Prognostic factors for patients with advanced stage serous borderline tumours of the ovary

P. Morice1,+, S. Camatte1, A. Rey2, D. Atallah1, C. Lhommé3, P. Pautier3, C. Pomel1, J.-F. Coté4, C. Haie-Meder5, P. Duvillard4 and D. Castaigne1

Departments of 1 Surgery, 2 Biostatistics, 3 Medical Oncology, 4 Pathology and 5 Radiotherapy, Institut Gustave Roussy, Villejuif, France

Received 28 August 2002; revised 15 November 2002; accepted 3 December 2002


    Abstract
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Background:

The aim of this study was to determine the prognostic factors for patients with advanced stage, low malignant potential ovarian tumour (LMPOT).

Patients and methods:

A retrospective review of 80 patients with serous LMPOT and peritoneal implants treated at or referred to our institution was carried out.

Results:

Sixty-five patients had non-invasive implants. Fifteen patients had invasive implants. Twenty-nine patients had stage II and 51 patients had stage III disease. Three patients died of evolutive invasive disease and four of complications of treatment. The only prognostic factor of progression to ‘evolutive invasive disease’ is the pathologic subtype of peritoneal implants. The 5-year rates of evolutive invasive disease in patients with non-invasive implants and invasive implants were 2% and 31%, respectively (P <0.002).

Conclusions:

In this series, the only prognostic factor for patients with advanced stage borderline tumour is the type of peritoneal implant. More patients died of the treatment’s complications than of the disease itself. The patients’ prognosis with non-invasive implants seems to be excellent, and conservative management could be discussed in younger patients.

Key words: borderline tumour, invasive implants, non-invasive implants, ovary, peritoneal implants


    Introduction
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Taylor first described low malignant potential ovarian tumours (LMPOT) in 1929 [1]. This ovarian malignancy is defined by an epithelial tumour with a stratification of the epithelial lining, but with a lack of frank stromal invasion at pathologic examination. It has a less aggressive behaviour than invasive epithelial ovarian tumours [1]. The prognosis of patients with a disease limited to the ovary is excellent, but patients with extra-ovarian spread have an uncertain prognosis and evolution. The evolution of patients with advanced stage LMPOT depends on the histological subtype. Thus, the prognosis of patients with mucinous borderline tumour and peritoneal extension (pseudomyxoma peritonei) is different from patients with serous tumour [2], and treatment of this subgroup of patients should be specific. Therefore, data from patients with mucinous tumour and pseudomyxoma peritonei are not studied in the present series. We limited our report to serous LMPOT. Recent series reported that 30% of patients with serous borderline tumours with peritoneal implants had recurrences, most commonly in the form of serous carcinoma [3, 4].

We performed a current review of series including patients with advanced stage serous borderline tumour of the ovary, using a Medline search (PubMed system; terms: borderline tumour, low malignant potential, ovarian tumour, ovarian neoplasms, peritoneal implants; no limits of date or language). In this paper, we report the second largest series of patients with advanced stage LMPOT. This study aims at determining the prognosis of patients with serous LMPOT associated with peritoneal implants as well as proposing an adequate treatment for these patients.


    Patients and methods
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
We studied data from patients treated at, or referred to, the Gustave Roussy Institute, from January 1969 to February 1999, for LMPOT and peritoneal implants. Eighty patients were reviewed. Histopathological review of the ovarian tumour and peritoneal implants was performed by one of the authors (P.Duvillard) (with clinical course of the patient masked). LMPOT was defined as an ovarian tumour with: (i) stratification of the epithelial lining; (ii) formation of microscopic papillary projections; (iii) presence of nuclear atypia and, most importantly; (iv) absence of frank stromal invasion. The staging used was the 1987 FIGO (International Federation of Gynecology and Obstetrics) classification [5]. The stage was determined using macroscopic description during the surgical procedure and by reviewing the pathology records. Stage I was defined as a tumour limited to one or both ovaries. Stages II and III were defined by the presence of an ovarian borderline tumour and pelvic (stage II) or abdomino-pelvic (stage III) peritoneal implants. Peritoneal implants were classified as non-invasive or invasive, according to the criteria previously described by Bell and Scully [6]. A non-invasive implant was defined by a glandular or papillary proliferation, but with no stromal invasion. Peritoneal non-invasive implants could be subdivided into epithelial type (with a predominance of epithelial elements) and desmoplastic type (in which the epithelial elements lay in a predominant inflamed, desmoplastic stroma). On the other hand, an invasive implant was defined by a proliferation in the peritoneum with a stromal invasion [3, 6, 7].

