Borderline Ovarian Cancer Borderline Tumor Overview
- Author: Andrew E Green, MD; Chief Editor: Warner K Huh, MD more...
Borderline Tumor Overview
In 1929, Taylor first described a subset of epithelial ovarian tumors that he termed semimalignant. These lesions have a more favorable outcome than do other ovarian cancers, but they were not separately classified by the International Federation of Gynecology and Obstetrics (FIGO) and the World Health Organization (WHO) until the early 1970s.
Go to Ovarian Cancer for more complete information on this topic.
The 2 major histologic tumor subtypes are serous and mucinous, with serous being more common. Serous tumors are presumed to originate from the germinal epithelium. Mucinous tumors do not have a clearly defined origin. Substantial information indicates that many tumors may actually originate from the appendix; thus, this organ should be removed at the time of surgery.
Presenting symptoms of borderline ovarian tumors
Borderline tumors, as with other ovarian tumors, are difficult to detect clinically until they are advanced in size or stage. In one study, the most common presenting symptoms were abdominal pain, increasing girth or abdominal distention, and abdominal mass. Approximately 23% of patients were asymptomatic.
Etiology and associated factors
The etiology of this disease remains unclear because of the small number of cases and the lack of randomized, controlled studies. Based on molecular studies, some mucinous borderline tumors of the ovary may actually represent metastasis from the appendix.
Although none of these has been shown to be statistically significant, factors reportedly linked with borderline tumors include the following:
Oral contraceptive use
Age at first pregnancy
Age at first delivery
Family history of ovarian cancer
Incidence of borderline ovarian tumors
One woman in 55 (1.8%) develops some form of ovarian cancer in her lifetime. Approximately 90% of these cancers are tumors of epithelial origin. If benign lesions are included, epithelial tumors account for 60% of all ovarian tumors.
In the United States, borderline tumors make up approximately 15% of all epithelial ovarian tumors. The mean age of occurrence is approximately 10 years younger than that of women with frankly malignant ovarian cancer.
In Sweden, the incidence of borderline tumors may be increasing, according to a study. The frequency of these tumors in the population increased from 1 case to 5.3 cases per 100,000 women years from 1960-1964 to 2000-2005. The authors suggested that this change may have been due to a rise in diagnostic activity, as well as by “a lack of protective effect of oral contraception use.”
Comprehensive staging of borderline ovarian tumors is of significant prognostic value and is performed surgically. As opposed to its true malignant counterpart, epithelial ovarian carcinoma, borderline ovarian cancer is often found at an early stage.
Without comprehensive surgical staging, the prognosis for an individual patient is difficult to predict.
Borderline ovarian tumors are staged according to the FIGO classification of ovarian cancer. Many clinicians group stages II-IV together for prognostic consideration.
Another common component of staging is the description of the type of implants, as these have significant prognostic value.
Preoperatively, borderline tumors are often presumed to be either benign or malignant ovarian masses; however, as with other ovarian masses, staging is performed surgically. Many sources recommend complete staging if a borderline tumor is found. Current guidelines include biopsy specimens of the pelvic peritoneum (cul-de-sac, pelvic wall, and bladder peritoneum), abdominal peritoneum (paracolic gutters and diaphragmatic surfaces), omentum, intestinal serosa and mesentery, and retroperitoneal lymph nodes (pelvic and para-aortic).
One study found that only 12% of patients were adequately staged at initial operation. Of these patients, 78% were operated on by general obstetrician/gynecologists, 10% by gynecologic oncologists, and 6% by general surgeons. When gynecologic oncologists were asked about surgical staging for borderline tumors, 97% recommended some type of staging procedure, although opinions varied significantly about which samples should be taken.
In a study of stage I disease, all recurrences appeared in patients who were inadequately staged. Many, if not all, of these patients probably did not actually have stage I disease.
Pathologic diagnosis is difficult to confirm by frozen section. Borderline tumors are correctly diagnosed 58-86% of the time by frozen section, depending on the experience of the pathologist and the site of the operation (eg, tertiary care vs community hospital).
However, in one study at a tertiary referral center, benign disease was excluded in 94% of cases subsequently diagnosed as borderline tumor. Thus, the proper operation and staging procedures could have been performed during the initial operation in most cases, even though the diagnosis by frozen section was not completely accurate.
Staging versus laboratory findings
The diagnosis of borderline ovarian cancer is based on surgical staging. From the available data, there is no accurate way to predict the final pathology of ovarian tumors from laboratory or imaging studies alone.
