Winter, William E. III MD; Kucera, Paul R. MD; Rodgers, William MD, PhD; McBroom, John W. MD; Olsen, Cara MS; Maxwell, G. Larry MD
Initially described by Taylor in 1929,1 ovarian tumors of low malignant potential were first officially recognized by the International Federation of Gynecology and Obstetrics (FIGO) in 1971.2 These tumors make up approximately 15% of epithelial ovarian neoplasms, with 3000 cases annually in the United States. Low malignant potential tumors of the ovary are characterized by infrequent recurrences and improved survival. Recurrence rates range from 0–17% in early‐stage tumors of low malignant potential3–8 to 20–45% in more advanced tumors of low malignant potential.3,6,8–11 Overall 5‐year survival approaches 100% for stage I tumors and 80–90% for stage III disease.4,5,8,12,13
Approximately 50–70% of serous and 75–100% of mucinous low malignant potential tumors appear to be confined to the ovary at the time of diagnosis.4–8 Retro‐peritoneal lymph node dissection during surgical staging of low malignant potential tumors results in upstaging in 22% of patients with intraperitoneal disease confined to the ovary and 33% of patients with disease confined to the pelvis intraoperatively.12 Although adjuvant therapy of patients with advanced low malignant potential tumors does not significantly improve outcome,14 many gynecologic oncologists have advocated surgical staging of low malignant potential tumors because of the frequency of microscopic disease resulting in upstaging. Proponents of surgical staging argue that the diagnosis of low malignant potential on frozen section analysis of the tumor may be upgraded to invasive cancer by final pathologic analysis in approximately 6–27% of cases, depending upon the institution.15,16 Several retrospective studies3,5,7,17–19 have found favorable outcomes in patients who had limited and directed cytoreduction of ovarian tumors of low malignant potential. However, there are no reports specifically comparing the outcome of patients who had complete surgical staging with those who were not staged.
The objective of this study was to compare the outcome of patients who had complete surgical staging for ovarian tumors of low malignant potential with those who were not staged during their surgery in order to determine whether the recurrence or survival rate is different in patients who are completely surgically staged.
MATERIALS AND METHODS
Tumor registries and pathology review identified 110 consecutive patients with low malignant potential tumor of the ovary who had primary cytoreduction at Oregon Health Sciences University or Good Samaritan Hospital and Medical Center (Portland, OR) between 1979 and 1997. The Institutional Review Boards of both institutions approved this retrospective study. Seventeen patients were excluded from analysis because operative reports, final pathology reports, or disease status were unavailable. We conducted a retrospective chart review on the remaining 93 patients. Two cohorts of patients were identified—patients who had complete surgical staging (n = 48) and those who were unstaged, who had debulking and incomplete staging (n = 45). Frozen section analysis was performed on 83 of 93 tumors identified as low malignant potential tumors of the ovary on final pathology report. Approximately 95% of the patients in this review had surgery after the 1985 FIGO revision of surgical staging for ovarian cancers. Complete surgical staging was defined as an extrafascial hysterectomy and bilateral salpingo‐oophorectomy (unless surgically absent), bilateral pelvic or para‐aortic lymph node sampling, omental biopsy, peritoneal biopsies, and cytologic washings.
The original permanent slides were available to confirm the correct diagnosis in 58 of 93 cases. All histologic subtypes were included in the analysis. Demographic, intraoperative, and pathologic information and outcome data were recorded and compared between the two groups. During the time of this study, eight other patients were identified through pathology records, in addition to the 93 patients originally analyzed, who had a low malignant potential tumor on frozen section analysis but a diagnosis of invasive carcinoma on final pathology. We reanalyzed the outcomes data including five of these eight additional patients whose records were available. Statistical analysis was performed using SPSS software 10.1 (Chicago, IL). Fisher exact test was used to compare the outcomes between the two cohorts due to low frequency of recurrence and death.
