ORIGINAL STUDY
Risk Factors for Progression to Invasive Carcinoma in Patients With Borderline Ovarian Tumors Taejong Song, MD,* Yoo-Young Lee, MD, PhD,Þ Chel Hun Choi, MD,Þ Tae-Joong Kim, MD,Þ Jeong-Won Lee, MD, PhD,Þ Duk-Soo Bae, MD, PhD,Þ and Byoung-Gie Kim, MD, PhDÞ
Objective: The aim of this study was to identify risk factors for progression to invasive carcinoma in patients with borderline ovarian tumors (BOTs). Methods: We performed a retrospective review of all patients treated and followed for BOTs between 1996 and 2011. Multivariate Cox proportional hazards model analysis was performed to identify independent risk factors for progression to invasive carcinoma. Results: A total of 364 patients were identified. During the median follow-up of 53.8 months, 31 patients (8.5%) developed recurrent disease: 12 (3.3%) had recurrent disease with progression to invasive carcinoma, and 19 (5.2%) had recurrent disease with borderline histology. Disease-related deaths (7/364; 1.7%) were observed only in patients with progression to invasive carcinoma. The multivariate analysis showed that independent risk factors for progression to invasive carcinoma were advanced disease stage (hazard ratio [HR], 5.59; P = 0.005), age 65 years or older (HR, 5.13; P = 0.037), and the presence of microinvasion (HR, 3.71; P = 0.047). These 3 factors were also independently related to overall survival. Conclusions: Although patients with BOTs have an excellent prognosis, the risk of progression to invasive carcinoma and thereby death remains. Therefore, physicians should pay closer attention to BOT patients with these risk factors (ie, advanced disease stage, old age, and microinvasion), and more careful surveillance for progression to invasive carcinoma is needed. Key Words: Borderline ovarian tumors, Risk factors, Recurrence, Progression to invasive carcinoma Received March 4, 2014, and in revised form March 13, 2014. Accepted for publication June 12, 2014. (Int J Gynecol Cancer 2014;24: 1206Y1214)
1
ovarian tumors (BOTs), first described by Taylor B orderline in 1929, comprise about 10% to 20% of all epithelial ovarian malignancies.2,3 As opposed to patients with invasive ovarian
*Department of Obstetrics and Gynecology, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea; and †Department of Obstetrics and Gynecology, Samsung Medical Center, Sungkyunkwan University of School of Medicine, Seoul, Republic of Korea. Address correspondence and reprint requests to Byoung-Gie Kim, MD, PhD, Department of Obstetrics and Gynecology, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Irwon-Dong, Gangnam-gu, Seoul 135-710, Republic of Korea. E-mail: [email protected]. The authors declare no conflicts of interest. Copyright * 2014 by IGCS and ESGO ISSN: 1048-891X DOI: 10.1097/IGC.0000000000000216
1206
carcinomas, those with BOTs tend to present at an earlier stage and at a younger age.4Y6 Patients with BOTs have a much better prognosis than those with invasive carcinomas,4,6 and 5- and 10-year survival rates for stages I, II, and III disease are 99% and 97%, 98% and 90%, and 96% and 88%, respectively.7 Despite these favorable data, some patients relapse or die of the disease. Recurrences in patients with BOTs were subclassified into 2 types: (1) borderline-type recurrence and (2) invasive-type recurrence (namely, progression to invasive carcinoma). The prognosis of patients with borderline-type recurrence is excellent, and therefore, a second round of fertility-sparing surgery is undertaken in cases of those with a strong desire for future childbearing.8 In contrast, progression to invasive carcinoma could affect patient survival and is regarded as a cause of disease-related deaths in patients with BOTs.9 However, risk factors for progression to invasive carcinoma have not yet been established. Therefore, the aim of
International Journal of Gynecological Cancer
& Volume 24, Number 7, September 2014
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 24, Number 7, September 2014 Invasive Recurrence of Borderline Ovarian Tumors
this study was to determine the risk factors for progression to invasive carcinoma in patients with BOTs.
METHODS Using the tumor registry at the Samsung Medical Center (Seoul, Korea), we retrospectively identified patients with pathologically confirmed BOTs who were treated and followed up between January 1996 and December 2011. Patients who did not receive primary treatment or follow up for at least 6 months at the Samsung Medical Center were excluded from the analysis. The study was approved by the institutional review board. Surgical treatment was classified as either radical (bilateral salpingo-oophorectomy with or without hysterectomy) or fertility sparing (defined as conservation of the uterus and at least a portion of one ovary). The completeness of the surgical staging was evaluated according to the International Federation of Gynecology and Obstetrics (FIGO) guidelines10 but modified with respect to lymphadenectomy.5,11,12 Complete staging included removal of all visible tumor tissue, peritoneal washing cytology, omentectomy, random multiple peritoneal biopsies, and appendectomy (at least in mucinous tumor types). Extraovarian implants were divided histologically into noninvasive and invasive implants by the absence or presence of destructive stromal invasion into the underlying tissue. The ovarian stromal microinvasion was defined as the presence of microscopic foci of stromal tumor invasion, less than 3 mm in the longest linear dimension and 10 mm2 or less in area, by single cells and nests of moderately atypical cells.13 The criteria for the diagnosis of micropapillary pattern were (a) tumors showing a filigree pattern of small, uniform, elongated, stroma-poor, or stroma-free papillae, at least five times as long as wide, arising directly in a nonhierarchical fashion from large fibrotic, edematous, or myxoid papillary stalks or from cyst walls, and/or a cribriform pattern of epithelial cells lining stalks or cyst walls, and (b) the presence of either or both patterns on a confluent area at least 5 mm in greatest dimension on at least 1 slide.14,15 Following completion of the primary treatment, patients were examined every 3 months during the first 2 years, every 6 months during the next 3 years, and yearly thereafter. Posttreatment surveillance included physical examination, assessment of tumor markers, and imaging studies. Recurrence was defined as the detection of the same tumor cell histology after an apparently complete surgical resection. Statistical analysis was performed with SPSS software (version 12.0; SPSS, Inc, Chicago, IL). The tumor recurrence rate, progression to invasive carcinoma rate, and mortality rate were analyzed. Invasive recurrence-free survival (iRFS) was calculated from the date of surgery to the date of invasive recurrence (progression to invasive carcinoma) or last contact. Overall survival was calculated from the date of surgery to the date of disease-related death or last contact. For each characteristic, the distributions of time to event outcomes were described by the Kaplan-Meier method and compared using the long-rank test. Multivariate Cox proportional hazards models were fitted to compare survival experiences among risks factors while adjusting for the effects of other demographic and clinical characteristics. P G 0.05 was considered statistically significant.
RESULTS Patient Characteristics A total of 364 patients were identified. Clinical and pathologic characteristics for the cohort are listed in Table 1. The median age was 39 years (range, 10Y85 years). The most common reason for the hospital visit was symptoms due to tumors such as pain, palpable mass, distension, or bleeding (245 patients [67.3%]), followed by the diagnosis being the coincidental outcome of a medical checkup (71 patients [19.5%]). The median levels of serum CA-125 and CA-19-9 were 29.2 U/mL (range, 0.6Y7810.0 U/mL) and 19.3 U/mL (range, 0.1Y56,430.0 U/mL), respectively. All patients underwent surgery as the primary treatment. A total of 173 patients (47.5%) with a median age of 50 years underwent radical surgery, whereas 191 (52.5%) with a median age of 30 years underwent fertility-sparing surgery (144 unilateral salpingo-oophorectomy and 47 cystectomy). The staging procedure was performed in 223 patients (61.3%). Pelvic lymphadenectomy was performed in 51 patients (14.0%), whereas pelvic and para-aortic lymphadenectomy was performed in 6 patients (1.6%). At the completion of the primary surgery (and restaging procedure, if any), 361 patients (99.2%) had no residual tumor, and the remaining 3 (0.8%) had residual tumors with a diameter ranging from a few millimeters to 1 cm. The majority of patients had early-stage disease; 328 patients (90.1%) had stage I disease, 8 (2.2%) had stage II disease, and 28 (7.7%) had stage III disease. The most common sites of extraovarian implants were omentum, pelvic peritoneum, and fallopian tubes. Two patients had metastasis in the pelvic lymph node. The most common histology was mucinous (246 patients [67.6%]), followed by serous (109 patients [29.9%]), endometrioid (4 patients [1.1%]), clear cell (4 patients [1.1%]), and fibroma (1 patient, 0.03%). According to final pathologic diagnoses, there were 42 cases of microinvasion (11.5%), 21 cases of micropapillary pattern (5.8%), 27 cases of intraepithelial carcinoma (7.4%), 28 cases of noninvasive implants (7.7%), and 8 cases of invasive implants (2.2%). Of 42 cases with microinvasion, there were mucinous histology in 21 cases (50.0%), serous histology in 19 cases (45.2%), endometrioid histology in 1 case (2.4%), and clear cell histology in 1 case (2.4%). Postoperative chemotherapy was administrated to 37 patients in the form of platinum-based regimens; chemotherapy was mainly indicated in patients with advanced-stage disease. No chemotherapy-related morbidity or death was observed.
