Archived
This content is available here for research, reference, and/or recordkeeping.
Author ORCID Identifier
https://orcid.org/0009-0009-0933-652
Date Available
10-30-2026
Year of Publication
2026
Document Type
Doctoral Dissertation
Degree Name
Doctor of Philosophy (PhD)
College
Medicine
Department/School/Program
Toxicology and Cancer Biology
Faculty
Christine Brainson
Faculty
Yekaterina Zaytseva
Abstract
Non-small cell lung cancer (NSCLC) is currently the leading cause of cancer-related death in the United States. Immunotherapies, such as those targeting the PD1/PD-L1 axis to allow the patient’s own immune cells to target the tumor, have improved outcomes. However, sustained responses are observed in only a fraction of patients. Finding ways to boost immunotherapy responses is the central goal of this project. Many groups are testing additional drugs or dietary intervention to reprogram the tumor and tumor microenvironment (TME) interactions from one that allows tumor growth and resistance to one that has robust anti-tumor immunity. One attractive target for tumor and TME reprogramming is EZH2, a methyltransferase in the Polycomb Repressive Complex 2 that tri-methylates lysine 27 on histone 3 to silence gene expression. Several small molecule inhibitors targeting EZH2 can lead to epigenetic reprogramming, and dietary interventions that alter methionine metabolism can also change methyltransferase functions by altering the levels of methyl donors.
The first goal of my work was to test if the EZH1/2 inhibitor valemetostat could improve anti-PD1 responses in our syngeneic lung cancer model. Our published work demonstrated that the EZH2 inhibitors GSK126 and tazemetostat boost immunotherapy responses in lung cancer models, and these findings led to an industry-sponsored clinical trial with the next generation EZH1/2 inhibitor valemetostat. I observed that while each single therapy reduced tumor growth, only the combination therapy was able to significantly shrink tumors. The remaining tumors also had fewer tumor cells, more areas of tumor-associated immune and mesenchymal cells and reduced EZH2 and H3K27me3 levels. By flow cytometry, MHC Class I was modestly increased on tumor cells, but MHC Class II was highly increased, suggesting that MHC Class II is the major mediator of increased anti-tumor activity. I also observed significantly increased activated CD8+ T cells expressing PD1 and IFNy and decreased CD11b+ myeloid cells. Like with GSK126, I observed a drastic significant shift towards neutrophil maturity in the bone marrow of mice exposed to valemetostat. To examine the effects of EZH2 depletion on neutrophils, I used an EZH2 conditional knockout mouse and isolated bone marrow 7 days after Ezh2 deletion. I found that Ezh2-null neutrophils were more apoptotic, less migratory, less able to produce extracellular nets, but had similar ability to kill bacteria as Ezh2-WT neutrophils. Together, these data suggest that MHC Class II and systemic alteration of neutrophils are the major phenotypes of EZH2 depletion that drive anti-PD1 responses.
A second goal of my work was to establish a reproducible ex vivo model to test immunotherapy responses. My colleagues and I devised “multi-cultures” that combine tumoroids with lung mesenchymal cells, T cells, and bone marrow-derived myeloid cells in a three-dimensional air-liquid-interface Matrigel culture. We first observed that bone marrow from healthy mice or mice with an actively growing tumor significantly boosted tumoroid growth when compared to bone marrow-free cultures. This increase in tumoroids could be rescued by the addition of EZH2 inhibitor and anti-PD1 to the cultures, mimicking our in vivo results. Next, we tested bone marrow from mice that had either rejected a tumor or been treated with valemetostat, which both led to a dramatic depletion of tumoroids, with the highest reduction in tumoroids observed with valemetostat-treated bone marrow. By flow cytometry, CD8+ T cells were present at much higher proportions, suggesting that valemetostat-treated bone marrow led to increased survival of this anti-tumor population. Finally, to test the requirement of MHC Class II, we added MHC Class II blocking antibody to cultures with healthy bone marrow. MHC Class II block prevented the bone marrow-derived boost to tumoroid growth, most notably by reducing IL22, CCL3, and CCL4 levels, as well as CXCL9 and IFNγ. In addition to no increase in tumoroid growth, the EZH2 inhibitor and anti-PD1 treatment was not effective when the MHC Class II block was added. These cultures show that testing immunotherapy responses ex vivo is possible, and that myeloid cells can be both pro- and anti-tumor.
The final goal of my work was to understand whether anti-PD1 immunotherapy can be safely combined with methionine restriction (MR). Prior work from our lab showed that MR reduced lung tumor burden and increased carboplatin treatment relative to control chow. To understand these results better, I analyzed tumors using matrix-assisted laser desorption/ionization (MALDI) and observed glutathione and taurine were significantly lower in MR tumors. I also harvested T cells from MR and control chow mice and found that T cells from MR mice were just as proliferative as those from control chow mice, even when kept on low methionine in vitro. Finally, I used a syngeneic lung cancer model and found that MR effectively controlled tumors and slightly increased immunotherapy efficacy.
These studies demonstrate that inhibiting EZH2 in NSCLC is an effective way to boost immunotherapy response by MHC Class II up-regulation and alteration of the lymphoid-myeloid interactions. Implementing dietary changes into clinical practice can not only improve outcomes but also give patients the power to positively impact their own treatment. We also show, through the use of advanced “multi-culture” experiments, that we can successfully model in vivo environments to better predict cancer responses to treatment.
Digital Object Identifier (DOI)
https://doi.org/10.13023/etd.2026.232
Archival?
Archival
Funding Information
This work was supported in part by R01 HL170193, R01 CA237643, P20 GM121327, American Cancer Society Grants 133123-RSG-19-081-01-TBG, and the Markey Research Women Strong Distinguished Researcher Award (CFB), UL1 TR001998 Center for Clinical and Translational Science pilot award (CFB), NCI T32 CA165990 (DRP and CMG), K99 CA303792 (DRP), Markey STRONG Scholars Program through the American Cancer Society IRG-19-140-31 (EMK). This research was also supported by the Biostatistics & Bioinformatics, Cancer Research Informatics, Oncogenomics, Biospecimen Procurement & Translational Pathology, Flow Cytometry & Immune Monitoring Shared Resources, and pilot grant funding from the University of Kentucky Markey Cancer Center through P30 CA177558.
Recommended Citation
Childress, Avery, "Modulating EZH2 Activity and Methionine Metabolism to Improve Treatments in Non-Small Cell Lung Cancer" (2026). Theses and Dissertations--Toxicology and Cancer Biology. 71.
https://uknowledge.uky.edu/toxicology_etds/71
