YGYNO-976195; No. of pages: 8; 4C:
Gynecologic Oncology xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Gynecologic Oncology
journal homepage: www.elsevier.com/locate/ygyno
Performance of AGO score for secondary cytoreduction in a high-volume
U.S. center
Jo Marie Tran Janco a,1, Amanika Kumar b, Amy L. Weaver c, Michaela E. McGree c, William A. Cliby b,⁎
a
b
c
Division of Gynecologic Oncology, Scripps Clinic, La Jolla, CA, United States
Division of Gynecologic Surgery, Mayo Clinic, Rochester, MN, United States
Division of Biomedical Statistics, Mayo Clinic, Rochester, MN, United States
H I G H L I G H T S
• AGO score correlates with complete cytoreduction, but negative predictive value is low
• Disease-free interval, number of recurrence sites increase AGO score predictive value
• DFI, recurrence pattern, ECOG PS may predict OS after complete secondary cytoreduction
a r t i c l e
i n f o
Article history:
Received 25 September 2015
Received in revised form 18 January 2016
Accepted 29 January 2016
Available online xxxx
Keywords:
Secondary cytoreduction
Recurrent ovarian cancer
a b s t r a c t
Objectives. Determine the predictive value of the Arbeitsgemeinschaft Gynaekologische Onkologie (AGO)
score, and prognostic factors for survival, in patients undergoing secondary cytoreductive surgery (SCS) for recurrent ovarian cancer in a high-volume U.S. center.
Methods. Medical records of women undergoing SCS between 12/1/1998 and 12/31/2013 were reviewed.
Women with no gross residual disease (RD0) at primary surgery, ECOG performance status (PS) ≤1 at recurrence, and no ascites on CT at recurrence were classified as AGO score positive. Women with incomplete information to determine the AGO score were excluded. Overall survival (OS) and progression-free survival (PFS)
following SCS, respectively, were estimated from multivariable Cox proportional hazards models.
Results. 192 women met inclusion criteria. Median disease-free interval (DFI) was 1.9 years (IQR, 1.0–3.5). Of
the 102 (53.1%) AGO score positive cases, 84.3% (95% CI, 77.3–91.4%) achieved RD0 at SCS. However, 64.4% of
AGO score negative cases also reached RD0. Patients with RD0 after SCS survived longer (median OS,
5.4 years) vs. RD ≤1 cm (2.4 years) vs. RD N1 cm (1.3 years) (p b 0.001). Median PFS was also longer in patients
with RD0 (1.5 years) vs. RD ≤1 cm (0.9 years) vs. RD N1 cm (1.0 years) (p = 0.001). Among those with RD0 at
SCS, AGO score was not associated with survival benefit, however, number of disease sites at recurrence, ECOG PS
at recurrence, and DFI were associated with OS and PFS.
Conclusions. AGO score can identify patients with a high likelihood of complete secondary cytoreduction and
improved survival. However, most AGO score negative cases were also completed resected at SCS. Additional refinement of the score is needed to exclude women from SCS.
© 2016 Elsevier Inc. All rights reserved.
1. Introduction
The Arbeitsgemeinschaft Gynaekologische Onkologie (AGO) score
was introduced as a means of selecting appropriate candidates for secondary cytoreduction in the setting of recurrent ovarian carcinoma.
The AGO score utilizes three factors to categorize patients as “score
⁎ Corresponding author at: Division of Gynecologic Surgery, Mayo Clinic, 200 First St
SW, Rochester, MN 55905, United States.
E-mail address: [email protected] (W.A. Cliby).
1
Present address: Mayo Clinic, Rochester, Minnesota; Scripps Clinic Gynecologic
Oncology, La Jolla, CA, United States.
positive” which is associated with a high likelihood of successful removal of all gross disease during secondary cytoreduction. AGO score
positive patients (defined as having had complete cytoreduction at
their initial surgery for ovarian carcinoma, no ascites at the time of recurrence, and a good performance status at the time of recurrence (defined by Eastern Cooperative Oncology Group (ECOG) performance
status ≤1) were predicted to have a 79% chance of complete secondary
cytoreduction [1]. Subsequently, the same group prospectively validated the score, reporting a 76% rate of complete secondary
cytoreduction for score positive women with a first relapse [2]. While
of value, the clinical utility of the scoring system has been questioned
as patients who fell into the “AGO score negative” group still had
http://dx.doi.org/10.1016/j.ygyno.2016.01.027
0090-8258/© 2016 Elsevier Inc. All rights reserved.
