Endometrial carcinosarcomas have a different prognosis and pattern of spread compared to high-risk epithelial endometrial cancer

Gynecologic Oncology 98 (2005) 274 – 280 www.elsevier.com/locate/ygyno

Endometrial carcinosarcomas have a different prognosis and pattern of spread compared to high-risk epithelial endometrial cancer Frederic Amant a,*, Isabelle Cadron a, Luca Fuso a, Patrick Berteloot b, Eric de Jonge c, Gerd Jacomen d, Johan Van Robaeys e, Patrick Neven a, Philippe Moerman f, Ignace Vergote a a

Division of Gynecological Oncology, Department of Obstetrics and Gynecology, University Hospitals Leuven, Katholieke Universiteit Leuven, Herestraat 49, 3000 Leuven, Belgium b Obstetrics and Gynecology, St. Maarten Hospital, Duffel, Belgium c Obstetrics and Gynecology, Ziekenhuis Oost Limburg, Genk, Belgium d Department of Pathology, St. Maarten Hospital, Duffel, Belgium e Department of Pathology, Ziekenhuis Oost Limburg, Genk, Belgium f Department of Pathology, University Hospitals Leuven, Katholieke Universiteit Leuven, Belgium Received 21 January 2005 Available online 21 June 2005

Abstract Objective. The endometrial origin of uterine carcinosarcoma has recently been well established. The current study investigates whether uterine carcinosarcomas can be included in protocols on high-risk endometrial cancer, given the similarities in biologic behavior of both entities. Methods. Pathological and surgical notes of patients diagnosed with grade 3 endometrioid, carcinosarcoma, serous and clear cell endometrial cancer subtypes were retrospectively analyzed with special attention to the spread pattern of the different subtypes. Information on site of relapse and time to recurrence was obtained. Results. We traced 146 patients of which 9 patients were ineligible. Histological subtypes of the remaining 137 patients were as follows: 50 (37%) grade 3 endometrioid carcinoma, 54 (39%) serous or clear cell carcinoma (non-endometrioid carcinoma), and 33 (24%) carcinosarcomas. Distribution of early stage disease (I and II) was 67, 46, and 78% for grade 3 endometrioid, non-endometrioid, and carcinosarcoma, respectively. Although we could not trace differences in hematogenic and transperitoneal spread among the three subtypes, non-endometrioid and carcinosarcomas were more likely to spread to pelvic and paraaortic lymph nodes (P < 0.01). Using univariate analysis, both stage (P < 0.006, Wald statistic) and histological type appear to determine the outcome, whereas lymphovascular space infiltration (P < 0.25) and age (P < 0.07) were not significantly different between the three histological subtypes. Cox Regression multivariate analysis on 127 women suffering from the three histological subtypes suggested that both stage III – IV disease (P < 0.00001) and histological type (carcinosarcoma) (P < 0.003) were of prognostic significance [hazard ratio (CI 95%) were, respectively, 3.8 (2.1 – 7.0) and 3.2 (1.7 – 5.9)]. Analyzing cases limited to stage I – II endometrial cancer, 24/28 (86%) grade 3 endometrioid, 18/24 (75%) nonendometrioid, and 11/25 (44%) carcinosarcomas survived, suggesting a worse outcome for endometrial carcinosarcoma when compared to the other subtypes (P < 0.008, Log Rank). A higher incidence of pulmonary metastases explained the worse outcome for early stage carcinosarcoma (P < 0.006), whereas the incidence of liver metastasis, transperitoneal spread, or recurrences in lymph nodes or vagina were comparable between the three pathologic subtypes. Conclusions. Although endometrial carcinosarcoma originates from epithelial cancer, the intrinsic more aggressive tumor biology suggests that this subtype should not be incorporated in studies on high-risk epithelial endometrial cancer. D 2005 Elsevier Inc. All rights reserved. Keywords: Carcinosarcoma; Endometrial; Serous; Adjuvant; Chemotherapy

* Corresponding author. Fax: +32 16 34 46 29. E-mail address: [email protected] (F. Amant). 0090-8258/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.ygyno.2005.04.027

