These tumors have a very favorable prognosis, but treatment is often complicated by two factors: their unexpected discovery during surgery, and the patient's desire to preserve her fertility.
Gynecologists and obstetricians typically encounter borderline ovarian tumors in the course of managing adnexal masses. In contrast to invasive epithelial ovarian cancers, borderline tumors are more likely to occur in premenopausal women and are more likely to be stage I, or confined to the ovaries at presentation. When borderline ovarian tumors are confined to the ovaries, long-term survival approaches 100%. Even when the disease has spread beyond the ovaries, long-term survival exceeds 70%. Despite the generally favorable prognosis, the management of borderline tumors is often complicated by two factors: one, the unexpected nature of the diagnosis, which often occurs intraoperatively, and two, the desire of patients in the reproductive age group to maintain fertility. This article will review the clinical and surgical management of these tumors.
Approximately 3,000 American women are diagnosed with borderline ovarian tumors every year. In 1929, Howard Taylor was the first to recognize that there was a subset of epithelial ovarian neoplasms that had distinct histologic features and a more favorable prognosis. In the 1970s, FIGO and the World Health Organization recognized borderline tumors as a distinct category of epithelial ovarian neoplasmsan intermediate category between benign adenomas and frankly malignant carcinomas.
The histologic criteria for the diagnosis of borderline ovarian tumors include nuclear atypia, stratification of the epithelium, formation of microscopic papillary projections, and the absence of stromal invasion.1 In the same way that invasive ovarian cancer is divided into different histologies, borderline tumors are also divided into different histologic subtypes. Serous and mucinous borderline tumors account for 50% and 46%, respectively, of all such tumors, while endometrioid, clear cell, and Brenner borderline tumors are rare. Serous and mucinous borderline ovarian tumors differ in several ways (Figure 1). Mucinous tumors, which tend to be unilateral and are less likely to have peritoneal implants, are often larger than serous borderline tumors.
The average age at diagnosis for both serous and mucinous ovarian tumors is 45, in contrast to 61 for invasive ovarian cancer. Aside from this difference in age, studies have shown a similarity in risk factors for borderline ovarian tumors and invasive ovarian cancers.2 Pregnancy, breast-feeding, and the use of oral contraceptives appear to be protective. Several case reports linking fertility drugs to the development of borderline ovarian tumors have received significant public attention, but a more recent report from Mosgaard and colleagues has found that while infertility is a risk factor for the development of ovarian tumors, the additional use of fertility drugs is not.3 Longer prospective studies are needed to determine causality.
A family history of ovarian cancer is not generally believed to be a risk factor for the development of borderline tumors. While there are individual reports in the literature of women with borderline ovarian tumors and germline mutations of BRCA1 and BRCA2, other studies have not found these tumors to be part of the hereditary breast and ovarian cancer syndromes.4
Borderline ovarian tumors present in the same way as other adnexal masses: Patients may complain of pelvic pain or dyspareunia, although frequently they are asymptomatic. Adnexal masses may be palpated at the time of a pelvic exam. Occasionally they are picked up as an incidental finding during routine obstetrical U/S.
A borderline tumor will often show up as a complex ovarian mass on U/S. Attempts to use color-flow Doppler and MRI to distinguish a benign ovarian mass from a borderline ovarian tumor or a borderline tumor from an outright malignant one, however, have not been reliable. Similarly, CA-125 is not a good marker for early-stage disease, since 60% of stage I borderline ovarian tumors have CA-125 levels within the normal range.5 And since many benign conditions like endometriosis can elevate CA-125 levels, even when CA-125 levels are high, they are nonspecific in premenopausal women.
Because there are no reliable tests to identify borderline tumors preoperatively, and because these tumors appear fairly low on the differential diagnosis list for an adnexal mass in a woman of reproductive age, a borderline tumor is often misdiagnosed before surgery. With that in mind, surgeons who operate on reproductive age patients with adnexal masses should be aware of this potential diagnosis and be ready to formulate a surgical plan and counsel their patients preoperatively.
General obstetrician-gynecologists, rather than gynecologic oncologists, are more likely to encounter borderline ovarian tumors while operating for an adnexal mass. Because of the imprecision of basing a diagnosis on gross appearance, a frozen section should be obtained for all women undergoing surgery for an adnexal mass. Standard treatment for a woman who has completed childbearing includes total abdominal hysterectomy, bilateral salpingo-oophorectomy and surgical staging (Figure 2). Ideally, the extent of surgical staging will include cytologic washings, omental biopsy, and lymph node sampling. Thorough visualization of the entire pelvis and abdomen is important. Because the surgery may be performed laparoscopically, or the abdominal incision may be a Pfannenstiel, complete surgical staging is often not achieved. Make every effort to do as complete a staging procedure as possible in these cases.
