Date of Award

Spring 1-1-2019

Degree Type

Dissertation

Degree Name

Doctor of Philosophy Degree

Department

Regulatory Biology

First Advisor

Weyman, Crystal M.

Second Advisor

Dr. Valentin Boerner

Third Advisor

Dr. Alex Almasen

Abstract

Epithelial oncogenesis is believed to be generally associated with the accumulation over time of an increasing number of mitotic errors until a threshold number of mutations required for the initiation of cancer is achieved. Preemption of cancer through the morphologic detection of dysplastic cells, i.e. cells with a number of mitotic errors that are still below the threshold for cancer, followed by their surgical removal or eradication, has had an enormous impact on reducing the incidence of cancer of the uterine cervix, skin and colon worldwide, but this strategy has been much less successful with cancers in most other body sites. Inflammation is a relatively common occurrence in the epithelium and is far more common than cancer. A major current obstacle to the preemption of carcinoma is distinguishing morphologically atypical epithelial cells in the presence of inflammation (inflammatory atypia) that mimic dysplasia from morphologically atypical epithelial cells that are truly dysplastic. Formation of double stranded breaks in DNA (DSBs) is an accepted etiology for carcinoma and is, therefore, expected to be associated with dysplasia. Utilizing both algorithmic and artificial intelligence-based computer image analysis of IHC levels, we document the unexpected finding that phosphorylation of molecular markers associated with DSBs is consistently correlated with non-dysplastic iv inflammatory atypia in both squamous (oral cavity) and glandular (Barrett’s metaplasia) epithelia. Using these same image analysis methods, we further show that quantitative immunohistochemistry of the ratio of p-Chk2, a marker of DSB’s, and for mutational failure of the DNA damage repair pathway (p53) required for the proper response to DSBs can distinguish between inflammatory and dysplastic cellular atypia. The ability to use quantitative means to reliably distinguish between inflammatory and dysplastic atypia may facilitate the use of cytological screening for dysplasia to prevent cancer in numerous body sites.

COinS