Recurrence, Progression to Invasive Carcinoma, and Mortality Rates During the median follow-up of 53.8 months (range, 6.4Y206.5 months), 31 patients (8.5%) developed recurrent disease. Of these 31 patients, 12 (3.3%) had recurrent disease with progression to invasive carcinoma, and 19 (5.2%) had recurrent disease with borderline histology. When classified according to the histology of recurrent diseases, 13 patients had mucinous BOTs, 17 patients had serous BOT, and 1 patient had endometrioid BOT. Table 2 shows the characteristics of patients with progression to invasive carcinoma. At the time of the last follow-up, 337 patients (92.6%) had no evidence of disease, 3 patients
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1207
Song et al
International Journal of Gynecological Cancer
& Volume 24, Number 7, September 2014
TABLE 1. Characteristics of 364 patients with BOT No. Patients (%) or Median (Range) Age at diagnosis, y G65 y Q65 y Body mass index, kg/m2 Diagnosis reached through Checkup Symptoms Fortuitous Unknown CA-125,* U/mL G35 U/mL Q35 U/mL CA-19-9,* U/mL G37 U/mL Q37 U/mL Tumor size, cm Surgical approach Laparoscopy Laparotomy Surgical radicality Radical surgery Fertility-sparing surgery Staging procedure Histology Serous Mucinous Other Disease stage I II-III Microinvasion Micropapillary Intraepithelial carcinoma Bilaterality Extraovarian implant No implant Noninvasive implant Invasive implant Adjuvant chemotherapy Follow-up, mo Recurrence with borderline or invasive histology Recurrence with borderline histology Recurrence with invasive carcinoma histology Disease-specific mortality
39 326 38 22.0
(10Y85) (89.6%) (10.4%) (14.2Y35.5)
71 245 33 15 29.2 182 144 19.3 178 86 12.0
(19.5%) (67.3%) (9.1%) (4.1%) (0.6Y7810.0) (55.8%) (44.2%) (0.1Y56,430.0) (67.4%) (32.6%) (2.0Y39.0)
86 (23.6%) 278 (76.4%) 173 (47.5%) 191 (52.5%) 223 (61.3%) 109 (29.9%) 246 (67.6%) 9 (2.5%) 328 36 42 21 27 37
(90.1%) (9.9%) (11.5%) (5.8%) (7.4%) (10.2%)
328 (90.1%) 28 (7.7%) 8 (2.2%) 37 (10.2%) 58.3 (6.4Y206.5) 31 (8.5%) 19 (5.2%) 12 (3.3%) 7 (1.9%)
*Preoperative CA-125 and CA-19-9 were determined in 326 and 264 patients, respectively.
1208
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
45
68
29
79
77
18 50
76
53
35
74
2
3
4
5
6
7 8
9
10
11
12
IIIC, S
IIC, E
IB, M
IA, M
IA, M IIIC, S
IA, S
IA, M
IIIC, S
IIIC, S
IA, M
IIIB, S
USO
V
Invasive
TH, BSO, staging, chemotherapy
TH, BSO, chemotherapy Noninvasive USO
V
V USO Noninvasive TH, BSO, staging, chemotherapy V TH, BSO, staging
V
Invasive
TH, BSO, staging, chemotherapy TH, BSO, staging, chemotherapy BSO
TH, BSO, staging
V Invasive
TH, BSO, staging
Noninvasive
Primary Treatment
20.7
7.6
18.0
6.0
38.6 76.9
13.4
30.3
32.9
20.9
13.8
20.3 Debulking, chemotherapy Chemotherapy
LAR, chemotherapy
Treatment at Progression
Contralateral ovary, peritoneal spread Pelvis, abdomen
Pelvis, peritoneal spread Omentum
Debulking, chemotherapy TH, USO, debulking, chemotherapy Chemotherapy
Chemotherapy
Debulking, chemotherapy TH, staging, chemotherapy Liver surface, pelvis, Refusal of treatment abdomen Contralateral ovary TH, USO, staging Pleura VATS, chemotherapy
Pelvic lymph node, peritoneal spread Vagina, peritoneal spread Pelvis
Rectosigmoid surface Pelvis, omentum
Interval to Progression, mo Site of Progression
DOD
DOD
20.6 36.0
DOD
DOD
NED AWD
DOD
AWD
DOD
DOD
AWD
NED
49.1
17.3
63.9 132.3
18.0
66.9
86.5
53.7
43.2
28.5
Follow-up Current Period, mo Status
AWD, alive with disease; DOD, died of disease; E, endometrioid; LAR, lower, anterior resection; M, mucinous; NED, no evidence of disease; S, serous; TH, total hysterectomy; BSO, bilateral salpingo-oophorectomy; USO, unilateral salpingo-oophorectomy; VATS, video-assisted thoracoscopic surgery.
43
1
Age at Stage and Extraovarian Patient No. Diagnosis, y Histology Implant
TABLE 2. Main characteristics of BOT patients with progression to carcinoma
International Journal of Gynecological Cancer
& Volume 24, Number 7, September 2014 Invasive Recurrence of Borderline Ovarian Tumors
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1209
International Journal of Gynecological Cancer
Song et al
(0.8%) were alive with disease, 7 patients (1.9%) were dead of disease, and 17 patients (4.7%) were dead of other causes. All 7 patients who died of disease had progression to invasive carcinoma. No disease-related death was observed in patients who had only borderline-type recurrence. The median time of progression to invasive carcinoma was 20.5 months, with a range of 6.0 to 76.9 months. Treatment at progression was as follows: 7 patients received surgery and chemotherapy, 3 received only salvage chemotherapy, 1 received only surgical treatment, and the remaining 1 refused any treatment after biopsy confirmation. The
& Volume 24, Number 7, September 2014
median iRFS was 57.9 months, and the 5- and 7-year iRFS rates were 96.5% and 95.7%, respectively.
Risk Factors for Progression to Invasive Carcinoma Univariate and multivariate analyses were performed to identify risk factors for progression to carcinoma. These are outlined in Table 3. In the univariate analysis, advanced FIGO stage, age 65 years or older, CA-125 35 U/mL or greater, the presence of invasive implants, microinvasion, micropapillary pattern, and adjuvant chemotherapy were significant risk
TABLE 3. Univariate and multivariate analysis of progression to carcinoma
Variable FIGO stage I II-III Age G65 y Q65 y CA-125 G35 U/mL Q35 U/mL CA-19-9 G37 U/mL Q37 U/mL Surgical approach Laparotomy Laparoscopy Surgical radicality Radical surgery Fertility-sparing surgery Staging procedure Yes No Histology Serous Mucinous Invasive implants Yes No Microinvasion Micropapillary patterns Intraepithelial carcinoma Adjuvant chemotherapy Yes No
Univariate
Multivariate*
HR
95% CI
1 9.45
3.05Y29.34
1 7.06
HR
95% CI
P
G0.001
5.59
1.70Y18.36
0.005
2.24Y22.30
0.001
5.13
1.10Y23.93
0.037
1 3.75
1.02Y13.85
0.047
1.77
0.38Y8.24
0.464
1 1.01
0.25Y4.02
0.994
1 1.89
0.57Y6.29
0.303
2.23
0.67Y7.41
0.191
1.09
0.35Y3.45
0.880
2.68 1
0.82Y8.79
0.104
15.94 1 8.42 4.85 2.69
4.29Y59.27
G0.001
2.71Y26.13 1.31Y17.92 0.59Y12.29
G0.001 0.018 0.202
3.71 1.18
1.02Y13.53 0.23Y5.99
0.047 0.845
1.87Y18.63
0.002
5.90 1
P
*Multivariate analysis was applied only for P G 0.05 at the univariate analysis.
