A new prognostic model for FIGO stage 1 epithelial ovarian cancer

Gynecologic Oncology 104 (2007) 607 – 611 www.elsevier.com/locate/ygyno

A new prognostic model for FIGO stage 1 epithelial ovarian cancer Andreas Obermair a,⁎, Arlan Fuller b , Elisa Lopez-Varela b , Toon van Gorp c , Ignace Vergote c , Lynne Eaton d , Jeff Fowler d , Michael Quinn e , Ian Hammond f , Donald Marsden g , Anthony Proietto h , Jonathan Carter i , Margaret Davy j , Lee Tripcony a , Nadeem Abu-Rustum k a

d

Queensland Centre for Gynaecological Cancer, Medical School University of Queensland, Royal Brisbane Hospital, Brisbane, Australia b Gillette Center for Women’s Cancer, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA c University Hospitals Leuven, Katholieke Universiteit Leuven, Dept. Gynaecological Oncology, Leuven, Belgium Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, The James Cancer Hospital and Solove Research Institute, The Ohio State University, Columbus, OH, USA e Department of Gynaecological Oncology, Royal Women’s Hospital, Melbourne, Australia f Western Australian Gynaecological Cancer Service, King Edward Memorial Hospital, Subiaco, Western Australia, School of Women’s and Infant’s Health, University of Western Australia, Australia g Gynaecological Cancer Centre, Royal Hospital for Women, Randwick, New South Wales, Australia h Hunter Centre for Gynaecological Cancer, John Hunter Hospital, New South Wales, Australia i Sydney Gynaecological Oncology Group, Royal Prince Alfred Hospital, Missenden Rd, Camperdown, New South Wales, Australia j Department of Gynaecological Oncology, Royal Adelaide Hospital, Adelaide, Australia k Memorial Sloan-Kettering Cancer Center, New York, NY, USA Received 22 May 2006 Available online 7 November 2006

Abstract Background. No consensus exists which patients with surgical stage 1 epithelial ovarian should receive postoperative chemotherapy. The purpose of this study was to evaluate the prognostic impact of preoperative CA-125 and to establish a prognostic index to identify patients in different risk categories. Methods. Data of 600 surgically staged patients with FIGO stage 1 EOC treated in eleven gynecological cancer centers in Australia, the USA and Europe were analyzed. Eligible patients include those with invasive EOC where a preoperative CA-125 was obtained and standard surgical staging performed. Overall survival (OS) was chosen as study endpoint. Preoperative CA-125 values were compared with other prognostic factors, and univariate and multivariate Cox models were calculated. Results. Two hundred and one patients (33.5%) had preoperative CA-125 ≤30 U/ml and CA-125 levels ≤30 U/ml were associated with lower grade, substage 1A and mucinous histologic cell type. Patients with elevated CA-125 levels were more likely to receive chemotherapy. OS probability was 95% and 85% for patients with pretreatment CA-125 ≤ 30 U/ml and > 30 U/ml, respectively (p 0.003). Multivariate analysis confirmed preoperative serum CA-125 > 30 U/ml (OR 2.7) and age at diagnosis >70 years (OR 2.6) as the only independent predictors for overall survival. Conclusion. Pretreatment of CA-125 ≤ 30 U/ml dominates over histologic cell type, substage and grade to identify a subgroup of FIGO stage 1 patients with a genuinely good prognosis with extremely good survival and who could possibly be spared with adjuvant chemotherapy. © 2006 Elsevier Inc. All rights reserved. Keywords: Ovarian cancer; Prognosis; CA-125; Staging

Introduction ⁎ Corresponding author. QCGC-Research Gynaecological Oncology, Ned Hanlon Building, 6th Floor, Royal Women’s Hospital, Australia. Fax: +61 7 3636 5289. E-mail address: [email protected] (A. Obermair). 0090-8258/$ - see front matter © 2006 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2006.09.021

Fifteen to twenty five percent of patients with epithelial ovarian cancer (EOC) present with disease confined to the ovaries [1]. There is no consensus as to whether these patients would benefit from postoperative chemotherapy. Patients are

