Although the simultaneous presentation of endometrial and ovarian carcinomas of the endometrioid type is well described, little is known about a similar phenomenon involving the endometrium and fallopian tube (FT). We present the clinicopathologic features of 13 such cases seen in the Department of Pathology at The University of Texas M.D. Anderson Cancer Center over an 8 year period (1995 to 2002). FT tumors that could have represented luminal extension of the endometrial carcinoma or that represented an unequivocal metastasis to the FT were excluded. The patients' ages ranged from 34 to 77 years (median 54). The most common symptom was abnormal uterine/vaginal bleeding (11) and all of the patients were considered overweight or obese (mean body mass index was 41). The size of the endometrial carcinomas ranged from 0.3 to 8 cm. According to the FIGO grading of the endometrial endometrioid carcinomas, the cases were distributed as follows: Grade 1 (3) and Grade 2 (10). In 2 cases, there were also small areas of other histologic types, papillary serous carcinoma (1 case), and papillary endometrial carcinoma of intermediate grade (another case). The size of the fallopian tube carcinomas ranged from 0.2 to 17.5 cm. Seven of these tumors were located in the distal/fimbriated end of the FT. There was bilateral involvement in 2 cases. Three of the FT tumors were in situ. The grades of the fallopian tube carcinomas were as follows: Grade 1 (2), Grade 2 (6), and Grade 3 (2). In situ components were seen in all cases. One fallopian tube carcinoma was mixed with serous carcinoma. In 4 cases, there was also an endometrioid carcinoma involving the ovary, all of them with an intact ovarian capsule. Patients were treated as follows: total abdominal hysterectomy and bilateral salpingo-oophorectomy (TAH/BSO) (4), TAH/BSO/chemotherapy (chemo) (4), TAH/BSO/radiation (3), and TAH/BSO/chemo/radiation (2). Follow-up ranging from 6 to 54 months was available in 10 patients: 1 patient died of disease (at 38 mo), 1 patient is alive with disease (at 9 mo), 7 patients have no evidence of disease (6 to 54 mo), and 1 patient died of metastatic endometrial carcinoma (at 9 mo). Simultaneous endometrioid carcinomas of the uterus and FT are unusual and occur primarily in obese perimenopausal women. The tumors are predominantly well or moderately differentiated with dissimilar endometrial and FT grades. The FT carcinoma is usually unilateral and located at the distal end of the tube.
Summary: A 2-tier system that classifies ovarian serous carcinoma (OSC) as low grade or high grade is gaining acceptance. Women with low-grade OSC generally have higher 5-year survival rates than do women with high-grade OSC. We examined the expression of various markers to further understand the molecular differences between low-grade and high-grade OSCs: the potential therapeutic targets or prognostic markers Her-2/neu, estrogen receptor, and progesterone receptor (PR); the metastasis-associated markers cyclin D1 (BCL1), E-cadherin, matrix metalloproteinase (MMP) 2, and MMP-9; and the cell proliferation-associated markers BCL1, Ki-67 antigen (Ki-67), and p53. For this immunohistochemical analysis, we used paraffin-embedded specimens from 47 patients with advanced-stage low-grade OSC and from 49 patients with advanced-stage high-grade OSC. Our results showed that low-grade tumors expressed significantly higher levels of estrogen receptor, PR, and E-cadherin than did high-grade tumors, suggesting the involvement of gonadal steroid hormones, especially in the pathogenesis of low-grade OSC; the PR positivity was also observed in the stromal component of these low-grade tumors. On the other hand, high-grade tumors trended toward increased expression of MMP-9, BCL1, p53, and Ki-67, and robust MMP-9 positivity was observed in the stromal component of these high-grade tumors. These differences may lead to the development of different therapeutic strategies for women with either the low-grade or the high-grade form of OSC.