The surgical treatment was either radical (bilateral salpingo-oophorectomy with or without hysterectomy), or conservative (defined as conservation of the uterus and at least a portion of one ovary). Additional surgical procedures were occasionally performed, such as: peritoneal washings, omentectomy, multiple peritoneal biopsies and pelvic and para-aortic lymphadenectomy. Carrying out these surgical procedures depended on the time period, the surgeon’s preference, the availability of pathological information during the operation and stage of the disease.

Until 1985, adjuvant treatment (chemotherapy and/or external radiation therapy) was delivered. The type of adjuvant treatment depended on the characteristics of the patients (persistence of residual tumour at the end of initial surgery, patients’ age and general status), the stage of disease at which the treatment was given and on the date of treatment. External radiation therapy (with or without chemotherapy) was performed up to 1975. Since this time, adjuvant therapy has consisted only of chemotherapy. Until 1985, a systematic second-look laparotomy was performed in patients without signs of recurrent disease at the end of adjuvant therapy.

The follow-up of patients included clinical examination and blood tests (CA 125) every 3 months during the first year following the procedure, then every 6 months for 2 years and then yearly afterwards. An ultrasound examination was performed on patients followed-up since 1985 (whatever the type of surgical treatment: conservative or radical). The time of follow-up was evaluated from the date of the surgical procedure. The patients were considered as lost to follow-up when the duration of follow-up was <24 months following surgical treatment.

For the statistical analysis, two outcomes were the end points: the rate of recurrence (both as a borderline tumour and as evolutive invasive disease) and the number of patients with ‘evolutive invasive disease’. ‘Evolutive invasive disease’ is defined as the occurrence during follow-up of invasive peritoneal implants in patients with initial non-invasive implants or the occurrence of recurrent invasive peritoneal implants needing an iterative treatment (surgery and/or chemotherapy) in patients with initial invasive implants. The following characteristics were studied: stage, type of surgery, persistence of residual tumour, nodal status, characteristics of peritoneal implants (non-invasive or invasive), stromal micro-invasion and treatment by adjuvant therapy (Table 1). Overall survival rates (Table 2) were determined by using the Kaplan–Meier method, rates of evolutive invasive disease (Table 3) were determined by using 1-Kaplan–Meier and 95% confidence intervals were calculated by the Rothman method [8]. Exact confidence intervals were determined for the 0% or 100% survival rates in Tables 2 and 3. The log-rank test was used to compare the curves and to determine the P value. A P value of <0.05 was considered as significant. To evaluate the rates of evolutive invasive disease, patients without evolutive invasive disease were censored (at the time of last follow-up).


View this table:
[in this window]
[in a new window]
 
Table 1. Characteristics of 80 patients with peritoneal implants
 

View this table:
[in this window]
[in a new window]
 
Table 2. Prognostic factors for overall survival in 80 patients
 

View this table:
[in this window]
[in a new window]
 
Table 3. Prognostic factors for evolution to invasive disease in 80 patients
 

    Results
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
The median age of the 80 patients at the time of surgical procedure was 32 years (range 14–72). Preoperative CA 125 levels were available in 27 patients. Twenty-five of them had CA 125 levels >35 U/l with a mean level of 183 ± 229 U/l. Twenty-nine patients had stage II disease and 51 had stage III disease. The median number of implants examined per patient during pathologic analysis was three per patient (range 1–9). Sixty-five patients had non-invasive implants and 15 had invasive implants. The median age of patients with invasive and non-invasive implants was 28 years (range 14–72) and 35 years (range 16–74), respectively. Twenty-seven patients underwent pelvic and/or para-aortic lymphadenectomy. Eight patients had nodal involvement with lymph node histological features similar to those of the primary ovarian serous borderline tumour. Five of these patients also had endosalpingiosis present in the lymph nodes. Three out of eight patients with positive nodes had invasive implants and five had non-invasive implants. Both pelvic and para-aortic lymph nodes were involved in three patients, pelvic lymph nodes in four and para-aortic lymph nodes alone in one. The characteristics of these patients are given in Table 1.

