Autofluorescence imaging can identify preinvasive or clinically occult lesions in fallopian tube epithelium: a promising step towards screening and early detection.

BACKGROUND: Optical imaging systems are robust, portable, relatively inexpensive, and have proven utility in detecting precancerous lesions in the lung, esophagus, colon, oral cavity and cervix. We describe the use of light-induced endogenous fluorescence (autofluorescence) in identifying preinvasive and occult carcinomas in ex vivo samples of human fallopian tube (FT) epithelium. METHODS: Women undergoing surgery for an i) ovarian mass, ii) a history suggestive of hereditary breast-ovarian cancer, or iii) known serous ovarian cancer following neoadjuvant chemotherapy (NAC) were approached for informed consent. Immediately following surgery, FT's were photographed in reflectance and fluorescence at high resolution. Images included: (1) white-light reflectance of luminal/epithelial surface; (2) narrow-band green reflectance (570 nm) (3) green autofluorescence (405/436 nm excitation); and (4) blue autofluorescence (405 nm excitation). Areas revealing a loss of natural tissue fluorescence or marked increase in tissue microvasculature were recorded and compared to final histopathologic diagnosis (SEE-FIM protocol). RESULTS: Fifty-six cases involving one or both fallopian tubes underwent reflectance and fluorescence visualization. Nine cases were excluded, either secondary to non-ovarian primary pathology (7) or excessive trauma (2) rendering tissue interpretation impossible. Of the 47 cases remaining, there were 11 high grade serous (HGS) and 9 non-serous ovarian carcinomas undergoing primary debulking surgery, 5 serous carcinomas having received NAC, 8 benign ovarian tumors, and 14 women undergoing risk-reducing bilateral salpingo-oophorectomy (RRBSO). Methodology was feasible, efficient, and reproducible. TIC or carcinoma was identified in 7/11 HGS, 3/5 NAC, and 1/14 RRBSO. Optical images were reviewed to determine test positive or negative based on standardized criteria. Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were calculated for the entire cohort (73%; 83%; 57%; 91%) and in a subgroup that excluded non-serous histology (87.5%; 92%; 78%; 96%). CONCLUSIONS: Abnormal FT lesions can be identified using ex vivo optical imaging technologies. With this platform, we will move towards genomic interrogation of identified lesions, and developing in vivo screening modalities via falloposcopy.