The immunohistochemical expression pattern of p16 in biopsy samples has been useful as part of a panel to distinguish adenocarcinomas arising from the endometrium from those arising from the endocervix. However, no information is available on the expression of p16 in uterine serous carcinoma (USC) or ovarian high-grade serous carcinoma that could be used for diagnostic purposes. Here, we retrospectively analyzed the immunohistochemical expression of p16 in 11 cases of USC (5 pure and 6 mixed with endometrioid adenocarcinoma) and 10 cases of ovarian high-grade serous carcinoma and compared p16 expression with that of p53 in the same samples. p16 was strongly expressed by 100% of tumor cells in all 11 uterine specimens and in 5 of the 10 ovarian specimens; of the other 5 ovarian specimens, 4 showed strong positivity in 20% to 80% of tumor cells, and 1 case showed only weak expression. Positivity for p53 was strong and diffuse (100% of tumor cells) in 5 uterine tumors and in 3 ovarian tumors. p53 expression in 6 of the uterine specimens and 7 of the ovarian specimens was present in fewer tumor cells, of weak intensity, or both. We also performed human papilloma virus (HPV) DNA in situ hybridization in 4 uterine pure serous carcinomas; all 4 were negative. The negative results were confirmed by reverse transcriptase in situ polymerase chain reaction. We conclude that p16, owing to its diffuse expression in USC, should not be interpreted as indicating cervical origin or HPV-induced carcinogenesis; however, p16 may be a better marker (albeit unspecific) than p53 for identifying USC. The overexpression of p16 in USC is unrelated to HPV. Further studies are necessary to determine whether p16 expression is useful in the differential diagnosis of ovarian high-grade serous carcinoma.
Recently some investigators have proposed abandoning the term ovarian serous tumor of low malignant potential (SLMP) and dividing the tumors in this category into two new groups, micropapillary serous carcinoma and atypical proliferative serous tumor, based on the presence or absence of marked epithelial proliferation with a micropapillary or cribriform pattern (MP/CP). We reviewed 99 cases of advanced stage SLMP (FIGO stages II and III) to determine whether the presence or absence of MP/CP predicts the clinical course, thus justifying the proposed change in terminology. Eighteen cases of MP/CP and 81 cases of typical SLMP were identified. The patients with MP/CP ranged from 23 to 59 years of age at the time of diagnosis (median 35 years), whereas those with typical SLMP were 17 to 67 years old (median 38 years). Bilateral ovarian involvement by SLMP was more frequent in the MP/CP cases, 13 of 18 (72%), as compared with the cases with typical SLMP, 46 of 81 (57%). There was a trend toward a greater frequency of invasive implants in MP/CP cases, 3 of 18 (17%) MP/CP versus 5 of 81 (6%) typical. The mean follow-up period was 125 months for MP/CP patients and 132 months for the typical group. Differences in the frequency of recurrence and the progression-free survival between the groups were found to be significant. Fourteen (78%) of the MP/CP patients experienced either progression or recurrence of disease, whereas 25 (31%) of the typical SLMP patients had a recurrence (p = 0.001). The progression-free survival ranged from 3 to 208 months for MP/CP patients versus 15 to 233 months for typical SLMP patients (p <0.0001). The majority of the recurrences in both groups were low-grade serous carcinoma, 11 of 14 (79%) patients with progression/recurrence in the MP/CP group and 17 of 25 (68%) patients with recurrence in the typical group. The overall survival of the patients in the two groups, however, was not significantly different. Five (28%) MP/CP patients and 12 (15%) patients with typical SLMP died of disease (p = 0.11). All 17 of these patients developed serous carcinoma and died secondary to tumor progression. Four of eight (50%) patients with invasive implants and 13 of 91 (14%) patients with noninvasive implants died of disease. Our findings of more frequent bilateral ovarian involvement, more frequent recurrence with a shorter progression-free interval, and the trend toward a more frequent association with invasive implants support the contention that MP/CP SLMP are a distinct subgroup of serous tumors. However, the overall survival of patients with MP/CP SLMP is similar to that of patients with typical SLMP and justifies the retention of tumors with MP/CP within the LMP category. Additionally, our long-term follow-up of patients with typical SLMP indicates that a number of these tumors do not follow a benign course and supports their continued designation as borderline neoplasms.
