PROJECT SUMMARY This proposal describes a career development program to prepare Dr. Lu for an independent research career that focuses on developing computational and experimental methods to improve cancer detection, diagnosis, and treatment. This program will provide Dr. Lu with new expertise in single-cell spatial omics, integrating with her background in machine learning-based image computation (gained as a graduate student) and clinical single- cell drug imaging (gained as a postdoctoral researcher) to advance our understanding of the mechanism that drives drug resistance of pancreatic cancer. Dr. Lu will be mentored by Dr. Garry Nolan, who invented the CODEX technology for highly multiplexed single-cell imaging, and co-mentored by Dr. Eben Rosenthal, a physician-scientist who pioneered the first-in-human clinical studies for fluorescence-guided cancer surgery, and Dr. Robert West, who developed the Smart-3SEQ technology for spatial transcriptomics. The K99 phase of Dr. Lu’s training will consist of (i) structured mentorship by the primary mentor and co-mentors, (ii) close interactions with advisory committee and collaborators, (iii) technical and academic training, (iv) a provocative research project, and (v) a program of career transition. Elucidating the role of the tumor microenvironment (TME) in drug resistance is critical to developing effective cancer therapies, but quantifying the drug delivery and action together with host environment factors within clinical tumors remains technically challenging. Antibody-based therapeutics, such as antibody-drug conjugates (ADCs) and immune checkpoint inhibitors (ICIs), are especially susceptible to blockade by TME barriers. The overall objective of this project is to identify the TME factors driving drug resistance in pancreatic ductal adenocarcinoma (PDAC) by integrating single-cell geospatial mapping of therapeutic antibodies with the deep spatial profiling of the TME. The central hypothesis is that periostin and tumor-associated macrophages (TAMs) play a key role in inhibiting drug delivery and response in PDAC. The central hypothesis will be tested by pursuing three aims: (Aim 1) establish a computational spatial omics platform by integrating CODEX and Smart-3SEQ to chart the baseline architecture of PDAC TME in an unbiased way; (Aim 2) combine single-cell drug imaging with spatial omics to determine the impact of stromal barriers to antibody delivery into PDAC and evaluate whether inhibiting periostin improves the delivery of anti-EGFR antibodies and ADCs in patient-derived xenograft mouse models; and (Aim 3) examine the role of chemotherapy in altering the phenotype and function of TAMs in human and mouse PDAC; identify chemo-induced alterations in TAM-ICI interactions in PDAC patients infused with a fluorescent anti-PD-L1 antibody; and validate whether inhibiting TAM-ICI interactions improves response to ICI plus chemotherapy in a transgenic mice model of PDAC. This project will provide novel computational tools to quantify cell-cell and cell-drug interactions in clinical tumors, offer new mechanistic insights on drug resistance in pancreatic cancer, and lead to new treatment strategies to improve patient survival.

PROJECT NARRATIVE Many anti-cancer drugs are ineffective in solid tumors, but the underlying mechanism is not fully understood. This proposal aims to identify the factors in the tumor microenvironment that drive drug resistance in pancreatic cancer by integrating single-cell drug imaging and spatial multi-omics. This project will result in computational tools to chart the spatial landscape of the cell-cell and drug-cell interactions within the tumor microenvironment, revolutionize our understanding of the drug resistance mechanisms in pancreatic cancer, and facilitate the development of new treatment strategies to improve patient outcomes with this devastating disease.