Among the 80 patients, five were considered for the purposes of the analysis as lost to follow-up: one patient died of myocardial infarction after the surgical procedure and four patients were lost to follow-up (at 4 months for two patients and at 24 months for the two others). The median time of follow-up was 94 (range 24–348) months. Fifteen patients out of 80 underwent a conservative treatment (12 patients with non-invasive implants and three with invasive implants ): 13 had unilateral salpingo-oophorectomy (with controlateral cystectomy in five patients); one had unilateral cystectomy and one bilateral cystectomy.

Adjuvant treatment was given in 46 patients (34 with non-invasive implants and 12 with invasive implants). Abdomino-pelvic external radiation therapy was delivered to six patients (treated before 1975) and chemotherapy to 32 patients. In eight patients both treatments were given. The doses of radiation therapy depended on the period of treatment. Doses were higher at the beginning of this study (between 20 and 30 Gy to the abdomen and between 20 and 50 Gy to the pelvic cavity). For patients who underwent chemotherapy, regimens consisted of combined chemotherapy (without a platinum component in six patients treated before 1979). Since that date, all patients received regimens based on platinum (with paclitaxel in one patient). From 1979, the number of cycles varied between four and six. Twelve patients had major complications following treatment: chronic radiation enteritis following external radiation therapy (four patients), chronic radiation cystitis (two patients), retroperitoneal fibrosis with chronic renal failure (one patient), toxic neuropathy (three patients), radiation-induced pelvic rhabdomyosarcoma (one patient) and acute myeloblastic leukaemia after chemotherapy (one patient). Four patients died of these complications and will be discussed later.

Twenty-one patients underwent second-look laparotomy. At the end of initial surgery, three patients had absence of residual disease, 17 had a residual tumour <=2 cm and one a tumour >2 cm. All these patients underwent adjuvant chemotherapy (with radiation therapy in five). In seven of those patients who underwent second-look surgery, a residual tumour was found. All seven patients had residual tumour at the time of initial surgery (<=2 cm in six patients and >2 cm in one patient). None of the patients with absence of macroscopic residual disease at the time of initial surgery had residual disease at the time of second-look surgery.

Fifteen patients had recurrences. These recurrences were not diagnosed at the time of second-look surgery in asymptomatic patients. All these patients had abnormal clinical examination, CA 125 levels or ultrasound examination, and underwent an iterative surgical procedure with histological confirmation of their recurrent disease. Four patients had elevated CA 125 levels with abnormal ultrasound or computed tomography scan at the time of recurrence. None of the patients with recurrence had isolated elevation of CA 125 levels. The median delay for recurrences was 23 months (range 5–84). Nine patients relapsed with the same borderline histology treated previously (with iterative conservative treatment in three patients). Two of them were observed shortly (5 and 8 months) after initial cystectomy (unilateral salpingo-oophorectomy with contralateral cystectomy in one patient and bilateral cystectomy in the other) and so could be considered as real recurrent disease of the initial tumour. In the other seven patients, six ‘relapses’ were observed later and/or on the ovary contralateral to the initial tumour and so could be considered as a new ovarian tumour independent from the primary tumour. Only six patients had peritoneal recurrences with evolutive invasive disease. Among these six evolutive invasive recurrences, four were observed in the 15 patients with invasive implants and two in the 65 patients with non-invasive implants (P <0.002; Table 3). Two patients treated conservatively recurred with invasive peritoneal disease (associated with ovarian recurrence under the form of borderline tumour in one patient). None of the patients treated conservatively recurred with transformation into serous invasive ovarian cancer.