1210
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 24, Number 7, September 2014 Invasive Recurrence of Borderline Ovarian Tumors
FIGURE 1. Kaplan-Meier plot. A, Progression to invasive carcinoma by stage. B, Progression to invasive carcinoma by age. C, Progression to invasive carcinoma by microinvasion. D, Overall survival by stage. E, Overall survival by age. F, Overall survival by microinvasion. factors for progression to invasive carcinoma (Figs. 1AYC). Univariate analysis for residual disease was not performed as the number of patients with residual disease was too small. Risk factors that had significance upon the univariate analysis were included in the multivariate analysis. Because most patients who received adjuvant chemotherapy had an advanced stage of disease, the iRFS of patients who received adjuvant chemotherapy was lower, and this factor was therefore excluded from the multivariate analysis. Because an extraovarian invasive implant is a surrogate for advanced-stage disease, this was also excluded from the multivariate analysis. Multivariate analysis showed that independent risk factors for progression to invasive carcinoma were advanced FIGO stage (hazard ratio [HR], 5.59; 95% confidence interval [CI], 1.70Y 18.36; P = 0.005), age 65 years or older (HR, 5.13; 95% CI, 1.10Y 22.30; P = 0.037), and the presence of microinvasion (HR, 3.71; 95% CI, 1.02Y13.53; P = 0.047) (Table 3). Moreover, in the multivariate analysis of overall survival, advanced FIGO stage (HR, 11.27; 95% CI, 2.36Y53.71; P = 0.002), age 65 years or older (HR, 6.90; 95% CI, 1.33Y35.95; P = 0.022), and the presence of microinvasion (HR, 5.27; 95% CI, 1.18Y23.64; P = 0.031) remained as significant risk factors (Table 4; Figs. 1DYF).
DISCUSSION This study was performed to determine the risk factors for progression to invasive carcinoma in 364 patients with BOTs. We found that the progression rate was 3.3%, and the independent risk factors for progression were advanced FIGO stage, age 65 years or older, and the presence of microinvasion. These 3 factors were also independently related to overall survival. We also found that no patient with borderline-type recurrence died of the disease, and disease-related death was observed only in patients with progression to invasive carcinoma. Identification of patients with high-risk factors (ie, advanced stage, old age, and microinvasion) is essential for offering more selective treatment or monitoring to prevent progression to invasive carcinoma. This study is the first to determine the risk factors for progression to invasive carcinoma in patients with BOTs. In our study, the risk of progression to invasive carcinoma was estimated as 3.3%. This rate was similar with 2 previous studies.9,12 Zanetta et al12 reported a 2.1% progression rate in 339 BOT patients with a median follow-up of 70 months. Morice et al9 reported a 2% to 3% progression rate in their review article in which 47 cases of progression to
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1211
International Journal of Gynecological Cancer
Song et al
& Volume 24, Number 7, September 2014
TABLE 4. Univariate and multivariate analysis of overall survival
Variable FIGO stage I II-III Age G65 y Q65 y CA-125 G35 U/mL Q35 U/mL CA-19-9 G37 U/mL Q37 U/mL Surgical approach Laparotomy Laparoscopy Surgical radicality Radical surgery Fertility-sparing surgery Staging procedure Yes No Histology Serous Mucinous Invasive implants Yes No Microinvasion Micropapillary patterns Intraepithelial carcinoma Adjuvant chemotherapy Yes No
Univariate
Multivariate*
HR
95% CI
P
HR
95% CI
P
1 11.69
2.61Y52.32
0.001
11.27
2.36Y53.71
0.002
1 14.35
3.12Y64.42
0.001
6.90
1.33Y35.95
0.022
1 3.01
0.58Y15.52
0.188
1 2.08
0.23Y18.52
0.511
1 3.29
0.73Y14.91
0.123
1 2.68
0.52Y13.53
0.239
1 1.13
0.25Y5.07
0.870
4.07 1
0.74Y22.28
0.106
31.39 1 11.03 2.47 2.01
7.00Y140.74
G0.001
2.47Y49.33 0.29Y20.95 0.29Y14.00
0.002 0.407 0.480
5.27
1.18Y23.64
0.031
2.20Y44.18
0.003
9.85 1
*Multivariate analysis was applied only for P G 0.05 at the univariate analysis.
invasive carcinoma reported in 21 articles published since 2009 were analyzed. In our study, one interesting observation was that serous BOTs seemed to be more prone to progress to invasive carcinoma (5.5% vs 2.0%; P = 0.104). However, this observation is inconsistent with other studies.16,17 Koskas et al16 reported that the 10-year cumulative risk of recurrence in the form of invasive carcinoma was 13% in 97 patients with mucinous BOT and concluded that, unlike serous BOT, mucinous BOT does not appear to be such a ‘‘safe’’disease.16 We cannot determine why this discrepancy exists. However, Zanetta et al12 reported similar results to ours, in that they
1212
observed that the rate of progression to invasive carcinoma in serous BOT was higher than that of the mucinous counterpart (2.4% vs 1.6%). In this study, advanced FIGO stage was an independent risk factor for progression to invasive carcinoma and diseasespecific mortality. Our finding was supported by previous studies. The study by Ren et al18 evaluated 234 patients with BOTs and showed that advanced FIGO stage was independently related to recurrence-free survival (P G 0.001). Cusido et al19 analyzed 457 patients with BOTs from 27 centers of Spain and concluded that advanced FIGO stage was an independent risk * 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 24, Number 7, September 2014 Invasive Recurrence of Borderline Ovarian Tumors
factor for recurrence. Sornsukolrat and Tuipae20 evaluated 210 patients with BOTs and concluded that advanced FIGO stage was a significant risk factor for disease-specific mortality as well as for recurrence. We believe that FIGO stage is of paramount importance for the prognosis of patients with BOTs. Although old age has been widely known as a risk factor for recurrence or disease-specific mortality in patients with cervical cancer,21 endometrial cancer,22 and ovarian cancer,23 no data are available on the effect of old age on BOT recurrence or survival. To date, a large number of studies on risk factors for recurrence or survival in patients with BOTs have been conducted.4,8,9,11,16,18,24 Among them, some studies included age as a variable8,18,25,26; however, they all failed to reach a conclusion that old age was a risk factor for recurrence or survival. The main reasons for that were that age was treated as a continuous variable,8,11 and an age cutoff of 30 years,25 35 years,26 or 40 years18 was used as a categorical variable. However, 2 studies8,11 showed that age as a continuous variable was a risk factor for recurrence, which was similar with our finding. A controversy exists with regard to the prognostic significance of microinvasion.27 Analysis of the effect of microinvasion on the rate of progression to invasive carcinoma or overall survival is especially difficult because of potential confounding factors, such as micropapillary patterns and presence of extraovarian implants. Therefore, a large sample size and long follow-up period are essential for the study of microinvasion in BOTs. According to a study that analyzed 12 series of serous BOTs with microinvasion (n = 133),9 20 patients had recurrence (15%), including 35% (7/20) with invasive carcinoma at recurrence. The rate of disease-specific mortality was 6% (8/133). In accordance with the findings of previous studies, our data indicate that microinvasion is a risk factor for progression to invasive carcinoma (HR, 3.71; 95% CI, 1.02Y13.53) and disease-specific mortality (HR, 5.27; 95% CI, 1.18Y23.64). With regard to the use of postoperative chemotherapy, we noticed that in our study population the incidence of recurrence, progression to invasive carcinoma, or death was not improved and was rather higher in patients who received chemotherapy when compared with those who were not treated with chemotherapy. This observation seems to confirm the results of 2 other studies8,28 including a report of the M. D. Anderson Cancer Center.28 In other studies,29,30 the rate of recurrence or progression was not different between patients who received chemotherapy and those who did not. Researchers performing a Cochrane meta-analysis of interventions for the treatment of BOTs concluded that evidence did not support the use of any specific type of adjuvant treatment.31 Therefore, adjuvant chemotherapy is currently not believed to play a role in the treatment of BOTs. This study has some limitations. First, this study had a retrospective, single-center design, therefore possibly introducing some degree of bias. Prospective, multicenter trials are needed to confirm our results. Second, the staging procedure was not considered beforehand in all cases so that our data may be biased in this respect. However, Camatte et al24 reported that the absence of a staging procedure in patients with an ‘‘apparent stage I’’ BOT does not modify survival. Nevertheless, this study
has the advantages of including a large sample size and long follow-up periods; this is very important, considering the rarity and late recurrence of BOTs. Moreover, this study is the first study to determine the risk factors for progression to invasive carcinoma in patients with BOTs. In conclusion, although BOTs have an excellent prognosis, they are not exempt from the risk of progression to invasive carcinoma. Identification of patients with high-risk factors is essential in order to prevent their disease-specific mortality. Our results indicate that the risk factors for progression to invasive carcinoma in patients with BOTs are advanced FIGO stage, age 65 years or older, and microinvasion Therefore, physicians should pay closer attention to BOT patients with these risk factors and conduct more careful surveillance for progression to invasive carcinoma.