Please cite this article as: J.M.T. Janco, et al., Performance of AGO score for secondary cytoreduction in a high-volume U.S. center, Gynecol Oncol
(2016), http://dx.doi.org/10.1016/j.ygyno.2016.01.027
2
J.M.T. Janco et al. / Gynecologic Oncology xxx (2016) xxx–xxx
reasonable rates of complete secondary cytoreduction in both studies, at
42% and 63%. Other studies have proposed alternate criteria to consider
when selecting patients for secondary cytoreduction. Tian, et al. proposed a scoring system based on initial stage, residual disease after primary cytoreduction, progression-free interval, and performance status,
CA-125, and presence of ascites at recurrence [3]. However, in external
validation studies, the predictive value of scores to determine noncandidacy for surgical cytoreduction continues to be poor. For both
the AGO and Tian models, those who were categorized by the scoring
systems to be unlikely to achieve complete secondary cytoreduction
still had a 56–71% rate of complete secondary cytoreduction [4, 5].
AGO score and similar systems have also been suggested to have
value in predicting survival in the recurrent setting. Harter's group proposed the possibility of the AGO score being used not only as a predictor
of complete secondary cytoreduction but also as an independent prognostic factor for survival - an important factor when considering options
in recurrence [6]. As there is a strong correlation between complete secondary cytoreduction and survival [1, 7], AGO score positive may simply
reflect the survival benefit of complete cytoreduction. Other reports
have also found that prognostic scoring for surgery (such as the Tian
risk score) even after adjusting for secondary surgical outcome, may
be useful for survival prognosis for patients undergoing surgery for recurrent ovarian carcinoma [4].
Complete secondary cytoreduction for recurrent ovarian carcinoma
appears to be advantageous to patient survival in selected patients.
The ideal triage system will accurately identify women likely to benefit
from surgery, but must not be so stringent to exclude a group of women
who are nearly equally likely to benefit. Additionally, it appears that certain characteristics are also predictive of survival outcomes aside from
Table 1
Patient characteristics at the time of the primary cytoreduction surgery and recurrence, by
AGO score.
AGO score
positive
(N = 102)
Characteristic
AGO score
negative
(N = 90)
Age at primary surgery (years), mean (SD)
Result of primary surgery, N (%)
Complete (no residual)
Optimal (1 cm residual or less)
Suboptimal (N1 cm residual)
Initial stage, N (%)
Stage I
Stage II
Stage III
Stage IV
Unknown stage
Time from therapy to first recurrence
(years)b, mean (SD)
ECOG performance status at recurrence, N
(%)
0
1
2
Ascites at recurrence, N (%)
No
Yes
Number of disease sites based on
preoperative imaging, N (%)
Single
Multiple
Quadrants involved, N (%)
Lower only
Upper only
Upper and lower
56.4 (10.0) 55.5 (9.8)
12 (13.3)
64 (71.1)
14 (15.6)
102 (100.0)
0 (0.0)
0 (0.0)
2 (2.2)
2 (2.2)
74 (82.2)
12 (13.3)
0 (0.0)
1.9 (2.2)
21 (20.6)
16 (15.7)
55 (53.9)
8 (7.8)
2 (2.0)
3.1 (2.1)
pa
0.50
b0.001
b0.001
surgical result. We sought to evaluate the previously developed AGO
score in a high volume U.S. referral center. Our overall objective was
to improve upon currently available decision tools to help clinicians
and patients when considering surgery at the time of recurrence of
ovarian cancer. Additionally, in order to improve the tools available for
patient counseling, we also evaluated various factors, including the
AGO score, and their influence on patient survival.