608

A. Obermair et al. / Gynecologic Oncology 104 (2007) 607–611

usually considered at high risk for recurrence if they have not had comprehensive staging and have moderately or poorly differentiated tumors, adhesions to surrounding structures, stage 1C and/or high risk histology. Adjuvant chemotherapy is often offered to these patients, although the evidence of a survival benefit as compared to observation is sparse [2]. CA-125 is an antigenic determinant of a high molecular weight glycoprotein recognized by the murine monoclonal antibody OC-125 as performed by a routine blood test [3]. It has an established role in monitoring treatment and detecting recurrence of ovarian cancer and has been advocated as a prognostic marker for advanced ovarian cancer [4–6]. Recently, we have reported that CA-125 was prognostic in a series of 518 patients with stage 1 EOC. Five-year survival was 95% compared with 82% for patients with preoperative CA-125 ≤ 30 U/ml or > 30 U/ml, respectively [7]. The prognostic effect was limited to patients who were adequately surgically staged (n = 346). In order to obtain sufficient statistical power, we pooled data from large international gynecological cancer research groups to identify prognostic groupings within a population of patients with surgical stage 1 EOC. The main focus of this work was to identify patients at low and at high risk of relapse. Ideally, adjuvant therapy could be withheld after adequate staging of tumors with a low risk of recurrence.

followed based on clinical, radiological, biochemical and imaging techniques. Disease status was assessed by clinical examination, radiology and laboratory studies.

Patients and methods

Results

Between 1990 and 2002, six hundred patients underwent primary surgery for surgical stage 1 EOC at seven gynecological cancer centers in Australia (Queensland Centre for Gynaecological Cancer, n = 103; The King Edward Memorial Hospital, Perth, n = 73; Royal Women’s Hospital, Melbourne, n = 60; Royal Hospital for Women Gynaecological Cancer Unit, Sydney, n = 45; Hunter Centre for Gynaecological Cancer, n = 40; Royal Prince Alfred Hospital, n = 24; Royal Adelaide Hospital, n = 10), at three centers in the United States (Memorial Sloan-Kettering Cancer Center, New York, n = 94; Massachusetts General Hospital, Boston, n = 92; The Ohio State University, Columbus, n = 24) and in one center in Europe (Catholic University Hospital Leuven, Belgium, n = 35). After Institutional Review Board approval, data for this retrospective analysis were obtained from the respective cancer registry databases and additional information was retrieved from chart reviews. Patients had to have histologically proven EOC, surgical stage 1 disease and the preoperative serum CA-125 level had to be available. We excluded patients with non-epithelial tumors, patients with higher than stage 1 disease, patients in whom CA-125 was not recorded, patients who had not undergone adequate surgical staging and patients with tumors of low malignant potential. Histologic grading was performed as outlined by Day et al. [8], with highly differentiated tumors classified as grade 1, moderately differentiated tumors as grade 2 and poorly or undifferentiated tumors as grade 3. The histological type was classified as defined by the World Health Organization [9]. Histologic slides were reviewed by specialized gynecological pathologists at each institution at the time of diagnosis, but there was no central review process. The CA-125 levels were measured by a commercially available radioimmunoassay [10]. Standard surgical treatment consisted of total abdominal hysterectomy, bilateral salpingo-oophorectomy, omentectomy and pelvic and para-aortic lymphadenectomy. Peritoneal washings, diaphragmatic scrapings and thorough exploration of the peritoneal cavity, including biopsies, were performed as part of the surgical staging process. All patients had to have a standard surgical staging. Postoperative chemotherapy commenced approximately from 5 to 10 days postoperatively and consisted of a combination of cisplatin along with cyclophosphamide until 1998 and thereafter carboplatin as single agent or in combination with paclitaxel. Patients were reviewed every 3 months for the first 2 years, then 6 monthly for up to 5 years and annually thereafter. Patients were

