Mucinous epithelial ovarian carcinoma

Abstract

Mucinous tumours involving the ovary may be benign, borderline, or malignant. Malignant tumours may be primary or metastatic. Differentiation between primary and metastatic involvement of the ovary is critical for optimal patient management. Even among skilled pathologists, this distinction can be problematic, as can the distinction between borderline ovarian tumour of intestinal type and well-differentiated invasive primary mucinous ovarian carcinoma. Primary invasive mucinous ovarian carcinoma and mucinous carcinoma metastatic to the ovary do have distinct patterns of macroscopic and microscopic involvement which will reveal the correct diagnosis in many cases. There are also well-recognized patterns of immunohistochemical staining that can further assist in this differentiation. As a result of the application of these histopathological techniques, the incidence of primary invasive mucinous epithelial carcinoma has fallen over recent years from ∼12% to ∼3%. However, even in recent multicentre clinical trials such as GOG 182, expert pathological review suggests that ∼60% of tumours originally classified as primary invasive mucinous carcinomas were in fact metastatic tumours to the ovary. Review of outcome data for patients with mucinous carcinoma entered into multicentre trials suggests that this subtype of disease has a particularly poor prognosis in comparison with other subtypes of ovarian carcinoma. Historically, patients with mucinous epithelial ovarian carcinoma (mEOC) have been treated in the same way as other subtypes of ovarian carcinoma. While there is undoubtedly a response rate to platinum-based chemotherapy, retrospective reviews of individual centre experience suggest that this is substantially lower than for high-grade papillary serous carcinoma and in the order of only 30%-40%. The mEOC trial was established to investigate the possibility that the combination of capecitabine and oxaliplatin (chemotherapy drugs more commonly used in colorectal carcinoma) may be superior to conventional carboplatin and paclitaxel chemotherapy. In a 2 × 2 factorial design, there was also a randomization to bevacizumab. Unfortunately, this trial closed early, 5 years after initiation having recruited just 50 of a proposed 322 patients. mEOC is now characterized as a type I tumour with an identifiable stepwise progression from a premalignant lesion, through non-invasive, to invasive malignancy. Molecular characterization of mEOC reveals it to be distinct from other subtypes of the disease with a KRAS mutation occurring in 40%-50% of patients. Other gene abnormalities including HER2 amplification in ∼19% also occur. This raises the possibility of the use of targeted molecular therapies which with molecular analysis of individual patient tumours could form the basis of a future clinical trial. It is, however, clear that if trials are to be conducted in this rare subtype of disease, they will need to be truly international in nature and carefully designed, possibly using an adaptive stepwise approach and will require an appropriate level of funding with a realistic assessment of likely recruitment. Associated translational research will clearly be essential.