Molecular pathogenesis of serous Fallopian tube and ovarian carcinoma

Abstract

Serous type Fallopian tube carcinoma (FTC) and ovarian carcinoma (OVCA) are gynaecological malignancies with a poor prognosis. Although most of the encountered tumours are sporadic, a positive family history is an important risk factor. Both tumour types have been linked to mutations in the BRCA1 and BRCA2 genes and share several cellular, molecular and clinical features. However, a detailed and unbiased comparison at the molecular level had been lacking. In this thesis the molecular pathogenesis of serous FTC and serous OVCA was further studied. After an introduction in Chapter 1, we describe a genome-wide array comparative genomic hybridization (array CGH) study of 14 serous FTCs (Chapter 2). All FTCs showed a high frequency of copy number aberrations, while these were remarkably homogeneous. Recurrent regions of amplification suggested known oncogenes MYC, CCNE1 and AKT2 to be driver genes. In Chapter 3 we describe the immunohistochemical analysis of p53, HER-2/neu and p27Kip1 on 28 serous FTCs and we correlated expression levels with DNA copy number. The studies suggested that p53 accumulation and p27Kip1 down-regulation are early events, whereas HER-2/neu overexpression may be involved in progression. Furthermore, overexpression of HER-2/neu may be caused partly by amplification. In Chapter 4 we were the first to report on a direct comparison with array CGH of serous FTCs and OVCAs. The analysis suggested that these display (quantitative) differences in their genomic profiles, besides shared features. Multiplex ligation-dependent probe amplification (MLPA) directly identified EIF2C2 as a new possible driver gene. Gains/amplifications of CCNE1 and MYC, often in conjunction with changes of EVI1 and PTK2, seemed to be common in earlier stages, whereas changes of HER2 were associated with advanced stages (see Chapter 3). In Chapter 5 we report on the development of dedicated MLPA gain-probe-sets, specifically tailored at high-grade serous FTC and OVCA and primarily based on our previous array CGH (see Chapter 4). We identified, besides known genes PIK3CA, AKT2, FGFR1 and PDGFB, new putative genes MYCBP, LIMK1, SOCS1, SMARCA4, DPF1, BCL2L1, NCOA3, PTPN1, NFATC2 and KCNQ2. As an application, we performed logistic regression, resulting in a model that could discriminate 76% of the FTCs and 80% of the OVCAs on basis of 3 markers. Since serous endometrial carcinomas (ECs) share their aggressive clinical behaviour and their histological appearance with serous OVCA and FTC, it could be hypothesized that these serous cancers share common denominators. In Chapter 6, we examined 9 serous ECs with the same array as before and compared the profiles with those described in Chapter 4. We could show that serous ECs can be discriminated from serous OVCAs. In Chapter 7 the results of the studies are discussed. We conclude that despite their histomorphological and clinical resemblance, high-grade serous FTC and OVCA show differences in their genomic profiles, besides also shared features, and should be regarded as distinct entities. We have identified new candidate genes for carcinogenesis and validated certain known genes, which has resulted in a more detailed knowledge on the pathogenesis of this class of tumours.