The results of this study are based on the data of 600 patients, with 245 patients from the United States (Memorial SloanKettering Cancer Center, New York; Massachusetts General Hospital, Boston, MA; The Ohio State University, Columbus, OH;) and Europe (Leuven, Belgium) and 355 patients from five states in Australia. From those 600 patients, 201 (33.5%) had preoperative CA-125 ≤ 30 U/ml and 399 patients (66.5%) had CA-125 levels >30 U/ml. Patients with moderately and poorly differentiated tumors, substages IB and IC and endometrioid tumors were more likely to have elevated CA-125 levels, and patients with mucinous and clear-cell tumors were less likely to express CA-125. Patients with elevated CA-125 were more likely to receive chemotherapy. Detailed patients characteristics are given in Table 1. Within a median follow-up of 48 months (95% confidence interval, 45.3 to 50.7 months), 57 patients (9.5%) died. For all patients, the survival probability at 5 years was 88.0%. On univariate analysis, preoperative CA-125 level > 30 U/ml and patient age at diagnosis > 70 years were significantly associated with impaired survival, whereas grade, substage and histologic type failed to be prognostically significant (Table 2). Chemotherapy was associated with an adverse prognostic effect on survival. The 5-year overall survival rate was 85% (95% confidence interval, 81% to 89%) for patients with CA-125 levels >30 U/ml and 95% (95% confidence interval, 92% to 99%) for patients with CA-125 of 30 U/ml or less (p = 0.003) (Fig. 1). Within the group of patients with elevated CA-125, the magnitude of CA-125 expression was not associated with OS rates (data not shown). Of the 57

Statistical analysis The Chi-square test was used to examine patient characteristics for discrete categorical variables or factors. Overall survival was the endpoint for the study with a statistical event defined as death from any cause. Survival time was calculated using date of first diagnosis and date last seen, or date of death. In the absence of an established cut-off value for defining serum CA-125 as a prognostic factor, preoperative serum CA-125 levels greater than 30 U/ml were considered elevated and values less than or equal to 30 U/ml normal [7,11,12]. Actuarial survival plots or time to event curves were constructed using the Product-limit method of Kaplan and Meier with the logrank test applied to detect differences between groups. A generalized linear interactive modeling package (GLIM-4) was used to explore the data. All p-values are the result of 2-sided tests with p-values less than 0.05 considered statistically significant. Univariate and multivariate proportional-hazards models (Cox) were fitted to the data to determine the importance of recognized explanatory variables. Factors that were significant in univariate analysis and factors that showed a trend towards significance were included in the multivariate model. Factors included in the final multivariate model were serum CA-125 (≤30 U/ml vs. >30 U/ml), histological grade (G1, G2, G3), FIGO substage (1a, 1b, 1c), chemotherapy (nil vs. yes) and age at diagnosis (≤40 years, 40 to 49 years, 50 to 69 years, ≥70 years). The addition of histological type to this model did not result in a reduction of deviance and consequently did not contribute to a significant improvement to the model fit. Therefore, the best model isolated the main effects of grade, substage, age at diagnosis, chemotherapy and CA-125.

A. Obermair et al. / Gynecologic Oncology 104 (2007) 607–611

609

Table 1 Patient characteristics Factor

Grade

FIGO substage

Histological cell type

Chemotherapy

Age

G1 G2 G3 Unknown IA IB IC Unknown Serous Mucinous Endometrioid Clear Cell Unknown/other Nil Yes Unknown <40 40–49 50–69 70+

CA-125 ≤ 30 U/ml n = 201 (100%)

CA-125 > 30 U/ml n = 399

56 (40.0%) 45 (27.6%) 41 (26.4%) 59 (41.5%) 116 (39.0%) 16 (29.6%) 69 (27.9%) 0 30 (28.3%) 39 (42.4%) 36 (23.7%) 43 (35.5%) 53 (41.0%) 109 (42.5%) 92 (26.9%) 0 27 (36.5%) 43 (32.5%) 104 (34.3%) 27 (29.6%)

84 (60.0%) 118 (72.4%) 114 (73.6%) 83 (58.5%) 181 (61.0%) 38 (70.4%) 178 (72.1%) 2 (100%) 76 (71.7%) 53 (59.6%) 116 (72.3%) 78 (64.5%) 76 (59.0%) 147 (57.5%) 250 (73.1%) 2 (100%) 47 (63.5%) 89 (67.5%) 199 (65.7%) 64 (70.4%)

