Background: Patients with BCR-ABL+ acute lymphoblastic leukemia (Ph+ ALL) can experience significant disease reduction after induction chemotherapy, but subsequently incur adverse long-term outcomes, due to the presence of minimal residual disease (MRD). This suggests a need for novel therapeutic strategies to eliminate MRD. Checkpoint blockade has been considered of little benefit in ALL due to the overall low frequency of somatic mutations. However, the protein product of the BCR-ABL translocation region forms a tumor associated antigen (TAA). Prior studies have demonstrated that patients with Ph+ ALL harbor CD8+ and CD4+ T-cells specific for peptides derived from this junctional region, presented in MHCI or MHCII contexts. This suggests that relapse in Ph+ ALL may be countered by immunotherapeutic strategies that expand sufficient numbers of BCR-ABL-specific effector T-cells. The current study uses a murine model of Ph+ ALL to characterize effector cell types and mechanisms responsible for leukemia control and eradication. We also investigated the impact of commonly used chemotherapy agents on effector cell function, to determine the viability of a combination immunotherapy and vaccination approach at eliminating MRD after induction.
Methods: Bone marrow cells from p19ARF-null C57BL/6 mice were transduced with a BCR-ABL expressing retrovirus. Transformed cells were injected into immunocompetent B/6 mice, resulting in a uniformly fatal leukemia in 20-25 days. We used a peptide: MHC-II tetramer to label endogenous CD4+ T-cells specific for a BCR-ABL-derived peptide ("BAp") presented in an MHC II I-Ab context (hereafter referred to as "BAp"-specific T-cells).
Results: Previously, we found that BAp-specific CD4+ T-cells were elicited during leukemia development but were limited by regulatory-T-cells (Tregs) that were cross-reactive with endogenous ABL protein. Nonetheless, a heterologous vaccination strategy using BAp peptide, combined with dual CTLA-4 and PD1 checkpoint blockade, was able to extend survival and eradicate leukemia in a subset of mice. Prolonged survival correlated with the robust expression of an ensemble of antigen-presentation molecules by host leukemia cells, as well as the presence of polyfunctional, cytotoxic CD4+ BAp-specific T-cells.
To formally evaluate the relative contributions of CD4+ vs. CD8+ T-cells, we next established leukemia and allowed it to progress after the selective depletion of CD4+, CD8+, or both CD4+ and CD8+ T-cells. All mice were also treated with nilotinib and PD-L1 checkpoint blockade. Depletion of CD4+ T-cells led to markedly shortened survival. In contrast, depletion of CD8+ T-cells had no effect, and the majority of mice survived long-term. As this confirmed that CD4+ T-cells were critical for leukemia control, we next characterized the impact of common induction chemotherapy regimens on this subset. Combinations of cytotoxic chemotherapies mimicking either a standard intensity Hyper-CVAD-like regimen or a reduced-intensity regimen were administered to naïve mice, followed by vaccination with BAp. Mice treated with a Hyper-CVAD-like regimen generated fewer numbers of BAp-specific CD4+ T-cells in response to subsequent vaccination. Moreover, scRNAseq analysis of activated CD4+ T-cells demonstrated that standard intensity regimens induced dysfunctional gene expression signatures. These changes were not observed after treatment with a reduced intensity regimen. Dysfunctional signatures were characterized by the upregulation of senescence-associated genes including p21, as well as decreased expression of T-stem-central-memory (Tscm) genes including TCF7 and Tox2. Altogether these findings suggest that Hyper-CVAD-like chemotherapy inhibits proliferation as well as the self-renewal of CD4+ T-cells responding to leukemia antigens.
Conclusions:
CD4+ T-cells specific for peptides derived from the BCR-ABL protein are responsive to PD-L1 checkpoint blockade, and are critical for Ph+ ALL clearance. Standard intensity induction chemotherapy leads to significant dysfunction of these cells. Ongoing experiments are exploring the mechanisms underlying this dysfunction, and methods to overcome it via BCR-ABL-based peptide vaccination combined with checkpoint blockade. Validation experiments utilizing human samples are also underway and will be reported.
Disclosures
No relevant conflicts of interest to declare.
Image Enhancement using Recursive Separate Standard Intensity Deviation Based Clipped Sub Image Histogram Equalization