2202

Prognostic Factors for High-Risk Early-Stage

Epithelial Ovarian Cancer
A Gynecologic Oncology Group Study

John K. Chan, MD1

Chunqiao Tian, MS2
Bradley J. Monk, MD3
Thomas Herzog, MD4
Daniel S. Kapp, MD, PhD5
Jeffrey Bell, MD6
Robert C. Young, MD7

BACKGROUND. The purpose was to identify the factors predictive of recurrence

and survival in patients with high-risk (stage I, grade 3; stage IC, stage II, or clear
cell) epithelial ovarian cancer after adjuvant therapy.
METHODS. Data was extracted from patients who underwent primary surgery followed by adjuvant therapy in 2 randomized trials by the Gynecologic Oncology
Group (Protocols 95 and 157). Kaplan-Meier survival estimates and Cox proportional hazards model adjusted for covariates were used for analyses.

RESULTS. Of 506 patients (median age 5 56.2 years), 347 (68.6%) had stage I and
1

Department of Obstetrics, Gynecology, and

Reproductive Sciences, University of California,
San Francisco School of Medicine, UCSF Helen
Diller Family Comprehensive Cancer Center, San
Francisco, California.
2

GOG Statistical & Data Center, Roswell Park

Cancer Institute, Buffalo, New York.
3

Department of Obstetrics and Gynecology, Chao

Family Comprehensive Cancer Center, University
of California, Irvine Medical Center, Orange,
California.
4

Department of Obstetrics and Gynecology, Columbia University, New York, New York.
5

Department of Radiation Oncology, Stanford

University School of Medicine, Stanford Cancer
Center, Stanford, California.
6

Department of Obstetrics and Gynecology, Ohio

State University, Riverside Methodist Hospital,
Columbus, Ohio.
7
Department of Medical Oncology, Fox Chase
Cancer Center, Philadelphia, Pennsylvania.

This study was supported by National Cancer

Institute grants to the Gynecologic Oncology
Group Administrative Office (CA27469), the
Gynecologic Oncology Group Statistical and Data
Center (CA37517), and Gynecologic Oncology
Group new investigator award to JKC.
The following Gynecologic Oncology Group
member institutions participated in this study:
University of Alabama at Birmingham, Oregon
Health Sciences University, Duke University
Medical Center, Abington Memorial Hospital,
University of Rochester Medical Center, Walter
Reed Army Medical Center, Wayne State
University, University of Minnesota Medical

2008 American Cancer Society

159 (31.4%) had stage II cancers. The 5-year recurrence-free (RFS) and overall survivals (OS) were 75.5% and 81.7%, respectively. On multivariate analysis, older age,
higher stage, higher grade, and malignant cytology were independent prognostic
factors predictive for recurrence and poorer survival. The risk of recurrence was
higher for those !60 versus < 60 years (hazards ratio [HR] 5 1.57, 95% confidence
interval [CI], 1.122.19), stage II (stage II: HR 5 2.70, 95% CI, 1.415.16) versus
stage IA or IB, grade 2 (HR 5 1.84, 95% CI, 1.043.27) and grade 3 (HR 5 2.47, 95%
CI, 1.394.37) versus grade 1, and positive versus negative cytology (HR 5 1.72,
95% CI, 1.212.45). By using these factors in a prognostic index, those with lowrisk (no or 1 risk factor), intermediate-risk (2 factors), and high-risk (34 risk factors) disease had survivals of 88%, 82%, and 75%, respectively (P < .05).
CONCLUSIONS. Age, stage, grade, and cytology are important prognostic factors in highrisk early-stage epithelial ovarian cancer. This information may be used in the design
of future clinical trials. Cancer 2008;112:220210. ! 2008 American Cancer Society.

KEYWORDS: ovarian cancer, early-stage, prognosis, survival.

School, University of Southern California at Los
Angeles, University of Mississippi Medical Center,
Colorado Gynecologic Oncology Group P.C.,
University of California at Los Angeles, University
of Pennsylvania Cancer Center, University of
Miami School of Medicine, Milton S. Hershey
Medical Center, Georgetown University Hospital,
University of Cincinnati, University of North
Carolina School of Medicine, University of Iowa
Hospitals and Clinics, University of Texas
Southwestern Medical Center at Dallas, Indiana
University School of Medicine, Wake Forest
University School of Medicine, Albany Medical
College, University of California Medical Center at
Irvine, Tufts-New England Medical Center, RushPresbyterian-St. Lukes Medical Center, SUNY
Downstate Medical Center, University of
Kentucky, Eastern Virginia Medical School, The
Cleveland Clinic Foundation, Johns Hopkins
Oncology Center, State University of New York at
Stony Brook, Eastern Pennsylvania GYN/ONC

DOI 10.1002/cncr.23390
Published online 17 March 2008 in Wiley InterScience (www.interscience.wiley.com).