F. Amant et al. / Gynecologic Oncology 98 (2005) 274 – 280

Introduction Uterine sarcomas encompass leiomyosarcoma, carcinosarcoma (CS), and endometrial stromal sarcoma according to traditional classification systems. The biphasic cellular population found in CS entered the sphere of interest of many researchers, who focused on the question whether this tumor originated from one (monoclonal theory) or two cell populations. A wide spectrum of methods has been elaborated to answer this question. During the last 15 years, epidemiological, clinicopathogical, immunohistological, in vitro, in vivo, and molecular genetic research have found arguments to support the monoclonal nature of CS that pointed towards an endometrial origin. Recent textbooks therefore classify CS as a subtype of endometrial cancer [1]. Consequently, the designation endometrial CS corresponds best to its tissue origin and should be used instead of uterine CS. Endometrial CS is an aggressive tumor and better treatment modalities are necessary. The relatively rare occurrence of CS adds to the observation that only two randomized trials currently are available. Also, many reports include a variety of tumors and conclusions for endometrial CS are often difficult to draw. From the clinical point of view, it would be interesting to know whether the endometrial origin of CS is a sufficient strong argument to incorporate CS in trials on endometrial serous/clear cell carcinoma. These endometrial carcinomas share two characteristics with CS in that they are rare and notorious for their aggressive behavior. Moreover, the need for trials in serous endometrial cancer has recently been underscored [2]. The inclusion of both endometrial CS and serous/clear cell carcinoma in a single trial might facilitate the conductance of such a trial. Therefore, to test whether CS and epithelial high-risk endometrial carcinoma could be incorporated in the same studies on high-risk endometrial cancer, we looked at patterns of metastatic spread and prognosis of these tumor types.

Patients and methods Three centers with special interest in gynecologic oncology collaborated in this study. Pathology records were retrospectively searched from 1990 to 2004 for grade 3 endometrioid endometrial carcinoma (G3 EC), non-endometrioid carcinoma (NEC), and CS. Pure clear cell carcinoma, pure serous carcinoma, and mixed forms, irrespective of the percentage of each component, were included in the NEC. CS slides were reviewed by one pathologist (P.M.), using the criteria as defined by Ronnett et al. [1], whereas all other slides were reassessed by the local pathologist. All hospital records were retrospectively analyzed for clinical, surgical, and histopathological data with special

275

attention to the spread pattern of the different subtypes. Furthermore, localization of recurrent disease (hematogenic versus local and lymphatic) and survival was noted and compared between the different subtypes. Patients were staged according to the 1988 FFe´de´ration International de Gyne´cologie et Obste´trie_ staging system (FIGO) for endometrial cancer: stage I disease confined to the uterus; stage II confined to uterus and cervix; stage III confined to the pelvis; and stage IV extra pelvic disease. Staging information was derived from surgical notes and pathology reports. The main clinical outcome measures included follow-up in months, age, hormonal treatment at diagnosis, menopausal status, primary and adjuvant treatment, site of recurrence, progression free interval, treatment of recurrence, actuarial survival, and current status. The main histological outcomes included: histological subtype, lymphovascular invasion, cervical involvement, involvement of intra-abdominal organs, lymph node status, and cytological examination of peritoneal fluid. Pre-operative examinations consisted of blood sampling (full blood count, liver function tests, kidney function, and electrolytes), chest X-ray, transvaginal ultrasound, and whole abdominal and pelvic computer tomography. The surgical procedure included sampling of peritoneal cytology, total hysterectomy, bilateral salpingo-oophorectomy, and pelvic lymph node dissection. Paraaortic lymphadenectomy inferior to the inferior mesenteric artery was performed when the pelvic lymph nodes were macroscopically positive, or in case of macroscopic metastasis to the ovaries or invasion of the uterine serosa. In case of NEC or CS, we performed an omentectomy. However, in a multicentric setting and given the 14-year period studied, some women have received a less stringent surgical staging procedure. Adjuvant treatment was mainly based on surgical and histopathological findings according to the following guidelines. In case of positive lymph nodes, external radiotherapy was given in fractions of 2 Gy, totaling 50 Gy. In case of early stage NEC, three cycles of adriamycin (60 mg/m2)– cisplatin (50 mg/m2) were administered post-operatively. All patients were followed every 3 months during the first 2 years, every 6 months during the following 3 years and yearly after 5 years. Statistical analysis Fisher’s exact or Chi-square test were used for pairs and categorical variables whereas the Independent-Samples t test was used for variables with a continuous distribution. Overall survival was calculated from the date of diagnosis until death from any cause, or censored at last follow-up using the Kaplan– Meier product-limit method. Comparison of survival curves was performed through the LogRank test. The prognostic effect of each factor was first tested in a univariate setting and then a multivariate