Although differential WT-1 expression between ovarian and uterine serous carcinoma has been discussed in the literature, there have been no studies of WT-1 expression in serous carcinomas in the peritoneum with or without concurrent serous carcinoma in an endometrial polyp. This study addresses this issue and includes a small series of uterine and ovarian serous carcinomas for comparison. Nine peritoneal serous carcinomas with coexistent serous carcinoma in an endometrial polyp, 10 peritoneal serous carcinomas without serous carcinoma in an endometrial polyp, 9 uterine serous carcinomas, and 12 ovarian serous carcinomas were stained with antibody to WT-1. Ninety-two percent of ovarian serous carcinomas and 80% of peritoneal serous carcinomas without serous carcinoma involving an endometrial polyp expressed WT-1. In contrast, 12% of peritoneal serous carcinomas with serous carcinoma in an endometrial polyp expressed WT-1 with the serous carcinoma in the endometrial polyp staining concordantly. For uterine serous carcinoma without an endometrial polyp, only 11% expressed WT-1. Peritoneal serous carcinomas without coexistent serous carcinoma in an endometrial polyp have a WT-1 expression pattern similar to ovarian serous carcinoma while peritoneal serous carcinomas with coexistent serous carcinoma in an endometrial polyp have a staining pattern similar to uterine serous carcinoma. The difference in WT-1 expression among serous carcinomas suggests a difference in biology based on the site of origin. Additionally, the difference in WT-1 expression between peritoneal serous carcinomas with and without coexistent serous carcinoma in endometrial polyps suggests that peritoneal serous carcinoma may have different pathogenetic pathways.
Genetic aberrations, such as loss of heterozygosity (LOH) and mutations leading to functional inactivation of the PTEN tumor suppressor gene, located on chromosome 10q23.3, have been shown to be associated with approximately one third of ovarian adenocarcinomas. In addition, microsatellite instability (MSI) leading to the functional inactivation of the PTEN gene has also been reported for ovarian adenocarcinomas with frequencies varying from 6 to 37%. However, the frequency of PTEN gene abnormalities has not been well studied or evaluated in lesions such as typical and atypical endometriosis. The aim of this study was to investigate a possible sequential progression from endometriosis through atypical endometriosis to ovarian carcinoma by assessing LOH at 10q23.3 and MSI in those entities. Genomic DNA was analyzed for LOH and MSI at 3 loci on chromosome 10, using polymerase chain reaction amplification. Significant differences in LOH were seen between endometriosis (4.3%) and ovarian carcinoma (23.5%) at D10S608. The differences at the other 2 loci were not significant. A high frequency of MSI was found in endometriosis (82.6%) and atypical endometriosis (75%); however, the differences were not significant. These results suggest that LOH at D105608 may possibly be an important molecular event in the progression of endometriosis to carcinoma. This study highlights that endometriosis and atypical endometriosis might act as precursor lesions that have the potential to progress into ovarian adenocarcinoma.
The cyclin-dependent kinase inhibitor p27(kip1) regulates cellular progression from G(1) to S phase. Several studies have shown that loss of p27(kip1) protein expression is associated with disease progression in various malignancies. The purpose of this study was to evaluate the subcellular localization of this cyclin-dependent kinase inhibitor in a large cohort of primary ovarian carcinomas and compare the results with clinicopathologic variables and overall survival.Subcellular localization of p27(kip1) was first assessed by Western blotting in nuclear and cytoplasmic extract from 13 cases of ovarian carcinoma. Subcellular localization of the p27(kip1) protein was evaluated using tissue microarrays containing 421 cases of ovarian carcinoma.The presence of p27(kip1) in the cytoplasm regardless of the nuclear stain correlated strongly with late-stage disease (P < 0.03), extent of cytoreduction (P = 0.03), and shorter disease-specific survival (P < 0.0001).Cytoplasmic localization of p27(kip1) predicts poorer prognosis in ovarian carcinoma, particularly in late-stage disease.