Eight patients died. Three of these patients (two with invasive implants) died of tumour progression at 9, 60 and 156 months following the end of treatment. Four patients died of complications: one of a myocardial infarction immediately after surgery, two (13 and 240 months after the end of treatment) of bowel complications related to external radiation therapy (chronic enteritis) and the fourth one died of acute myeloblastic leukaemia 75 months after the end of treatment of the borderline tumour. The last patient died of pancreatic cancer 205 months after treatment for borderline tumour of the ovary.

In the statistical analysis, the histological characteristic of the peritoneal implants (invasive or non-invasive) was the only prognostic factor for recurrence under the form of evolutive invasive disease (Tables 2 and 3). The actuarial 5-year rates of evolutive invasive disease in patients with non-invasive implants and invasive implants were 2% and 31%, respectively (P <0.002). Other studied factors did not reach statistical significance (Tables 2 and 3). Overall and disease-free survival curves are depicted in Figure 1. Overall survival according to the histological type of implants is shown in Figure 2.



View larger version (17K):
[in this window]
[in a new window]
 
Figure 1. Overall survival and disease-free survival curves in 80 patients.

 


View larger version (16K):
[in this window]
[in a new window]
 
Figure 2. Overall survival curves according to the characteristics of peritoneal implants.

 

    Discussion
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
Since the first description of LMPOT 70 years ago, the prognosis of advanced stage borderline tumour of the ovary is unclear and the treatment remains debated. In fact, one of the problems is that the literature shows few reports focusing on the evolution of these patients. Most studies included small numbers of patients (<30) [915]. Few reports included >30 patients, and the characteristics of peritoneal implants were not studied [2, 16]. Only four studies reported an appreciable number of cases (>50 patients) [3, 4, 1719] with determination of histological characteristics of peritoneal implants. The interpretation of the results also depended on the criteria used to perform the statistical analysis. Some authors used the overall and progression-free survival time [3]; others analysed the survival of patients who died of their tumour, who died of other causes or were still alive [12, 17]; others used the rate of development of invasive carcinoma [18]. In our series, only three patients died of their tumour. A statistical analysis based on tumour-specific mortality is impossible, due to this small number of patients. As a result, our statistical analysis was carried out using the rate of recurrence (including recurrence with borderline histology and evolutive invasive disease) and the rate of evolutive invasive disease.

Several prognostic factors were discussed in the literature, the first one being the stage of the disease. In the series by Bell et al. [17], 4% and 20% of stage II and III patients, respectively, died of tumour. In the series by Manchul et al. [14], the 10-year survival rates were, respectively, 75% and 50% in patients with stage II and III disease. In the series by Leake et al. [16], the rates of recurrence were, respectively, 54% and 17% in patients with stage III and II disease. In Michael and Roth’s study [12], the prognostic significance of disease stage was much greater than other factors. On the other hand, our study, like that of Gershenson et al. [4], shows that the stage of the disease (II versus III) did not appear as a relevant prognostic factor.

Another clinical factor that was examined was the existence of residual tumour. In several series, persistence of residual disease is an independent prognostic factor [4, 17]. In our series though, this factor was important but did not reach statistical significance (P = 0.11). We conclude that the surgical procedure for borderline tumours with peritoneal disease, as in ovarian cancer, should include a resection of all macroscopic peritoneal implants. Optimal surgery is crucial because it allows complete removal of all peritoneal tumours and invasive implants. Thus, we may obtain a correct pathological diagnosis based on the entire tumour tissue.

In our series, the only prognostic factor predicting evolution to invasive disease is the histological characteristics of the implants. Previous studies debated the role of histological subtypes of peritoneal implants (invasive or non-invasive ones) as a prognostic factor [915, 1719]. The most recent studies are those published by Gershenson et al. [3, 4], who reported the largest series of patients with serous borderline tumour and peritoneal implants. This series consisted of 73 patients with non-invasive implants and 39 with invasive implants [3, 4]. Thirty per cent of patients developed evolutive or recurrent tumour [3, 4]. It is interesting to note that in these series [3, 4], the rate of recurrence observed in patients with invasive and non-invasive peritoneal implants was similar. Most of them recurred with evolutive invasive disease. In the 73 patients with non-invasive disease, 22 recurred [4]. In 20 of them, histological analysis of the recurrence was possible. In six patients, borderline recurrent tumour was observed. In 14 patients (20%), invasive serous carcinoma was observed [3]. In our series, only two patients out of 65 with non-invasive implants had evolution to invasive disease. The 5-year rate of progression to invasive disease was 2% (Table 3). Such progression remains rare. On the other hand, our study shows that the rate of evolutive invasive disease is significantly different in patients with invasive peritoneal implants compared to patients with non-invasive implants (31% versus 2%; P <0.002). Other reports suggest a very good survival in patients with peritoneal implants. In the series by Leake et al. the 10-year survival of these patients was 95% [16]. Kennedy and Hart [15], Seidman and Kurman [18] and Zanetta et al. [19] reported similar survival rates.