Center, P.C., Southwestern Oncology Group,

Washington University School of Medicine,
Cooper Hospital/University Medical Center,
Columbus Cancer Council, North Central
Cancer Treatment Group, University of Massachusetts Medical School, Fox Chase Cancer
Center, Medical University of South Carolina,
Womens Cancer Center, University of Oklahoma,
University of Chicago, and Tacoma General
Hospital.
Address for reprints: John K. Chan, MD, Department
of Obstetrics, Gynecology, and Reproductive Sciences,
University of California, San Francisco School of
Medicine, UCSF Helen Diller Family Comprehensive
Cancer Center, 1600 Divisadero St., Box 1702, San
Francisco, CA 94143-1702; Fax: 415-885-3586;
E-mail: chanjohn@ obgyn.ucsf.edu
Received September 25, 2007; revision received
November 9, 2007; accepted November 19, 2007.

Stage I-II Prognostic Factors/Chan et al.

n 2006 there will be an estimated 20,180 new

epithelial ovarian cancers diagnosed in the US,
with approximately one-third having FIGO (International Federation of Obstetrics and Gynecology) stage
I and II disease.1 Although the survival of early-stage
disease is significantly higher than those with advanced cancers, approximately 20% to 30% of these
patients will die of their disease.27
The clinical and pathologic prognostic factors
that have been previously described for patients with
early-stage epithelial ovarian cancers include: age,
stage, tumor rupture, cell type, tumor grade, large
volume ascites, and dense adhesions.816 The limitations of many of these prior studies include the small
sample size, inadequacy of surgical staging, inclusion
of borderline tumors, stage III cancers with minimal
residual disease, lack of central pathology review,
and variation in adjuvant therapies.
Young et al.10 suggested that women with lowrisk cancers, defined as stage IA, IB, grade 1 or 2,
nonclear-cell histologies, do not need further adjuvant therapy. Patients with high-risk early-stage
epithelial ovarian cancer, defined as stage I, grade 3;
stage IC, stage II, and clear-cell cancers, were felt to
require postsurgical adjuvant treatment. Over the
past 20 years, the Gynecologic Oncology Group
(GOG) has conducted 2 large prospective clinical
trials on this population. Because both clinical trials
had the same eligibility criteria for patient entry,
these studies provide a unique opportunity to investigate the prognostic factors for high-risk early-stage
ovarian cancer. The results of this analysis can
potentially allow us to assess factors that are predictive for recurrence and survival in these women.
More important, it can help identify subgroups at
significant risk for recurrence after chemotherapy
treatments who may warrant novel therapies and
more aggressive treatment.

MATERIALS AND METHODS

In all, 506 women diagnosed with high-risk early-stage
epithelial ovarian cancer patients enrolled in 2 prospective randomized clinical trials conducted by
the GOG, protocol 95 (n 5 205) and protocol 157
(n 5 301). High-risk early-stage epithelial ovarian cancer was defined as stage IA or IB (grade 3), stage IC or
II (any grade), and stage I or II clear-cell epithelial
ovarian cancer. Of these, 150 patients with incomplete
staging information were excluded from this analysis.
Patients provided written informed consent consistent
with all federal, state, and local requirements before
enrolling in the protocols. Details regarding eligibility
criteria, treatment, and outcome for each particular
study have been previously published.17,19 On the

2203

basis of the study entry criteria, a complete surgical

staging procedure was required. In summary, all peritoneal surfaces, including the undersurfaces of both
diaphragms, serosa, and mesentery, were to be visually
inspected and palpated for evidence of implants. If
there was no evidence of disease beyond the ovary or
pelvis, biopsies of the cul-de-sac, vesico uterine peritoneum, bilateral pelvic side walls, paracolic gutters, and
undersurface of the diaphragm, and sampling of the
pelvic and para-aortic nodes, were to be performed.
All patients who underwent surgical staging were operated on by gynecologic oncologists mostly from academic institutions. In addition, all tumors underwent
central pathology review by expert gynecologic oncology pathologists.
Baseline performance status before initiating
chemotherapy was defined according to GOG criteria
as: 0 5 normal activity; 1 5 symptomatic, fully ambulatory; 2 5 symptomatic, in bed less than 50% of
the time. The primary endpoints for both studies
were disease recurrence-free survival (RFS) and overall survival (OS). RFS was calculated from the date of
study enrollment to the date of disease recurrence
(confirmed on physical, serologic, or radiologic exam),
or most recent follow-up visit. OS was calculated
from the date of study enrollment to the date of
death regardless of cause or last follow-up.
Kaplan-Meier survival analyses were performed
initially to estimate RFS and OS by each variable, using
a log-rank test to compare the differences in survival
functions. Multivariate analysis was then conducted to
identify the independent prognostic factors as well as
to estimate their effects on RFS and OS adjusted for
covariates. In survival analysis, patients with a GOG
performance status of 1 or 2 were combined because
of comparable associations with prognosis. In addition,
those with suspicious positive washings were considered positive for cytology. Furthermore, stage I patients
were further divided into 2 subgroups (stage IA/IB and
stage IC). We elected to use a categorical variable for
age, defined as < 60 years versus !60 years old based
on preassessment. Multivariate analysis was conducted
using a stepwise Cox proportional hazards model, stratified by type of treatment to control for potential confounding effects of the 2 study protocols and type of
treatment. All statistical tests were 2-tailed with a significance level set at 5%. Statistical analyses were performed using Statistical Analysis Software (SAS) v. 9.1
(SAS Institute, Cary, NC).