For patients who wish to preserve their fertility, a unilateral oophorectomy may be sufficient if the tumor is confined to one ovary. The contralateral ovary should be examined: If it appears normal, we do not recommend doing a wedge biopsy. Ovarian cystectomy has also been performed in women who have borderline ovarian tumors and wish to retain fertility. In a report by Lim-Tan and colleagues, recurrence in the ipsilateral or contralateral ovary was approximately 10% following cystectomy.6 After additional surgery, these women remained disease-free. Even in women with bilateral borderline tumors, attempts at preserving ovarian tissue are reasonable. Counsel patients that they may be at risk for additional surgery, but their life is not jeopardized by retaining ovarian tissue.
In patients with clearly advanced stage borderline ovarian tumor, all visible disease should be surgically removed. Aggressive debulking surgery is important, since studies have shown that patients with gross residual disease are at greater risk for recurrence and death. Furthermore, borderline tumors are less sensitive to chemotherapy, perhaps because the cells are not as rapidly dividing as frankly malignant cells.
In patients with a mucinous tumor, perform appendectomy and a careful exploration of the GI organs. There is increasing evidence that many of these tumors are actually metastases from a primary tumor of the appendix or GI tract. This is particularly true when associated with pseudomyxoma peritonei, a condition in which the entire abdominal cavity is filled with a mucinous, gel-like material. In cases of pseudomyoma peritonei, make an effort to remove all visible tumor.
Accurate pathologic diagnosis is crucial for postoperative counseling. To avoid missing a focus of outright carcinoma, serial sectioning of the tumor is necessary, preferably with a section taken at every 1 cm of the tumor. In hospitals where relatively few borderline tumors are seen or in cases where there may be uncertainty about diagnosis, seek a second opinion.
Counseling a patient about postoperative treatment and prognosis requires appropriate staging, in addition to accurate pathologic diagnosis. Thorough inspection of the pelvic and abdominal organs and peritoneum should be documented in the operative report. At a minimum, we recommend cytologic washings, omental biopsy or infracolic omentectomy, and pelvic peritoneal biopsy.
Lack of appropriate staging in the treatment of borderline ovarian tumors is common, and in these situations the patient and physician are faced with a choice between additional surgery for staging purposes or having to settle on a treatment plan with incomplete information. There is a definite risk of advanced borderline tumor in a patient who has not been adequately staged, but surgery might not be the first choice. We recommend that the patient discuss the risks and benefits of re-staging with a clinician experienced in managing these tumors. Close follow-up with serial exams and U/S or CT scans may be an alternative to surgical staging.
Stage I borderline ovarian tumors. Patients with stage I borderline ovarian tumors should be reassured that their chance for long-term cure approaches 100%. Compiling several studies, Barnhill and colleagues found that only seven patients of 988 (0.7%) died of their tumor.7 No additional postoperative chemotherapy or radiation is indicated in patients with stage I borderline ovarian tumors.8
Long term follow-up is essential for these patients, however. We recently reported that 11 of 160 patients with reasonably well-staged stage I serous borderline tumors had recurrences 7 to 39 years after initial diagnosis, with a mean of 16 years.9 Ten of the 11 patients developed low-grade serous carcinomas and one had a borderline tumor. The long interval raises the question of whether tumors we considered recurrences were actually new primary tumors. Regardless of this issue, long-term follow-up is needed for early diagnosis of recurrent or new disease.
Advanced-stage borderline tumors. In approximately 30% of women with serous borderline ovarian tumors, disease is present beyond the ovary. These patients with stage III borderline disease have a long-term survival rate that approximates 70%, in sharp contrast to the 15% 5-year survival rate for women with advanced-stage invasive epithelial ovarian cancer.10
How can we determine which women with advanced-stage borderline tumors are at risk for relapse and death? In several series, including our own, we have noted that the type of peritoneal implant may reflect the clinical behavior of the tumor.11,12 Implants can be categorized as invasive or noninvasive. Invasive implants consist of clusters of cells that are seen infiltrating the surrounding stroma (Figure 3). By contrast, noninvasive implants resemble the primary borderline tumor and consist of groups of cells exhibiting glandular or papillary proliferations (Figure 3). Noninvasive implants can also exhibit desmoplastic fibrosis. In studies of patients with advanced borderline tumors and implants that were categorized as invasive or noninvasive, the relapse rate for patients with invasive implants was 45% and the death rate 37%.13 But for patients with noninvasive implants, the relapse rate was 11% and the death rate was only 6%.14
Kurman and colleagues have proposed that advanced-stage borderline ovarian tumors with invasive implants be considered carcinomas. Furthermore, they have proposed histologic criteria to re-classify a subset of borderline tumors that may have a more aggressive clinical course as "micropapillary serous carcinoma."15 While their recommendations have not been fully accepted by pathologists, further histologic and molecular studies are needed to identify those patients at highest risk for recurrence and poor outcome.16
For women with advanced-stage borderline ovarian tumors, there is controversy as to whether postoperative chemotherapy is beneficial. Several studies conducted in the 1970s and 1980s showed no benefit from postoperative treatment for advanced-stage borderline ovarian tumors, while documenting substantial side effects, including deaths from acute nonlymphocytic leukemia related to prolonged chemotherapy with alkylating agents. As we are better able to identify the cohort of women with borderline tumors who are at highest risk for relapse and death, further studies will be needed to determine optimal postoperative treatment.