The average age of the patients who had surgical staging and those who were unstaged was 51 and 48 years, respectively. The average duration of follow‐up was 6.5 ± 4.2 years and was similar in the two groups. Mean length of follow‐up, CA‐125 levels, and the average size of the ovarian mass were similar in both groups (Table 1). The types of surgical procedures performed are listed in Table 2. Forty‐five of 48 (94%) patients who were surgically staged had total abdominal hysterectomy and bilateral salpingo‐oophorectomy. Also, three patients had unilateral salpingo‐oophorectomy with complete staging because of prior surgery involving removal of the other reproductive organs. In the unstaged group, approximately 67% had a bilateral salpingo‐oophorectomy, 27% had a unilateral salpingo‐oophorectomy, and 4% had an ovarian cystectomy. When information was available, gynecologic oncologists (46 of 72) were more likely to completely stage their patients compared with general gynecologists (two of 19), even with a general surgeon present (P = .001). In addition, the presence of an adnexal mass with complex radiologic features (41 of 67 cases), such as septations, papillations, or excrescences, versus simple radiologic features (one of 10 cases; P = .003) and a diagnosis of at least a low malignant potential tumor (43 of 63) versus benign (three of 20) on frozen section analysis (P < .001) were both predictive of which patients had surgical staging. Finally, 39 of 66 (59%) tumors diagnosed since 1990 were surgically staged compared with eight of 26 (31%) diagnosed before 1990 (P = 0.024).
Early stage (I or II) disease was diagnosed in 31 of 48 (64%) patients who had surgical staging and 42 of 45 (93%) patients who were unstaged (P = .001). In the staged group, eight of 48 (17%) patients were upstaged on the basis of microscopic disease in the retroperitoneal lymph nodes (n = 3), omentum (n = 2), appendix (n = 2), and uterine serosa (n = 1). In contrast, six of 48 (13%) patients were downstaged on the basis of histologically negative enlarged retroperitoneal lymph nodes (n = 5) and pelvic peritoneal implants (n = 1).
Histologic analysis revealed that 47% of the tumors were serous and 48% were mucinous. There were 52% serous and 46% mucinous tumors in the staged group compared with 40% serous and 51% mucinous tumors in the unstaged group (P = .04). Only one patient had invasive peritoneal implants, and there were no patients with micropapillary architecture.
Adjuvant chemotherapy was given to seven patients with a low malignant potential tumor of the ovary. In the staged cohort, six patients each received 12 courses of melphalan. The other staged patient received six courses of cyclophosphamide and cisplatin. In the unstaged group, only one patient with a stage IIIA tumor was treated. She received six courses of cyclophosphamide, doxorubicin, and cisplatin. Overall 5‐year survival was approximately 93% for all stages. The overall 5‐year survival was 100% and 97% for stage I tumors in the staged and unstaged groups, respectively (P = .35). The time to recurrence was 4.5–6.3 years for the completely staged group versus 0.5–8.0 years for the unstaged group. There were three recurrences in both the staged and unstaged groups. Details of these recurrences and treatment are listed in Table 3. Likewise, there were two deaths from disease in both groups (Table 4). Kaplan‐Meier survival curves were constructed from actuarial tables for overall and progression‐free survival for the staged and unstaged cohorts. The curves (not shown) were essentially identical in both cases.
Intraoperative frozen section analysis was performed in 83 of 93 patients with a diagnosis of ovarian tumor of low malignant potential on final pathology. Of these 83 frozen sections, 21 (25%) were reported as benign cystadenomas, eight (10%) as invasive carcinomas, and 54 (65%) as ovarian low malignant potential tumors. Therefore, the sensitivity of the frozen section analysis for low malignant potential tumors was 65% (95% confidence interval 55%, 75%).
During the time of this study, eight additional patients who had a low malignant potential tumor on frozen section analysis but a diagnosis of invasive carcinoma on final pathology were identified through pathology records. This translated to an 8% incidence of upgrading to invasive carcinoma from a low malignant potential frozen section diagnosis during the period of collection. Complete follow‐up data were available for only five of the eight patients. Four cancer patients were surgically staged, three of whom had stage I cancers (two serous and one mucinous) and one of whom had a stage IIIC mucinous adenocarcinoma. Two of the staged patients received chemotherapy, whereas the other two patients (both stage IA) were observed. Only the patient with stage IIIC mucinous adenocarcinoma had cancer recurrence and died of causes secondary to the tumor. Only one patient with cancer on final pathology was unstaged. She had an apparent stage IC clear cell carcinoma of the ovary and received adjuvant chemotherapy, but the tumor recurred 35 months after her original diagnosis and she died of the disease. The inclusion of these five occult cancers on final pathology in the staged (n = 4) and unstaged (n = 1) groups did not change our results and conclusions regarding the minimal effects of staging on outcome. There was one additional recurrence and death from disease in both the staged and unstaged cohorts, with inclusion of the outcome data from these occult cancer patients.