p-value

χ2 = 7.7 p = 0.02

χ2 = 7.9 p = 0.02

χ2 = 10.8 p = 0.01 χ2 = 16.1 p = 0.0006

χ2 = 1.04 p = 0.8

deaths among the 600 patients in this study, 8 (3.9%) and 49 (12.3%) were attributed to the group of CA-125 ≤ 30 U/ml (n = 201) and >30 U/ml (n = 399), respectively. Multivariate analysis revealed that CA-125 level >30 U/ml and patient age at diagnosis > 70 years remained the only statistically significant predictors of survival (Table 3). A trend towards impaired survival for patients with stage 1C and patients with moderately or poorly differentiated tumors was noted. In the subgroup of patients with stage 1A disease (n = 297), 116 patients (39.0%) had low and 181 patients (61.0%) had high preoperative CA-125 levels. Fifty-five patients had poorly differentiated (G3) 3 tumors and 106 patients received chemotherapy postoperatively. Within this group of stage 1A patients, OS probability at 5 years was 98% and 87% for patients with low and elevated CA-125, respectively (Fig. 1). Patients with elevated CA-125 had a 3.5-fold increased risk to die from any cause than patients with normal CA-125 (p 0.003). We recorded 3 deaths among the 116 patients with low CA-125 and 19 deaths among 181 patients with elevated CA-125. Survival probabilities at 1, 3 and 5 years for substages 1A, 1B and 1C, grades 1 through 3, CA-125 ≤ 30 U/ml and >30 U/ml Table 2 Univariate Cox models Factor Grade FIGO CA-125 Histology Chemotherapy Age

I vs. II I vs. III IA vs. IB IA vs. IC ≤30 vs. >30 Clear cell vs. Non-clear cell No vs. Yes 50–69 vs. <40 50–69 vs. 40–49 50–69 vs. 70+

Fig. 1. Overall survival of patients with surgical stage 1A epithelial ovarian cancer with preoperative CA-125 ≤30 U/ml versus >30 U/ml.

Hazard

95% CI

p-value

1.67 1.92 1.45 1.53 2.94 0.81 2.24 0.91 1.50 2.28

0.75, 3.71 0.86, 4.28 0.59, 3.56 0.88, 2.67 1.42, 6.07 0.42, 1.59 1.22, 4.10 0.35, 2.36 0.78, 2.86 1.17, 4.41

0.20 0.11 0.40 0.13 0.004 0.60 0.01 0.80 0.23 0.02

for patients’ age at diagnosis and for different histological types are detailed in Table 4. Discussion Preoperative serum CA-125 values rather than FIGO substage, histologic cell type and tumor grade identifies patients with FIGO stage 1 ovarian cancer with very good prognosis. Only a quarter of patients with epithelial ovarian cancer present with disease confined to the ovaries after adequate surgical staging. Overall, patients with surgical stage 1 disease have a survival probability of 81% to 92% [13]. To date, no prospective study has shown a survival benefit with adjuvant chemotherapy in surgical stage 1 epithelial ovarian cancer. Traditionally, chemotherapy has been withheld for patients with surgical stage 1A, well or moderately differentiated tumors but has been advocated for patients with poorly differentiated tumors, stage 1C disease or tumor rupture and for patients with clear-cell histological type [2,14]. The largest series of patients is the result of a joint analysis of two prospective randomized trials, ICON1 and ACTION [15]. A total of 925 patients who had surgery for early-stage ovarian cancer were randomized to receive platinum-based adjuvant chemotherapy or observation until chemotherapy was indicated. Chemotherapy was associated with a hazard ratio of 0.67 (p = 0.008) with an 8% improvement of survival in favor of chemotherapy. However, Table 3 Multivariate Cox models Factor Grade FIGO CA-125 Chemotherapy Age

I vs. II I vs. III I vs. IB IA vs. IC ≤ 30 vs. >30 Nil vs. Yes 50–69 vs. <40 50–69 vs. 40–49 50–69 vs. 70+

Hazard

95% CI

p-value

1.39 1.82 1.36 1.50 2.71 2.17 0.96 1.43 2.60

0.61, 3.17 0.80, 4.14 0.54, 3.40 0.85, 2.67 1.28, 5.76 1.1, 4.3 0.36, 2.57 0.74, 2.75 1.31, 5.16

0.42 0.16 0.50 0.20 0.01 0.03 0.92 0.30 0.007

610

A. Obermair et al. / Gynecologic Oncology 104 (2007) 607–611

Table 4 Overall survival at 1, 3 and 5 years for 456 patients with surgical stage 1 epithelial ovarian cancer Covariates FIGO stage 1a 1b 1c Grade G1 G2 G3 CA-125 ≤ 30 U/ml >30 U/ml Histology Clear cell Non-clear cell

n (%)

OS% 1 year

95% CI

OS% 3 years

95% CI

OS% 5 years

95% CI

223 (49.9) 46 (10.1) 187 (41.0)