2. Methods
This was a retrospective single institution study, and institutional review board approval was obtained. Patients undergoing secondary
cytoreduction surgery at Mayo Clinic Rochester from 12/1/1998 to 12/
31/2013 were identified and surgical records were reviewed. Patients
were included if they had been diagnosed with recurrent ovarian,
fallopian tube, or primary peritoneal carcinoma, had undergone secondary cytoreduction surgery at our institution, and had information available in the record on the initial surgery and recurrence to calculate the
AGO score.
The following information was abstracted from the medical record:
dates of surgeries (both primary and secondary), stage at initial diagnosis, result of initial cytoreduction surgery, whether primary adjuvant
chemotherapy was received, dates of recurrence, ECOG performance
status and major comorbidities at recurrence, number and location of
disease sites as well as presence of ascites on preoperative imaging (at
the time of recurrence), whether chemotherapy was received prior to
surgery for recurrence, result and procedures performed at secondary
cytoreduction, and postoperative therapy after secondary
cytoreduction. Initial cytoreduction was defined as complete if there
was no gross residual disease, and as optimal if b 1 cm of residual disease
remained. The AGO score was considered to be positive (as previously
reported) based on the following factors: complete resection at first surgery; ECOG performance status of 0 or 1 at the time of the recurrence;
and, absence of ascites at recurrence.
Patient characteristics at the time of the initial surgery and recurrence, along with frequency of surgical procedures performed concomitant with the secondary cytoreduction, were compared between AGO
score positive and AGO score negative patients using the two-sample
t-test for continuous variables and the chi-square or Fisher's exact test
for categorical variables. Positive and negative predictive values of scoring algorithms, for the outcome of complete resection at secondary
Table 2
Surgical procedures performed at secondary cytoreduction, by AGO score.
b0.001
Surgical procedure, N (%)
AGO score
negative
(N = 90)
AGO score
positive
(N = 102)
pa
Laparoscopy
Laparotomy
Hysterectomy
Pelvic lymphadenectomy
Paraaortic lymphadenectomy
Upper abdominal
lymphadenectomy
Liver resection
Diaphragm stripping or resection
Small or large bowel resection
Splenectomy
Omentectomy
Appendectomy
Diverting colostomy
Diverting ileostomy
Pelvic peritoneal resection
Urinary tract resection
USO/BSO
Low anterior
resection/reanastomosis
2 (2.2)
86 (95.6)
1 (1.1)
19 (21.1)
22 (24.4)
6 (6.7)
3 (2.9)
100 (98.0)
0 (0.0)
13 (12.7)
15 (14.7)
5 (4.9)
0.99
0.42
0.47
0.12
0.09
0.60
10 (11.1)
21 (23.3)
24 (26.7)
19 (21.1)
25 (27.8)
7 (7.8)
2 (2.2)
0 (0.0)
29 (32.2)
4 (4.4)
0 (0.0)
8 (8.9)
12 (11.8)
17 (16.7)
32 (31.4)
14 (13.7)
24 (23.5)
3 (2.9)
1 (1.0)
0 (0.0)
21 (20.6)
8 (7.8)
1 (1.0)
14 (13.7)
0.89
0.25
0.47
0.18
0.50
0.19
0.60
–
0.07
0.33
0.99
0.29
0.12
60 (66.7)
29 (32.2)
1 (1.1)
79 (77.5)
23 (22.5)
0 (0.0)
67 (74.4)
23 (25.6)
102 (100.0)
0 (0.0)
b0.001
0.03
27 (30.0)
63 (70.0)
46 (45.1)
56 (54.9)
28 (31.1)
33 (36.7)
29 (32.2)
54 (52.9)
17 (16.7)
31 (30.4)
0.002
Abbreviations: ECOG, Eastern Cooperative Oncology Group; SD, standard deviation.
a
t-Test p values presented for continuous variables and chi-square or Fisher's exact p
values presented for categorical variables.
b
Time from the date of completion of first chemotherapy to date of first recurrence for
those with primary chemotherapy and time from date of surgery to date of first recurrence
for those without primary chemotherapy.
a
Chi-square or Fisher's exact p values.