RESULTS
Of the 506 patients included in this analysis, the median age at diagnosis was 56 years (range, 2288

2204

CANCER

May 15, 2008 / Volume 112 / Number 10

TABLE 1
Patient and Clinicopathologic Characteristics (N 5 506)

Age, y
< 50
5059
6069
!70
Median [range]
Race
White
Black
Hispanic
Other
Performance status
0
1
2
FIGO stage
IA
IB
IC
IIA
IIB
IIC
Histology
Serous
Endometrioid
Clear cell
Mucinous
Other
Tumor grade*
1
2
3
Not graded, clear cell
Presence of Ascites
Yes
No
Cytology
Positive
Suspicious
Negative
Unknown
Ruptured tumor
Yes
No
Treatmenty
Protocol 95 32P
Protocol 95 CP
Protocol 157 PC (3)
Protocol 157 PC (6)

No. of patients

152
146
123
85
56

30.0
28.6
24.3
16.8
[2588]

451
20
21
14

89.1
4.0
4.2
2.8

264
222
20

52.2
43.9
4.0

69
10
268
43
28
88

13.6
2.0
53.0
8.5
5.5
17.4

108
134
137
50
77

21.3
26.5
27.1
9.9
15.2

95
127
147
137

18.8
25.1
29.1
27.1

153
353

30.2
69.8

125
23
340
12

25.0
4.6
68.0
2.4

219
287

43.3
56.7

98
107
155
146

19.4
21.1
30.6
28.9

* Clear cell not graded.

y 32
P indicates intraperritoneal phosphate; CP, cyclophosphamide 1 cisplatin; PC (3), paclitaxel 1
carboplatin for 3 courses; PC (6), paclitaxel 1 carboplatin for 6 courses.

years) (Table 1). The majority (89.1%) of these

patients were white and the remainder were characterized as Black (4%), Hispanic (4.2%), and Others
(2.8%). The baseline GOG performance status of
these women was 0, 1, and 2 in 52.2%, 43.9%, and
4.0%, respectively. All women underwent primary
surgery based on GOG standards. In all, 347 (68.6%)
patients had stage I disease, with stage IA in 13.6%,
IB in 2.0%, and 1C in 53.0%; 159 (31.4%) had stage II
cancers with stage IIA in 8.5%, IIB in 5.5%, and IIC
in 17.4%. Histologic cell types were distributed as
follows: clear cell (27.1%), endometrioid (26.5%),
serous (21.3%), mucinous (9.9%), and other cell types
(15.2%). Tumor grades were distributed as: grade 1
(18.8%), grade 2 (25.1%), grade 3 (29.1%), and not
graded (for clear cell) (27.1%). Of all patients, 153
(30.2%) were found to have ascites, and 219 (43.3%)
had tumor rupture found on surgery. On final pathology review of all washings and ascites, 148 (29.6%)
were cytologically positive (n 5 125) or suspicious
positive (n 5 23). The presence of ascites during surgery was greater in stage IC (33.6%) and stage II
(32.7%) compared with stage IA/IB (13.9%). The presence of ascites was also associated with positive
cytology: 46.4% of patients with malignant cells cytologically had ascites and 21.8% of patients without
ascites had positive cytology. Furthermore, most
patients with stage IC (61.9%) or mucinous (62.0%)
tumors had tumor rupture at the time of surgery.
Stage of disease was associated with tumor histology
and grade. Patients with mucinous or clear-cell
histologies were more likely to have stage I rather
than stage II disease compared with other histologies
(81.0% for clear cell, 94.0% for mucinous vs 67.7%
for other histologies). Univariate analysis of prognostic factors for RFS and OS are demonstrated in
Table 2.
All of the patients on these 2 trials (GOG 157 and
95) were treated with adjuvant platinum-based
chemotherapy or intraperitoneal radioactive chromic
phosphate. Nineteen percent of patients were treated
with intraperitoneal phosphate (32P), 21% with cyclophosphamide/cisplatin (CP), 31% with carboplatin/
paclitaxel (PC) for 3 courses, and 29% with carboplatin/paclitaxel for 6 courses.
With a median follow-up of 98 months (136
months for protocol 95 and 92 months for protocol
157), 140 recurrences (28%) and 151 (30%) deaths
were observed. The estimates of RFS and OS by
patient characteristics are shown in Table 2. Overall,
5-year RFS and OS were predicted to be 76% and
82%, respectively. Patients with age !60 years, stage
II, tumor grade 2 or 3, with the presence of ascites or
positive cytology had significantly worse RFS. There

Stage I-II Prognostic Factors/Chan et al.