There is still considerable debate about the surgical treatment of ovarian tumors of low malignant potential. The current practice in many institutions is surgical debulking, rather than complete staging, followed by observation. However, many surgeons still advocate formal surgical staging of patients with low malignant potential ovarian tumors. Investigators have reported that patients with localized intraperitoneal disease can be upstaged in 22–27% of cases based on retroperitoneal involvement found at staging procedures in patients with presumed stage I and II tumors.12,21 In contrast, we found that 17% of patients were upstaged on the basis of surgical staging, with only 6% of these cases (three of 48 staged patients) upstaged because of retroperitoneal involvement. Although the total number of patients in our report is small (n = 48), this series of staged patients is larger than that of either series previously reported.12,21 Because of the low risk of retroperitoneal involvement suggested by our data, we question the utility of lymph node sampling for low malignant potential tumors of the ovary even though morbidity is apparently limited when the procedure is performed by a gynecologic oncologist.22 Our data on intraoperative assessment of disease also indicated that early‐stage disease was significantly more common in patients who did not have complete surgical staging. The high percentage of apparent early‐stage disease in this group might not only reflect unrecognized extraovarian disease but also might result from selection bias by the surgeon, who chose to defer surgical staging in patients with tumors that had more grossly reassuring features.
Because of the indolent nature of borderline ovarian tumors, adjuvant therapy has a limited role in the primary treatment of patients, even patients with advanced low malignant potential ovarian tumors on final pathology.5,23 Chemotherapeutic treatment of patients with low malignant potential tumors reportedly has led to more treatment‐related deaths than the number of patients who have progressed from low malignant potential disease to invasive carcinoma.14 Typically, secondary debulking is performed for recurrence of low malignant potential tumors, and salvage chemotherapy is reserved for recurrences that are suboptimally debulked or have histologic features consistent with invasive cancer. Each of the six patients who had histologic confirmation of low malignant potential recurrence received salvage chemotherapy after secondary debulking.
Although many investigators have acknowledged the lack of data showing a survival benefit associated with surgical staging for patients with low malignant potential ovarian tumors, this diagnostic procedure has been advocated in order to provide the patient and family with more accurate prognostic information regarding recurrence. In a previous analysis of 255 patients with tumors of low malignant potential from MD Anderson Cancer Center, Lin et al15 found that only 27 (12%) patients had staging, defined by the investigators as sampling of the omentum, pelvic and abdominal peritoneum, and retroperitoneal lymph nodes (cytologic washings not included). The authors advocated surgical staging for patients with low malignant potential tumors even though they did not report an increased risk of recurrence in the 88% of patients with tumors that were not staged. Based on the data from our series, the ability of surgical staging to identify microscopic extraovarian low malignant potential disease did not appear to affect recurrence or survival in patients with low malignant potential tumors of the ovary.
Proponents of surgical staging argue that the diagnosis of low malignant potential on frozen section analysis of the tumor can be upgraded to invasive cancer by final pathologic analysis in approximately 6–27% of cases.15,16 To address this issue, we identified eight additional patients with low malignant potential tumors on frozen section, who had invasive cancer on final pathology during the study period. We obtained outcome data on five of the eight cases. The inclusion of these occult cancers and subsequent recurrences in the staged (n = 1) and unstaged (n = 1) groups did not change our statistical results and conclusions regarding the minimal effects of staging on outcome.