REFERENCES 1. Taylor HC. Malignant and semimalignant tumors of the ovary. Surg Gynecol Obstet. 1929;48:204Y230. 2. Trope CG, Kristensen G, Makar A. Surgery for borderline tumor of the ovary. Semin Surg Oncol. 2000;19:69Y75. 3. Lim MC, Moon EK, Shin A, et al. Incidence of cervical, endometrial, and ovarian cancer in Korea, 1999Y2010. J Gynecol Oncol. 2013;24:298Y302. 4. Sherman ME, Mink PJ, Curtis R, et al. Survival among women with borderline ovarian tumors and ovarian carcinoma: a population-based analysis. Cancer. 2004;100:1045Y1052. 5. Morice P. Borderline tumours of the ovary and fertility. Eur J Cancer. 2006;42:149Y158. 6. Morris CR, Liu L, Rodriguez AO, et al. Epidemiologic features of borderline ovarian tumors in California: a population-based study. Cancer Causes Control. 2013;24:665Y674. 7. Trimble CL, Kosary C, Trimble EL. Long-term survival and patterns of care in women with ovarian tumors of low malignant potential. Gynecol Oncol. 2002;86:34Y37. 8. Park JY, Kim DY, Kim JH, et al. Surgical management of borderline ovarian tumors: the role of fertility-sparing surgery. Gynecol Oncol. 2009;113:75Y82. 9. Morice P, Uzan C, Fauvet R, et al. Borderline ovarian tumour: pathological diagnostic dilemma and risk factors for invasive or lethal recurrence. Lancet Oncol. 2012;13:e103Y115. 10. Benedet J, Bender H, Jones H, III, et al. FIGO Committee on Gynecologic Oncology. FIGO staging classifications and clinical practice guidelines in the management of gynecologic cancers. Int J Gynecol Obstet. 2000;70:209Y262. 11. Shih KK, Zhou Q, Huh J, et al. Risk factors for recurrence of ovarian borderline tumors. Gynecol Oncol. 2011;120:480Y484. 12. Zanetta G, Rota S, Chiari S, et al. Behavior of borderline tumors with particular interest to persistence, recurrence, and progression to invasive carcinoma: a prospective study. J Clin Oncol. 2001;19:2658Y2664. 13. Scully RE. International Histological Classification of Tumors: Histological Typing of Ovarian Tumours. 2nd ed. Spring Verlag: New York; 1999. 14. Burks RT, Sherman ME, Kurman RJ. Micropapillary serous carcinoma of the ovary: a distinctive low-grade carcinoma related to serous borderline tumors. Am J Surg Pathol. 1996;20:1319Y1330. 15. Eichhorn JH, Bell DA, Young RH, et al. Ovarian serous borderline tumors with micropapillary and cribriform patterns: a
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1213
Song et al
16.
17.
18.
19.
20.
21.
22.
23.
International Journal of Gynecological Cancer
study of 40 cases and comparison with 44 cases without these patterns. Am J Surg Pathol. 1999;23:397Y409. Koskas M, Uzan C, Gouy S, et al. Prognostic factors of a large retrospective series of mucinous borderline tumors of the ovary (excluding peritoneal pseudomyxoma). Ann Surg Oncol. 2011;18:40Y48. Benito V, Lubrano A, Arencibia O, et al. Serous and mucinous borderline ovarian tumors: are there real differences between these two entities?. Eur J Obstet Gynecol Reprod Biol. 2010;153:188Y192. Ren J, Peng Z, Yang K. A clinicopathologic multivariate analysis affecting recurrence of borderline ovarian tumors. Gynecol Oncol. 2008;110:162Y167. Cusido M, Balaguero L, Hernandez G, et al. Results of the national survey of borderline ovarian tumors in Spain. Gynecol Oncol. 2007;104:617Y622. Sornsukolrat S, Tuipae S. Prognostic factors and survival of borderline ovarian tumors in Rajavithi Hospital between 1979-2006 A.D. J Med Assoc Thai. 2012;95:1141Y1148. Sharma C, Deutsch I, Horowitz DP, et al. Patterns of care and treatment outcomes for elderly women with cervical cancer. Cancer. 2012;118:3618Y3626. Ahmed A, Zamba G, DeGeest K, et al. The impact of surgery on survival of elderly women with endometrial cancer in the SEER program from 1992-2002. Gynecol Oncol. 2008;111:35Y40. Chereau E, Ballester M, Selle F, et al. Ovarian cancer in the elderly: impact of surgery on morbidity and survival. Eur J Surg Oncol. 2011;37:537Y542.
1214
& Volume 24, Number 7, September 2014
24. Camatte S, Morice P, Thoury A, et al. Impact of surgical staging in patients with macroscopic ‘‘stage I’’ ovarian borderline tumours: analysis of a continuous series of 101 cases. Eur J Cancer. 2004;40:1842Y1849. 25. Azuar AS, Bourdel N, Ferrarrese G, et al. Laparoscopic restaging of borderline ovarian tumours (BLOT): a retrospective study of 142 cases. Eur J Obstet Gynecol Reprod Biol. 2013;168:87Y91. 26. Tsai HW, Ko CC, Yeh CC, et al. Unilateral salpingo-oophorectomy as fertility-sparing surgery for borderline ovarian tumors. J Chin Med Assoc. 2011;74:250Y254. 27. Longacre TA, McKenney JK, Tazelaar HD, et al. Ovarian serous tumors of low malignant potential (borderline tumors): outcome-based study of 276 patients with long-term (9or =5-year) follow-up. Am J Surg Pathol. 2005;29: 707Y723. 28. Gershenson DM, Silva EG, Levy L, et al. Ovarian serous borderline tumors with invasive peritoneal implants. Cancer. 1998;82:1096Y1103. 29. Kane A, Uzan C, Rey A, et al. Prognostic factors in patients with ovarian serous low malignant potential (borderline) tumors with peritoneal implants. Oncologist. 2009;14:591Y600. 30. Trope C, Kaern J, Vergote IB, et al. Are borderline tumors of the ovary overtreated both surgically and systemically? A review of four prospective randomized trials including 253 patients with borderline tumors. Gynecol Oncol. 1993;51:236Y243. 31. Faluyi O, Mackean M, Gourley C, et al. Interventions for the treatment of borderline ovarian tumours. Cochrane Database Syst Rev. 2010:CD007696.
