Abstract AP07: PROTEIN AND IMAGING MARKERS FOR EARLY DETECTION OF EPITHELIAL OVARIAN CANCER

Abstract

Abstract A central problem in ovarian cancer is late diagnosis, where the 5-year survival rate plummets well below 50%. It is now believed that a significant fraction of ovarian high grade serous carcinoma (HGSC) may actually start in the fallopian tubes (FTs), and that precancerous changes are detectable before metastasis to the ovary and peritoneal cavity occurs. A recently published molecular evolutionary analysis identifies a window of 7 years between development of a serous tubal intraepithelial carcinoma (STIC) and initiation of ovarian carcinoma, with metastases following rapidly thereafter. The authors are working to address the unmet clinical need for a minimally invasive test for early (pre-metastatic) ovarian cancer. Existing whole-body imaging techniques and transvaginal ultrasound have thus far displayed inadequate resolution and sensitivity, however optical imaging methods could fulfill the criteria for detection of primary HGSC when it is still located in the FTs and before it has spread to the ovaries. We have shown that two optical imaging modalities, optical coherence tomography (OCT) and multispectral fluorescence imaging (MFI) have great promise for detection of ovarian and FT abnormalities. In a small ex-vivo study, we found that 5 MFI measurements were sufficient to distinguish between normal, cancerous, and benign FT and ovary tissue with 100% sensitivity and specificity. We used laparoscopic OCT in an in-vivo study of 17 women and revealed distinct image features for normal, cancerous, and benign abnormalities of the ovary. For minimally invasive imaging of the fallopian tubes, a sub-millimeter diameter, flexible, steerable endoscope is necessary in order to navigate through the natural orifice of the ostium in the uterus. We have recently prototyped the smallest (0.8 mm diameter) OCT/MFI imaging endoscope with a forward-viewing imaging channel. The system as tested has approximately 70 degree full field of view, a working distance from 1 mm to infinity, and resolution that varies from 35 µm at 1 mm to 180 µm at 10 mm working distance. Such an imaging test is minimally invasive however not appropriate for wholesale screening of the general population. Therefore, we propose to utilize it as an adjunct confirmatory test after an initial positive or suspicious blood test. Our preliminary data indicate that there are significant changes in serum protein biomarkers in HGSC cases more than 18 months before diagnosis. We have identified FBG and PF4 as pre-diagnostic biomarkers of HGSC with 18-84 months lead time, and have generated and validated a 3-biomarker panel (FBG, PF4, and CA125) that classifies HGSC with >80% sensitivity/>70% specificity and 18-84 months lead time. We have identified overexpression of several glycolytic enzymes in STIC lesions and in human pre-diagnostic serum samples, indicating potential usefulness of these proteins as screening biomarkers and potential utility of glycolysis as imaging biomarker of STIC lesion. Our continuing work is focused on refining the serum protein biomarker sensitivity and specificity, creating a second-generation falloposcope for a first-in-women feasibility study, and creating an integrated model that incorporates both sets of data into an early EOC detection method. Citation Format: Jennifer K. Barton, Anna E. Lokshin, David A. Fishman, John F. Black. PROTEIN AND IMAGING MARKERS FOR EARLY DETECTION OF EPITHELIAL OVARIAN CANCER [abstract]. In: Proceedings of the 12th Biennial Ovarian Cancer Research Symposium; Sep 13-15, 2018; Seattle, WA. Philadelphia (PA): AACR; Clin Cancer Res 2019;25(22 Suppl):Abstract nr AP07.