276

F. Amant et al. / Gynecologic Oncology 98 (2005) 274 – 280

Table 1 Characteristics of the patients according to histological subtype Grade 3 EC, n = 50 Age in years, mean (range) Hormonal treatment at diagnosis, n (%) Primary treatment, n (%) Surgery Chemotherapy Radiotherapy Palliation Stage, n (%) I – II III – IV Median survival, months (%) Dead of disease, n (%) Duration of follow-up, months (CI 95%)

NEC, n = 54

66 (44 – 86) 4/26 (15.4)

CS, n = 33

67 (31 – 87) 1/35 (2.9)

All, n = 137

68 (50 – 86) 1/23 (4.3)

41/50 5/50 2/50 2/50

(82) (10) (4) (4)

38/54 11/54 3/54 2/54

(70.4) (20.4) (5.5) (3.7)

27/33 4/33 1/33 1/33

(81.8) (12.1) (3.0) (3.0)

32/48 16/48 60 14/50 25

(66.7) (33.3) (68.9) (28) (18 – 31)

24/52 28/52 40 23/54 28

(46.2) (53.8) (57.4) (42.6) (22 – 35)

25/32 7/32 23 19/33 28

(78.1) (21.9) (42.4) (57.6) (16 – 39)

67 (31 – 87) 6/84 (7.1) 106/137 20/137 6/137 5/137 81/132 51/132 38 56/137 27

P valuea 0.40 0.14

(77.4) (14.6) (4.4) (2.9)

0.43

(61.4) (38.6) (57.5) (40.9) (22 – 31)

0.009 0.14 0.01 0.82

EC, endometrioid carcinoma subtype; NEC, non-endometrioid carcinoma subtype; CS, carcinosarcoma subtype. a 2 v test.

analysis was performed on the variables with a P value lower than 0.05. Cox proportional hazards regression modeling was used for both univariate and multivariate analysis and P values were obtained by means of a Wald statistic. Factors included in these analyses were stage, age, and histological type. A two-sided P value of less than 0.05 was considered significant, and 95% Confidence Intervals (CIs) were quoted. All end points were updated in June 2004. Analysis was performed using SPSS package version 12 (SPSS Inc., IL) and S-Plus (Insightful Corp., WA).

Results We traced 146 patients of which 9 patients were ineligible because there were no follow-up data. Histological subtypes of the remaining 137 patients were as follows:

50 (37%) grade 3 endometrioid carcinoma, 54 (39%) serous or clear cell carcinoma (NEC), and 33 (24%) CS. The most important clinico-pathological characteristics of these subgroups are depicted in Table 1. Distribution of early stage disease (I and II) was 67, 46, and 78% for G3 EC, NEC, and CS, respectively, suggesting that endometrial CS is more likely to be diagnosed in an early stage when compared to NEC. Patterns of spread were obtained from both pre- and postoperative findings that are shown in Table 2. Although we could not trace differences in hematogenic and transperitoneal spread among the three subtypes, NEC and CS were more likely to spread to pelvic and paraaortic lymph nodes (P < 0.01, Fisher’s exact test). After a median followup of 24 months (range, 18 – 31) 14/50 (28%) patients with G3 EC died of disease. After a mean follow-up of 28 months, 23/54 (43%) of women with NEC and 19/33 (58%) with CS died of their disease (Table 1). Calculated by univariate analysis, both stage (P < 0.006, Wald statistic)