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
Risk Factors for Progression to Invasive Carcinoma in Patients With Borderline Ovarian Tumors Taejong Song, MD,* Yoo-Young Lee, MD, PhD,Þ Chel Hun Choi, MD,Þ Tae-Joong Kim, MD,Þ Jeong-Won Lee, MD, PhD,Þ Duk-Soo Bae, MD, PhD,Þ and Byoung-Gie Kim, MD, PhDÞ
Objective: The aim of this study was to identify risk factors for progression to invasive carcinoma in patients with borderline ovarian tumors (BOTs). Methods: We performed a retrospective review of all patients treated and followed for BOTs between 1996 and 2011. Multivariate Cox proportional hazards model analysis was performed to identify independent risk factors for progression to invasive carcinoma. Results: A total of 364 patients were identified. During the median follow-up of 53.8 months, 31 patients (8.5%) developed recurrent disease: 12 (3.3%) had recurrent disease with progression to invasive carcinoma, and 19 (5.2%) had recurrent disease with borderline histology. Disease-related deaths (7/364; 1.7%) were observed only in patients with progression to invasive carcinoma. The multivariate analysis showed that independent risk factors for progression to invasive carcinoma were advanced disease stage (hazard ratio [HR], 5.59; P = 0.005), age 65 years or older (HR, 5.13; P = 0.037), and the presence of microinvasion (HR, 3.71; P = 0.047). These 3 factors were also independently related to overall survival. Conclusions: Although patients with BOTs have an excellent prognosis, the risk of progression to invasive carcinoma and thereby death remains. Therefore, physicians should pay closer attention to BOT patients with these risk factors (ie, advanced disease stage, old age, and microinvasion), and more careful surveillance for progression to invasive carcinoma is needed. Key Words: Borderline ovarian tumors, Risk factors, Recurrence, Progression to invasive carcinoma Received March 4, 2014, and in revised form March 13, 2014. Accepted for publication June 12, 2014. (Int J Gynecol Cancer 2014;24: 1206Y1214)
1
ovarian tumors (BOTs), first described by Taylor B orderline in 1929, comprise about 10% to 20% of all epithelial ovarian malignancies.2,3 As opposed to patients with invasive ovarian
*Department of Obstetrics and Gynecology, Kangbuk Samsung Hospital, Sungkyunkwan University School of Medicine, Seoul, Republic of Korea; and †Department of Obstetrics and Gynecology, Samsung Medical Center, Sungkyunkwan University of School of Medicine, Seoul, Republic of Korea. Address correspondence and reprint requests to Byoung-Gie Kim, MD, PhD, Department of Obstetrics and Gynecology, Samsung Medical Center, Sungkyunkwan University School of Medicine, 50 Irwon-Dong, Gangnam-gu, Seoul 135-710, Republic of Korea. E-mail: [email protected]. The authors declare no conflicts of interest. Copyright * 2014 by IGCS and ESGO ISSN: 1048-891X DOI: 10.1097/IGC.0000000000000216
1206
carcinomas, those with BOTs tend to present at an earlier stage and at a younger age.4Y6 Patients with BOTs have a much better prognosis than those with invasive carcinomas,4,6 and 5- and 10-year survival rates for stages I, II, and III disease are 99% and 97%, 98% and 90%, and 96% and 88%, respectively.7 Despite these favorable data, some patients relapse or die of the disease. Recurrences in patients with BOTs were subclassified into 2 types: (1) borderline-type recurrence and (2) invasive-type recurrence (namely, progression to invasive carcinoma). The prognosis of patients with borderline-type recurrence is excellent, and therefore, a second round of fertility-sparing surgery is undertaken in cases of those with a strong desire for future childbearing.8 In contrast, progression to invasive carcinoma could affect patient survival and is regarded as a cause of disease-related deaths in patients with BOTs.9 However, risk factors for progression to invasive carcinoma have not yet been established. Therefore, the aim of
International Journal of Gynecological Cancer
& Volume 24, Number 7, September 2014
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 24, Number 7, September 2014 Invasive Recurrence of Borderline Ovarian Tumors
this study was to determine the risk factors for progression to invasive carcinoma in patients with BOTs.
METHODS Using the tumor registry at the Samsung Medical Center (Seoul, Korea), we retrospectively identified patients with pathologically confirmed BOTs who were treated and followed up between January 1996 and December 2011. Patients who did not receive primary treatment or follow up for at least 6 months at the Samsung Medical Center were excluded from the analysis. The study was approved by the institutional review board. Surgical treatment was classified as either radical (bilateral salpingo-oophorectomy with or without hysterectomy) or fertility sparing (defined as conservation of the uterus and at least a portion of one ovary). The completeness of the surgical staging was evaluated according to the International Federation of Gynecology and Obstetrics (FIGO) guidelines10 but modified with respect to lymphadenectomy.5,11,12 Complete staging included removal of all visible tumor tissue, peritoneal washing cytology, omentectomy, random multiple peritoneal biopsies, and appendectomy (at least in mucinous tumor types). Extraovarian implants were divided histologically into noninvasive and invasive implants by the absence or presence of destructive stromal invasion into the underlying tissue. The ovarian stromal microinvasion was defined as the presence of microscopic foci of stromal tumor invasion, less than 3 mm in the longest linear dimension and 10 mm2 or less in area, by single cells and nests of moderately atypical cells.13 The criteria for the diagnosis of micropapillary pattern were (a) tumors showing a filigree pattern of small, uniform, elongated, stroma-poor, or stroma-free papillae, at least five times as long as wide, arising directly in a nonhierarchical fashion from large fibrotic, edematous, or myxoid papillary stalks or from cyst walls, and/or a cribriform pattern of epithelial cells lining stalks or cyst walls, and (b) the presence of either or both patterns on a confluent area at least 5 mm in greatest dimension on at least 1 slide.14,15 Following completion of the primary treatment, patients were examined every 3 months during the first 2 years, every 6 months during the next 3 years, and yearly thereafter. Posttreatment surveillance included physical examination, assessment of tumor markers, and imaging studies. Recurrence was defined as the detection of the same tumor cell histology after an apparently complete surgical resection. Statistical analysis was performed with SPSS software (version 12.0; SPSS, Inc, Chicago, IL). The tumor recurrence rate, progression to invasive carcinoma rate, and mortality rate were analyzed. Invasive recurrence-free survival (iRFS) was calculated from the date of surgery to the date of invasive recurrence (progression to invasive carcinoma) or last contact. Overall survival was calculated from the date of surgery to the date of disease-related death or last contact. For each characteristic, the distributions of time to event outcomes were described by the Kaplan-Meier method and compared using the long-rank test. Multivariate Cox proportional hazards models were fitted to compare survival experiences among risks factors while adjusting for the effects of other demographic and clinical characteristics. P G 0.05 was considered statistically significant.
RESULTS Patient Characteristics A total of 364 patients were identified. Clinical and pathologic characteristics for the cohort are listed in Table 1. The median age was 39 years (range, 10Y85 years). The most common reason for the hospital visit was symptoms due to tumors such as pain, palpable mass, distension, or bleeding (245 patients [67.3%]), followed by the diagnosis being the coincidental outcome of a medical checkup (71 patients [19.5%]). The median levels of serum CA-125 and CA-19-9 were 29.2 U/mL (range, 0.6Y7810.0 U/mL) and 19.3 U/mL (range, 0.1Y56,430.0 U/mL), respectively. All patients underwent surgery as the primary treatment. A total of 173 patients (47.5%) with a median age of 50 years underwent radical surgery, whereas 191 (52.5%) with a median age of 30 years underwent fertility-sparing surgery (144 unilateral salpingo-oophorectomy and 47 cystectomy). The staging procedure was performed in 223 patients (61.3%). Pelvic lymphadenectomy was performed in 51 patients (14.0%), whereas pelvic and para-aortic lymphadenectomy was performed in 6 patients (1.6%). At the completion of the primary surgery (and restaging procedure, if any), 361 patients (99.2%) had no residual tumor, and the remaining 3 (0.8%) had residual tumors with a diameter ranging from a few millimeters to 1 cm. The majority of patients had early-stage disease; 328 patients (90.1%) had stage I disease, 8 (2.2%) had stage II disease, and 28 (7.7%) had stage III disease. The most common sites of extraovarian implants were omentum, pelvic peritoneum, and fallopian tubes. Two patients had metastasis in the pelvic lymph node. The most common histology was mucinous (246 patients [67.6%]), followed by serous (109 patients [29.9%]), endometrioid (4 patients [1.1%]), clear cell (4 patients [1.1%]), and fibroma (1 patient, 0.03%). According to final pathologic diagnoses, there were 42 cases of microinvasion (11.5%), 21 cases of micropapillary pattern (5.8%), 27 cases of intraepithelial carcinoma (7.4%), 28 cases of noninvasive implants (7.7%), and 8 cases of invasive implants (2.2%). Of 42 cases with microinvasion, there were mucinous histology in 21 cases (50.0%), serous histology in 19 cases (45.2%), endometrioid histology in 1 case (2.4%), and clear cell histology in 1 case (2.4%). Postoperative chemotherapy was administrated to 37 patients in the form of platinum-based regimens; chemotherapy was mainly indicated in patients with advanced-stage disease. No chemotherapy-related morbidity or death was observed.