In patients who have a recurrence, a second debulking surgery is typically performed. Careful pathologic review of the tumor is important, since we have found that a substantial number of patients with advanced-stage serous borderline tumors have low- grade serous carcinoma at the time of relapse. Because these cases are relatively rare, there are no large prospective studies to determine optimal management. We recommend referring these patients to centers that have experience with advanced-stage and recurrent borderline ovarian tumors.
Patients who want to preserve their fertility and who undergo conservative surgical treatment for borderline ovarian tumors will be anxious to know both their chances of having children and their risk for developing recurrent disease. On a positive note, a study of 39 women with stage I borderline ovarian tumors who had been treated with unilateral oophorectomy or cystectomy reported that 22 women had subsequently become pregnant.17 Of these 22, 19 had delivered healthy infants and three pregnancies were ongoing at the time of the report. There were no recurrences of tumor during the follow-up, but the duration of the study was short. Our experience has been similar, with successful pregnancies after conservative surgical treatment.
While the diagnosis of a borderline ovarian tumor may be unexpected for both the patient and her physician, appropriate intraoperative management and postoperative counseling can be achieved on the basis of an understanding of the disease. Because these tumors frequently occur in women of reproductive age, surgical management should emphasize both staging and preservation of fertility. As more data accumulate on the small subset of women who do poorly after surgery for these tumors, we will be able to design rational postoperative therapy.
1. Czernobilsky B. Common epithelial tumors of the ovary. In: RJ Kurman, ed. Blaustein's Pathology of the Female Genital Tract. 3rd ed. New York, NY: Springer-Verlag; 1987:560-606.
2. Harris R, Whittemore AS, Itnyre J, Collaborative Ovarian Cancer Group. Characteristics relating to ovarian cancer risk: collaborative analysis of 12 U.S. case control studies. III. Epithelial tumors of low malignant potential in white women. Am J Epidemiol. 1992;136:1204-1211.
3. Mosgaard BJ, Lidegaard O, Kjaer SK, et al. Ovarian stimulation and borderline ovarian tumors: a case-control study. Fertil Steril. 1998;70:1049-1055.
4. Gotlieb WH, Friedman E, Bar-Sade RB, et al. Rates of Jewish ancestral mutations in BRCA1 and BRCA2 in borderline ovarian tumors. J Natl Cancer Inst. 1998;90: 995-1000.
5. Rice LW, Lage JM, Berkowitz RS, et al. Preoperative serum CA-125 levels in borderline tumors of the ovary. Gynecol Oncol. 1990;39:195-198.
6. Lim-Tan SK, Cajigas HE, Scully RE. Ovarian cystectomy for serous borderline tumors: a follow-up study of 35 cases. Obstet Gynecol. 1988;72:775-781.
7. Barnhill DR, Kurman RJ, Brady MF, et al. Preliminary analysis of the behavior of stage 1 ovarian serous tumors of low malignant potential: a Gynecologic Oncology Group Study. J Clin Oncol. 1995;13:2752-2756.
8. Trope C, Kaern J, Vergote IB, et al. Are borderline tumors of the ovary overtreated both surgically and systemically? A review of four prospective randomized trials including 253 patients with borderline tumors. Gynecol Oncol. 1993;51:236-243.
9. Silva EG, Tornos C, Zhuang Z, et al. Tumor recurrence in Stage 1 ovarian serous neoplasms of low malignant potential. Int J Gynecol Pathol. 1998;17:1-6.
10. Leake JF, Currie JL, Rosenshein NB, et al. Long term follow up of serous ovarian tumors of low malignant potential. Gynecol Oncol. 1992;47:150-158.
11. Gershenson DM, Silva EG. Serous ovarian tumors of low malignant potential with peritoneal implants. Cancer. 1990;65:578-585.
12. Bell DA, Weinstock MA, Scully RE. Peritoneal implants of ovarian serous borderline tumors. Cancer. 1988;62:2212-2222.
13. Gershenson DM, Silva EG, Levy L, et al. Serous borderline tumors of the ovary with invasive peritoneal implants. Cancer. 1998;82:1096-1103.
14. Gershenson DM, Silva EG, Tortolero-Luna G, et al. Serous borderline tumors of the ovary with non-invasive peritoneal implants. Cancer. 1998;83:2157-2163.
15. Seidman JD, Kurman RJ. Subclassification of serous borderline tumors of the ovary into benign and malignant types. Am J Surg Pathol. 1996;20:1331-1345.
16. Eichhorn JH, Bell DA, Young RH, et al. Ovarian serous borderline tumors with micropapillary and cribiform patterns: a study of 40 cases and comparison with 44 cases without these patterns. Am J Surg Pathol. 1999;23:397-409.
17. Gotlieb WH, Flikker S, Davidson B, et al. Borderline tumors of the ovary: fertility treatment, conservative management, and pregnancy outcome. Cancer. 1998;82: 141-146.
David Gershenson. Clinical management of borderline ovarian tumors.