99 100 97

97.1, 100 n/a 94.9, 99.6

96 95 92

93.8, 99.0 87.3, 100 87.2, 95.8

92 85 88

87.7, 96.6 72.0, 97.1 82.3, 93.3

140 (30.7) 162 (35.5) 154 (33.8)

99 97 99

97.9, 100 94.1, 99.6 96.7, 100

95 95 93

90.1, 98.9 91.1, 98.3 88.4, 97.4

93 90 87

87.3, 97.6 83.9, 95.1 80.3, 93.6

142 (31.1) 314 (68.9)

100 97

n/a 95.6, 99.2

98 92

95.7, 100 89.2, 95.5

97 86

93.8, 100 81.8, 90.9

77 (19.6) 316 (80.4)

99 99

96.1, 100 97.9, 100

95 96

90.2, 100 93.6, 98.3

90 91

82.1, 98.7 87.4, 95.0

n, number of cases; OS%, overall survival probability; CI, confidence interval.

only one-sixth of those patients were known to have adequate staging. There remains no consensus how to separate patients with surgical stage 1 disease who are at risk of tumor recurrence and death from those who are not. Histologic grade is considered one of the most important prognostic factors in stage 1 epithelial ovarian cancer [16–18]. The survival ranges from 66% to 86% depending on grade [13]. However, its poor reproducibility is one of the major disadvantages [19]. Variable degrees of intra and interobserver variability in grading of ovarian cancer might be the reason why grade proved not prognostic in the present series but also failed to attain statistical significance in a multivariate analysis of stage 1 and stage 2 ovarian cancer in the 2003 FIGO Annual Report on the Results of Treatment in Gynecological Cancer [13]. FIGO substage has also been shown to be prognostic [13,18]. However, its prognostic significance in the literature is inconsistent especially when adjusted for other prognostic factors. Rupture of tumor, capsular penetration and dense adhesions are generally believed to be associated with worse prognosis [17,18]. In many gynecological cancer centers, the presence of adhesions is regarded as stage 2 disease. The presence of dense adhesions was not recorded for the aim of this study and rupture was regarded as stage 1C disease. Histologic cell type has been found to be of even less prognostic significance [20]. Some analyses have suggested that clear-cell carcinoma is more aggressive than non-clear-cell types [21,22]. In contrast, others were unable to confirm survival differences between clear-cell and non-clear-cell types [17,18,23]. In addition, complete surgical staging followed by combination chemotherapy seems to improve survival of patients with stage 1 ovarian clear-cell carcinoma [24]. In our cohort, the 5-year survival probability was similar for clear-cell and non-clearcell cancers (Table 4). Patient age at diagnosis is a broadly accepted prognostic factor for advanced stage epithelial ovarian cancer [13,18,25], and age was also prognostic in the present series of surgical stage 1 ovarian cancer with overall survival as the main endpoint. However, this effect can also be explained by an increased risk of death by natural causes in older age groups.

Recently, we have reported that preoperative CA-125 levels >30 U/ml were associated with impaired survival in stage 1 EOC from seven gynecological cancer centers in Australia [7]. CA-125 was only prognostic in patients who had a complete surgical staging (n = 346) but failed to be prognostic in incompletely staged patients (n = 172). In the group of surgically staged patients, the survival was 82% and 95% in patients with preoperative CA-125 > or ≤ 30 U/ml (p = 0.028). This sample of surgically staged patients was underpowered to analyze subgroups of patients. Therefore, we extracted the Australian subset of surgically staged patients and merged this data set with another 254 patients from the United States and Europe. Within the latter subset, we found a trend towards impaired survival in patients with preoperative CA-125 >30 U/ml (p = 0.10). Hence, we pooled data in order to create a data set of 600 patients with surgical (FIGO) stage 1 EOC. Preoperative serum CA-125 dominates over traditional prognostic factors, such as FIGO substage, histologic cell type and grade, to explain the variance in survival. Interestingly, CA-125 was also a powerful prognostic factor in the subgroup of patients with disease confined to within one ovary (Fig. 1). CA-125 may reflect biological issues and is generally believed to reflect volume of disease, is widely available and its prognostic impact was consistent for the data set from Australia and from the United States/Europe. In contrast, the current FIGO staging classification is rather descriptive than prognostic. The prognostic significance of histologic grade, cell type and substage has been disputed but no alternatives have been available so far. Patients with chemotherapy did worse than patients without chemotherapy. This was true for patients with CA-125 ≤ 30 U/ml and for patients with CA-125 > 30 U/ml (data not shown). Patients with and without chemotherapy were analyzed separately in univariate and multivariate Cox models. While there was a trend towards impaired prognosis in patients with high CA-125 in both groups, this did not reach statistical significance (data not shown). However, the aim of this study was not to determine the value of postoperative chemotherapy in these patients. As with every retrospective analysis, the possibility of selection bias also applies to this study. While

