ORCID ID
Graduation Date
Spring 5-7-2022
Document Type
Dissertation
Degree Name
Doctor of Philosophy (PhD)
Programs
Cancer Research
First Advisor
Michael (Tony) Hollingsworth
Abstract
Pancreatic cancer is currently the third leading cause of cancer death and projected to be the second by 2030. Metastatic pancreatic cancer, the most common form of the disease, has a dismal 3% five-year survival rate. However, understanding of the metastatic disease and particularly the metastatic tumor microenvironment (TME), which includes all non-cancerous cells in and around the tumor, has remained limited. The well-documented impact of the TME on cancer cell proliferation, chemoresistance, and patient survival in the primary tumors, indicates that the study of the microenvironment in metastatic cancer is integral to treating advanced patients. To better comprehend this complex network of cells we performed advanced multiplex-immunofluorescent staining on fifteen different pancreatic cancer patients and an immunofluorescent analysis on 48 pancreatic cancer and normal patient tissue samples collected from the pancreas, liver, and lung.
This research enabled discovery of organotrophic trends in the microenvironments at each of the sites. Studying the liver metastasis, we noted pronounced hypovascularization and immunosuppression. This immunosuppression appears to be driven by regulatory B cells and dendritic cells, which have a high spatial correlation to cancer cells. The pancreatic tumors had less exaggerated but still present hypovascularization and immunosuppressive features than the liver metastasis. By studying pre-malignant pancreatic lesions, we were further able to identify that the hypovascularization of the pancreas began to develop prior to transformation. The lung metastasis, on the other hand presented a more inflammatory microenvironment, with large populations of mast cells and limited spatial interactions with tumor cells. These organotrophic trends to the microenvironment can provide additional targets to personalized treatment to each patient’s unique metastatic profile. Furthermore, by identifying cells and cellular architectures in the TME that are similar between organs, we can distinguish pan-metastatic targets for potential cellular therapies. The techniques and pipelines developed herein can also be used for the in-depth study of many different tissue types. Finally, our dataset on the makeup of the TME, which to our knowledge is the largest currently extant, can be mined by other researchers to explore interactions of their cells of interest. We present here only the beginning of the story.
Recommended Citation
Vance, Krysten, "Use of Machine Learning Algorithms and Highly Multiplexed Immunohistochemistry to Perform In-Depth Characterization of Primary Pancreatic Tumors and Metastatic Sites" (2022). Theses & Dissertations. 644.
https://digitalcommons.unmc.edu/etd/644
Approval to use my previously published work