TABLE 2
Recurrence-free Survival (RFS) and Overall Survival (OS) by Patient
Characteristics (N 5 506)
RFS
No. of
patients
Age group, y
< 60
!60
Race
White
Other
GOG performance
0
1 or 2
Stage
IA or IB
IC
II
Histology
Serous
Endometrioid
Clear cell
Mucinous
Others
Tumor grade*
1
2
3
Not graded, clear cell
Ascites
Yes
No
Cytologyy
Positive
Negative
Ruptured tumor
Yes
No
Type of treatment{
32
P
CP
PC(3)
PC(6)

% 5-year
RFS

% 5-year
OS

Disease recurrence

80.1
68.5

.004

84.8
77.1

<.001

451
55

74.9
80.9

.29

82.6
73.7

.96

264
242

76.4
74.2

.47

81.7
81.6

.24

87.1
77.8
65.9

.001

108
134
137
50
77

66.4
75.6
79.6
85.9
73.9

.16

95
127
147
137

85.1
73.4
67.4.
79.6

153
353

85.9
83.7
76.2

Death

HR

95% CI

HR

95% CI

1.0
1.57

1.122.19

.009

1.0
1.96

1.412.71

<.001

1.0
1.74
2.70

0.913.33
1.415.16

.003

1.0
1.54
2.36

0.852.79
1.304.27

.005

1.0
1.84
2.47
1.66

1.043.27
1.394.37
0.913.04

.02

1.0
1.23
1.86
1.46

0.722.09
1.103.15
0.852.50

.09

1.0
1.72

1.212.45

.003

1.0
1.53

1.092.16

.02

302
204

79
268
159

TABLE 3
Multivariate Analysis of Prognostic Factors for Recurrence-free
Survival (RFS) and Overall Survival (OS) (N 5 506)

OS

2205

.009

79.3
83.7
81.5
83.7
80.5

.47

.01

85.9
81.6
79.2
81.5

.10

71.4
77.3

.05

81.3
81.9

.58

148
358

67.0
78.9

<.001

75.5
84.2

.01

219
287

76.8
74.5

.22

83.1
80.6

.59

98
107
155
146

66.8
77.2
76.3
79.2

.08

77.3
84.0
80.3
84.5

.44

Age, y
< 60
!60
Stage
IA or IB
IC
II
Tumor grade*
1
2
3
Not graded, clear cell
Cytology
Negative
Positive

HR indicates hazard ratio; CI, confidence interval.

* Hazard ratio estimated by Cox model adjusted for age group, stage, tumor grade, and cytology, as
well as stratified with type of treatment.

GOG indicates Gynecologic Oncology Group.

* Clear cell not graded.
y
Suspicious positive (n 5 23) classified as positive and unknown classified as negative.
{ 32
P indicates intraperitoneal phosphate; CP, cyclophosphamide 1 cisplatin; PC(3), paclitaxel 1 carboplatin for 3 cycles; PC(6), paclitaxel 1 carboplatin for 6 cycles. Median survival time was estimated
by Kaplan-Meier procedure and a log-rank test was used to compare survival functions.

is also a suggestion that patients treated by PC and

CP had comparable RFS, but both of them had
improved RFS compared with patients treated by 32P.
Multivariate analysis identified 4 factors (age,
stage, tumor grade, and cytology) independently predictive of disease recurrence (Table 3). The relative
risk of disease recurrence for patients at age !60

years versus age < 60 years was 1.57 (95% confidence interval [CI], 1.122.19). The risk for recurrence was increased with advanced stage of disease
(stage II: hazard ratio [HR] 5 2.70, 95% CI, 1.415.16,
relative to stage IA or IB). In addition, patients with
grade 2 (HR 5 1.84, 95% CI, 1.043.27) and grade 3
(HR 5 2.47, 95% CI, 1.394.37) tumors were at
increased risk for disease recurrence compared with
grade 1 cancers. Women with positive cytology had a
significantly elevated risk for disease recurrence compared with those with negative cytology (HR 5 1.72,
95% CI, 1.212.45). Because the clear-cell patients
were not graded and all stage IA or IB patients
selected had grade 3 tumors, the results on tumor
grade above may not well reflect the effect of tumor
grade on prognosis. Further analysis restricted to
stage IC, stage II, excluding clear-cell patients was
conducted. The 5-year RFS was estimated to be
85.1%, 73.4%, and 62.6% for grade 1, grade 2, and
grade 3, respectively. The adjusted HR was 1.89 (95%
CI, 1.063.35) for grade 2 and 2.55 (95% CI, 1.42
4.59) for grade 3 compared with tumor grade 1, confirming the significance of tumor grade on disease
recurrence. All other variables (race, performance
status, histology, tumor rupture, and ascites) were
not significantly associated with RFS. The results on
OS were similar. These findings were consistent in
the OS analyses.
Figures 1 to 4 demonstrate the Kaplan-Meier
estimate of RFS and OS by age group, stage of dis-

2206

CANCER

May 15, 2008 / Volume 112 / Number 10

FIGURE 1. (A) Kaplan-Meier recurrence-free survival by age group

(P 5 .004). (B) Kaplan-Meier overall survival by age group (P < .001).

