as driver mechanism for treatment escape in refractory T-ALL. Aims: To characterize cell states and the transcriptional changes mediating state transitions as an explorative strategy to develop targeted therapies overcoming drug resistance in T-ALL. Methods: We performed full-length single-cell RNA sequencing of 3,188 leukemia cells collected from the blood of 2 sensitive and 3 refractory PDX models carrying activating NOTCH1 mutations, before and after treatment with the NOTCH inhibitor DBZ (Dipenzazepine). To characterize transcriptional states, we assessed developmental potential using CytoTRACE and identified state-specific gene-regulatory networks (GRN) driving state transition using SCENIC. Alterations in 3D chromatin structures differentiating sensitive and refractory models were assessed by evaluating enhancer rewiring and mapping of the associated loops using Hi-ChIP with an antibody against H3K27ac. Apoptotic priming and functional validation of anti-apoptotic dependencies were determined by BH3 profiling. Results: Analysis of single cell profiles of refractory T-ALL revealed dramatic transcriptional reprogramming resulting in expression of immature hematopoietic signatures and lineage infidelity co-existing within individual leukemia cells. Assessment of developmental programs prior to treatment indicated that deranged lineage commitment can predict response to NOTCH inhibition in vivo. Upon treatment, we observed expansion of 2 distinct populations that critically differ in differentiation stage and developmental trajectory. Fast-cycling cells express progenitor signature and genes associated with RNA processing and slow-cycling cells are enriched in lymphoid differentiation programs. Further analysis of developmental potential revealed that loss of active regulons and gain of transcriptional dependencies underlie the differentiation state transitions and treatment refraction. We identified ATF4 as a key transcriptional dependency in immature cell states. Assessment of epigenetic rewiring underlying cell state transitions demonstrated that immature states present increased numbers of enhancer-promoter interactions driving transcriptional plasticity. Enhancer rewiring results in the emergence of transcriptional dependencies leading to perturbed apoptotic signaling, including a dependency switch from BCL2/BCLxL to MCL. Summary/Conclusion: In conclusion, single-cell transcriptomic analysis identified cell states with high plasticity that are characterized by aberrant differentiation trajectories, distinct transcriptional circuitries and enhancer rewiring resulting in treatment escape. Definition of state-specific transcriptional dependencies combined with BH3 profiling could predict response to NOTCH inhibition and allowed the identification of potential therapeutic targets to overcome Notch-inhibitor resistance in T-ALL. Keywords: Drug resistance, Plasticity, T-ALL, RNA-seq