Please cite this article as: J.M.T. Janco, et al., Performance of AGO score for secondary cytoreduction in a high-volume U.S. center, Gynecol Oncol
(2016), http://dx.doi.org/10.1016/j.ygyno.2016.01.027
J.M.T. Janco et al. / Gynecologic Oncology xxx (2016) xxx–xxx
cytoreduction surgery, were estimated along with their 95% confidence
intervals (CI). Logistic regression models were fit; in addition, recursive
partitioning methods were utilized to identify combination of factors
that would further discriminate or classify patients based on their predicted probability of having a complete secondary cytoreduction. The
predictive performance of each scoring algorithm was measured using
the area under the Receiver Operating Characteristic (ROC) curve
(AUC), which is equivalent to the c-index. The AUC estimates were compared between models using the DeLong, DeLong, and Clarke-Pearson
non-parametric method for comparing the AUC between correlated
ROC curves. Overall survival (OS) and progression-free survival (PFS)
following secondary cytoreduction were estimated using the KaplanMeier method. The duration of follow-up of surviving patients was censored at their date of last follow-up. Among the subset with a complete
secondary cytoreduction, Cox proportional hazards regression models
were fit to evaluate factors for their association with OS and PFS, respectively. A penalized smoothing spline was used to model the nonlinear
relationship of length of the disease-free interval with each long-term
outcome. All calculated p-values were two-sided and p-values b 0.05
were considered statistically significant. The statistical analysis was performed using SAS version 9.3 and R version 3.0.2. The recursive
partitioning analysis was performed using the rpart package (rpart: Recursive Partitioning. http://CRAN.R-project.org/package=rpart).
3. Results
Among 209 patients with a secondary cytoreduction surgery performed at our institution, 17 patients were excluded from the study
since they either had insufficient information from the initial surgery
to confirm residual disease (n = 16) or they lacked information needed
3
to calculate the AGO score (n = 1). The results herein are based on the
remaining 192 patients who underwent secondary cytoreduction surgery at a mean age of 59.3 (SD, 10.0) years with a median disease-free
interval of 1.9 (interquartile range (IQR), 1.0–3.5) years. At the time of
the initial surgery, 114 (59.4%) had a complete cytoreduction, 64
(33.3%) had RD ≤1 cm with gross residual disease, and 14 (7.3%) had
a suboptimal resection. Nearly all patients (97.4%) received platinumbased adjuvant chemotherapy as first-line therapy. Of the cohort, 102
(53.1%) patients were categorized as AGO score positive leaving a
large percentage of patients undergoing secondary surgery who were
AGO score negative. The specific AGO scoring characteristics are
shown in Table 1. Most of the 90 AGO score negative patients were
score negative due solely to incomplete cytoreduction at their initial
surgery (n = 66). In addition, there was a small group that had ascites
at the time of recurrence (n = 11) or both ascites and incomplete initial
cytoreduction (n = 12). The remaining AGO score negative had ECOG
performance status of N1.
Both groups underwent similarly complex surgical procedures, with
a rate of bowel resection of 26.7% in the AGO score negative group and
31.4% in the AGO score positive group. Diaphragmatic procedures were
performed in 23.3% of the AGO score negative group, and 16.7% of the
AGO score positive group (not statistically significant), and a diverting
colostomy in 1–2% of both groups (Table 2). The majority of patients
in both groups received postoperative chemotherapy (67.8% of AGO
score negative, 76.5% of AGO score positive); a small number in both
groups (10.0% of AGO score negative, 12.7% of AGO score positive)
also received postoperative radiation therapy.
Overall, 144 (75.0%) patients had a complete resection at the time of
secondary cytoreduction. Of the 102 AGO score positive patients, 86
(84.3%, 95% CI 77.3–91.4%) had complete resection at the time of
Fig. 1. Predictive value of factors in addition to AGO score for predicting complete secondary cytoreduction.
Please cite this article as: J.M.T. Janco, et al., Performance of AGO score for secondary cytoreduction in a high-volume U.S. center, Gynecol Oncol
(2016), http://dx.doi.org/10.1016/j.ygyno.2016.01.027
4
J.M.T. Janco et al. / Gynecologic Oncology xxx (2016) xxx–xxx
Fig. 2. Predictive value of factors for predicting complete secondary cytoreduction.