ease, tumor grade, and cytology. Among the prognostic parameters, these independent prognostic factors
were used to develop a prognostic index for potential
clinical application. The prognostic model for RFS is
based on the 4 risk factors: age !60 years (vs
age < 60 years), stage II disease (vs stage I disease),
positive cytology (vs negative cytology), and grade 2
3 tumors or clear cell (vs grade 1 disease). Low-risk
patients were defined as those with no or 1 risk factor; intermediate-risk patients as those with any 2
risk factors; and high-risk patients as women with
any 3 or 4 risk factors. The 5-year RFS of the low-,
intermediate-, and high-risk groups was estimated to
be 88%, 71%, and 62%, respectively. On the basis of
the number of risk factors, patients in the low-, intermediate-, and high-risk groups had corresponding
OSs of 88%, 82%, and 75% (P < .05) (Fig. 5A,B).

DISCUSSION
Early-stage ovarian cancer patients constitute a heterogeneous group with respect to risk of recurrence
and survival. Prior reports have shown that patients
with early-stage disease have overall survivals ran-

FIGURE 2. (A) Kaplan-Meier recurrence-free survival by stage (P 5 .001).

(B) Kaplan-Meier overall survival by stage (P 5 .009).
ging from 60% to 100%.1,2,46,20 Thus, stratifying this
heterogeneous group of patients can potentially
identify subgroups of high-risk patients for individualized novel therapies in an attempt to improve
outcome. Likewise, it is important to identify a lowrisk group that may not require further cytotoxic
treatment. In this current analysis of 506 women
diagnosed with high-risk stage I and II epithelial
ovarian cancer treated on 2 GOG prospective randomized trials, we found that older age, higher stage,
higher grade, and positive cytology are important
prognostic factors for recurrence and survival.
Earlier studies on the prognostic significance of
age in ovarian cancer have been inconclusive.
Although most reports have shown that younger
women are diagnosed with lower-stage and more
well-differentiated tumors, and have an improved
outcome compared with older women,2125 others
have found that age is not an independent prognostic factor after adjusting for stage and grade of disease.2628 In addition, because of the low prevalence
of young patients diagnosed with invasive ovarian
cancer, these studies have also been limited by small
numbers of patients, inclusion of low malignant

Stage I-II Prognostic Factors/Chan et al.

2207

FIGURE 3. (A) Kaplan-Meier recurrence-free survival by grade of disease

FIGURE 4. (A) Kaplan-Meier recurrence-free survival by cytology

(P 5 .004). (B) Kaplan-Meier overall survival by grade of disease (P 5 .01).

(P < .001). (B) Kaplan-Meier overall survival by cytology (P 5 .01).

potential tumors, germ cell or sex cord stromal

tumors, and unstaged cancers. In a recent analysis of
28,165 patients, Chan et al.28 identified 400 women
who were < 30 years (very young) (1.4%), 11,601
were 30 to 60 (young) (41.2%), and 16,164 were > 60
(older) (57.4%) years of age. Across all stages, very
young women had a significant survival advantage
over the young and older groups, with 5-year disease-specific survival estimates at 78.7% versus
58.8% and 35.3%, respectively (P < .001). In this current analysis of a well-characterized group of earlystage ovarian cancer patients with long follow-up,
younger age was an independent prognostic factor
for improved survival after controlling for surgery,
stage, grade, adjuvant therapy, and other clinicopathologic factors.
Previous studies have also demonstrated that
stage of disease is a prognostic factor in early-stage
ovarian cancers.13,16,2933 Patients in this current
study with stage I cancers have a 5-year disease-specific survival of 84% compared with 76% in those
with stage II disease. An analysis on the subgroups
of stage I cancers found that those with stage IA or
IB disease have a survival of 85.9%. Given the excel-

lent outcome of these patients and the potential

toxicities associated with adjuvant chemotherapy,17,18
future studies must be carefully structured to determine the risk and benefit of cytotoxic chemotherapy
in low-risk disease. However, the outcome for
patients with stage II is significantly poorer, with RFS
and OS of 65.9 and 76.2%, respectively. Prior studies
have included these patients in clinical trials along
with more advanced (stage III and IV) cancers.31,34
Although the survival of stage II patients is poorer
compared with stage I disease, these women still
have a distinct survival advantage over those with
more advanced cancers.
Tumor grade was also found to be an independent prognostic factor for progression-free and disease-specific survival in our study. Similarly, Vergote
et al.35 studied a group of 1545 patients with stage I
disease and found that grade of disease was an independent prognostic factor associated with diseasefree survival. These findings have also been confirmed by other, smaller studies.11,14,16,30,3644
In this current analysis, malignant cytology was
an independent prognostic factor for increased risk
of recurrence and poorer survival. Early studies have

2208

CANCER

May 15, 2008 / Volume 112 / Number 10

FIGURE 5. (A) Kaplan-Meier recurrence-free survival by number of risk

factors (age !60 years, stage II disease, positive cytology, and grade 2"3

tumors or clear cell). Low-risk: 0"1 risk factors; mid-risk: 2 risk factors;
high-risk: 3"4 risk factors (P < .001). (B) Kaplan-Meier overall survival by
number of risk factors (P < .001).