In our series, only one case of low malignant potential tumor with invasive implants was observed among the 93 low malignant potential cases. Our conclusions regarding surgical staging might not be applicable to cases involving low malignant potential tumors with invasive implants. This form of low malignant potential disease is associated with an increased likelihood of recurrence, prompting some surgeons to consider surgical staging, adjuvant therapy, or both. Retrospective series by Gershenson et al9,10 have found a median time from diagnosis to progression of 24 months for patients with invasive implants and 7.1 years for those with noninvasive implants. In contrast, the largest reported prospective series of low malignant potential tumors (n = 339) showed that the disease‐free interval might be better than that previously reported.3 Although recurrence occurred in five of 16 patients with invasive implants, all of the patients were without evidence of disease 21, 25, 30, 35, and 81 months after salvage therapy. Persistence of tumor was noted in four of 16 (25%) patients who were not given postoperative chemotherapy compared with five of 12 recurrences in patients who received platinum‐based treatment, illustrating that adjuvant therapy has limited value in the treatment of low malignant potential tumors with invasive implants.3 In institutions where adjuvant therapy is prescribed for patients with invasive implants, formal surgical staging could be considered until data in which the outcomes of staged and unstaged patients are available. Our understanding of the efficacy of adjuvant therapy on outcome in patients with low malignant potential tumors with invasive implants remains poorly understood.
Conservative surgery has been described in several case series of patients with ovarian low malignant potential tumors. Unilateral salpingo‐oophorectomy or cystectomy has been advocated as part of fertility‐sparing surgery for women with low malignant potential ovarian tumors. Retrospective data have revealed recurrence rates of 12–58% for cystectomy18,19 to 23% for unilateral oophorectomy.19 The prospective series reported by Zanetta et al3 likewise showed that, although the risk of recurrence is higher after fertility‐sparing surgery (35 of 189 cases) than after radical surgery (seven of 150 cases), all but one woman with recurrence of low malignant potential after conservative treatment were salvaged. Because future recurrence can be treated effectively in most cases, young women can have surgery designed to conservatively debulk the tumor allowing for future conception. Morris et al19 reported delivery of 16 viable infants among 12 of 24 patients who had conservative surgery for low malignant potential tumors. No fetal abnormalities in the offspring of these women were noted.
We acknowledge that the low incidence of recurrence of low malignant potential tumors makes the evaluation of recurrence and survival among patients with this disease suboptimal. However, because the proportion of patients with recurrence was almost identical in the staged and unstaged groups, the likelihood of a type II error is unlikely. To detect a 20% difference in recurrence rates between the staged and unstaged groups using the log rank test, a sample size of 2020 patients per group would be required to achieve 80% power, assuming a recurrence rate of 10% for early‐stage low malignant potential ovarian tumors. If the study were to analyze only advanced‐stage low malignant potential tumors, approximately 748 patients in each group would be required to detect a 20% difference in the recurrence rate (assumed to be 30% in unstaged patients). These sample sizes make the likelihood of a prospective trial with adequate power unlikely. However, our results suggest that the effects of surgical staging on outcome in patients with low malignant potential tumors are not clinically that staggering.
Physicians should consider the concordance of frozen section analysis and final pathology, at their respective institution, before determining whether a diagnosis based on frozen section analysis of a low malignant potential tumor should prompt surgical staging. In health care settings where frozen section analysis and final pathology are relatively consistent, surgery aimed at optimal debulking or complete excision of the tumor without formal surgical staging is an acceptable option. Alternatively, surgeons may chose to surgically stage patients in institutions where the frozen section false‐negative rate for low malignant potential tumors is unacceptably high or in practice settings where adjuvant therapy is prescribed for low malignant potential tumors based on the presence of invasive implants or advanced‐stage disease.
1. Taylor HC. Malignant and semimalignant tumors of the ovary. Surg Gynecol Obstet 1929;48:204–30.
2. International Federation of Gynecology and Obstetrics. Classification and staging of malignant tumors in the female pelvis. Acta Obstet Gynecol Scand 1971;50:1–7.
3. Zanetta G, Rota S, Chiara S, Bonazzi C, Bratina G, Mangioni C. Behavior of borderline tumors with particular interest to persistence, recurrence, and progression to invasive carcinoma: A prospective study. J Clin Oncol 2001; 19:2658–64.
4. Buttini M, Nicklin JL, Crandon A. Low malignant potential ovarian tumours: A review of 175 consecutive cases. Aust N Z Obstet Gynaecol 1997;37:100–3.