Table 2 Tumor spread in primary setting Grade 3 EC

NEC

CS

P valuea

All

Pre-operative findings Hematogenous metastatic spread Lung Liver Other

8/50 1/50 3/50 4/50

(16.0) (2.0) (6.0) (8.0)

9/54 3/54 1/54 5/54

(16.7) (5.6) (1.9) (9.3)

6/33 4/33 1/33 1/33

(18.2) (12.1) (3.0) (3.0)

23/137 8/137 5/137 10/109

(16.8) (5.8) (3.6) (7.3)

0.96 0.15 0.51 0.54

Post-operative findings Lymphatic metastatic spread Pelvic Paraaortic Transperitoneal metastatic spread Ovary Omentum Peritoneum

2/27 2/25 0/3 8/43 4/41 2/8 5/43

(7.4) (8.0) (0) (18.6) (9.8) (25) (11.6)

13/31 13/30 4/5 11/43 8/37 5/10 8/43

(41.9) (43.3) (80) (25.5) (21.6) (50) (18.6)

4/15 4/15 2/2 9/30 2/30 1/7 9/30

(26.7) (26.6) (100) (30) (6.7) (14.3) (30)

19/73 19/70 9/10 28/116 14/108 8/25 16/101

(26) (27.1) (90) (24.1) (13.0) (32) (15.8)

0.01 0.01 0.03 0.21 0.45 0.26 0.14

Results are presented as n (%). EC, endometrioid carcinoma subtype; NEC, non-endometrioid carcinoma subtype; CS, carcinosarcoma subtype. a 2 v test.

F. Amant et al. / Gynecologic Oncology 98 (2005) 274 – 280

and histological type (P < 0.06, Wald statistic) appear to determine the outcome, whereas lymphovascular space infiltration (P < 0.25) and age (P < 0.07) were not significantly different between the three histological subtypes (Table 3). In univariate analysis, a hazard ratio for recurrence free survival of 1.9 (CI 95% 0.96– 3.81) was calculated for women suffering from endometrial CS when compared to G3 EC and NEC (Table 3). Cox Regression multivariate analysis on 127 women suffering from the three histological subtypes suggested that both stage III – IV disease (P < 0.00001) and histological type (CS) (P < 0.003) were of prognostic significance [hazard ratio (CI 95%) were, respectively, 3.8 (2.1 –7.0) and 3.2 (1.7 – 5.9)]. When this analysis was repeated on a series (n = 78) after the exclusion of G3 EC, CS had a significantly worse outcome when compared to NEC (P < 0.003) (hazard ratio 3.1, CI 95% 1.5– 6.8) (Table 4). Interestingly, even for stage I– II disease, CS appeared to have a worse outcome when compared to the other subtypes (Fig. 1). According to this analysis limited to stage I– II endometrial cancer, 24/28 (86%) G3 EC, 18/24 (75%) NEC, and 11/25 (44%) CS survived, suggesting a worse outcome for CS when compared to the other subtypes (P < 0.008, Log Rank) (Fig. 1). Overall survival in all stages suggested a trend (both lines do not touch) towards a worse outcome of CS (n = 33) when compared to serous cancer (n = 34) (P < 0.25, Log Rank) (Fig. 2). To investigate the worse outcome of early stage CS further, we investigated patterns of relapse in cases that were diagnosed primarily as stage I– II disease. Since lung metastases were detected in 0/21 (0%) G3 EC, 0/ 21 (0%) NEC, and 5/22 (23%) CS, it appears that a higher incidence of pulmonary metastases explained the worse outcome for early stage CS (P < 0.006). The incidence of liver metastasis, transperitoneal spread, or recurrences in

Table 3 Univariate analysis of prognostic factors Hazard ratio Stage (n = 129) I – II III – IV LVSI (n = 46) Negative Positive Age in years (n = 131) <70 70 <60 60 Histological type (n = 132) Grade 3 EC NEC vs. Grade 3 EC + CS CS vs. Grade 3 EC + NEC

1 2.6 1 29.2 1 1.62 1 1.22 1 1.23 1.90

CI 95%

1.5 – 4.4

P valuea

0.006

0.08 – 1.0214

0.25

0.96 – 2.77

0.07

0.60 – 2.50

0.58

0.63 – 2.39 0.96 – 3.81

0.53 0.06

EC, endometrioid carcinoma subtype; NEC, non-endometrioid carcinoma subtype; CS, carcinosarcoma subtype; LVSI, lymphovascular space infiltration; vs., versus. a Wald statistic.