Prognosis in Borderline Ovarian Cancer
Patients with borderline tumors have an excellent overall prognosis. These women have a 60% chance of having stage I disease when diagnosed. Postoperative treatment for any stage is controversial; therefore, recommending reoperation for surgical staging alone is difficult.[2, 3] This being said, adequate staging is essential for determining the prognosis. One study showed that the only recurrences were noted in unstaged stage 1 borderline ovarian cancer.
Approximately 95% of borderline ovarian tumors have diploid deoxyribonucleic acid (DNA). This finding is almost always associated with an excellent prognosis. If the tumor is aneuploid, the recurrence rate is high. Some authors suggest treating these as low-grade invasive carcinoma (also called micropapillary carcinoma).
Conflicting data exist with respect to overexpression of various oncogenes and tumor suppressor genes. Although TP 53 positivity and HER2 overexpression in invasive cancer have been associated with a worse prognosis, the same gene profile has conferred a survival advantage in borderline tumors.
Recurrence and Survival
Patients with stage I disease confirmed by comprehensive staging have a recurrence rate of approximately 15%. The 5-year survival rate for such patients approaches 100%. However, the 10-year survival rate is 90-95%, depending on histologic findings.
In patients with stage II-IV disease, the prognosis is different; an increased stage is associated with a worse prognosis and only age at diagnosis and the presence of invasive implants are shown to influence prognosis.
In reported series, women who had serous tumors with noninvasive peritoneal implants demonstrated a mean 20% recurrence rate and a mean 7% death rate. In patients with recurrence, a median time to diagnosis of 3.1 years was reported if the recurrence was of the borderline type. In patients whose recurrence was invasive carcinoma, the median time to diagnosis was 8.3 years. It is believed that the former was a recurrence but that the latter was probably a new primary tumor. The cancer antigen 125 (CA-125) level was normal in 65% of the recurring cases (see the section Biomarkers and DNA Cytometry). Death was noted only when invasive carcinoma was noted in the recurrence.
In patients with serous borderline tumors with invasive implants, the relapse rate was 31-45%, according to a study by Gershenson et al. The median time from diagnosis to recurrence was 24 months, although the time to progression of disease was significantly longer in patients who had no macroscopic disease remaining at the time of initial operation. Additionally, patients who received postoperative platinum-based chemotherapy had a significantly worse progression-free survival rate. However, the authors of this study believed that this finding might have been due to selection bias.
Gershenson and colleagues' research indicated that the following factors had no effect on progression-free survival:
Type of surgery
Coexistence with noninvasive implants
Number of invasive implants
No statistically significant differences are found in survival between mucinous and serous tumors. Mucinous tumors are most often stage I at time of diagnosis, and it is quite unusual to find extraovarian disease in tumors of mucinous origin.
Given the excellent prognosis of patients with stage I disease and its occurrence in women of reproductive age, fertility-sparing surgery is of great interest.
In patients diagnosed with stage I disease who were treated with fertility-sparing surgery of any type, a higher recurrence was found in patients who had a cystectomy, with or without contralateral oophorectomy (58%), as opposed to patients treated with oophorectomy (23%). However, only half of these patients underwent complete staging. When comprehensive staging was performed, no statistical difference was found in recurrence in confirmed cases of stage I disease. Thus, fertility-sparing surgery is an acceptable option in confirmed stage I disease. This again emphasizes the need for comprehensive staging in all cases.
Of patients who attempted pregnancy after fertility-sparing operations, a 50% conception rate was achieved among 24 patients who were studied. At the endpoint of the study, 16 live births, 4 spontaneous abortions, 3 ectopic pregnancies, and 2 ongoing pregnancies were found. No fetal anomalies were reported.
All authors, nevertheless, indicate that this is an area that needs further research because, in the literature, only 48 patients were found to have conceived after having conservative surgery for borderline tumors.
In another study, which looked at 25 women who underwent fertility-sparing surgery, no recurrences took place in the study period, although the range of follow-up varied widely (4-157 mo). Of the 6 patients who attempted to become pregnant, 5 were successful, resulting in 5 live births, as well as 1 patient who underwent assisted reproductive techniques but had a miscarriage.
A separate review found 254 pregnancies in 206 patients; the investigators calculated the overall pregnancy rate at 48%, based on the published numbers of women trying to conceive.
Biomarkers and DNA Cytometry
Cancer antigen 125
CA-125 levels are not shown to aid in the diagnosis or follow-up care of patients with borderline tumors. However, preoperatively, it can be useful in counseling the patient as to what to expect in the operating room.