A. Obermair et al. / Gynecologic Oncology 104 (2007) 607–611

every effort is made to adjust for confounding covariates in a multivariate Cox model, patients with worse prognostic features might have been more likely to receive adjuvant chemotherapy. Unfortunately, data on tumor rupture, the presence of adhesions and the cause of death were not consistently recorded among the 11 study centers and therefore cannot be presented. We emphasize that only patients with surgical stage 1 disease have been included into this analysis, and this therefore represents a highly selected group of patients. A number of previous publications have addressed the issue about prognostic factors for stage 1 epithelial ovarian cancer but suffered from lack of adequate surgical staging, inclusion of heterogeneous groups of patients (i.e. borderline tumors or non-epithelial cancers) and lack of a sufficient number of cases making the studies underpowered. For incompletely staged tumors, current evidence suggests that the administration of adjuvant chemotherapy may improve outcomes [15] while in surgically staged tumors, no consensus exists as to the best approach to adjuvant therapy. Preoperative CA-125 identifies patients with a very good prognosis (i.e. > 95% at 5 years), which very unlikely can be improved further by additional treatment measures, such as chemotherapy. The relative rarity of FIGO stage 1 epithelial ovarian cancer and its generally good outcome make this group of patients suboptimal candidates for the conventional methodology of a randomized clinical trial. A clinical trial comparing adjuvant chemotherapy versus observation in stage 1 ovarian cancer patients with normal preoperative CA-125 score would require enrolment of an enormous number of patients to show a difference in outcome, and such a trial would very likely not be feasible. In contrast, a trial comparing treatment to no treatment in patients with a high CA-125 may be feasible provided ethical issues could be solved. Clinicians might be reluctant to enroll patients with high risk features into a trial with half of the patients receiving no treatment. A prospective and multi-institutional validation of the factors associated with low risk (substage, cell type, histologic grade, CA-125) should substantiate the findings of this retrospective study. References [1] Young RC, Decker DG, Wharton JT, Piver MS, Sindelar WF, Edwards BK, et al. Staging laparatomy in early ovarian cancer. JAMA 1983; 250:3072–6. [2] Ozols RF, Rubin SC, Thomas GM, Robboy SJ. Epithelial ovarian cancer. In: Hoskins WJ, Perez CA, Young RC, editors. Principles and practice of gynecologic oncology. 3rd ed. Philadelphia, PA: Lippincott Willimas & Wilkins; 2000. p. 981–1057. [3] Bast RC, Klug T, John E, Jenison E, Niloff JM, Lazarus H, et al. A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer. N Engl J Med 1983;309:883–7. [4] Rustin GJ, Nelstrop AE, Mclean P, Brady MF, McGuire WP, Hoskins WJ, et al. Defining response of ovarian carcinoma to initial chemotherapy according serum CA 125. J Clin Oncol 1996;14:1545–51. [5] Rustin GJ, Nelstrop AE, Tuxen MK, Lambert HE. Defining progression of ovarian carcinoma during follow-up according to CA 125. A North Thames Ovary Group study. Ann Oncol 1996;7:361–4. [6] Ind T, Iles R, Shepherd JH, Chard T. Serum concentration of cancer antigen 125, placental alkaline phosphatase, cancer associated serum antigen and

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14] [15]

[16]

[17]

[18]

[19] [20]

[21] [22]

[23]

[24]

[25]

