Interferon gamma brings the heat to cold tumors

From the Pollack Lab, Clinical Research Division

T cells have long been studied for their ability to target and eliminate tumor cells. In order for the T cells to do so, the tumors must express antigens not presented by healthy tissues, and the tumors must process and display the antigens using major histocompatibility complex (MHC) proteins. One group of abnormally expressed antigens is known as cancer testis antigens (CTA) because they are normally found only expressed in male germ cells. The Pollack laboratory (Clinical Research Division) studied two types of cancers that express CTA, synovial sarcoma (SS) and myxoid/round cell liposarcoma (MRCL), in a phase 0 clinical trial.  Phase 0 trials are used to define the scope of a disease, cancer in this case, and set the stage for therapeutic trials.  

Dr. Seth Pollack explained why they chose to study SS and MRCL: “These two tumors should be great targets for immunotherapy because they have very high levels of a target [CTA] called NY-ESO-1. However, we have found that they can be resistant to these immunotherapies because of low expression of MHC molecules and a lack of infiltrating T cells.” The absence of T cells and inflammation in tumors like SS and MRCL has prompted researchers to classify them as “cold tumors.” Dr. Shihong Zhang, a postdoc in the Pollack lab, led a study they published in Cancer Immunology Research on heating up these cold tumors by treating patients with an inflammatory cytokine, interferon  gamma (IFN-γ).

A diagram showing the processing and presentation of protein on MHC.
A diagram showing the processing and presentation of protein on MHC. Image by Scray [CC BY-SA 3.0 (]

IFN-γ has the ability to increase surface expression of MHC and the processing and presenting of antigens. The authors hypothesized that administering IFN-γ to patients with SS or MRCL would allow T cells to better recognize and eliminate tumor cells. They administered systemic IFN-γ to six SS patients and two MRCL patients and compared pre and post treatment tumor cells. They found an increase in the frequency of tumor cells expressing MHC-I in seven of the patients. The authors used a technique known as gene set enrichment analysis (GSEA) to compare expression of antigen processing and presentation genes in this patient and found them upregulated after IFN-γ treatment. 

As these results suggested that the tumor cells would be more vulnerable to T cells, the authors next investigated tumor-infiltrating T cells. They found that before treatment with IFN-γ, there were very few T cells in the tumor, but the number of T cell infiltrates increased in six out of the seven patients after treatment. All except one of these patients had cytotoxic CD8+ T cells as the majority T cell infiltrate. Further investigation of these T cells with GSEA showed that before IFN-γ treatment they expressed an exhausted phenotype. After treatment, the T cells expressed genes indicative of effector T cells. These results suggested that the few T cells in the tumor before IFN-γ treatment were ineffective against tumor cells, but treatment promoted infiltration of more cytotoxic T cells. Further supporting this idea, the post-treatment tumor of one patient contained CD4+ T cells which were specific for NY-ESO-1. In addition, three patients developed antibodies against more tumor antigens after IFN-γ treatment.

Despite these promising results, the authors found an increase in the frequency of tumor- infiltrating myeloid cells in all patients expressing a protein known as PD-L1 after IFN-γ treatment. PD-L1 binds to PD-1 expressed on some T cells, which promotes T cell apoptosis and inhibits T cell proliferation. Additionally, two of the patients had an increase in frequency of tumor cells expressing PD-L1. Dr. Zhang explained the significance of these results: “We did find that in our cohort IFN-γ treatment upregulates PD-L1 expression in the tumor microenvironment. But many studies have shown that the inhibition of T cell functionality by PD-L1 can be overcome by the well-studied checkpoint inhibitors.” 

Checkpoint inhibitors like anti-PD-1 have shown promise in “hot tumors” like melanoma and lung cancer, but have been ineffective in “cold tumors,” likely because of the lack of T cell infiltration and MHC expression. The Pollack lab is continuing studies based on the data showing IFN-γ treatment increasing MHC expression and T cell infiltration in SS and MRCL tumors. Dr. Pollack said, “We launched a national multi-center study with the Cancer Immunotherapy Trials Network (CITN) combining interferon gamma with pembrolizumab (anti-PD-1) for these patients.”

This work was supported by The Gilman Sarcoma Foundation and the National Cancer Institute.

UW/Fred Hutch Cancer Consortium members Venu Pillarisetty, Lee Cranmer, Robert Pierce, Stanley Riddell, and Seth Pollack contributed to this work.

Zhang S, Kohli K, Black RG, Yao L, Spadinger SM, He Q, Pillarisetty VG, Cranmer LD, Van Tine BA, Yee C, Pierce RH, Riddell S, Jones RL, Pollack SM. 2019. Systemic Interferon-γ Increases MHC Class I Expression and T-cell Infiltration in Cold Tumors: Results of a Phase 0 Clinical Trial. Cancer Immunol Res. 2019 Jun 6;. doi: 10.1158/2326-6066.CIR-18-0940. [Epub ahead of print] PubMed PMID: 31171504.