Background T cell Acute Lymphoblastic Leukemia (T-ALL) is a devastating malignancy found primarily in pediatric populations. Standard of care for T-ALL has not progressed from intensive regimens of chemotherapy. Another therapeutic strategy for treating T-ALL is to harness anti-leukemia T cells by immunotherapy. Currently, whether T-ALL is sufficiently immunogenic to generate anti-leukemia T cells is unknown. Furthermore, it is unclear how differences in the immune milieu of distinct tissue types (lymphoid vs non-lymphoid) that become infiltrated by T-ALL impacts T cell interactions with leukemia.

Methods These studies utilized primary T-ALL cells from a murine model that were transplanted into immune-competent, congenic (CD45.1) recipient mice. Tissues were evaluated by flow cytometry at distinct stages of disease to help determine if T cells respond to T-ALL. In addition, frozen tissue sections were analyzed using NanoString’s GeoMX Digital Spatial Profiling platform to evaluate T cells in specific regions of varying proximity to T-ALL.

Results Drastic changes to the composition of the TME were found at distinct stages of tumor burden. Evaluation of changes to the hosts’ (CD45.1+) T cells revealed a higher frequency of CD8 T cells with an activated phenotype. Furthermore, this increase largely correlated with tumor burden (figure 1). As this may represent anti-leukemia T cell responses, we next determined if they could be harnessed with immunotherapies directed against T cell co-signaling receptors. Although PD1 and OX40 monotherapies had no discernable effect, the combination of anti-PD1 with anti-OX40 led to a drastic reduction in T-ALL burden. Importantly, control of tumor growth was accompanied by a concomitant increase in cytotoxic CD8 T cells actively undergoing proliferation specifically in response to combination therapy. To gain better insight into T cell interactions with T-ALL, frozen tissue sections were used for comprehensive digital spatial profiling using NanoString’s GeoMX platform. This analysis revealed strong correlations between immune markers indicative of anti-leukemia responses as well as suppressive factors. Interestingly, regions enriched for activation markers were largely constrained to certain regions indicating the formation of ‘immunological hotspots’ in the context of T-ALL.

• Download figure
• Open in new tab
• Download powerpoint

Abstract 816 Figure 1

Increase in memory CD8+ T cells in response to T-ALLChanges to the T cell compartment were evaluated by transplanting primary T-ALL cells (CD45.2+) into immune-competent CD45.1 congenic recipient mice. T cells were then evaluated in the spleens at distinct stages of disease. As shown below, an increase in the frequency of CD8+ T cells that are memory (CD44+) and effector memory largely correlated with tumor burden in the spleens of transplanted mice that could indicate anti-leukemia T cell responses. Data is representative of a cohort from 1 of 3 independent experiments

Conclusions The results from these studies suggest that T-ALL is recognized by T cells. As immune responses were not uniform within an organ, it will be important to specifically evaluate these ‘immunological hotspots’ in order to identify targets to activate T cells found in these regions. Ongoing studies are therefore aimed at comparing T cell interactions with T-ALL and their responses to immunotherapy between tissue types.

Acknowledgements Analysis of tissue section was supported in part by a SITC-NanoString Technologies Spatial Profiling Award given to T.A.T.