GYNECOLOGIC

40, 230-236 (1991)

ONCOLOGY

Primary Peritoneal Papillary Serous Adenocarcinoma: Clinical and Management Aspects MARCOM. ALTARAS, M.D. ,*,’ RAMI AVIRAM, M.D. ,* ILAN COHEN, M.D. ,* MARIO CORDOBA,M.D. ,t ELI WEISSM.D.,+ ANDYORAM BEYTH, M.D.* *Division and *Medical

of Gynecological Laboratories,

Oncology, Sapir Medical

Department of Obstetrics and Gynecology “A”, Center, Kfar Saba, Sackler School of Medicine,

TDepartment University

of Pathology, of Tel-Aviv, Israel

Received July 26, 1990

From January 1, 1984 to April 30, 1990, 38 patients were surgically found to have an intraabdominaldiseaseresembling epithelial ovarian cancer. This diagnosiswas confirmed in 31 patients; the remaining 7 met the criteria of primary peritoneal papillary serouscarcinoma.Five of thesewere diagnosedretrospectively and two during surgery. The mean age at diagnosis was 61.2 years. Tumor histology revealed papillary serouscarcinoma in six and mixed (papillary serousand papillary clear cell carcinoma)in one patient. Optimal debulking was achieved in three of seven cases(42.8%). Cisplatin-basedcombination chemotherapywasadministeredto all in the study group. Complete responsewasobtained in four patients, with one surviving for 76 months. The median survival in these patients was 34.5 months(range6-76 months).Currently, three patientswith complete responseare alive with clinically undetectabledisease.CA125assayswere available in three casesand blood levelscorroborated the clinically determined status of the disease.Tumor steroid hormonereceptor statuswas determinedin one caseand revealed low levels of estrogenand progesteronereceptors. To the best of our knowledge,the usefulnessof CA-125 in the diagnosis,management,follow-up, and determination of tumor steroidhormonereceptor status,mixed papillary serousand clear cell subtypehistologicalpatterns for primary peritonealpapillary serouscarcinomaare first describedin this report. It seemsthat this neoplasmmay be treated and followed up as in epithelial ovarian cancer, obtaining long-term survival; however, the biologic behavior and managementproblemsof this relatively new entity deservefurther clinical experience. o 1~1hdemic PRSS, IIIC.

sizing to it, similar to widespread or localized tumors seen in epithelial ovarian cancer [l]. This condition occurs in the absence of concurrent ovarian tumor and is associated with peritoneal papillary neoplasms, which morphologically resemble serous ovarian tumors. Neoplasms like primary peritoneal papillary serous carcinoma, various forms of peritoneal mesotheliomas, extraovarian epithelial tumors or mesenchymal neoplasms, and metastatic tumors of the peritoneum may all express themselves clinically as normal-sized ovary carcinoma syndrome [l-7]. Since the prognosis and treatment modalities of these various entities differ, their precise diagnosis and perioperative management would contribute to the true recognition of the actual disease and its optimal treatment and prognosis. Primary peritoneal papillary serous carcinoma is the most common entity contributing to the development of normal-sized ovary carcinoma syndrome and is associated with a seemingly more favorable outcome than the others [2,3,5,7]. The aim of the study is to present our accumulated clinical experience with a modest group of patients diagnosed as having this neoplasm, to discuss aspects of its diagnostic management and prognostic factors, and to add our data to the existing scant literature for this condition. MATERIALS

INTRODUCTION

In the 76-month period January 1984 to April 1990, 38 consecutive patients underwent explorative laparotomy for suspicion of epithelial ovarian cancer. Thirty-six (94.7%) of these were diagnosed as International Federation of Gynecology and Obstetrics (FIGO) stage IIIV invasive epithelial ovarian carcinomas. In the remaining two, operative findings were highly suspicious for pri-

Normal-sized ovary carcinoma syndrome is a term which defines and encompasses various malignant conditions, either arising from the peritoneum or metasta’ To whom requests for reprints and correspondence addressed.

should be 230

0090.8258/91 $1.50 Copyright 0 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.