A new entity of patients with peritoneal implants associated with borderline tumour with micropapillary pattern [micropapillary serous carcinoma (MPSC)] was recently described, in order to identify a subgroup of patients with non-invasive peritoneal implants and poor prognosis [18]. In this study of about 65 cases of advanced stage serous borderline tumour, three groups of patients were identified: patients with non-invasive implants, patients with invasive implants and patients with micropapillary pattern (whatever the type of implant: invasive or non-invasive) [18]. Micropapillary pattern was more commonly associated with invasive than non-invasive implants [18]. The 10-year actuarial survival rates of patients with non-invasive implants, invasive implants or MPSC were 98%, 33% and 71%, respectively [18]. In this initial report, MPSC had higher recurrence and mortality rates than borderline tumour with non-invasive implants [18]. However, recently Eichhorn et al. reported that the evolution of patients with non-invasive implants associated with MPSC is similar to patients with non-invasive implants without MPSC [20]. Therefore, the prognostic value of MPSC in patients with non-invasive implants remains debatable. This histological pattern was not studied in the present study. Stromal micro-invasion was also discussed as a prognostic factor in a series of Buttin et al. [21]. In this study, the rate of recurrence is increased in patients with borderline tumour associated with micro-invasion (23% versus 3.5%; P = 0.023) [21]. Nevertheless, in other series, like in our study, presence of stromal micro-invasion is not an adverse prognostic factor (Tables 2 and 3) [22]. In Seidman and Kurman’s literature review, the overall survival of patients with micro-invasion is 100% [22].

The place of conservative treatment in the surgical management of patients with peritoneal implants is another important issue. Several series have studied the results of conservative treatment in early stages of the disease. The rate of ovarian recurrence following conservative treatment is higher than in patients undergoing a radical treatment [2325]. This rate is even higher following cystectomy [23, 25, 26]. We have found that the type of surgery (conservative or radical) is not a prognostic factor (Tables 2 and 3). Most of the patients who recurred following conservative treatment had recurrence as a borderline tumour. Such recurrence was easily treated with a new conservative surgical approach [25]. Only two patients (one of them with invasive implants) treated conservatively recurred with invasive peritoneal implants. However, none of these patients recurred with an ovarian cancer. Therefore, initial radical surgery would probably not change the evolution of these two patients. Thus, performing conservative surgery did not affect survival. Indeed, these two patients are currently alive (one of them without disease). The 5- and 10-year survival rates of the patients treated conservatively is 100% (Table 2). Based on these good results, in the case of younger patients with a good prognosis (borderline ovarian tumour with non-invasive implants), we perform at our institution conservative surgery (if surgically feasible) and total resection of the peritoneal implants. In such cases, an increased rate of borderline ovarian recurrence is anticipated, but without an increased risk of ovarian carcinoma and without an adverse effect on survival. In patients with invasive implants, we prefer to perform a radical surgery.

