The Natural Killer – Dendritic Cell Axis Regulates Responses to Immunotherapy

We are interested in understanding how the immune system recognizes and responds to cancer. We are particularly interested in understanding how the innate immune system acts to regulate immune responses to cancer and responses to immunotherapies. Immunotherapies, such as anti-PD-1 immunotherapy, aim to reactivate T cell responses to cancer. While anti-PD-1 immunotherapy provides long lasting protection to some patients, a large number of patients have no responses to this treatment. Thus, our lab is focused on understanding how to increase patient responses to current immunotherapies, as well as identifying components of the immune system that can be targeted to generate the next generation of immunotherapies.

It is well established that dendritic cells (DCs) are found in the tumor microenvironment and a subset of these cells, termed cDC1s, are required for T cell mediated anti-tumor responses. Our previous work has found that cDC1s correlate with increased survival and responses to anti-PD-1 immunotherapy in metastatic melanoma patients (Figure 1). We further discovered that cDC1s in the tumor make close contacts with natural killer (NK) cells (Movie 1). We went on to show that NK cells correlate with better overall survival, responses to anti-PD-1 immunotherapy, and control cDC1 abundance in the tumor by producing the cytokine FLT3LG. Our previous findings suggest a model in which NK cell production of FLT3LG in the tumor microenvironment regulates cDC1 abundance, subsequent activation of T cell-dependent killing of tumor cells, and augments reactivation of the immune system by anti-PD-1 immunotherapy (Figure 2).

We hypothesize that targeting NK cells may lead to increased levels of protective cDC1s in the tumor, better tumor directed T cell-responses, and increased efficacy of current immunotherapies. Future research in the lab will focus on understanding the cellular and molecular mechanisms regulating NK cells and DCs in the tumor. We aim to identify key pathways that can be targeted to harness the NK-DC axis to shape anti-tumor immunity and responses to immunotherapy.

Figure 1

cDC1s and NK cells correlate with response to anti-PD-1 immunotherapy.
cDC1s and NK cells correlate with response to anti-PD-1 immunotherapy. a-b. Quantified frequency of cDC1s among HLA-DR+ cells (a) and NK cells among total immune cells (CD45+) (b) in the tumors of metastatic melanoma patients. c. Heat map of 33 immune cell populations defined from flow cytometric analysis of human metastatic melanoma samples. Each column shows data from one patient sample. For each immune cell population, two-tailed P values were calculated using the Wilcoxon rank-sum test and data for each row were logged and mean-centered. Resp./Responder = partial or complete responses. NonResp./Non-responder = stable disease or progressive disease. Data from Barry et al., Nat Med. 2018 Aug;24(8):1178-1191. PMID: 29942093

Figure 2

Diagram of NK cell – cDC1 axis.
Diagram of NK cell – cDC1 axis.

NK Cell Video 

Live two-photon imaging of the stable interaction between NK cells (Ncr1-GFP; green cells) and cDC1s (anti-XCR1; white cells) in a murine ectopic melanoma tumor slice. Data from Barry et al., Nat Med. 2018 Aug;24(8):1178-1191. PMID: 29942093

The immune response to fibrolamellar hepatocellular carcinoma

Fibrolamellar hepatocellular carcinoma (FLC) is a rare form of liver cancer (<1% of all liver cancers) that primarily affects adolescents and young adults with otherwise normal livers. FLC tumors tend to be refractory to chemotherapy and other targeted therapies approved for conventional hepatocellular carcinoma. The current treatment strategy for FLC is surgical resection, but curative surgical resection is often not possible because the lack of underlying liver disease in FLC patients typically delays diagnosis, leading to advanced disease at the time of surgery. Clearly there is a need to further understand the mechanism of FLC tumorigenesis and to develop new treatments.

FLC expresses a characteristic fusion gene (DNAJB1-PRKACA) that has been shown to drive tumorigenesis in murine models. We are using human immunology and, in collaboration with Dr. Julien Sage at Stanford University, a novel model of FLC to study the immune response to this disease and to lay the foundation for the use of immunotherapies as novel treatments for FLC.

Immune profiling of human tumors

The tumor microenvironment contains a wide range of cells: including tumor, stromal, and immune cells. In partnership with clinicians we collect valuable human tumor samples in order to define the immune composition, immune cell gene expression, and spatial organization of the immune system in a variety of human tumors (e.g. melanoma, liver cancer, breast cancer, etc.). Each tumor sample is analyzed by multiparameter flow cytometry, transcriptional profiling, and fluorescent multiparameter immunohistochemistry (Figure 3). Importantly, patient outcomes (response to immunotherapy, disease progression, etc.) is also tracked allowing for the identification of immune components that correlate with protection to cancer patients. These studies will allow us to define the immune axes and components of the immune system that regulate protective immune responses to cancer, leading the way to new treatments that can re-invigorate immune responses to cancer.

Figure 3

Diagram of immune profiling experiments.
Diagram of immune profiling experiments.