277

Table 4 Multivariate prognostic model as calculated by Cox Regression analysis

Stage I – II III – IV Histological type (n = 78) Non-endometrioid Carcinosarcoma a

Hazard ratio

CI 95%

P valuea

1 2.9

1.4 – 6.4

0.0052

1 3.1

1.5 – 6.8

0.0032

Wald statistic.

lymph nodes or vagina was comparable between the three pathologic subtypes (data not shown).

Discussion The current study was conducted to test the hypothesis that endometrial CS could be incorporated together with serous/clear cell carcinomas in the same trials. The rationale to do so is the endometrial origin of CS and the advantage of this policy would be the higher accrual rate for studies in uncommon disease entities. However, the results point towards a more aggressive behavior for CS. Limiting the analysis to early stage disease, CS had a worse outcome when compared to NEC (Fig. 1) and this seemed mainly due to a higher incidence of pulmonary metastases after primary surgery. The intrinsic more aggressive behavior of CS suggests not to include this subtype in trials on high-risk endometrial cancer. One previous report addresses a similar study, although numbers of NEC were somewhat lower (n = 11 and 14 for serous and clear cell subtype, respectively) [3]. In contrast to the current series, involvement of the upper abdomen frequently contributed to a more advanced stage in CS. However, similar to the current report, the authors also encountered a worse outcome for women suffering from the CS subtype, even after controlling for stage [3]. The tendency for a sarcomatous predominance that we encountered during ovarian CS tumor progression might contribute to a worse outcome in CS [4]. We were able to determine the composition of metastatic ovarian carcinosarcoma since surgery is performed in selected cases of recurrent ovarian cancer whereas this rarely is the case for recurrent endometrial cancer. It is unlikely therefore that we will have a series of biopsies from recurrent endometrial carcinosarcoma in order to compare the composition to the primary biopsy. However, the observation that, also in the ovarian counterpart, CS arises from epithelial cancer [5] enables us to extrapolate these findings to the endometrial subtype. In this study, we have incorporated mixed and pure serous and clear cell endometrial cancer in one group, since three studies previously pointed towards similar outcomes for mixed and pure serous endometrial cancer [6 –8]. In the largest study, no cutoff percentage of the serous component was used (personal communication by Dr. B. Slomovitz),

278

F. Amant et al. / Gynecologic Oncology 98 (2005) 274 – 280

Fig. 1. Actuarial survival in stage I – II disease according to different histotypes ( P < 0.008, Log Rank).

suggesting that even the presence of a small serous component might dictate an aggressive outcome. Although we currently perform systematic lymphadenectomy in high-risk endometrial cancer, only 15/33 (45%) and 30/54 (56%) pelvic lymphadenectomies for CS and NEC were, respectively, performed in this retrospective multicentric study. A multicentric setting and a relative long period of study (14 years) have contributed to an overall suboptimal surgical staging (non-standardized surgical approach). However, a similar percentage (45 vs. 56) of cases not undergoing full lymphadenectomy in both subtypes suggests that the inclusion of suboptimally surgically staged cases will not significantly alter the results. Little is known on effective treatment modalities in endometrial CS since the only two randomized trials [9,10] cannot provide a gain in survival of neither chemotherapy

nor radiotherapy. Of note, these and observational studies in CS are hampered by the inclusion of other uterine sarcomas and an incomplete surgical staging. In a pilot study aiming for both local and systemic control in 38 patients with clinical stage I – II CS, a promising outcome was demonstrated recently. Women completing both radiotherapy and chemotherapy after surgery appeared to have a better outcome when compared to women who did not receive the multimodality protocol [11]. In another pilot study, a comparable approach was used in serous endometrial cancer. Ten women receiving both surgical staging and chemotherapy were compared to 21 historical controls. After a follow-up of 32 months, all study patients survived and the difference was statistically significant [12]. These preliminary results suggest both local and systemic control might improve the long-term outcome of NEC and CS.