The initial treatment of patients with peritoneal implants is mainly surgical and includes removal of all macroscopic disease with appropriate surgical staging (including omentectomy). However, the effects of adjuvant therapy on peritoneal implants of borderline tumours remains controversial. At the beginning of this study, patients were treated similarly to patients with ovarian cancer and so received adjuvant treatment. Recently, some authors reported responses following chemotherapy in patients with advanced stage borderline tumour of the ovary [27, 28]. Similarly to other authors, we observed that adjuvant chemotherapy does not improve survival [24, 17]. In a recent series, Crispens et al. observed few responses to non-surgical therapy [29]. In the present series, 12 major complications were observed, resulting in four deaths. A further four patients died from their tumour. Thus, in the present study, as in that by Kurman and Trimble, more patients died from complications of the adjuvant treatment than from progression of the disease [30]. This underlines the need for treatment to be based on optimal surgery with removal of all macroscopic disease. In patients with non-invasive implants, this surgical procedure should be the exclusive therapy. However, in view of the rate of recurrent invasive disease observed in patients with invasive implants, therapeutic abstention following surgical procedure is difficult in such cases. Adjuvant therapy could be discussed in these cases. In our institution, such therapy (platinum-based chemotherapy) is currently given only to patients with invasive peritoneal implants.

In conclusion, in the present series the most important prognostic factor for patients with advanced stage borderline tumour is the histological subtype of peritoneal implants. The prognosis for patients with non-invasive implants is excellent and conservative management can be discussed in younger patients. The treatment of patients with advanced stage borderline tumour is based on optimal surgery and should include removal of all macroscopic disease. Adjuvant chemotherapy could be discussed in patients with invasive implants.


    Acknowledgements
 
The authors express their thanks to Gilles Charrot for editing the manuscript.


    Footnotes
 
+ Correspondence to: Dr P. Morice, Institut Gustave Roussy, 39 rue Camille Desmoulins, 94805 Villejuif, France. Tel: +33-142-11-44-39; Fax: +33-142-11-52-13; E-mail: morice{at}igr.fr Back


    References
 Top
 Abstract
 Introduction
 Patients and methods
 Results
 Discussion
 References
 
1. Taylor HC. Malignant and semi-malignant tumors of the ovary. Surg Gynecol Obstet 1929; 48: 204–230.

2. Kaern J, Trope CG, Abeler VM. A retrospective study of 370 borderline tumors of the ovary treated at the Norwegian Radium Hospital from 1970 to 1982. A review of clinicopathologic features and treatment modalities. Cancer 1993; 71: 1810–1820.[ISI][Medline]

3. Gershenson D, Silva EL, Tortolero-Luna G et al. Serous borderline tumors of the ovary with non invasive peritoneal implants. Cancer 1998; 83: 2157–2163.[CrossRef][ISI][Medline]

4. Gershenson D, Silva E, Levy L et al. Ovarian serous borderline tumors with invasive peritoneal implants. Cancer 1998; 82: 1096–1103.[CrossRef][ISI][Medline]

5. International Federation of Gynecology and Obstetrics. Changes in definitions of clinical staging for carcinoma of the cervix and ovary. Am J Obstet Gynecol 1987; 156: 263–264.[Medline]

6. Bell DA, Scully RE. Serous borderline tumors of the peritoneum. Am J Surg Pathol 1990; 14: 230–239.[ISI][Medline]

7. Duvillard P. Tumeurs ovariennes à la limite de la malignité. Ann Pathol 1996; 16: 396–405.[ISI][Medline]

8. Rothman RJ. Estimation of confidence limits for the cumulative probability of survival in life table analysis. J Chron Dis 1978; 31: 557–560.[ISI][Medline]

9. McCaughey WT, Kirk ME, Lester W, Dardick I. Peritoneal epithelial lesions associated with proliferative serous tumours of ovary. Histopathology 1984; 8: 195–208.[ISI][Medline]

10. Russell P. Borderline epithelial tumours of the ovary: a conceptual dilemma. Clin Obstet Gynaecol 1984; 11: 259–277.[ISI][Medline]

11. Kliman L, Rome RM, Fortune DW. Low malignant potential tumors of the ovary: a study of 76 cases. Obstet Gynecol 1986; 68: 338–344.[Abstract]

12. Michael H, Roth LM. Invasive and noninvasive implants in ovarian serous tumors of low malignant potential. Cancer 1986; 57: 1240–1247.[ISI][Medline]

13. De Nictolis M, Montironi R, Tommasoni S et al. Serous borderline tumors of the ovary. A clinicopathologic, immunohistochemical, and quantitative study of 44 cases. Cancer 1992; 70: 152–160.[ISI][Medline]