Fig. 3. Comparison of the ROC curves for three algorithms for predicting complete
secondary cytoreduction.
secondary cytoreduction, 9 (8.8%, 95% CI 3.3–14.3%) had optimal
cytoreduction, and 7 (6.9%, 95% CI 2.0–11.8%) had a suboptimal resection. Of the 90 AGO score negative patients, 58 (64.4%, 95% CI 54.6–
74.3%) achieved complete cytoreduction at the time of secondary surgery, 23 (25.6%, 95% CI 16.5–34.6%) had optimal cytoreduction, and 9
(10.0%, 95% CI 3.8–16.2%) had a suboptimal resection. This translates
to a false negative rate of 40.3% (95% CI, 32.3–48.3%) when using the
AGO score to predict the ability to achieve a complete secondary
cytoreduction. Of note, 48 patients received chemotherapy for ovarian
cancer recurrence at some point in their disease course before their secondary cytoreductive surgery. 21 (20.6%) of the 102 AGO score positive
patients received chemotherapy before their secondary cytoreduction
surgery compared to 27 (30.0%) of the 90 AGO score negative patients.
However, this difference was not statistically significant (p = 0.13). We
repeated the analysis of AGO score predictive value for complete secondary cytoreduction after excluding these cases and the results were
similar: we observed a complete cytoreduction rate of 82.7% (67/81)
among the AGO score positive patients and 68.3% (43/63) among the
AGO score negative patients. Therefore, we included these patients
who received interval chemotherapy for recurrence in the subsequent
analyses. We also evaluated whether restricting the ECOG performance
status from 0 or 1 to just 0 for defining AGO score positive would alter
the results. Using this revised criteria, we obtained a complete
cytoreduction rate of 84.8% (67/79) among the revised score positive
patients and 68.1% (77/113) among the revised score negative patients.
Given the similar positive predictive value, we chose to proceed with an
ECOG performance status of 0 or 1 in defining the AGO score as this
more accurately reflects clinical practice and practical candidacy for surgery at our institution.
Please cite this article as: J.M.T. Janco, et al., Performance of AGO score for secondary cytoreduction in a high-volume U.S. center, Gynecol Oncol
(2016), http://dx.doi.org/10.1016/j.ygyno.2016.01.027
J.M.T. Janco et al. / Gynecologic Oncology xxx (2016) xxx–xxx
As one of the objectives of the present study was to improve upon
the AGO score, we utilized recursive partitioning methods to identify
variables that would further discriminate or classify patients based on
their predicted probability of having a complete secondary
cytoreduction. The additional variables considered included stage at initial surgery (I–IIIA, IIIB–IIIC, IV), number of disease sites at recurrence on
preoperative imaging (single versus multiple), abdominal quadrant involved (lower, upper, both), and length of disease-free interval from initial therapy. Among AGO score positive patients, none of the additional
variables further discriminated between the patients. However, among
the AGO negative patients, further discrimination was achieved based
on 2 factors; number of disease sites at recurrence and length of the disease free-interval (Fig. 1). Specifically, AGO score negative patients
could be further stratified into complete secondary cytoreduction
rates of 85.2%, 66.7%, and 43.3%. We utilized recursive partitioning
methods to derive an alternative algorithm considering the three components of the AGO score individually along with the four variables
listed above; this algorithm is shown in Fig. 2. ROC curves for the
three algorithms are shown in Fig. 3. The overall predictive ability of
the AGO score, as measured by the area under the ROC curve or cindex, was 0.63 (95% CI, 0.55–0.71). In contrast the c-indices were
0.69 (95% CI, 0.60–0.78) and 0.70 (95% CI, 0.62, 0.79) for the algorithms
5
depicted in Figs. 1 and 2, respectively. Although the latter two c-indices
were higher than the c-index for AGO score alone, the differences were
not statistically significant (p = 0.17 and p = 0.13, respectively).