Recurrence, Progression to Invasive Carcinoma, and Mortality Rates During the median follow-up of 53.8 months (range, 6.4Y206.5 months), 31 patients (8.5%) developed recurrent disease. Of these 31 patients, 12 (3.3%) had recurrent disease with progression to invasive carcinoma, and 19 (5.2%) had recurrent disease with borderline histology. When classified according to the histology of recurrent diseases, 13 patients had mucinous BOTs, 17 patients had serous BOT, and 1 patient had endometrioid BOT. Table 2 shows the characteristics of patients with progression to invasive carcinoma. At the time of the last follow-up, 337 patients (92.6%) had no evidence of disease, 3 patients
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1207
Song et al
International Journal of Gynecological Cancer
& Volume 24, Number 7, September 2014
TABLE 1. Characteristics of 364 patients with BOT No. Patients (%) or Median (Range) Age at diagnosis, y G65 y Q65 y Body mass index, kg/m2 Diagnosis reached through Checkup Symptoms Fortuitous Unknown CA-125,* U/mL G35 U/mL Q35 U/mL CA-19-9,* U/mL G37 U/mL Q37 U/mL Tumor size, cm Surgical approach Laparoscopy Laparotomy Surgical radicality Radical surgery Fertility-sparing surgery Staging procedure Histology Serous Mucinous Other Disease stage I II-III Microinvasion Micropapillary Intraepithelial carcinoma Bilaterality Extraovarian implant No implant Noninvasive implant Invasive implant Adjuvant chemotherapy Follow-up, mo Recurrence with borderline or invasive histology Recurrence with borderline histology Recurrence with invasive carcinoma histology Disease-specific mortality
39 326 38 22.0
(10Y85) (89.6%) (10.4%) (14.2Y35.5)
71 245 33 15 29.2 182 144 19.3 178 86 12.0
(19.5%) (67.3%) (9.1%) (4.1%) (0.6Y7810.0) (55.8%) (44.2%) (0.1Y56,430.0) (67.4%) (32.6%) (2.0Y39.0)
86 (23.6%) 278 (76.4%) 173 (47.5%) 191 (52.5%) 223 (61.3%) 109 (29.9%) 246 (67.6%) 9 (2.5%) 328 36 42 21 27 37
(90.1%) (9.9%) (11.5%) (5.8%) (7.4%) (10.2%)
328 (90.1%) 28 (7.7%) 8 (2.2%) 37 (10.2%) 58.3 (6.4Y206.5) 31 (8.5%) 19 (5.2%) 12 (3.3%) 7 (1.9%)
*Preoperative CA-125 and CA-19-9 were determined in 326 and 264 patients, respectively.
1208
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
45
68
29
79
77
18 50
76
53
35
74
2
3
4
5
6
7 8
9
10
11
12
IIIC, S
IIC, E
IB, M
IA, M
IA, M IIIC, S
IA, S
IA, M
IIIC, S
IIIC, S
IA, M
IIIB, S
USO
V
Invasive
TH, BSO, staging, chemotherapy
TH, BSO, chemotherapy Noninvasive USO
V
V USO Noninvasive TH, BSO, staging, chemotherapy V TH, BSO, staging
V
Invasive
TH, BSO, staging, chemotherapy TH, BSO, staging, chemotherapy BSO
TH, BSO, staging
V Invasive
TH, BSO, staging
Noninvasive
Primary Treatment
20.7
7.6
18.0
6.0
38.6 76.9
13.4
30.3
32.9
20.9
13.8
20.3 Debulking, chemotherapy Chemotherapy
LAR, chemotherapy
Treatment at Progression
Contralateral ovary, peritoneal spread Pelvis, abdomen
Pelvis, peritoneal spread Omentum
Debulking, chemotherapy TH, USO, debulking, chemotherapy Chemotherapy
Chemotherapy
Debulking, chemotherapy TH, staging, chemotherapy Liver surface, pelvis, Refusal of treatment abdomen Contralateral ovary TH, USO, staging Pleura VATS, chemotherapy
Pelvic lymph node, peritoneal spread Vagina, peritoneal spread Pelvis
Rectosigmoid surface Pelvis, omentum
Interval to Progression, mo Site of Progression
DOD
DOD
20.6 36.0
DOD
DOD
NED AWD
DOD
AWD
DOD
DOD
AWD
NED
49.1
17.3
63.9 132.3
18.0
66.9
86.5
53.7
43.2
28.5
Follow-up Current Period, mo Status
AWD, alive with disease; DOD, died of disease; E, endometrioid; LAR, lower, anterior resection; M, mucinous; NED, no evidence of disease; S, serous; TH, total hysterectomy; BSO, bilateral salpingo-oophorectomy; USO, unilateral salpingo-oophorectomy; VATS, video-assisted thoracoscopic surgery.
43
1
Age at Stage and Extraovarian Patient No. Diagnosis, y Histology Implant
TABLE 2. Main characteristics of BOT patients with progression to carcinoma
International Journal of Gynecological Cancer
& Volume 24, Number 7, September 2014 Invasive Recurrence of Borderline Ovarian Tumors
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1209
International Journal of Gynecological Cancer
Song et al
(0.8%) were alive with disease, 7 patients (1.9%) were dead of disease, and 17 patients (4.7%) were dead of other causes. All 7 patients who died of disease had progression to invasive carcinoma. No disease-related death was observed in patients who had only borderline-type recurrence. The median time of progression to invasive carcinoma was 20.5 months, with a range of 6.0 to 76.9 months. Treatment at progression was as follows: 7 patients received surgery and chemotherapy, 3 received only salvage chemotherapy, 1 received only surgical treatment, and the remaining 1 refused any treatment after biopsy confirmation. The
& Volume 24, Number 7, September 2014
median iRFS was 57.9 months, and the 5- and 7-year iRFS rates were 96.5% and 95.7%, respectively.
Risk Factors for Progression to Invasive Carcinoma Univariate and multivariate analyses were performed to identify risk factors for progression to carcinoma. These are outlined in Table 3. In the univariate analysis, advanced FIGO stage, age 65 years or older, CA-125 35 U/mL or greater, the presence of invasive implants, microinvasion, micropapillary pattern, and adjuvant chemotherapy were significant risk
TABLE 3. Univariate and multivariate analysis of progression to carcinoma
Variable FIGO stage I II-III Age G65 y Q65 y CA-125 G35 U/mL Q35 U/mL CA-19-9 G37 U/mL Q37 U/mL Surgical approach Laparotomy Laparoscopy Surgical radicality Radical surgery Fertility-sparing surgery Staging procedure Yes No Histology Serous Mucinous Invasive implants Yes No Microinvasion Micropapillary patterns Intraepithelial carcinoma Adjuvant chemotherapy Yes No
Univariate
Multivariate*
HR
95% CI
1 9.45
3.05Y29.34
1 7.06
HR
95% CI
P
G0.001
5.59
1.70Y18.36
0.005
2.24Y22.30
0.001
5.13
1.10Y23.93
0.037
1 3.75
1.02Y13.85
0.047
1.77
0.38Y8.24
0.464
1 1.01
0.25Y4.02
0.994
1 1.89
0.57Y6.29
0.303
2.23
0.67Y7.41
0.191
1.09
0.35Y3.45
0.880
2.68 1
0.82Y8.79
0.104
15.94 1 8.42 4.85 2.69
4.29Y59.27
G0.001
2.71Y26.13 1.31Y17.92 0.59Y12.29
G0.001 0.018 0.202
3.71 1.18
1.02Y13.53 0.23Y5.99
0.047 0.845
1.87Y18.63
0.002
5.90 1
P
*Multivariate analysis was applied only for P G 0.05 at the univariate analysis.