5. Barnhill DR, Kurman RJ, Brady MF, Omura GA, Yorda E, Given FT, et al. Preliminary analysis of the behavior of stage I ovarian serous tumors of low malignant potential: A Gynecologic Oncology Group study. J Clin Oncol 1995; 13:2752–6.
6. Hinke V, Martin MC. Ovarian carcinoma of low malignant potential treated at the Jewish General Hospital, Montreal, between 1973 and 1997. Can J Surg 1999;42:253–7.
7. Darai E, Teboul J, Walker F, Benifla JL, Meneux E, Guglielmina JN, et al. Epithelial ovarian carcinoma of low malignant potential. Eur J Obstet Gynecol Reprod Biol 1996;66:141–5.
8. Martin-Jimenez A, Miralles RM, Sanchez AE, Martinez LG, Mundo EC, Llado LB. Ovarian tumors of low malignant potential (borderline): A retrospective study of 31 cases. Eur J Gynaecol Oncol 1994;15:300–4.
9. Gershenson DM, Silva EG, Tortolero-Luna G, Levenback C, Morris M, Tornos C. Serous borderline tumors of the ovary with noninvasive implants. Cancer 1998;83:2157–63.
10. Gershenson DM, Silvia EG, Levy L, Burke TW, Wolf JK, Tornos C. Ovarian serous borderline tumors with invasive peritoneal implants. Cancer 1998;82:1096–103.
11. Goldstein NS, Ceniza N. Ovarian micropapillary serous borderline tumors: Clinicopathologic features and outcome of seven surgically staged patients. Am J Clin Pathol 2000;114:380–6.
12. Leake JF, Rader JS, Woodruff JD, Rosenshein NB. Retro-peritoneal lymphatic involvement with epithelial ovarian tumors of low malignant potential. Gynecol Oncol 1991; 42:124–30.
13. Kurman RJ, Trimble CL. The behavior of serous tumors of low malignant potential: Are they ever malignant? Int J Gynecol Pathol 1993;12:120–7.
14. Sutton GP, Bundy BN, Omura GA, Yordan EL, Beecham JB, Bonfiglio T. Stage III ovarian tumors of low malignant potential treated with cisplatin combination therapy (a Gynecologic Oncology Group study). Gynecol Oncol 1991;41:230–3.
15. 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;85:905–11.
16. Houck K, Nikrui N, Duska L, Chang Y, Fuller AF, Bell D, et al. Borderline tumors of the ovary: Correlation of frozen and permanent histopathologic diagnosis. Obstet Gynecol 2000;95:839–43.
17. Miller DM, Ehlen TG, Saleh EA. Successful term pregnancy following conservative debulking for a stage IIIA low-malignant-potential tumor of the ovary: A case report. Gynecol Oncol 1997;66:535–8.
18. Lim-Tan SK, Cajigas HE, Scully RE. Ovarian cystectomy for serous borderline tumors: A follow-up study of 35 cases. Obstet Gynecol 1988;72:775–81.
19. Morris RT, Gershenson DM, Silva EG, Follen M, Morris M, Wharton JT. Outcome and reproductive function after conservative surgery for borderline ovarian tumors. Obstet Gynecol 2000;95:541–7.
20. Darai E, Teboul J, Fauconnier A, Scoazec JY, Benifla JL, Madelenat P. Management and outcome of borderline ovarian tumors incidentally discovered at or after laparoscopy. Acta Obstet Gynaecol Scand 1998;77:451–7.
21. Yazigi R, Sandstad J, Munoz AK. Primary staging in ovarian tumors of low malignant potential. Gynecol Oncol 1988;31:402–8.
22. Moore DH, Fowler WC, Walton LA, Droegmueller W. Morbidity of lymph node sampling in cancers of the uterine corpus and cervix. Obstet Gynecol 1989;74:180–4.
23. Barakat RR, Benjamin I, Lewis JL, Saigo PE, Curtin JP, Hoskins WJ. Platinum-based chemotherapy for advanced-stage serous ovarian carcinoma of low malignant potential. Gynecol Oncol 1995;59:390–3.
24. Sykes PH, Quinn MA, Rome RM. Ovarian tumors of low malignant potential: A retrospective study of 234 patients. Int J Gynecol Cancer 1997;7:218–26.
© 2002 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.