Among the 192 patients, 108 deaths have occurred; the median duration of follow-up for the remaining patients was 3.1 years (IQR, 1.4–
5.3 years). Progression has been documented in 137 patients. Not surprisingly, overall survival (OS) following secondary cytoreduction, as
well as progression-free survival (PFS), was significantly different between patients with complete, optimal, or suboptimal secondary
cytoreduction (Fig. 4). Patients with complete secondary cytoreduction
survived longer compared to those with optimal or suboptimal secondary cytoreduction (median OS, 5.4 vs. 2.4 vs. 1.3 years, respectively;
p b 0.001). Likewise the median PFS was also longer in patients with
complete versus optimal or suboptimal secondary cytoreduction (median PFS, 1.5 vs. 0.9 vs. 1.0 years, respectively; p = 0.001). To control
for the effect of residual disease at SCS on OS and PFS, we restricted
the remaining analyses to the 144 patients who achieved complete secondary cytoreduction. Among these 144 patients, AGO score was not associated with OS or PFS (Fig. 5). Therefore, we investigated whether any
of the three individual components of the AGO score itself, stage at initial surgery (I–IIIA, IIIB–IIIC, IV), number of disease sites at recurrence on
preoperative imaging (single versus multiple), abdominal quadrant
Fig. 4. Overall survival and progression-free survival, by result of secondary cytoreduction.
Please cite this article as: J.M.T. Janco, et al., Performance of AGO score for secondary cytoreduction in a high-volume U.S. center, Gynecol Oncol
(2016), http://dx.doi.org/10.1016/j.ygyno.2016.01.027
6
J.M.T. Janco et al. / Gynecologic Oncology xxx (2016) xxx–xxx
Fig. 5. Overall survival and progression-free survival, by AGO score, among the 144 patients with complete secondary cytoreduction.
involved (lower, upper, both) or length of the disease-free interval were
associated with OS and PFS. ECOG performance status at recurrence,
number of disease sites at recurrence, and length of the disease-free interval were all significantly (p b 0.05) associated with OS based on multivariable analysis. The same three variables were significantly
associated with PFS based on multivariable analysis with the exception
that number of disease-sites was of borderline significance (p = 0.09).
Estimates of 3- and 5-year OS and 1- and 3-year PFS predicted from
the multivariable Cox models are depicted in the panels of Fig. 6 according to the ECOG performance status at recurrence, number of disease
sites at recurrence, and length of the disease-free interval.
4. Discussion
There is no clear standard for management of patients with recurrent ovarian carcinoma. However, for the appropriate candidate, surgery can be of significant benefit to survival. There may even be
benefit in the platinum-resistant setting [8], although studies in this
patient population are largely limited to those with platinum-sensitive
disease. Those who are able to be cytoreduced to no residual disease
have a marked advantage to those with residual disease after surgery
for recurrence [9]. Appropriate selection of candidates who will benefit
from surgery is critical; surgery may be quite complex given the distribution of disease and prior extensive surgery in many cases. Several
groups have attempted to define scoring systems to determine which
patients are most likely to undergo complete secondary cytoreduction,
of which the AGO score is probably the most widely recognized and utilized. Several other factors, including time to recurrence, preoperative
CA-125, and number of sites of disease at recurrence, have been used
in attempts to strengthen the ability of the factors included in the AGO
score to predict complete secondary cytoreduction [3, 7, 10]. However,
recent external validation studies have confirmed the limitations of
these models. Positive predictive value is quite good for both the AGO
score and the Tian model, which incorporates some of the additional
variables mentioned above. In the validation studies, patients in the
AGO score positive group or in the Tian low-risk group for incomplete
Please cite this article as: J.M.T. Janco, et al., Performance of AGO score for secondary cytoreduction in a high-volume U.S. center, Gynecol Oncol
(2016), http://dx.doi.org/10.1016/j.ygyno.2016.01.027
J.M.T. Janco et al. / Gynecologic Oncology xxx (2016) xxx–xxx
7
Fig. 6. Survival after SCS is correlated with performance status, disease free interval and number of recurrence sites. The predicted probabilities of OS for (A) 3-year, (B) 5-year or PFS for
(C) 1-year, (D) 3-year are shown. Each curve represents a combination of ECOG PS and number of sites of recurrence plotted against disease free interval on the X-axis. Each point on the
curve represents the probability of survival for the indicated interval of a specific panel.
cytoreduction had a 74–82% probability of complete secondary
cytoreduction. Conversely, though, those patients in the AGO score negative or Tian high-risk groups still had a 56–71% rate of complete secondary cytoreduction [4, 5].