611

free beta human chorionic gonadotrophin as prognostic markers for epithelial ovarian cancer. Br J Obstet Gynaecol 1997;104:1024–9. Paramasivam S, Tripcony L, Crandon A, Quinn M, Hammond I, Marsden D, et al. Prognostic importance of preoperative CA-125 in International Federation of Gynecology and Obstetrics Stage I Epithelial Ovarian Cancer: an Australian Multicenter Study. J Clin Oncol 2005; 23:5938–42. Day TG, Gallager HS, Rutledge FN. Epithelial carcinoma of the ovary. The prognostic importance of histologic grade. Natl Cancer Inst Monogr 1975;42:15–21. Serov SF, Skully RE, Sorbin LH. Histological typing of ovarian tumours. In: World Health Organisation, editor. International histological classification of tumours, vol. 17. Geneva: World Health Organisation; 1973. Einhorn N, Sjovall K, Knapp RC, Hall P, Scully RE, Bast Jr RC, et al. Prospective evaluation of serum CA 125 levels for early detection of ovarian cancer. Obstet Gynecol 1992;80:14–8. Jacobs IJ, Skates S, Davies AP, Woolas RP, Jeyerajah A, Weidemann P, et al. Risk of diagnosis of ovarian cancer after raised serum CA 125 concentration. A prospective cohort study. BMJ 1996;313:1355–8. Jacobs I, Davies AP, Bridjes J, Stabile I, Fay T, Lower A, et al. Prevalence screening for ovarian cancer in postmenopausal women by CA 125 measurement and ultrasonography. BMJ 1993;306:1030–4. 25th FIGO Annual Report. The Results of Treatment in Gynecological Cancer [Annual Report on CD_Rom]. Milano, Italy, Reed Elsevier Nederland B.V., 2003. McGuire WP. CA-125 and early ovarian cancer: does this help the clinician or further muddy the water? J Clin Oncol 2005;23:5862–4 [editorial]. International Collaborative Ovarian Neoplasm 1 (ICON1), European Organisation for Research and Treatment of Cancer CollaboratorsAdjuvant ChemoTherapy in Ovarian Neoplasm (EORTC-ACTION). International Collaborative Ovarian Neoplasm Trial 1 and Adjuvant ChemoTherapy In Ovarian Neoplasm Trial: two parallel randomized phase III trials of adjuvant chemotherapy in patients with early-stage ovarian carcinoma. J Natl Cancer Inst 2003;95:105–12. Young RC, Walton LA, Ellenberg SS, Homesley HD, Wilbanks GD, Decker DG, et al. Adjuvant therapy in stage I and II epithelial ovarian cancer. Results of two prospective randomized trials. N Engl J Med 1990; 322:1021–7. Dembo AJ, Davy M, Stenwigg AE, Berle EJ, Bush RS, Kjortstad K. Prognostic factors in patients with stage I epithelial ovarian cancer. Obstet Gynecol 1990;75:263–73. Vergote I, De Brabanter J, Fyles A, Bertelsen K, Einhorn N, Sevelda P, et al. Prognostic importance of degree of differentiation and cyst rupture in stage I invasive epithelial ovarian carcinoma. Lancet 2001;375:176–82. Hernson DE. The histologic grading of neoplasms. Arch Pathol Lab Med 1988;112:1091–6. Russel P. Surface epithelial-stromal tumors of the ovary. In: Kurman RJ, editor. Blaustein’s pathology of the female genital tract. 4th ed. New York, NY: Springer; 1994. p. 705–82. Guthrie D, Davy M, Philips PR. A study of 656 patients with “early” ovarian cancer. Gynecol Oncol 1984;17:363–9. Ludescher C, Weger AR, Lindholm J, Oefner D, Hausmanninger H, Reitsamer R, et al. Prognostic significance of tumour cell morphometry, histopathology, and clinical parameters in advanced ovarian carcinoma. Int J Gynecol Pathol 1990;9:343–51. Skirnisdottir I, Seidal T, Karlsson MG, Sorbe B. Clinical and biological characteristics of clear cell carcinomas of the ovary in FIGO stages I–II. Int J Oncol 2005;26:177–83. Ho CM, Chien TY, Shih BY, Huang SH. Evaluation of complete surgical staging with pelvic and para-aortic lymphadenectomy and paclitaxel plus carboplatin chemotherapy for improvement of survival in stage I ovarian clear cell carcinoma. Gynecol Oncol 2003;88:393–4. Wimberger P, Lehmann N, Kimmig R, Burges A, Meier W, Hoppenau B. Impact of age on outcome in patients with advanced ovarian cancer treated within a prospectively randomized phase III study of the Arbeitsgemeinschaft Gynaekologische Onkologie Ovarian cancer Study Group (AGO-OVAR). Gynecol Oncol 2006;100:300–7.