Fig. 2. Overall survival in all stages according to the subtype. Although there is a trend of carcinosarcoma (n = 33) to have a worse outcome when compared to serous cancer (n = 34), the difference is not significant ( P < 0.25, Log Rank).

F. Amant et al. / Gynecologic Oncology 98 (2005) 274 – 280

The current study was performed in order to evaluate whether both NEC and CS could be incorporated in the same studies. Surgical staging provides both prognostic information as well as local control and appears to be important in both entities. Prospective studies in NEC and CS should therefore best address the clinical benefit of adjuvant chemotherapy. Although the literature data on spread patterns originate from different studies with limited numbers in some, there appears to be a similarity in spread pattern between CS and NEC. A summary of the literature data is presented in Table 5. From this table, it appears that spread patterns as measured by adnexal, omental, and lymphatic metastasis as well as positive peritoneal cytology are comparable between CS and serous and clear cell endometrial cancer. The tendency for lymphatic and transperitoneal spread underscores the importance of adequate surgical staging in these malignancies. G3 EC appears to have a lesser tendency towards omental and adnexal metastasis (Table 5) [11,13– 22]. Since the spread pattern resembles that of ovarian cancer, it seems logical to offer women suffering from high-risk endometrial cancer (NEC + CS) a similar staging procedure. Therefore, a comprehensive surgical staging is advised including adnexectomy, omentectomy, peritoneal washings, biopsies of suspect lesions, and blind biopsies from the peritoneum in the pelvis, paracolic gutter, and the right hemidiaphragm [23]. Given the propensity for lymphatic spread [8,15,18], a thorough lymphatic dissection is necessary in high-risk endometrial cancer. The importance of performing a formal lymph node dissection for local control in serous endometrial cancer has been suggested recently, since 7 of 40 node-negative stage I patients not receiving adjuvant therapy failed, none to the pelvic sidewall, but 6 ultimately experienced distal failures [2,24]. With regard to the systemic treatment, phase II studies that included all types of endometrial cancer indicate that cisplatin and doxorubicin are the most active agents in endometrial cancer. Carboplatin given in dosages of 300 to 400 mg/m2 every 4 weeks has been associated with response rates similar to cisplatin. Limiting to serous endometrial cancer, small non-randomized trials reported response rates to cisplatin-based therapy of 10– 33%, whereas the reports demonstrate a response rate of only 0– 30% in salvage settings for doxorubicin. In light of the poor response to platinum- and doxorubicin-based therapies, objective response rates of 77% to single agent paclitaxel are appealing [25], and confirm earlier reported activity [26].

279

Furthermore, neo-adjuvant combination of paclitaxel and carboplatin successfully has been administrated in serous endometrial cancer, since all 5 stage IV cases were left without residual disease after interval debulking surgery [27 – 29]. The comparable biologic behavior of serous endometrial and ovarian cancer might explain the successful results of paclitaxel in serous endometrial cancer that is in line with the broad experience in ovarian cancer. The combination of paclitaxel and carboplatin is standard treatment in ovarian cancer and is well tolerated because of an acceptable toxicity profile. Future trials therefore might test the clinical benefit after complete surgical staging of adjuvant paclitaxel and carboplatin in stage I –II NEC. Recently, the need for such a trial was again stressed since all cases of stage I serous endometrial cancer after complete surgical staging (n = 7) and exposure to chemotherapy were without recurrences and death [24]. Systemic treatment modalities for CS can be adapted from NEC since pathologists agree that the epithelial component is predominant in early stage CS [22,30,31]. This knowledge is a strong argument for the inclusion of drugs used in high-risk endometrial cancer. In the presence of sarcomatous cells, however, these patients might further benefit from agents against mesenchymal malignant cells including doxorubicin and ifosfamide. However, these drugs are notorious for their toxicity and special attention should be paid to their toxicity when used in combination. The use of epirubicin might deserve special interest since its use is associated with less toxicity and since it was successfully used in endometrial CS in the promising pilot study from Manolitsas et al. [11]. Therefore, the value of paclitaxel –epirubicin – carboplatin polychemotherapy in the adjuvant treatment of completely staged CS appears to be a reasonable study arm in future trials. To summarize, the current study was conducted to test the hypothesis that CS could be included in the highly needed studies in high-risk endometrial cancer. These trials frequently are hampered by low accrual rates. The endometrial origin of CS enables one to use a similar trial profile for CS and NEC. Such a trial would test for the value of systemic control by administrating adjuvant chemotherapy, after providing local control by a thorough staging procedure. However, the current observation of a more aggressive behavior of early stage CS suggests that both subtypes should be tested in separate trials whereas the mesenchymal component might necessitate an adaptation of the cytotoxic regimen in CS.

