Tumor progression in epithelial tissues, the sites of >90% of human malignancies, involves interactions at the tumor-stroma interface. During progression from in situ neoplasia to invasive cancer, this interface is the epithelial basement membrane (BM). Cutaneous squamous cell carcinoma (SCC), as the second most common cancer in the U.S., is a devastating burden on our veteran patients, and serves as a prototype epithelial malignancy. This proposal will characterize the role of tumor-stroma interactions in early SCC tumor progression in vivo and in a new organotypic model of human neoplasia. Without local invasion, epithelial neoplasms fail to metastasize, yet the proteins controlling tumor- stroma interactions at the BM during early tumorigenesis are undefined. Prevalent cancer progression models don't include human cells within architecturally faithful human tissue on an intact BM, and thus can provide misleading results. During the prior cycle, we developed a human skin xenograft model of inducible epidermal cancer and used it to identify disruption of an integrin-centered gene network in human tissue tumor progression. Consistent with this, expression of the major epidermal integrin subunits, 21 and 24 is abnormal in spontaneous patient SCCs. A modest functional role for 21 in early human tumor progression was confirmed, however, the role of other integrins in this setting is uncharacterized. We postulate that 21 acts cooperatively with the other major epidermal 2 integrin subunit, 24, in human epithelia to enable neoplastic invasion.
Aim I is thus designed to characterize a potentially cooperative role of BM integrins in epidermal tumor progression in human skin tissue in vivo. In vitro organotypic models of human tissue tumor progression are attractive complements to in vivo models by virtue of their speed, cost, and tractability to use with in vitro drug inhibitors as well as their potential to recreate the tumor microenvironment from defined elements. During the last funding cycle, we refined an organotypic model incorporating inducibly neoplastic normal human epithelial cells, intact stroma with BM, and stromal cells to generate such models for skin, esophageal, cervical and oropharyngeal tissues. Normal stromal fibroblasts enhanced tumor progression to deep invasion, consistent with their potential secretion of an array of integrin ligands, including fibronectin, collagens, and laminins. Because it can be generated from defined cell populations without potentially confounding infiltration by mouse host cells seen in xenograft models, organotypic neoplasia models enable direct assessment of stromal cell populations implicated in accelerating tumor progression, including cancer-associated fibroblasts (CAFs).
Aim II is designed to study both the potentially cooperative integrins roles in Aim I above using a distinct human tissue platform and to begin to define the functional impact of CAFs versus normal fibroblasts in tumor progression. This proposal aims to characterize mediators of tumor-stroma interactions at the BM as a foundation for future strategies for cancer prevention and treatment.

Public Health Relevance

Epithelial cancers represent approximately 90% of malignancies plaguing our veterans and the U.S. population as a whole. Current cancer research models don't use human cells within intact human tissue and thus produce results that don't always translate into clinical advances. The current proposal uses entirely human tissue models of invasive cancer in multiple different tissues to more accurately study how cancer progresses in humans in order to assist development of future strategies for cancer prevention and treatment.

National Institute of Health (NIH)
Veterans Affairs (VA)
Non-HHS Research Projects (I01)
Project #
Application #
Study Section
Oncology A (ONCA)
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Veterans Admin Palo Alto Health Care Sys
Palo Alto
United States
Zip Code