C-Myc Driven Tolerogenic Tumor Microenvironment (TME) in Mantle Cell Lymphoma (MCL) Is Overcome By Bromodomain Inhibition

Remarkable clinical efficacy and durable responses to antibodies that block the programmed death-1 (PD-1)-programmed death-ligand 1 (PD-L1) pathway have been observed in patients with multiple cancers, including classical Hodgkin lymphomas (cHL). However, the responses in the majority of Non-Hodgkin lymphoma patients, including mantle cell lymphoma (MCL), treated with anti- PD1/PDL1 antibodies have been modest to date. It has been postulated that the immune suppressive nature of the tumor microenvironment (TME) may play a role in limiting the efficacy of checkpoint blockade strategies. As such, identification of critical molecules in TME required for driving response and resistance is key to improve lymphoma immunotherapy.

We have therefore generated in vivo and ex-vivo MCL lymphoma-stroma co-culture models and capitalize this model with primary human MCL cells as well. First, we found that co-injection of murine Fc-muMCL1 cells with stromal cells significantly promote lymphoma growthas compared to Fc-muMCL1 cells injected alone. This aggressive growth was associated with less tumor infiltrating cytotoxic T-cells in the TME. Second, to identify the tolerogenic mechanism(s) that drive immunosuppression in TME, we co-cultured MCL cells with stroma cells ex-vivo and found an increased translation and transcription of PD-L1 via upregulation of c-Myc. Furthermore, co-culture of patient primary lymphoma cells with stromal cell dramatically increases PD-L1 expression in both stromal cells and lymphoma cells. Tumor infiltrating T cells also induce PD-L1 expression in stromal cells. Of note, just by knocking down c-Myc in stromal cell we were able to block co-culture-induced PD-L1 expression, highlighting a critical role for c-Myc in driving this tolerogenic process in the TME.

In lieu of the above findings, next we treated murine MCL in vitro with a bromodomain inhibitor (JQ1) and observed a significant decrease in c-Myc/PD-L1 expression which was associated with increased immunogenicity of malignant B-cells leading to a better T-cell activation. More importantly, treatment of MCL-bearing mice with a combination of a bromodomain inhibitor with anti-PD1 antibody resulted in enhanced inhibition of MCL growth, increased effector memory T cells and improved function of tumor infiltrating T cells in vivo. No such effects were observed in MCL-bearing mice treated with either agent alone.

Taken together, we have identified the c-Myc/PD-L1 axis in stromal cells that by creating a tolerogenic/immunosuppressive TME imposes a significant barrier to the efficacy of checkpoint blockade therapy in lymphomas. This barrier seems not to be unsurmountable since the addition of a bromodomain inhibitor augmented the efficacy of checkpoint blockade by inducing a more immunogenic TME in MCL.