Go to Gynecologic Tumor Markers for more complete information on this topic.
Although not routine, static DNA cytometry can be performed on biopsy specimens. As previously stated, about 95% of borderline ovarian tumors have diploid DNA, which is almost always associated with an excellent prognosis.
With the advent of microarray technology, the characterization of the tumor genome can now be studied. Some preliminary work concerning the invasive forms of ovarian cancer is starting to emerge. However, little has been done on borderline ovarian cancer, mostly because of its low incidence and good prognosis.
Some data suggest that invasive tumors discovered at low stages and borderline tumors, especially those with invasive implants, share genetic expression profiles. However, the significance of this has not been determined.
Preoperative transvaginal color Doppler ultrasonography has been used to assess the possibility of malignancy in ovarian masses. The rate of detection of intratumoral blood flow in borderline tumors (90%) is similar to that of malignant neoplasms (92%). The resistance and pulsatility indexes are also significantly reduced in carcinoma and borderline ovarian tumors compared with those of benign tumors.
Although ultrasonography is useful in identifying the mass, this medium is not currently able to predict the final pathology of the tumor. It is neither sensitive nor specific enough to be used as a screening tool in the normal populations.
Computed tomography (CT) scanning should be considered preoperatively to identify possible foci of metastasis. CT scanning can also be useful when following the patient in the future. Again, as with ultrasonography, there are no distinguishing characteristics that clearly identify a borderline ovarian tumor.
In a retrospective study looking at MRI characteristics of known borderline tumors, Bent et al found that serous borderline tumors were significantly smaller than mucinous borderline tumors. However, the investigators were not able to identify any key imaging characteristics that would distinguish borderline tumors from other ovarian tumors.
Histology and Cytology
According to Dietel and Hauptmann, the histology of borderline tumors is characterized by the following features:
Epithelial multilayering of more than 4 cell layers
Not more than 4 mitoses per 10 high-power field (HPF)
Mild nuclear atypia
Increase in nuclear/cytoplasmic ratio
Slight to complex branching of epithelial papillae and pseudopapillae
Epithelial budding and cell detachment into the lumen
No destructive stromal invasion - A major component in differentiating malignant from borderline tumors
Mucinous tumors look grossly similar to their benign counterparts, having large, multilocular cysts with smooth surfaces. The epithelial layer is characterized by stratification of 2-3 layers. Nuclear atypia, enlarged nuclei, and mitotic figures are observed.
Approximately 25% of borderline tumors have cell proliferations on the outer surface of the lesion, with no evidence of growth from the inner surface. Of these, approximately 90% develop peritoneal implants. Only 4% of cases with peritoneal implants do not have surface proliferation.
Peritoneal implants are described as invasive or noninvasive. Noninvasive implants are glandular or papillary proliferations with cell detachments. Psammoma bodies, cellular atypia, and desmoplastic fibrosis are observed in some instances. The appearance of invasive implants is similar, but they have epithelial cells infiltrating the stroma.
The accepted initial treatment of borderline ovarian tumors is surgical removal of the tumor and the performance of biopsies. However, the postoperative management protocol is far from clear. To date, no medical therapy has been shown to clearly improve outcomes.
No consensus has been reached concerning treatment of patients with stage II-IV disease. Although these women still have high 5-year survival rates compared with their counterparts with true malignant ovarian cancer, an increased stage is associated with a worse prognosis. However, stage (II vs III vs IV), type of surgery, postoperative treatment, postoperative platinum-based chemotherapy, and even the number of noninvasive implants have no effect on progression-free survival. Only age at diagnosis and the presence of invasive implants are shown to influence prognosis.
Various chemotherapy regimens have been used, but evidence is insufficient to determine exactly which therapy is indicated for borderline ovarian tumors. Many authors have used platinum-based agents, but with varying results. Some authors recommend platinum-based therapy for patients with invasive peritoneal implants because of their worse prognosis. Standard chemotherapy regimens for invasive ovarian cancer are used if any medical therapy is given. In patients without metastatic disease, chemotherapy is not indicated.
An important area of research is postoperative chemotherapy. Little advantage has been reported after postoperative chemotherapy, but the number of patients studied has been small and the chemotherapeutic regimens used have been varied. The general consensus is that borderline tumors with noninvasive implants do not require any further therapy and should be observed. However, the benefit of treating tumors with invasive implants has been discussed. To date, no randomized data show a benefit.
