About 7% of all ovarian tumours are metastatic, predominantly from the gastrointestinal tract. Conversely, about 3.4% of women with colorectal carcinoma develop ovarian metastases2. In a few cases, the ovarian mass is the presenting feature.
Histopathology
The ovarian metastases from colorectal carcinomas resemble endometrioid carcinoma and the differentiation of primary ovarian endometrioid and mucinous carcinomas from colorectal metastases is problematic. Although the predominant pattern is that of large glands, there may be a small glandular or papillary architecture. Some cases show a mucinous pattern and goblet cells may be apparent. Deceptively, the mucinous component may appear of borderline malignancy or even focally appear benign: these foci do not prove that the tumour is an ovarian primary2. Dirty necrosis is suggested of a metastasis from colorectum. Metastases from the endocervix usually mimic endometrioid or mucinous tumours of gastrointestinal type, nor endocervical-like/seromucinous type3.
|
Primary ovarian mucinous tumours |
Mucinous tumours metastatic to ovary |
||
|
usually large (>15 cm) |
often small (<10 cm) |
||
|
unilateral, multicystic with a smooth capsule |
bilateral, forming nodules with surface disease |
||
|
usually no extra-ovarian disease |
there is extra-ovarian disease |
||
|
a borderline component is present |
usually absent, but may be simulated |
||
|
growth is confluent, glandular and expansive |
growth is infiltrative |
||
Immunohistochemistry
While cytokeratins may be informative, up to 70% of primary ovarian mucinous carcinomas express cytokeratin 20 and about 10% of colorectal adenocarcinomas express cytokeratin 7.
|
|
Primary ovarian adenocarcinoma |
Metastatic colorectal carcinoma |
||||
|
focal (11-40% of cells positive) |
diffuse (>40% of cells positive) |
focal (11-40% of cells positive) |
diffuse (>40% of cells positive) |
|||
|
usually positive |
5/302 |
|
||||
|
usually negative (except in mucinous tumours) |
|
29/292 |
||||
|
6/231 |
5/23 (positivity correlates with an infiltrative pattern)1 |
1/221 |
21/22 (independent of the tumour subtype, mucinous or pseudo-endometrioid)1, 6/62 |
|||
|
|
|
|
4/42 |
|||
|
|
|
|
6/72 |
|||
|
|
|
|
0/12 |
|||
|
1/231 |
0/231 |
2/221 |
7/221 |
|||
|
3/231 |
1/231 |
5/221 |
8/221 |
|||
|
positive for at least two of the markers CDX-2, P504S and b-catenin |
|
3/231 |
|
20/221 |
||
|
positive for all three of the markers CDX-2, P504S and b-catenin |
|
0/231 |
|
7/221 |
||
|
|
||||
|
Primary ovarian tumours |
atypical proliferative mucinous, gastrointestinal type |
0/30 |
0/30 |
|
|
atypical proliferative mucinous, seromucinous type |
11/11 |
7/11 |
||
|
invasive mucinous carcinoma of usual type |
0/11 |
0/11 |
||
|
Metastatic mucinous carcinoma |
gastric |
0/4 |
0/4 |
|
|
appendiceal |
0/4 |
0/4 |
||
|
colorectal |
0/24 |
0/24 |
||
|
pancreatic |
0/13 |
0/13 |
||
|
gallbladder/biliary tract |
0/4 |
0/4 |
||
|
endocervical |
3/8 |
0/8 |
||
|
unknown |
0/15 |
0/15 |
||
Hormone receptor expression is of no value in distinguishing atypical proliferative (borderline) mucinous tumours of gastrointestinal type and primary ovarian invasive mucinous carcinomas of usual type from the majority of mucinous carcinomas metastatic to the ovary3.
This page last revised 17.8.2006.
©SMUHT/PW Bishop