Allelic loss at 19q12 and Xq11–12 predict an adverse clinical outcome in patients with mucinous ovarian tumours of low malignant potential

Ovarian carcinoma is the most lethal gynaecological malignancy at present (Wingo et al, 1995). In 1929, Taylor first recognised the existence of a group of ovarian carcinomas that were associated with an improved prognosis; he termed these lesions ‘semimalignant’ tumours (Taylor, 1929). In 1961, the International Federation of Gynaecology and Obstetrics (FIGO) proposed a classification of ovarian tumours that included low malignant potential (LMP) lesions; this designation became effective in 1971 (FIGO, 1987) and was incorporated into the World Health Organisation classification in 1973 (Serov et al, 1973). Characterised by complex branching papillae, epithelial stratification, nuclear atypia, mitotic activity, and absence of stromal invasion, these tumours are considered to be of LMP. Serous and mucinous tumours comprise the vast majority of cases, with endometrioid, clear cell, and Brenner-type tumours also described (Russel et al, 1994). Low malignant potentials (30–50%) are mucinous type (Lee and Scully, 2000) and the incidence of mucinous LMPs is high in Japan compared to Western countries (Nakashima et al, 1990; Tamakoshi et al, 1997). Mucinous LMPs present at earlier stages than invasive ovarian carcinomas and have a more indolent long-term course and excellent prognosis after surgical management alone (Tazelaar et al, 1985; Casey et al, 1993; Trimble and Trimble, 1994; Siriaunkgul et al, 1995; Gershenson, 1999; Lee and Scully, 2000). Mucinous LMP patients (60–85%) who present with stage I disease can be managed conservatively with unilateral salpingo-oophorectomy and complete staging. The postoperation 10-year survival rate is 93–99% for stage I disease (Casey et al, 1993; Trimble and Trimble, 1994; Gershenson, 1999; Lee and Scully, 2000). For the 15–30% of patients who present with stage II–III disease, TAH-BSO and surgical staging are the treatments of choice. Approximately 10–30% of these patients will relapse and 10% will die of the tumour. The criteria for adjuvant chemotherapy for LMP patients are still undetermined. An estimated 10% of patients will develop ovarian carcinoma (Lee and Scully, 2000). Whether these invasive secondary tumours represent progression of LMP to invasive disease or new secondary tumours arising in at-risk populations is still controversial (Siriaunkgul et al, 1995; Lee and Scully, 2000). However, a subset of these tumours can progress and become lethal. This poor outcome is largely due to a lack of sensitive diagnostic tools for monitoring patients after surgical operation, with a reliance upon clinical and pathological parameters that are often difficult to assess for therapeutic decision making. In the search for more reliable prognostic indicators, a few investigators have focused on biological markers that might be predictive of LMP development (Kaern et al, 1990; Padberg et al, 1991). For example, Kaern et al (1990) reported that DNA ploidy pattern was the most important prognostic marker in their study. However, until now it has not been known which molecular markers influence the fate of these patients, because little is known about genetic alterations in LMPs. Among the various types of genetic alterations involved in ovarian carcinoma development and progression, loss of heterozygosity (LOH) of particular chromosomal regions is thought to indicate the deletion of a normally resident tumour suppressor gene (Callahan and Campbell, 1989; Knudson, 1993). Thus, it is possible that the loss of specific alleles may act as diagnostic markers of prognosis. Recently, we reported that allelic loss could be a predictor of poor outcome for ovarian carcinoma patients (Nakayama et al, 2003b). In order to seek potential predictive markers in the current study, we performed a genome-wide scan for LOH in 41 mucinous LMPs using 91 polymorphic microsatellite markers at an average interval of 50 cM throughout all of the human chromosomes (screening markers) and 25 LOH markers reportedly associated with ovarian carcinoma (hotspot markers). In addition, we assessed whether various clinicopathological parameters, microvessel density, Ki-67 labeling index, apoptotic index (AI), or p53 overexpression would be useful for predicting the postoperative outcome of mucinous LMP patients.