demonstrated that patients with positive washing have

a poorer prognosis.45,46 Creasman and Rutledge47
reported that 60% of 98 patients with ovarian cancer
who underwent surgery had abnormal peritoneal
cytologic specimens. Likewise, a more recent report
also found positive peritoneal washing cytology at
initial surgery in 90 (80.4%) of 112 patients with
ovarian carcinomas.48 Those authors also showed
that positive cytology portends a poorer prognosis.
The prognostic significance of clear-cell histology
compared with other subtypes of epithelial ovarian
cancers remains controversial. In a recent review of
54 studies on ovarian clear-cell carcinoma, Pectasides et al.49 showed that clear-cell cancers have a
significantly poorer survival compared with other
histologic subtypes of epithelial ovarian cancer. In
patients with more advanced stage cancer, the
response rate to platinum-based chemotherapy and
survival were significantly lower than those with serous tumors. However, others have not been able to

find an association between cell type and prognosis

in early-stage disease. Contributing factors for these
conflicting results may include the lack of central
pathology review and intraobserver variabilities on
determining cell type50,51 and various treatment regimens.49 In this current study of early-stage ovarian
cancer where the majority of patients were uniformly
staged and treated on 2 standardized protocols, we
were unable find a statistically significant survival
difference between clear-cell and other histologies.
Consistent with prior reports, our data did not
reveal that tumor rupture was associated with a
poorer outcome.15,30,39 However, Vergote et al.35
found a deleterious effect of rupture either during or
before surgery on disease-free survival. The interaction between tumor rupture and more early-stage
cancers in our study may have influenced our ability
to determine the true significance of tumor rupture.
One of the shortcomings of our study is that there is
a lack of information regarding the time of rupture,
eg, preoperatively or intraoperatively. Some prior studies have found that preoperative rupture may carry
a worse prognosis compared with intraoperative rupture.14,52 In addition, this current analysis did not
find ascites to be a significant prognostic factor. This
may be explained by the interaction between ascites,
cytology, and stage of disease. For, after adjusting for
these factors, ascites was no longer prognostically
important.
Our study was limited by the potential selection
bias inherent in randomized trials. This study cohort
may comprise a subset of high-risk patients treated
at research centers that may not represent the experience in the general population. Furthermore, given
that these patients were enrolled in these 2 large
trials ranging from 1986 to 1998, there may exist significant differences related to cancer supportive care
and treatment of recurrences over this time period.
Moreover, there was a lack of complete information
regarding the extent of the comprehensive staging
procedures on all patients. In fact, 29.5% of patients
in the GOG 157 trial had incomplete or inadequately
documented surgical staging information.17 It is important to note that the descriptive statistics and
results of this study were based on a selected group
of patients with high-risk early-stage cancers defined
by the eligibility criteria from 2 randomized clinical
trials of the GOG. Thus, in this specific subset of
patients there is a significantly higher proportion
(27.1%) of clear-cell histologies compared with other
cell types. Therefore, these results may not apply to
the overall group of stage I and II patients, particularly those with stage I, grade 1 or 2 disease, and
nonclear-cell type. Although it is important to iden-

Stage I-II Prognostic Factors/Chan et al.

tify a low-risk group that may not require further

cytotoxic treatment, we are unable to define such a
low-risk group of patients from this report because
all of the women in these clinical trials received adjuvant therapy. Thus, a prospective trial designed to
analyze the benefits of adjuvant therapy versus observation is warranted in this low-risk group defined
by this current study.
The strengths of our study include the high
number of patients reported from 2 randomized prospective trials with defined selection criteria and over
80 months of follow-up. Furthermore, these patients
underwent staging by gynecologic oncologists mostly
from academic institutions. In addition, these tumors
underwent central pathology review by expert gynecologic oncology pathologists.
In summary, our findings suggest that age, stage,
grade, and malignant cytology are important prognostic factors in early-stage epithelial ovarian cancer.
This information may be considered in the design of
future clinical trials.

REFERENCES
1.

