Histological and molecular analysis of Fallopian tube precursors in pelvic serous carcinogenesis

Abstract

Epithelial ovarian cancer is the second most common gynaecological cancer, but has the highest fatality-to-case rate, which can be primarily attributed to diagnosis delay due to rapid disease progression and location. This is especially true for the serous subtype, which shows some form of pelvic spread in approximately 96% of diagnosed women, resulting in a 5-year survival rate of 37-25%. As the name implies, ovarian cancer has, regardless of subtype, been thought to arise from the ovary. However, recent studies have discovered non-invasive lesions sharing many characteristics with serous invasive cancer arising from the Fallopian tube rather than the ovary, thereby suggesting the Fallopian tube as a plausible source of many serous cancers. The first lesions to be described resemble invasive cancer in both morphology and protein expression, their non-invasive nature being the only differentiating characteristic, and were named serous tubal intra-epithelial carcinoma (STIC). This was later followed by the discovery of morphologically benign lesions, which showed aberrant protein (p53) expression, termed p53 signatures. Both lesions preferentially arise from the fimbriated end of the Fallopian tube, which lies in closest proximity to the ovary and coincides with the region in which many early invasive carcinomas occur. This thesis examines the molecular characteristics of STICs and p53 signatures by methylation and microRNA analysis, in an attempt to further elucidate their link to invasive cancer. MicroRNAs are recently discovered non-coding RNAs that are promising candidates for use as future biomarkers. We show that STICs share many microRNAs with invasive serous cancer and can be distinguished from benign Fallopian tube epithelium. These results may provide the first steps towards early detection through microRNA profiles. Regarding p53 signatures, molecular analysis of promoter hypermethylation of tumour suppressor genes did not show an increase from benign epithelium, indicating that p53 signatures may be too early in carcinogenesis to display major aberrations. However, we did find higher methylation levels in women at high risk of developing serous carcinoma compared to levels in the general population, a finding that may in part explain their increased cancer risk. Second, this thesis encompasses some of the first works describing a novel putative precursor lesion, secretory cell outgrowths (SCOUTs), which may precede the previously mentioned p53 signature and STIC and further expands the precursor sequence playing out in the Fallopian tube. Third, we have focused on the transition from non-invasive (STIC) to invasive disease and provide the first direct evidence of a metastasizing STIC in the absence of invasion, through the proposed mechanism of epithelial surface shedding of malignant cells. These findings suggest a re-evaluation of STIC characteristics is warranted to better understand their risk. We describe a polarization marker (p-ERM) which may aid in distinguishing STICs from their more benign counterparts. Lastly, we review the current state of using epigenetic biomarkers (including methylation and microRNAs) as diagnostic tools in clinical practice. As of yet their use is limited to a research setting, but continuing developments may ultimately contribute to pushing back diagnosis delay associated mortality rates.