Our study had similar findings. Use of the AGO score to predict complete secondary cytoreduction had an excellent positive predictive
value, at 84.3%, but 64.4% of score negative patients in our population
still achieved complete secondary cytoreduction. We were able to demonstrate that surgical effort was similar in both groups, with similar
rates of upper abdominal procedures and bowel resection. Our attempt
to improve upon the AGO score for predicting complete secondary
cytoreduction, by incorporating number of sites of disease and length
of disease-free interval into the algorithm, was modestly successful.
The c-index, as an overall measure of the algorithm's predictive ability
increased from 0.632 to 0.689. The positive predictive value of the algorithm decreased from 84.3% to 80.9% (131/162), and the proportion of
algorithm negative patients achieving complete secondary
cytoreduction decreased from 64.4% to 43.3% (13/30), highlighting the
difficulty in using these models to triage patients away from surgery.
We propose that these factors (number of sites of disease and length
of disease-free interval, as well as the components of the AGO score)
should be noted when assessing the individual patient for consideration
of secondary cytoreductive surgery, with individualized decision making regarding the risk/benefit of surgical exploration and intraoperative
evaluation of feasibility of cytoreduction.
What our study adds to the literature is a more comprehensive assessment of the factors that confer a survival advantage apart from the
effect of surgery, and this may be more important in deciding who is a
good candidate for surgery vs. additional chemotherapy. The panels of
Fig. 6 nicely illustrate that even among the patients who undergo complete secondary cytoreduction, there is a continuum of benefit according to length of disease-free interval, with longer time to recurrence
yielding greater survival benefit. Additionally, ECOG performance status
and number of sites of disease at recurrence also contribute to the effect
on survival. This information might contribute to patients' decisions to
undergo potentially complex and morbid surgical procedures, when
weighing the benefits of surgery versus chemotherapy for recurrence.
Strengths of our study include a patient population drawn from a
high-volume referral center, surgery by a single group of gynecologic
oncologists with similar surgical philosophy and effort in the setting of
recurrent ovarian carcinoma, and a representative patient population.
Forty-eight patients (25%) had previously undergone chemotherapy
for recurrence, and the majority of patients were initially diagnosed at
an advanced stage. Weaknesses include the retrospective nature of the
study and the highly selected nature of the patient population, as
these were patients selected for surgery and felt by their surgeons to
be able to undergo successful secondary cytoreduction. Additionally,
the study is limited to a single institution and may not be reflective of
other practices, and also was unable to adequately address the role
and timing of chemotherapy in relationship to cytoreduction for ovarian
cancer recurrence. We look forward to the results of several prospective
trials in process, such as GOG 213, DESKTOP III, and the SOCceR trial,
which may help to elucidate some of these issues. We offer, though, a
more comprehensive analysis of survival predictors than previous
Please cite this article as: J.M.T. Janco, et al., Performance of AGO score for secondary cytoreduction in a high-volume U.S. center, Gynecol Oncol
(2016), http://dx.doi.org/10.1016/j.ygyno.2016.01.027
8
J.M.T. Janco et al. / Gynecologic Oncology xxx (2016) xxx–xxx
studies, and affirm the importance and survival benefit of complete secondary cytoreduction for recurrent ovarian carcinoma. As many patients undergo complex surgery for secondary cytoreduction,
information regarding the extent of benefit can be useful for decisionmaking as well as for referral to expert centers able and willing to perform complex surgery to achieve complete cytoreduction. Use of the
models and survival curves generated by our study may be beneficial
in counseling patients regarding their options for management of recurrent ovarian carcinoma.
Conflict of interest
[4]
[5]
[6]
[7]
The authors declare that there are no conflicts of interest.
References
[1] P. Harter, A. d B, M. Hahmann, A. Hasenburg, A. Burges, S. Loibl, M. Gropp, J. Huober,
D. Fink, W. Schroder, K. Muenstedt, B. Schmalfeldt, G. Emons, J. Pfisterer, K.
Wollschlaeger, H.G. Meerpohl, G.P. Bretibach, B. Tanner, J. Sehouli, Surgery in recurrent ovarian cancer: the Arbeitsgemeinschaft Gynaekologische Onkologie (AGO)
DESKTOP OVAR trial, Ann. Surg. Oncol. 13 (2006) 1702–1710.
