Gliomatosis peritonei: a series of eight cases and review of the literature

The metastatic implantation of mature glial tissue on the surface of the peritoneum, omentum, and abdominal lymph nodes is defined as GP. In a clinical setting, these widespread peritoneal, grayish tan-colored, tiny nodules encountered intraoperatively may be misdiagnosed as ovarian carcinomas or peritoneal tuberculosis. Ideal optimal resection may not be achieved in such cases with widespread implantation. However, GP has a favorable prognosis. In the report by Yoon et al., the overall survival did not differ between immature ovarian teratomas with GP and immature ovarian teratomas without GP, though patients with GP presented more frequent recurrence and shorter recurrence-free survival [2]. All patients (15 cases with GP) except one are currently alive when the articles were published.

GP is often diagnosed on HE-stained tissue sections, and its differentiation with low-grade epithelial ovarian tumors was difficult. A positive staining reaction of glial tissue for the neural marker, GFAP, is helpful. GFAP immunostain confirmed the glial nature of the tissue. A strong expression often suggests tumor cells are mature and well differentiated [3].

The etiology of GP is largely unknown. According to previous reports, there are two theories about the development of GP. One relates to capsular defects of the primary teratoma or dissemination via angiolymphatic channels. In some cases, GP within the omentum immediately adjacent to a capsular defect supports the mechanism of spread in its inception [4]. In 11 of the 12 cases in Robboy’s report, the capsule either had a tear or was adherent to the omentum or adnexal structure [5]. In support of lymphatic dissemination, mature glial tissue has been presented in para-aortic and pelvic lymph nodes with or without the presence of GP (Table 2). In some reports, GP was only found in the second surgery (as in case 4 in the present study). Kim et al. reviewed about 100 cases in the literature and found 9 cases of nodal gliomatosis in the pelvic or para-aortic lymph nodes with or without the presence of GP [6]. However, in their literature review, there were three cases of nodal gliomatosis presented as grades 1 to 3. As the nodal implantation contained immature elements, these should not be considered as nodal gliomatosis but as metastasis of the immature teratoma.

The other theory suggests that glial foci are genetically unassociated with teratoma. In Ferguson’s study [7], they use polymorphic microsatellite (MS) loci in two cases to address the origin of GP. They found that glial implants and normal tissue showed heterozygosity, while the teratoma showed homozygosity at the same MS loci. Their findings indicated that glial implants in GP were unrelated to the ovarian teratoma and arise from normal cells such as pluripotent Müllerian stem cells. It is possible that peritoneal stem cells can differentiate into glial cells under the stimulation of some factors secreted by teratomas [8]. The occurrence of GP after ventriculoperitoneal shunt operations when glial tissue is transported from cerebrospinal fluid into peritoneal cavity via shunt further support this theory [9]. However, the detailed mechanisms by which subperitoneal cells develop into glial follicles remain to be determined.

The stage and grade of the primary teratoma and the grade of its metastatic tumor are related to the prognosis of teratoma. Robboy and Scully reviewed the literature and presented a large series of cases [5]. They found that the prognoses in metastasized ovarian teratomas, or in ovarian teratomas leading to peritoneal implants, are favorable when the implants are composed of fully mature glial tissue. Müller reviewed all cases of GP published between 1906 and 2002. Eleven cases showed adverse outcomes [1]. They found that the recurrence of disease was associated with lack of extensive histological sampling at the first surgery. Thus, all specimens need to be adequately sampled and multiple biopsies should be taken to exclude immature glial tissue or teratoma elements. Once the presence of immature teratoma is confirmed in the metastatic tissue, the treatment scheme and prognosis may change.

Most of the papers on GP are cases. At present there are only 5 articles which report more than five cases (Table 3) [2, 8, 10‐12]. Most cases of GP are associated with ovarian teratoma, especially with immature ovarian teratoma. In Yoon’s report, there were a total of 16 ovarian teratomas associated with GP [2]. Among them, 15 cases were of immature ovarian teratomas of various grades (4 cases were grade 1; 11 cases were grade 2 or 3). Liang reported the largest series of GP cases so far [8]. Sixteen of 23 cases were associated with immature ovarian teratomas. However, immature ovarian teratoma with GP showed better prognosis than would be expected based on the grading of the primary immature teratoma. In Norris’s report, the survival of patients with grades 1, 2, and 3 were 82 %, 63 %, and 30 %, respectively [10]. In Yoon’s report, all but one case of immature teratoma with GP remained alive when their report was published, although GP showed more frequent recurrences [2]. In Robboy’s report, 12 patients were alive and well [5]. More over, the survival of 8 patients in our series may also indicate that the presence of mature glial implants does not affect adversely the prognosis of ovarian teratoma.

Presence of glial tissue in lymph nodes is rare. Thus far, 14 cases have been reported in nine articles [6, 8, 13‐18] (Table 2). Based on previous reports, adjuvant chemotherapy is unnecessary for patients with retroperitoneal lymph node metastasis of mature glial tissue. The prognosis of patients with nodal gliomatosis is favorable. In our series, patients with nodal gliomatosis remained alive and well during the follow-up period.

Sometimes immature ovarian teratoma can be associated with miliary spread of immature implants. After surgery and chemotherapy, the immature tissue may transfer into mature tissue. Chemotherapeutic retroconversion has also been called growing teratoma syndrome (GTS) [19]. However, there are some differences between GP and GTS. First, GP is composed of pure mature implants in the peritoneum without other mature nonglial tissues which can be seen in GTS. Second, GP could be encountered in the first surgery without prior chemotherapy. At last, the management of these two terms is quite different. An optimal cytoreduction is recommended for GTS to avoid complications such as bowel obstruction and perforation [20]. Because the lesions in GP are extensive, complete resection of GP is usually difficult. Luckily residual peritoneal disease in GP can be asymptomatic and quiescent over a long period [12]. Thus, residual implants can be ignored and the therapy is mainly depended on the stage and grade of the primary ovarian teratoma. Seven cases in our series presented macroscopic residual disease at postoperatively, and all of these patients remained alive and asymptomatic during the median follow-up of 60.5 months.

Bentivegna reported 10 cases of GP, while six of them were managed by conservative surgery [12]. All of their patients were asymptomatic at the time of the last consultation, and half had incomplete resected macroscopic GP. These authors found residual peritoneal disease could be totally quiescent over a long period without any impact on patient outcomes. As ovarian teratoma mostly occurs in young women who wish to preserve fertility, it is important to reduce the surgical scope and reduce surgical trauma without compromising cure rate. As residual implants of GP can be ignored, a more conservative surgical approach may be carried out in patients with massive peritoneal spread after exclusion of the presence of metastatic immature elements.

In the review of literature undertaken by Müller (86 cases plus their own two cases) [1], three progressions were observed: GP was asymptomatic and detected by a secondary surgery; GP may undergo “fibroblastic transformation” and gradually disappear [21]; GP may transform into malignant glial neoplasms such as glioblastoma which can result in patient death [22]. In most of the cases, patients with massive peritoneal spread cannot be completely resected, but these lesions can be totally quiescent over a long period. GP is associated with frequent recurrence in patients with immature ovarian teratoma [2]. In rare cases, GP showed potential for malignant transformation after several years. Thus, a long period of careful monitoring for patients with GP may be needed.