Abstract: The CD8 T-cell response to cancer is critical for disease control and response to checkpoint inhibitors.
A prevailing feature of all human cancer is the extreme variability of the magnitude of this response between
patients, and the factors that cause this variability are unknown. We previously reported that a critical component
of the tumor microenvironment in many cancer types is an immune niche that provides support for a CD8 T-cell
with stem-like qualities. This niche is a small area of antigen presenting cells localized with the tumor boarder
where TCF1+ stem-like CD8 T-cells reside. These TCF1+ CD8 T-cells are specific for tumor antigens and act
as a progenitor to continuously give rise to anti-tumor effector CD8 T-cells. Most importantly, if the niche is not
present in the tumor, these TCF1+ cells have nowhere to survive in the tumor microenvironment, and
consequently the CD8 response against the cancer is halted. Loss of this immune niche in kidney tumors is
associated with rapid progression of disease after surgery, and significantly worse response to immunotherapy.
In this proposal, we aim to investigate what genetic features cancer cells acquire that help them limit the
formation of the immune niche in tumors. To do this we have developed a novel method that allows sorting of
pure cancer cells from human RCC tumors. We performed RNAseq on cancer cells isolated from 99 T3 clear
cell RCC patients and found two very strong correlates of immune response to the cancers that will be
investigated further here. Firstly, cancer cells with active cell death and NFkB pathways had highly active antigen
presenting cells, bigger immune niches, and many more effector cells. Secondly, cancer cells with high levels of
oxidative phosphorylation and glycolysis were far less able to generate effectors from the stem-like cells in their
tumor, immune niches were far smaller and patients progressed more rapidly after surgery and did not respond
to later PD1 blockade. This proposal will expand upon these findings by completing the following aims.
Aim 1 will first study what genetic changes occur in cancer cells that correlate with transcriptional signaling, and
how heterogeneity within the tumor microenvironment might differ across regions of the primary tumor and
metastasis. In Aim 2, we present data that cold tumors have mutations in cell death pathways and we will
investigate how loss of particular inflammatory cell death pathways might limit antigen presenting cell activation
and CD8 effector generation in the TME using both PDX and syngeneic mouse models. Aim 3 will learn how
different energy metabolism systems in cancer cells influence the formation of immune niches and regulate CD8
effector differentiation in the TME. Together, these data will provide new insight into how cancer cells limit the
immune response against a tumor.
While much of this work will focus on kidney cancer, the mechanism we described were also found in prostate,
bladder, brain, and head and neck tumors. For this reason, the basic mechanisms that we describe in this work
will almost certainly have implications for understanding the CD8 T-cell response to many other cancers.