[2] P. Harter, J. S, A. Reuss, A. Hasenburg, G. Scambia, D. Cibula, S. Mahner, I. Vergote, A.
Reinthaller, A. Burges, L. Hanker, M. Polcher, C. Kurzeder, U. Canzler, K.U. Petry, A.
Obermair, E. Petru, B. Schmalfeldt, D. Lorusso, A. du Bois, Prospective validation
study of a predictive score for operability of recurrent ovarian cancer: the Multicenter Intergroup Study DESKTOP II, Int. J. Gynecol. Cancer 21 (2011) 289–295.
[3] W.J. Tian, D. C, J. Sehouli, C.G. Trope, R. Jiang, A. Ayhan, G. Cormio, Y. Xing, G.P.
Breitbach, E.I. Braicu, C.A. Rabbitt, H. Oksefjell, C. Fotopolou, H.G. Meerpohl, A. du
[8]
[9]
[10]
Bois, J.S. Berek, R.Y. Zang, P. Harter, A risk model for secondary cytoreductive surgery in recurrent ovarian cancer: an evidence-based proposal for patient selection,
Ann. Surg. Oncol. 19 (2012) 597–604.
R. Van de Laar, L. M, T. Van Gorp, J. IntHout, P.L. Zusterzeel, F.Y. Kruitwagen, External
validation of two prediction models of complete secondary cytoreductive surgery in
patients with recurrent epithelial ovarian cancer, Gynecol. Oncol. 137 (2015)
210–215.
E. Laas, M. L, M. De Cuypere, F. Selle, E. Darai, D. Querleu, R. Rouzier, E. Chereau, Secondary complete cytoreduction in recurrent ovarian cancer: benefit of optimal patient selection using scoring system, Int. J. Gynecol. Cancer 24 (2014) 238–246.
P. Harter, B. Bbb, P.F. Alesina, D. Lorenz, A. Boergers, F. Heitz, R. Hils, C. Kurzeder, A.
Traut, A. du Bois, Prognostic and predictive value of the Arbeitsgemeinschaft
Gynaekologische Onkologie (AGO) score in surgery for recurrent ovarian cancer,
Gynecol. Oncol. 132 (2014) 537–541.
D.S. Chi, K. M, J.P. Diaz, J. Huh, S. Schwabenbauer, A.J. Hummer, E.S. Venkatraman, C.
Aghajanian, Y. Sonoda, N.R. Abu-Rustum, R.R. Barakat, Guidelines and selection
criteria for secondary cytoreductive surgery in patients with recurrent, platinumsensitive epithelial ovarian carcinoma, Cancer 106 (2006) 1933–1939.
A. Musella, C. M, I. Palaia, G. Perniola, M. Giorgini, F. Lecce, L. Vertechy, R. Iadarola, F.
De Felice, M. Monti, L. Muzii, R. Angioli, P. Benedetti Panici, Secondary cytoreduction
in platinum-resistant recurrent ovarian cancer: a single-institution experience, Ann.
Surg. Oncol. 2015 (2015).
A. Rawahi, A. L, R.E. Bristow, A. Bryant, A. Elattar, S. Chattopadhyay, K. Galaal, Surgical cytoreduction for recurrent epithelial ovarian cancer, Cochrane Database Syst.
Rev. 28 (2013).
R. Angioli, S. C, A. Aloisi, R. Ricciardi, G. Scaletta, S. Lopez, A. Miranda, A. Di Pinto, C.
Terranova, F. Plotti, A predictive score for secondary cytoreductive surgery in recurrent ovarian cancer (SeC-Score): a single-centre, controlled study for preoperative
patient selection, Ann. Surg. Oncol. (2015).
Please cite this article as: J.M.T. Janco, et al., Performance of AGO score for secondary cytoreduction in a high-volume U.S. center, Gynecol Oncol
(2016), http://dx.doi.org/10.1016/j.ygyno.2016.01.027