Given the excellent prognosis for borderline ovarian tumors, hysterectomy and contralateral oophorectomy are not necessary (if the ovary appears normal) if the patient wishes to preserve fertility. If the patient is beyond childbearing age, then hysterectomy is a reasonable option. Removal of a normal, contralateral ovary should be based on existing data regarding ovarian physiology.
Results released in 2015 from a nationwide Danish register-based case-control study between 1982 and 2011 indicate that tubal ligation is effective in reducing the risk of epithelial ovarian cancer, especially endometrioid cancer. The study included 13,241 women with epithelial ovarian cancer and 3,605 women with borderline ovarian tumor; the tumors were stratified on the basis of their histology.
Tubal ligation reduced the overall epithelial ovarian cancer risk by 13%; bilateral salpingectomy reduced epithelial ovarian cancer risk by 42%. The investigators did not find an association between tubal ligation and risk of borderline ovarian tumors.
When a complex ovarian mass is discovered, surgery is often, if not always, indicated. Complete excision of the disease must be achieved if at all possible. Comprehensive staging should be a part of every operation. Although stage may or may not affect future treatment, it is of significant prognostic value and therefore is of value to the clinician and to the patient.
In one study, 77% of patients with invasive peritoneal implants also had noninvasive implants. Comprehensive debulking and staging decreases the chance of a sampling error that could result in an inaccurate diagnosis and prognosis.
In most instances, surgery is curative for patients with confirmed stage I disease. If the tumor is unilateral and adjacent to normal tissue, unilateral cystectomy can be performed; however, inspection of the capsule for signs of rupture should be performed before resection. If no normal adjacent tissue is present, oophorectomy or salpingo-oophorectomy should be performed. If the contralateral ovary is normal in appearance, a biopsy should not be performed on the adjacent ovary because of the risk of ovarian failure (if fertility is an issue).
Owing to the high association between surface proliferations and peritoneal implants, exploration of the peritoneum should be extensive and thorough. If possible, carefully evaluate and remove the implants. The type of implant (ie, invasive, noninvasive) should be noted by pathology, as it has significant prognostic value.
Contraindications to surgery include medical reasons (ie, the patient is too great a surgical risk secondary to other medical problems) or patient refusal. Otherwise, the masses should be surgically removed.
Complications of Borderline Ovarian Cancer
Most of the complications of borderline ovarian cancer are caused by the operation itself, subsequent therapy, or recurrence.
In one series of 28 borderline tumor–related deaths, 2 patients died of radiation-associated complications, 9 of chemotherapy-associated complications, 8 of bowel obstruction, and 8 of invasive carcinoma. This led the authors to suggest that most patients with borderline tumors died with the disease rather than from the disease.
Skírnisdóttir I, Garmo H, Wilander E, Holmberg L. Borderline ovarian tumors in Sweden 1960-2005: trends in incidence and age at diagnosis compared to ovarian cancer. Int J Cancer. 2008 Oct 15. 123(8):1897-901. [Medline].
Lin PS, Gershenson DM, Bevers MW, Lucas KR, Burke TW, Silva EG. The current status of surgical staging of ovarian serous borderline tumors. Cancer. 1999 Feb 15. 85(4):905-11. [Medline].
Cadron I, Leunen K, Van Gorp T, Amant F, Neven P, Vergote I. Management of borderline ovarian neoplasms. J Clin Oncol. 2007 Jul 10. 25(20):2928-37. [Medline].
Madsen C, Baandrup L, Dehlendorff C, Kjaer SK. Tubal ligation and salpingectomy and the risk of epithelial ovarian cancer and borderline ovarian tumors: a nationwide case-control study. Acta Obstet Gynecol Scand. 2015 Jan. 94(1):86-94. [Medline].
Bent CL, Sahdev A, Rockall AG, Singh N, Sohaib SA, Reznek RH. MRI appearances of borderline ovarian tumours. Clin Radiol. 2009 Apr. 64(4):430-8. [Medline].
Gershenson DM, Silva EG, Levy L, Burke TW, Wolf JK, Tornos C. Ovarian serous borderline tumors with invasive peritoneal implants. Cancer. 1998 Mar 15. 82(6):1096-103. [Medline].
Moore RG, Hawkins DM, Miller MC, et al. Combining clinical assessment and the Risk of Ovarian Malignancy Algorithm for the prediction of ovarian cancer. Gynecol Oncol. 2014 Dec. 135(3):547-51. [Medline].
Swanton A, Bankhead CR, Kehoe S. Pregnancy rates after conservative treatment for borderline ovarian tumours: a systematic review. Eur J Obstet Gynecol Reprod Biol. 2007 Nov. 135(1):3-7. [Medline].