Jemal A, Siegel R, Ward E, et al. Cancer statistics, 2006. CA

Cancer J Clin. 2006;56:106130.
2. Partridge EE, Phillips JL, Menck HR. The National Cancer
Data Base report on ovarian cancer treatment in United
States hospitals. Cancer. 1996;78:22362246.
3. Hoskins PJ, Swenerton KD, Manji M, et al. Moderate-risk
ovarian cancer (stage I, grade 2; stage II, grade 1 or 2) treated with cisplatin chemotherapy (single agent or combination) and pelvi-abdominal irradiation. Int J Gynecol Cancer.
1994;4:272278.
4. Heintz AP, Odicino F, Maisonneuve P, et al. Carcinoma of
the ovary. Int J Gynaecol Obstet. 2003;83(Suppl 1):135166.
5. Averette HE, Janicek MF, Menck HR. The National Cancer
Data Base report on ovarian cancer. American College of
Surgeons Commission on Cancer and the American Cancer
Society. Cancer. 1995;76:10961103.
6. Kosary CL. FIGO stage, histology, histologic grade, age and
race as prognostic factors in determining survival for cancers of the female gynecological system: an analysis of
197387 SEER cases of cancers of the endometrium, cervix,
ovary, vulva, and vagina. Semin Surg Oncol. 1994;10:3146.
7. Nguyen HN, Averette HE, Hoskins W, et al. National survey
of ovarian carcinoma. VI. Critical assessment of current
International Federation of Gynecology and Obstetrics staging system. Cancer. 1993;72:30073011.
8. Dembo AJ, Davy M, Stenwig AE, et al. Prognostic factors in
patients with stage I epithelial ovarian cancer. Obstet Gynecol. 1990;75:263273.
9. Sevelda P, Vavra N, Schemper M, et al. Prognostic factors
for survival in stage I epithelial ovarian carcinoma. Cancer.
1990;65:23492352.
10. Young RC, Walton LA, Ellenberg SS, et al. Adjuvant therapy
in stage I and stage II epithelial ovarian cancer. Results of
2 prospective randomized trials. N Engl J Med. 1990;322:
1021107.

2209

11. Finn CB, Luesley DM, Buxton EJ, et al. Is stage I epithelial
ovarian cancer overtreated both surgically and systemically? Results of a 5-year cancer registry review. Br J Obstet
Gynaecol. 1992;99:5458.
12. Vergote IB, Kaern J, Abeler VM, et al. Analysis of prognostic
factors in stage I epithelial ovarian carcinoma: importance
of degree of differentiation and deoxyribonucleic acid ploidy
in predicting relapse. Am J Obstet Gynecol. 1993;169:4052.
13. Bertelsen K, Holund B, Andersen JE, et al. Prognostic factors and adjuvant treatment in early epithelial ovarian cancer. Int J Gynecol Cancer. 1993;3:211218.
14. Sjovall K, Nilsson B, Einhorn N. Different types of rupture
of the tumor capsule and the impact on survival in early
ovarian carcinoma. Int J Gynecol Cancer. 1994;4:333336.
15. Ahmed FY, Wiltshaw E, AHern RP, et al. Natural history
and prognosis of untreated stage I epithelial ovarian carcinoma. J Clin Oncol. 1996;14:29682975.
16. Trope C, Kaern J, Hogberg T, et al. Randomized study on
adjuvant chemotherapy in stage I high-risk ovarian cancer
with evaluation of DNA-ploidy as prognostic instrument.
Ann Oncol. 2000;11:281288.
17. Bell J, Brady MF, Young RC, et al. Randomized phase III
trial of 3 versus 6 cycles of adjuvant carboplatin and paclitaxel in early stage epithelial ovarian carcinoma: a Gynecologic Oncology Group study. Gynecol Oncol. 2006;102:432439.
18. Travis LB, Holowaty EJ, Bergfeldt K, et al. Risk of leukemia
after platinum-based chemotherapy for ovarian cancer.
N Engl J Med. 1999;340:351357.
19. Young RC, Brady MF, Nieberg RK, et al. Adjuvant treatment
for early ovarian cancer: a randomized phase III trial of
intraperitoneal 32P or intravenous cyclophosphamide and
cisplatina gynecologic oncology group study. J Clin
Oncol. 2003;21:43504355.
20. Nguyen HN, Averette HE, Hoskins W, et al. National survey
of ovarian carcinoma. Part V. The impact of physicians
specialty on patients survival. Cancer. 1993;72:36633670.
21. Chan JK, Loizzi V, Lin YG, et al. Stages III and IV invasive
epithelial ovarian carcinoma in younger versus older
women: what prognostic factors are important? Obstet
Gynecol. 2003;102:156161.
22. Plaxe SC, Braly PS, Freddo JL, et al. Profiles of women age
3039 and age less than 30 with epithelial ovarian cancer.
Obstet Gynecol. 1993;81:651654.
23. Rodriguez M, Nguyen HN, Averette HE, et al. National survey of ovarian carcinoma XII. Epithelial ovarian malignancies in women less than or equal to 25 years of age.
Cancer. 1994;73:12451250.
24. Smedley H, Sikora K. Age as a prognostic factor in epithelial
ovarian carcinoma. Br J Obstet Gynaecol. 1985;92:839842.
25. Thigpen T, Brady MF, Omura GA, et al. Age as a prognostic
factor in ovarian carcinoma. The Gynecologic Oncology
Group experience. Cancer. 1993;71:606614.
26. Duska LR, Chang YC, Flynn CE, et al. Epithelial ovarian
carcinoma in the reproductive age group. Cancer. 1999;
85:2623269.
27. Massi D, Susini T, Savino L, et al. Epithelial ovarian tumors
in the reproductive age group: age is not an independent
prognostic factor. Cancer. 1996;77:1131116.
28. Chan JK, Urban R, Cheung MK, et al. Ovarian cancer in
younger vs older women: a population-based analysis. Br J
Cancer. 2006;95:13141320.
29. Nagele F, Petru E, Medl M, et al. Preoperative CA 125: an
independent prognostic factor in patients with stage I
epithelial ovarian cancer. Obstet Gynecol. 1995;86:259264.

