Commentary (Scully): Extraovarian Primary Peritoneal Carcinoma
Commentary (Scully): Extraovarian Primary Peritoneal Carcinoma
The distribution of abdominal serous carcinoma in the female ranges from ovarian carcinoma with no tumor involvement of the peritoneum to peritoneal carcinoma with no evidence of carcinoma in the ovary. For the purposes of investigation and patient care, it has been necessary to formulate criteria to distinguish tumors that are most probably primary ovarian carcinomas from those that are most likely primary peritoneal cancers.
The presently accepted criteria for making this distinction are based on minimal scientific evidence. It is well-known that in organs other than the ovary (for example, the breast, thyroid gland, lung, and stomach), tiny primary tumors may be associated with large metastases or extensive tumor spread. Therefore, it is likely that some tumors designated as primary peritoneal carcinoma according to current criteria are actually small ovarian cancers that find the peritoneum a more hospitable site for growth than the ovary. Possible reasons why the peritoneum may provide a more favorable environment for carcinoma than the ovary include the greater density of ovarian tissue, which may inhibit invasion of tumor cells originating within its superficial layers; and the production of a tumor-inhibitory substance by the ovarian stroma (which has been demonstrated in vitro).
We may never be able to distinguish, on scientific grounds, between primary ovarian and primary peritoneal serous carcinomas in every case. Clonality studies on a limited number of presumably primary peritoneal tumors have suggested a multifocal origin, in contrast to the usual finding of clonality in cases of typical ovarian carcinoma with peritoneal spread. More extensive studies of this type should be performed to confirm these results. At present, such studies cannot be performed routinely, however, and the criteria currently used to distinguish these two forms of serous neoplasia are the only ones available in the great majority of cases.
In addition to the Gynecologic Oncology Group (GOG) criteria cited by the authors, it is important to review the slides of a previously removed ovary, fallopian tube, or uterus in cases of suspected primary peritoneal serous carcinoma. This is particularly crucial since there have been cases in which tiny tumors were present in an ovary at a time when no peritoneal disease was recognized, and the patient returned subsequently with peritoneal involvement. In such cases, it may be difficult to determine without clonality studies whether the peritoneal tumor is a new primary tumor or a metastasis from the ovary.
Interesting Related Phenomena
The review by Eltabbakh and Piver provides a succinct summary of the clinicopathologic features of primary peritoneal serous carcinomas. It does not mention a number of related phenomena, which are of equal scientific interest. For example, one school of thought holds that what most investigators accept as peritoneal metastases of typical ovarian serous cancers may, in some cases, reflect an independent primary peritoneal cancer. Genetic studies to date, however, have not supported this speculation.
Also, not discussed in the review is the observation that primary peritoneal serous carcinomas are only part of the spectrum of primary peritoneal serous lesions. The benign counterpart is endosalpingiosis, in which gland-like structures and/or papillae, lined by tubal-type epithelium, are found just beneath or on the peritoneal surface. This disorder is associated most often with chronic salpingitis and ovarian serous borderline tumors. The association with chronic salpingitis suggests that shedding of tubal epithelial cells onto the peritoneum is a route of development of endosalpingiosis.
Primary serous borderline tumors of the peritoneum have also been reported, albeit in fewer numbers than carcinomas.[2-4] These primary serous borderline tumors have an excellent prognosis, although rare cases have been reported in which transformation to carcinoma has been observed on follow-up examination.[2,4] It is interesting that 41% to 99% of primary serous borderline tumors of the peritoneum are accompanied by endosalpingiosis, suggesting an origin therein.[2-4]
Endosalpingiosis is also a legitimate candidate for a precursor to primary peritoneal serous carcinoma. In one series, 2 of 14 carcinomas of this type were associated with this disorder. The much lower reported frequency of endosalpingiosis in the carcinoma cases may be attributable to a failure to record the lesion in the various reported series, most of which have limited pathologic data, or to biopsy of only obviously malignant lesions, which might have obliterated underlying endosalpingiosis. A careful search for endosalpingiosis in future cases may provide a clue as to whether it is a precursor to carcinoma.
Roswell Park Studies May Provide New Insights
The authors do not provide any details about papers currently in press from their institution that pertain to several aspects of primary peritoneal serous carcinomas. One cited conclusion is that the grade of the tumor is not a prognostic factor. However, a few reports of small series in the literature suggest that patients with grade 1 carcinomas have a better outcome than do those with higher-grade tumors. Indeed, the most highly differentiated form of serous carcinoma, the psammocarcinoma, which arises most often in the ovary but also on the peritoneum, appears to have such an excellent prognosis that some authors have concluded that chemotherapy is not indicated even in the presence of residual disease.[4,5]
If low-grade serous carcinoma of the peritoneum is shown to be associated with the same prognosis as high-grade disease in large series, the finding may not reflect the natural history of these two subtypes, but rather, the much better response to chemotherapy of the high-grade tumors. Details about the impact of grading on prognosis are eagerly awaited.
Another as yet unpublished study from Roswell Park Cancer Institute that should prove of interest is the epidemiologic investigation. The review by Eltabbakh and Piver does not mention the possible roles that parity and oral contraceptive intake may play in the development of primary peritoneal serous carcinoma. These prominent factors in the background of ovarian serous carcinoma have led to wide acceptance of the incessant ovulation theory. More detail on the authors’ study may be illuminating in this respect.
1. Karlan BY, Baldwin RL, Cirisano ED, et al: Secreted ovarian stromal substance inhibits ovarian epithelial cell proliferation. Gynecol Oncol 59:67-74, 1995.
2. Bell DA, Scully RE: Serous borderline tumors of the peritoneum. Am J Surg Pathol 14:230-249, 1990.