Obesity Attenuates T-Cell Function Which Impacts the Efficacy of T-Cell Based Immunotherapies in Acute Lymphoblastic Leukemia

The incidence of obesity continues to rise with over 50% of the world's population predicted to be overweight/obese by the year 2030. The global health impact of this trend is concerning given that obesity is a risk factor for developing cancers of varying etiologies. Alarmingly, the survival outcomes for obese patients with cancer are lower than those observed in lean patients. Obesity is characterized by the accumulation of adipocytes which alters drug dynamics and impacts the function of cancer and immune cells in the tumor microenvironment. Obesity-induced immune defects are troubling given the increasing use of immunotherapy in the treatment of malignancies. Here we show that adipocyte-secreted factors upregulate immunosuppressive mechanisms on human B-cell acute lymphoblastic leukemia (B-ALL) cells, attenuate the function of endogenous T-cells, and compromise the efficacy of T-cell based immunotherapies.

To study the impact of adiposity on T-cell function, CD4⁺ and CD8⁺ T-cells were purified from the spleens of C57BL/6 mice and activated with PMA/Ionomycin for 72 hours in unconditioned media (UCM), bone marrow stromal-cell conditioned media (SCM), and adipocyte-conditioned media (ACM) followed by flow cytometry analysis for surface marker expression, cytokine production, and the induction of cytolytic mediators. Interestingly, T-cells activated in ACM, but not UCM or SCM, showed an attenuated phenotype highlighted by decreased CD44 and PD-1 expression, diminished cytokine production (IFN-γ/TNF-α) and reduced induction of cytolytic mediators (granzyme B/perforin). These observations were also true in obese, relative to lean, patients with B-ALL where we found that T-cells purified from the peripheral blood mononuclear cells (PBMCs) failed to produce significant levels of TNF-α when stimulated with PMA/Ionomycin. In all, these results demonstrate that adipocyte-secreted factors directly compromise the function of endogenous T-cells, which phenocopies T-cell defects observed in obese relative to lean pediatric patients with B-ALL.

We next assessed the impact of adiposity on malignant cells by culturing human B-ALL cell lines in the conditioned mediums described above and performed flow cytometric analysis to assess their surface expression of the B-cell lineage antigen CD19 and proteins that modulate immunity. In addition to being a marker for B-cells, CD19 is the primary target of the T-cell based immunotherapies Blinatumomab and CAR T-cells directed against B-ALL cells. Surprisingly, when human B-ALL cells were co-cultured with adipocytes, every cell line tested (n=6) exhibited lower surface CD19 expression with 5 out 6 reaching statistical significance. Furthermore, adipocyte-secreted factors alone were sufficient to reduce CD19 surface levels on B-ALL cells in 2 of the 6 cell lines tested. Human B-ALL cells cultured in ACM, but not UCM or ACM, also upregulated their surface expression of the immunoinhibitory proteins HVEM, PD-L1, and PD-L2. These results demonstrate that adipocytes directly induce the downregulation of CD19 on human B-ALLs and increase their immune evasive capacity.

Given these observations, we hypothesized that adipocyte-secreted factors would compromise T-cell-based immunotherapies targeting CD19-expressing B-ALL cells. To this end, primary human T-cells were engineered to express a CD19-directed chimeric antigen receptor (CAR). CAR T-cells and human B-ALL cells were separately pre-treated for 24 hours in UCM, SCM or ACM followed by co-culture for cytolytic analysis using Annexin-V/PI staining. Adipocyte-secreted factors significantly inhibited CAR T-cell mediated killing of CD19-expressing B-ALL cells at 4 hours. In addition to CAR T-cells, we tested the leukemia killing efficacy of the bispecific T-cell engager, Blinatumomab. After 3 days of culture, we found that Blinatumomab significantly increased the killing capacity of endogenous T-cells with 60-80% of B-ALL cells being killed after 3 days of culture in UCM and SCM. In contrast, we found that ACM significantly compromised the efficacy Blinatumomab with only 30% of B-ALL cells being killed over 3 days when co-cultured with human T-cells.

Our pre-clinical data highlights the negative impact of an adipose-rich microenvironment on T-cell function and B-ALL immunogenicity, which subsequently compromises the efficacy of multiple classes of immunotherapies targeting CD19.