14. Manchul LA, Simm J, Levin W et al. Borderline epithelial ovarian tumors: a review of 81 cases with an assessment of the impact of treatment. Int J Radiat Oncol Biol Phys 1992; 22: 867–874.[ISI][Medline]

15. Kennedy AW, Hart WR. Ovarian papillary serous tumors of low malignant potential (serous borderline tumors). A long term follow-up study, including patients with microinvasion, lymph node metastasis, and transformation to invasive serous carcinoma. Cancer 1996; 78: 278–286.[CrossRef][ISI][Medline]

16. Leake JF, Currie JL, Rosenshein NB, Woodruff JD. Long-term follow-up of serous ovarian tumors of low malignant potential. Gynecol Oncol 1992; 47: 150–158.[CrossRef][ISI][Medline]

17. Bell DA, Weinstock MA, Scully RE. Peritoneal implants of ovarian serous borderline tumors. Histologic features and prognosis. Cancer 1988; 62: 2212–2222.[ISI][Medline]

18. Seidman JD, Kurman RJ. Subclassification of serous borderline tumors of the ovary into benign and malignant types. A clinicopathologic study of 65 advanced stage cases. Am J Surg Pathol 1996; 20: 1331–1345.[CrossRef][ISI][Medline]

19. Zanetta G, Rota S, Chiari S et al. Behavior of borderline tumors with particular interest to persistence, recurrence, and progression to invasive carcinoma: A prospective study. J Clin Oncol 2001; 19: 2656–2664.

20. Eichhorn JH, Bell DA, Young RH, Scully RE. Ovarian serous borderline tumors with micropapillary and cribriform patterns: a study of 40 cases and comparison with 44 cases without these patterns. Am J Surg Pathol 1999; 23: 397–409.[CrossRef][ISI][Medline]

21. Buttin BM, Herzog TJ, Powell MA et al. Epithelial ovarian tumors of low malignant potential: The role of micro-invasion. Obstet Gynecol 2002; 99: 11–17.[Abstract/Free Full Text]

22. Seidman JD, Kurman RJ. Ovarian serous borderline tumors: A critical review of the literature with emphasis on prognostic indicators. Hum Pathol 2000; 31: 539–557.[ISI][Medline]

23. Darai E, Teboul J, Fauconnier A et al. Management and outcome of borderline ovarian tumors incidentally discovered at or after laparoscopy. Acta Obstet Gynecol Scand 1998; 77: 451–457.[CrossRef][ISI][Medline]

24. Tazelaar H, Bostwick D, Ballon S. Conservative treatment of borderline ovarian tumors. Obstet Gynecol 1985; 66: 417–421.[Abstract]

25. Morice P, Camatte S, El Hassan J et al. Clinical outcomes and fertility after conservative treatment of ovarian borderline tumors. Fertil Steril 2001; 75: 92–96.[CrossRef][ISI][Medline]

26. Lim-Tan S, Cajigas H, Scully R. Ovarian cystectomy for serous borderline tumors: a follow-up study of 35 cases. Obstet Gynecol 1988; 72: 775–780.[Abstract]

27. Barakat RR, Benjamin I, Lewis JL Jr et al. Platinum-based chemotherapy for advanced-stage serous ovarian carcinoma of low malignant potential. Gynecol Oncol 1995; 59: 390–393.[CrossRef][ISI][Medline]

28. Eltabbakh GH, Srivastava A, Garafano LL. Surgical response of advanced-stage borderline ovarian tumors to paclitaxel and cisplatin chemotherapy. Gynecol Oncol 2000; 77: 343–344.[CrossRef][ISI][Medline]

29. Crispens M, Bodurka D, Deavers M et al. Response and survival in patients with progressive or recurrent serous ovarian tumors of low malignant potential. Obstet Gynecol 2002; 99: 3–10.[Abstract/Free Full Text]

30. Kurman RJ, Trimble CL. The behavior of serous tumors of low malignant potential: are they ever malignant? Int J Gynecol Pathol 1993; 12: 120–127.[ISI][Medline]