1210
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 24, Number 7, September 2014 Invasive Recurrence of Borderline Ovarian Tumors
FIGURE 1. Kaplan-Meier plot. A, Progression to invasive carcinoma by stage. B, Progression to invasive carcinoma by age. C, Progression to invasive carcinoma by microinvasion. D, Overall survival by stage. E, Overall survival by age. F, Overall survival by microinvasion. factors for progression to invasive carcinoma (Figs. 1AYC). Univariate analysis for residual disease was not performed as the number of patients with residual disease was too small. Risk factors that had significance upon the univariate analysis were included in the multivariate analysis. Because most patients who received adjuvant chemotherapy had an advanced stage of disease, the iRFS of patients who received adjuvant chemotherapy was lower, and this factor was therefore excluded from the multivariate analysis. Because an extraovarian invasive implant is a surrogate for advanced-stage disease, this was also excluded from the multivariate analysis. Multivariate analysis showed that independent risk factors for progression to invasive carcinoma were advanced FIGO stage (hazard ratio [HR], 5.59; 95% confidence interval [CI], 1.70Y 18.36; P = 0.005), age 65 years or older (HR, 5.13; 95% CI, 1.10Y 22.30; P = 0.037), and the presence of microinvasion (HR, 3.71; 95% CI, 1.02Y13.53; P = 0.047) (Table 3). Moreover, in the multivariate analysis of overall survival, advanced FIGO stage (HR, 11.27; 95% CI, 2.36Y53.71; P = 0.002), age 65 years or older (HR, 6.90; 95% CI, 1.33Y35.95; P = 0.022), and the presence of microinvasion (HR, 5.27; 95% CI, 1.18Y23.64; P = 0.031) remained as significant risk factors (Table 4; Figs. 1DYF).
DISCUSSION This study was performed to determine the risk factors for progression to invasive carcinoma in 364 patients with BOTs. We found that the progression rate was 3.3%, and the independent risk factors for progression were advanced FIGO stage, age 65 years or older, and the presence of microinvasion. These 3 factors were also independently related to overall survival. We also found that no patient with borderline-type recurrence died of the disease, and disease-related death was observed only in patients with progression to invasive carcinoma. Identification of patients with high-risk factors (ie, advanced stage, old age, and microinvasion) is essential for offering more selective treatment or monitoring to prevent progression to invasive carcinoma. This study is the first to determine the risk factors for progression to invasive carcinoma in patients with BOTs. In our study, the risk of progression to invasive carcinoma was estimated as 3.3%. This rate was similar with 2 previous studies.9,12 Zanetta et al12 reported a 2.1% progression rate in 339 BOT patients with a median follow-up of 70 months. Morice et al9 reported a 2% to 3% progression rate in their review article in which 47 cases of progression to
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1211
International Journal of Gynecological Cancer
Song et al
& Volume 24, Number 7, September 2014
TABLE 4. Univariate and multivariate analysis of overall survival
Variable FIGO stage I II-III Age G65 y Q65 y CA-125 G35 U/mL Q35 U/mL CA-19-9 G37 U/mL Q37 U/mL Surgical approach Laparotomy Laparoscopy Surgical radicality Radical surgery Fertility-sparing surgery Staging procedure Yes No Histology Serous Mucinous Invasive implants Yes No Microinvasion Micropapillary patterns Intraepithelial carcinoma Adjuvant chemotherapy Yes No
Univariate
Multivariate*
HR
95% CI
P
HR
95% CI
P
1 11.69
2.61Y52.32
0.001
11.27
2.36Y53.71
0.002
1 14.35
3.12Y64.42
0.001
6.90
1.33Y35.95
0.022
1 3.01
0.58Y15.52
0.188
1 2.08
0.23Y18.52
0.511
1 3.29
0.73Y14.91
0.123
1 2.68
0.52Y13.53
0.239
1 1.13
0.25Y5.07
0.870
4.07 1
0.74Y22.28
0.106
31.39 1 11.03 2.47 2.01
7.00Y140.74
G0.001
2.47Y49.33 0.29Y20.95 0.29Y14.00
0.002 0.407 0.480
5.27
1.18Y23.64
0.031
2.20Y44.18
0.003
9.85 1
*Multivariate analysis was applied only for P G 0.05 at the univariate analysis.
invasive carcinoma reported in 21 articles published since 2009 were analyzed. In our study, one interesting observation was that serous BOTs seemed to be more prone to progress to invasive carcinoma (5.5% vs 2.0%; P = 0.104). However, this observation is inconsistent with other studies.16,17 Koskas et al16 reported that the 10-year cumulative risk of recurrence in the form of invasive carcinoma was 13% in 97 patients with mucinous BOT and concluded that, unlike serous BOT, mucinous BOT does not appear to be such a ‘‘safe’’disease.16 We cannot determine why this discrepancy exists. However, Zanetta et al12 reported similar results to ours, in that they
1212
observed that the rate of progression to invasive carcinoma in serous BOT was higher than that of the mucinous counterpart (2.4% vs 1.6%). In this study, advanced FIGO stage was an independent risk factor for progression to invasive carcinoma and diseasespecific mortality. Our finding was supported by previous studies. The study by Ren et al18 evaluated 234 patients with BOTs and showed that advanced FIGO stage was independently related to recurrence-free survival (P G 0.001). Cusido et al19 analyzed 457 patients with BOTs from 27 centers of Spain and concluded that advanced FIGO stage was an independent risk * 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
International Journal of Gynecological Cancer
& Volume 24, Number 7, September 2014 Invasive Recurrence of Borderline Ovarian Tumors
factor for recurrence. Sornsukolrat and Tuipae20 evaluated 210 patients with BOTs and concluded that advanced FIGO stage was a significant risk factor for disease-specific mortality as well as for recurrence. We believe that FIGO stage is of paramount importance for the prognosis of patients with BOTs. Although old age has been widely known as a risk factor for recurrence or disease-specific mortality in patients with cervical cancer,21 endometrial cancer,22 and ovarian cancer,23 no data are available on the effect of old age on BOT recurrence or survival. To date, a large number of studies on risk factors for recurrence or survival in patients with BOTs have been conducted.4,8,9,11,16,18,24 Among them, some studies included age as a variable8,18,25,26; however, they all failed to reach a conclusion that old age was a risk factor for recurrence or survival. The main reasons for that were that age was treated as a continuous variable,8,11 and an age cutoff of 30 years,25 35 years,26 or 40 years18 was used as a categorical variable. However, 2 studies8,11 showed that age as a continuous variable was a risk factor for recurrence, which was similar with our finding. A controversy exists with regard to the prognostic significance of microinvasion.27 Analysis of the effect of microinvasion on the rate of progression to invasive carcinoma or overall survival is especially difficult because of potential confounding factors, such as micropapillary patterns and presence of extraovarian implants. Therefore, a large sample size and long follow-up period are essential for the study of microinvasion in BOTs. According to a study that analyzed 12 series of serous BOTs with microinvasion (n = 133),9 20 patients had recurrence (15%), including 35% (7/20) with invasive carcinoma at recurrence. The rate of disease-specific mortality was 6% (8/133). In accordance with the findings of previous studies, our data indicate that microinvasion is a risk factor for progression to invasive carcinoma (HR, 3.71; 95% CI, 1.02Y13.53) and disease-specific mortality (HR, 5.27; 95% CI, 1.18Y23.64). With regard to the use of postoperative chemotherapy, we noticed that in our study population the incidence of recurrence, progression to invasive carcinoma, or death was not improved and was rather higher in patients who received chemotherapy when compared with those who were not treated with chemotherapy. This observation seems to confirm the results of 2 other studies8,28 including a report of the M. D. Anderson Cancer Center.28 In other studies,29,30 the rate of recurrence or progression was not different between patients who received chemotherapy and those who did not. Researchers performing a Cochrane meta-analysis of interventions for the treatment of BOTs concluded that evidence did not support the use of any specific type of adjuvant treatment.31 Therefore, adjuvant chemotherapy is currently not believed to play a role in the treatment of BOTs. This study has some limitations. First, this study had a retrospective, single-center design, therefore possibly introducing some degree of bias. Prospective, multicenter trials are needed to confirm our results. Second, the staging procedure was not considered beforehand in all cases so that our data may be biased in this respect. However, Camatte et al24 reported that the absence of a staging procedure in patients with an ‘‘apparent stage I’’ BOT does not modify survival. Nevertheless, this study
has the advantages of including a large sample size and long follow-up periods; this is very important, considering the rarity and late recurrence of BOTs. Moreover, this study is the first study to determine the risk factors for progression to invasive carcinoma in patients with BOTs. In conclusion, although BOTs have an excellent prognosis, they are not exempt from the risk of progression to invasive carcinoma. Identification of patients with high-risk factors is essential in order to prevent their disease-specific mortality. Our results indicate that the risk factors for progression to invasive carcinoma in patients with BOTs are advanced FIGO stage, age 65 years or older, and microinvasion Therefore, physicians should pay closer attention to BOT patients with these risk factors and conduct more careful surveillance for progression to invasive carcinoma.