AND METHODS

MANAGEMENT

OF PRIMARY

PERITONEAL

mary peritoneal papillary serous carcinoma, and this diagnosis was confirmed by examination of the surgical specimens. Using the same clinicopathological criteria, the charts and slides of the 36 patients previously diagnosed as having epithelial ovarian carcinoma were then reviewed. Five of thirty-six (13.8%) met the criteria of primary peritoneal papillary serous carcinoma. As a criterion for entering the study, the ovaries were to have a maximum diameter of 4 cm, since this is defined by FIG0 as a normal-sized ovary [9]. Slides and histopathological material were reviewed by the same pathologist (M.C.). No macroscopic ovarian involvement was noted in any of the cases. In four patients, microscopic examination of the ovaries revealed minimal involvement, and this was considered as involvement of the ovaries rather than as origin of the primary tumor. All seven patients were given planned cisplatin-based combined chemotherapy every 3-4 weeks. Clinical status of the disease with respect to remission, regression, stabilization, and progression was determined according to the criteria used in our previously published report [8]. Complete response was defined as the total disappearance of measurable disease for a period in excess of 4 weeks. Partial response was determined as more than 50% decrease in the tumor’s maximum diameter for a similar period. Survival was calculated from the date of diagnosis to the most recent visit of the patient. We used 20 U/ml as the upper limit of normal for CA-125 assay (81. RESULTS

Table 1 depicts the patients’ ages, clinical stages of disease as for epithelial ovarian carcinoma, surgical procedures used, residual disease, tumor type, grade, ovarian size, and degree of involvement. The mean age was 61.2 TABLE

Case

Age

Clinical stage (FIGO)

1 2 3 4

84 56 67 50

IV IIIC IV IIIC

BSO” BSO BSO BSO

5

42

IIIC

1. BSO 2. TAH BSO + STH +

6 7

69 61

IIIC IIIC

+ + + +

PAPILLARY

231

SEROUS CARCINOMA

FIG. 1. External surface of a dermoid cyst (case 7) showing several turmoral implants (i).

years (range 42-84 years). All women were menopausal, except for patient 5. In cases 1-6 both ovaries were available; in case 7 only the left ovary was available. Case 7’s right adnexa consisted of a 12 x lo-cm dermoid cyst with no identifiable ovarian tissue. Several tumor implants were observed on its external surface (Fig. 1). No malignancy was found on histological examination of the cyst contents. Case 4 had a mixed papillary tumor, composed of serous and predominantly clear cell components, similar to mesonephroid-type carcinoma of the ovary. Only case 6 was nulliparous. Abdominal swelling was the most common presenting symptom in each patient. Five of seven patients had the disease confined to the abdomen and two had abdominal spread with pleural effusion. Optimal debulking was 1

Surgical procedure

Residual disease

biopsies ICO PO omental biopsy

> > > >

+ PO + debulking ICO + debulking BSO + ICO + debulking

< 2 cm c: 2 cm < 2 cm

2 2 2 2

’ BSO, bilateral salpingo-oophorectomy; PSA, papillary serous adenocarcinoma; partial omentectomy; STH, subtotal hysterectomy.

cm cm cm cm

Tumor histology

Ovarian involvement

PSA grade III PB PSA grade III PSA grade 1 Mixed: PSA and clear cell carcinoma PB grade I PSA grade I PB

