Early Ovarian Cancer Detection Using Fluorescence Spectroscopy in the Ultraviolet-C through Visible
Keywordsfiber optic probe
MetadataShow full item record
PublisherThe University of Arizona.
RightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
AbstractWe evaluate the changes in fluorescence from endogenous fluorophores such as amino acids, structural proteins and enzymatic co-factors to predict malignancy and risk of developing ovarian cancer. 249 ovarian biopsies of the surface epithelium were interrogated in vitro, over 270-550 nm excitation, and fluorescence was collected from 290-700nm. Spectroscopic data was analyzed using parallel factor analysis (PARAFAC) to determine excitation and emission spectra of the underlying fluorophores that contribute to the total detected fluorescence intensity. Using multivariate normal distribution fits and cross-validation techniques, sensitivity (SN) and specificity (SP) of 88 and 93 percent, respectively, were achieved when classifying malignant samples versus others, while 88 and 80 percent, respectively, were achieved when classifying normal post menopausal patients as being either at low- or high-risk of developing ovarian cancer based on their personal and family history of cancer. Also, the performance of classifying cancer increases when the normal group excludes benign neoplasm and endometriosis samples, while the performance of low- v. high-risk decreases when both pre- and post-menopausal samples are included. These results could potentially be useful in screening women at increased risk of developing ovarian cancer. This motivated our study to investigate similar changes in ovarian autofluorescence in vivo. 40 patients were recruited and a total of 189 samples were imaged using a fiber optic bundle and biopsied. Using PARAFAC, the factors computed from in vitro data analysis and in vitro data as a training set; pathology from each in vivo site biopsied was compared to calibrated tissue-fluorescence. It resulted in a SN and SP of 100 and 94 percent, respectively, for classifying normal versus malignant. In the case of risk assessment, cross validated in vivo data gave a SN and SP of 68 and 93%. Results obtained were consistent with those obtained in vitro, except for the presence of blood absorption peaks which affected risk assessment. Assessing endogenous fluorescence has diagnostic potential and if adapted to trans-vaginal access, would make the screening procedure less costly and less invasive, and would be most useful and economical in women at increased risk of developing ovarian cancer and might determine the ideal time to undergo an oophorectomy.
Degree ProgramGraduate College