2210

CANCER

May 15, 2008 / Volume 112 / Number 10

30. Brugghe J, Baak JP, Wiltshaw E, et al. Quantitative prognostic features in FIGO I ovarian cancer patients without postoperative treatment. Gynecol Oncol. 1998;68:4753.
31. Du Bois A, Rochon J, Lamparter C, et al. Pattern of care and
impact of participation in clinical studies on the outcome in
ovarian cancer. Int J Gynecol Cancer. 2005;15:183191.
32. Mizuno M, Kikkawa F, Shibata K, et al. Long-term prognosis of stage I ovarian carcinoma. Prognostic importance of
intraoperative rupture. Oncology. 2003;65:2936.
33. Valverde JJ, Martin M, Garcia-Asenjo JA, et al. Prognostic
value of DNA quantification in early epithelial ovarian carcinoma. Obstet Gynecol. 2001;97:409416.
34. International Collaborative Ovarian Neoplasm Group. Paclitaxel plus carboplatin versus standard chemotherapy
with either single-agent carboplatin or cyclophosphamide,
doxorubicin, and cisplatin in women with ovarian cancer:
the ICON3 randomised trial. Lancet. 2002;360:505515.
35. Vergote I, De Brabanter J, Fyles A, et al. Prognostic importance of degree of differentiation and cyst rupture in stage
I invasive epithelial ovarian carcinoma. Lancet. 2001;357:
176182.
36. Skirnisdottir I, Seidal T, Sorbe B. A new prognostic model comprising p53, EGFR, and tumor grade in early stage epithelial
ovarian carcinoma and avoiding the problem of inaccurate
surgical staging. Int J Gynecol Cancer. 2004;14:259270.
37. Trimbos JB, Vergote I, Bolis G, et al. Impact of adjuvant
chemotherapy and surgical staging in early-stage ovarian
carcinoma: European Organisation for Research and Treatment of Cancer-Adjuvant ChemoTherapy in Ovarian Neoplasm trial. J Natl Cancer Inst. 2003;95:113125.
38. Le T, Adolph A, Krepart GV, et al. The benefits of comprehensive surgical staging in the management of early-stage
epithelial ovarian carcinoma. Gynecol Oncol. 2002;85:351
355.
39. Zanetta G, Rota S, Chiari S, et al. The accuracy of staging:
an important prognostic determinator in stage I ovarian
carcinoma. A multivariate analysis. Ann Oncol. 1998;9:
10971101.
40. Rubin SC, Wong GY, Curtin JP, et al. Platinum-based chemotherapy of high-risk stage I epithelial ovarian cancer after comprehensive surgical staging. Obstet Gynecol. 1993;82: 143147.

41. Villa A, Parazzini F, Acerboni S, et al. Survival and prognostic factors of early ovarian cancer. Br J Cancer. 1998;77:
123124.
42. Bolis G, Colombo N, Pecorelli S, et al. Adjuvant treatment
for early epithelial ovarian cancer: results of 2 randomised
clinical trials comparing cisplatin to no further treatment
or chromic phosphate (32P). G.I.C.O.G.: Gruppo Interregionale Collaborativo in Ginecologia Oncologica. Ann Oncol.
1995;6:887893.
43. Sainz de la Cuesta R, Goff BA, Fuller AF Jr, et al. Prognostic
importance of intraoperative rupture of malignant ovarian
epithelial neoplasms. Obstet Gynecol. 1994;84:17.
44. Paramasivam S, Tripcony L, Crandon A, 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;23:59385942.
45. Keettel WC, Pixley E. Diagnostic value of peritoneal washings. Clin Obstet Gynecol. 1958;1:592606.
46. Morton DG, Moore JG, Chang N. The clinical value of peritoneal lavage for cytologic examination. Am J Obstet Gynecol. 1961;81:11151125.
47. Creasman WT, Rutledge F. The prognostic value of peritoneal cytology in gynecologic malignant disease. Am J
Obstet Gynecol. 1971;110:773781.
48. Zuna RE, Behrens A. Peritoneal washing cytology in gynecologic cancers: long-term follow-up of 355 patients. J Natl
Cancer Inst. 1996;88:980987.
49. Pectasides D, Pectasides E, Psyrri A, et al. Treatment issues
in clear cell carcinoma of the ovary: a different entity?
Oncologist. 2006;11:10891094.
50. Baak JP, Langley FA, Talerman A, et al. Interpathologist
and intrapathologist disagreement in ovarian tumor
grading and typing. Anal Quant Cytol Histol. 1986;8:354
357.
51. Hernandez E, Bhagavan BS, Parmley TH, et al. Interobserver variability in the interpretation of epithelial ovarian
cancer. Gynecol Oncol. 1984;17:117123.
52. Kodama S, Tanaka K, Tokunaga A, et al. Multivariate analysis of prognostic factors in patients with ovarian cancer
stage I and II. Int J Gynaecol Obstet. 1997;56:147153.