Surface No Surface Surface

PSA grade 1 PB PSA grade III PB

No No (left) Dermoid cyst (right), 10 X 12 cm

Surface

PB, psammoma bodies; ICO. infracolic omentectomy; PO,

232

ALTARAS

achieved in cases 5-7. Omentpm was the most common site where the tumor was encountered (7/7). In two patients inguinal node involvement was confirmed with fineneedle aspiration. Table 2 summarizes the mode and response to chemotherapy, number of treatment courses administered, CA125 measurements, and outcome of the patients for the study period. The cytotoxic agents were cisplatin, Adriamycin, and cyclophosphamide. No drug toxicity other than transient myelosuppression was observed. All patients completed their planned chemotherapy treatments. Only case 1 was given two courses of neoadjuvant chemotherapy. Due to satisfactory clinical response, she underwent exploratory laparotomy. Second-line chemotherapy was not administered to any of the patients. In cases 3 and 5 reexploration was performed following three and six courses of cisplatin, adriamycin, and cyclophosphamide, to achieve optimal debulking, which eventually succeeded in case 5. The mean survival rate for all cases was 24.1 months at the time of this writing. The concentrations of tumor cytosolic steroid hormone receptors for estrogen and progesterone were 33 and 16 fmole/mg, respectively. DISCUSSION Peritoneal tumors of the serous type are often regarded as mesotheliomas or disseminated ovarian carcinoma of advanced stage, due to their rarity and to a general lack of awareness of their primary occurrence in peritoneum. The unusual location of these tumors is explained by the common coelomic ancestry of the peritoneal mesothelium and the miillerian duct epithelium. A woman’s retained embryologic potential for miillerian differentiation by the peritoneum may express itself in a variety of metaplastic and neoplastic ways, such as epithelial inclusions in the ovary and lymph nodes, ovarian tumor of common epithelial origin, endometriosis, endosalpingiosis, and primary peritoneal serous neoplasms [2]. The finding of diffuse involvement of the peritoneum by papillary carcinoma, in the absence of an obvious primary site and grossly normal ovaries, was first described by Swerdlow as “malignant mesothelioma” [ll]. In 1977, Kannerstein et al. were the first to separate primary peritoneal papillary serous carcinoma from mesotheliomas and from all extraovarian epithelial tumors, establishing histological criteria that were adopted by others [6,12,13]. Until 1977 and thereafter, various terms were used for these extraovarian epithelial tumors [3,4]. Some authors have ultrastructurally investigated the cells of mesotheliomas and cells of primary peritoneal papillary serous carcinoma, including those of low malignant potential, and have concluded that primary peritoneal papillary serous carcinoma is a distinct morphological and clinical en-

ET AL.

tity from mesothelioma [2,15]. However, it is also well established that primary peritoneal papillary serous carcinoma is a different clinical entity from papillary serous ovarian carcinoma, but is histologically indistinguishable from it, being characterized without an identified primary tumor in the ovary and with an undetermined primary origin of the neoplasm [2-71. Figure 2 shows the histological appearance of the tumor found in patient 2, in whom neither ovary was involved with the neoplasm. Despite the criteria accepted 13 years ago separating primary peritoneal papillary serous carcinoma from mesotheliomas and papillary serous ovarian carcinoma, very few data have been accumulated during this time to clarify the diagnostic problems of the tumor’s biological characteristics, management strategies, factors influencing outcome, and survival rates. Most studies were designed retrospectively [2-6,12,13,14]. In only very few cases was the diagnosis made during surgery or subsequent to it, as was done in our cases 6 and 7 [7,14]. The incidence of primary peritoneal papillary serous carcinoma was found to vary between 7 and 13.8% among the cases of epithelial ovarian cancer or papillary serous ovarian carcinoma reassessed retrospectively [3,5]. In this study, of the 36 patients treated for epithelial ovarian cancer, 5 had primary peritoneal papillary serous carcinoma-an incidence of 13.8%. The reported age for primary peritoneal papillary serous carcinoma ranges between 4 and 87 years, with the mean age varying from 57.4 to 67.6 years. The age factor at the time of diagnosis warrants further investigation into its possible clinical significance [2,3,5,6,12,15]. In the present study, we found it to be 61.2 years. Among 47 patients described in four reports (including ours), 43 (91.4%) were older than 50 years with a mean age of 66.4 years, in contrast with the mean age of 57.4 years (range 40-75 years) found by Fromm et al. [2,6,12,15]. Despite the lack of additional large series, the aforementioned figures may support the presumption that primary peritoneal papillary serous carcinoma occurs mostly in elderly women and very occasionally in women younger than 50 years. The mean age of 66.4 years, or even 63 years as reported by Lele et al., is markedly higher than the mean age of papillary serous ovarian carcinoma, ranging from 55.3 to 59.6 years for all stages reported by FIGO, and this difference may be considered as additional support for this assumption [5-91. In most cases, the clinical presentation of this entity does not differ from that in the advanced stages of epithelial ovarian cancer, nor from the presentation in conditions causing normal-sized ovarian carcinoma syndrome, including widespread metastatic tumors of the peritoneum arising from the breast, thyroid, gastrointestinal tract, endometrium, or pancreas [1,3,5]. No primary

MANAGEMENT

OF PRIMARY

PERITONEAL TABLE

PAPILLARY

233

SEROUS CARCINOMA

2

Chemotherapy Case

Regimen and number of courses

Response

CA-125 NA NA NA 5OO U/ml prior to surgery;