REFERENCES 1. Taylor HC. Malignant and semimalignant tumors of the ovary. Surg Gynecol Obstet. 1929;48:204Y230. 2. Trope CG, Kristensen G, Makar A. Surgery for borderline tumor of the ovary. Semin Surg Oncol. 2000;19:69Y75. 3. Lim MC, Moon EK, Shin A, et al. Incidence of cervical, endometrial, and ovarian cancer in Korea, 1999Y2010. J Gynecol Oncol. 2013;24:298Y302. 4. Sherman ME, Mink PJ, Curtis R, et al. Survival among women with borderline ovarian tumors and ovarian carcinoma: a population-based analysis. Cancer. 2004;100:1045Y1052. 5. Morice P. Borderline tumours of the ovary and fertility. Eur J Cancer. 2006;42:149Y158. 6. Morris CR, Liu L, Rodriguez AO, et al. Epidemiologic features of borderline ovarian tumors in California: a population-based study. Cancer Causes Control. 2013;24:665Y674. 7. Trimble CL, Kosary C, Trimble EL. Long-term survival and patterns of care in women with ovarian tumors of low malignant potential. Gynecol Oncol. 2002;86:34Y37. 8. Park JY, Kim DY, Kim JH, et al. Surgical management of borderline ovarian tumors: the role of fertility-sparing surgery. Gynecol Oncol. 2009;113:75Y82. 9. Morice P, Uzan C, Fauvet R, et al. Borderline ovarian tumour: pathological diagnostic dilemma and risk factors for invasive or lethal recurrence. Lancet Oncol. 2012;13:e103Y115. 10. Benedet J, Bender H, Jones H, III, et al. FIGO Committee on Gynecologic Oncology. FIGO staging classifications and clinical practice guidelines in the management of gynecologic cancers. Int J Gynecol Obstet. 2000;70:209Y262. 11. Shih KK, Zhou Q, Huh J, et al. Risk factors for recurrence of ovarian borderline tumors. Gynecol Oncol. 2011;120:480Y484. 12. Zanetta G, Rota S, Chiari S, et al. Behavior of borderline tumors with particular interest to persistence, recurrence, and progression to invasive carcinoma: a prospective study. J Clin Oncol. 2001;19:2658Y2664. 13. Scully RE. International Histological Classification of Tumors: Histological Typing of Ovarian Tumours. 2nd ed. Spring Verlag: New York; 1999. 14. Burks RT, Sherman ME, Kurman RJ. Micropapillary serous carcinoma of the ovary: a distinctive low-grade carcinoma related to serous borderline tumors. Am J Surg Pathol. 1996;20:1319Y1330. 15. Eichhorn JH, Bell DA, Young RH, et al. Ovarian serous borderline tumors with micropapillary and cribriform patterns: a
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
1213
Song et al
16.
17.
18.
19.
20.
21.
22.
23.
International Journal of Gynecological Cancer
study of 40 cases and comparison with 44 cases without these patterns. Am J Surg Pathol. 1999;23:397Y409. Koskas M, Uzan C, Gouy S, et al. Prognostic factors of a large retrospective series of mucinous borderline tumors of the ovary (excluding peritoneal pseudomyxoma). Ann Surg Oncol. 2011;18:40Y48. Benito V, Lubrano A, Arencibia O, et al. Serous and mucinous borderline ovarian tumors: are there real differences between these two entities?. Eur J Obstet Gynecol Reprod Biol. 2010;153:188Y192. Ren J, Peng Z, Yang K. A clinicopathologic multivariate analysis affecting recurrence of borderline ovarian tumors. Gynecol Oncol. 2008;110:162Y167. Cusido M, Balaguero L, Hernandez G, et al. Results of the national survey of borderline ovarian tumors in Spain. Gynecol Oncol. 2007;104:617Y622. Sornsukolrat S, Tuipae S. Prognostic factors and survival of borderline ovarian tumors in Rajavithi Hospital between 1979-2006 A.D. J Med Assoc Thai. 2012;95:1141Y1148. Sharma C, Deutsch I, Horowitz DP, et al. Patterns of care and treatment outcomes for elderly women with cervical cancer. Cancer. 2012;118:3618Y3626. Ahmed A, Zamba G, DeGeest K, et al. The impact of surgery on survival of elderly women with endometrial cancer in the SEER program from 1992-2002. Gynecol Oncol. 2008;111:35Y40. Chereau E, Ballester M, Selle F, et al. Ovarian cancer in the elderly: impact of surgery on morbidity and survival. Eur J Surg Oncol. 2011;37:537Y542.
1214
& Volume 24, Number 7, September 2014
24. Camatte S, Morice P, Thoury A, et al. Impact of surgical staging in patients with macroscopic ‘‘stage I’’ ovarian borderline tumours: analysis of a continuous series of 101 cases. Eur J Cancer. 2004;40:1842Y1849. 25. Azuar AS, Bourdel N, Ferrarrese G, et al. Laparoscopic restaging of borderline ovarian tumours (BLOT): a retrospective study of 142 cases. Eur J Obstet Gynecol Reprod Biol. 2013;168:87Y91. 26. Tsai HW, Ko CC, Yeh CC, et al. Unilateral salpingo-oophorectomy as fertility-sparing surgery for borderline ovarian tumors. J Chin Med Assoc. 2011;74:250Y254. 27. Longacre TA, McKenney JK, Tazelaar HD, et al. Ovarian serous tumors of low malignant potential (borderline tumors): outcome-based study of 276 patients with long-term (9or =5-year) follow-up. Am J Surg Pathol. 2005;29: 707Y723. 28. Gershenson DM, Silva EG, Levy L, et al. Ovarian serous borderline tumors with invasive peritoneal implants. Cancer. 1998;82:1096Y1103. 29. Kane A, Uzan C, Rey A, et al. Prognostic factors in patients with ovarian serous low malignant potential (borderline) tumors with peritoneal implants. Oncologist. 2009;14:591Y600. 30. Trope C, Kaern J, Vergote IB, et al. Are borderline tumors of the ovary overtreated both surgically and systemically? A review of four prospective randomized trials including 253 patients with borderline tumors. Gynecol Oncol. 1993;51:236Y243. 31. Faluyi O, Mackean M, Gourley C, et al. Interventions for the treatment of borderline ovarian tumours. Cochrane Database Syst Rev. 2010:CD007696.
* 2014 IGCS and ESGO
Copyright © 2014 by IGCS and ESGO. Unauthorized reproduction of this article is prohibited.