Table 5 Overview on spread pattern in different subtypes of endometrial cancer as reported in literature and including current data, n (%)

Grade 3 endometrioid [13 – 15] Carcinosarcoma [11,16,18,20 – 22] Serous [6,8,14,19] Clear cell [14]

Peritoneal cytology

Adnexal

Omental

Pelvic lymph nodes

86/668 72/373 17/57 7/20

41/721 75/512 27/125 3/32

3/25 15/96 47/202 3/6

78/734 80/423 72/244 9/20

(13) (19) (13) (35)

(6) (15) (22) (9)

(12) (16) (23) (50)

(11) (19) (30) (45)

280

F. Amant et al. / Gynecologic Oncology 98 (2005) 274 – 280

References [1] Ronnett B, Zaino R, Ellenson L, et al. Endometrial carcinoma. In: Kurman RJ, editor. Blaustein’s Pathology of the Female Genital Tract, 5th ed. New York’ Springer-Verlag; 2002. p. 460 – 501. [2] Podratz K, Mariani A. Uterine papillary serous carcinomas: the exigency for clinical trials. Gynecol Oncol 2003;91:461 – 2. [3] George E, Lillemoe T, Twiggs L, et al. Malignant mixed Mullerian tumor versus high-grade endometrial carcinoma and aggressive variants of endometrial carcinoma: a comparative analysis of survival. Int J Gynecol Pathol 1995;14:39 – 44. [4] Amant F, Vloeberghs V, Woestenborghs H, et al. Transition of epithelial towards mesenchymal differentiation during ovarian carcinosarcoma tumorigenesis. Gynecol Oncol 2003;90:372 – 7. [5] Sonoda Y, Saigo P, Federici M, et al. Carcinosarcoma of the ovary in a patient with a germline BRCA2 mutation: evidence for monoclonal origin. Gynecol Oncol 2000;76:226 – 9. [6] Goff B, Kato D, Schmidt R, et al. Uterine papillary serous carcinoma: patterns of spread. Gynecol Oncol 1994;54:264 – 8. [7] Sherman M, Sturgeon S, Brinton L, et al. Risk factors and hormone levels in patients with serous and endometrioid uterine carcinomas. Mod Pathol 1997;10:963 – 8. [8] Slomovitz B, Burke T, Eifel P, et al. Uterine papillary serous carcinoma (UPSC): a single institution review of 129 cases. Gynecol Oncol 2003;91:463 – 9. [9] Omura G, Blessing J, Major F, et al. A randomized clinical trial of adjuvant adriamycin in uterine sarcomas: a gynaecologic oncology group study. J Clin Oncol 1985;3:1240 – 5. [10] Reed N, Mangioni C, Malmstrom H, et al. First results of a randomized trial comparing radiotherapy versus observation postoperatively in patients with uterine sarcomas. An EORTC-GCG study. Int J Gynecol Cancer 2003;13S:4. [11] Manolitsas T, Wain G, Williams K, et al. Multimodality therapy for patients with clinical stage I and II malignant mixed Mu¨llerian tumors of the uterus. Cancer 2001;91:1437 – 43. [12] Rosenberg P, Boeryd B, Simonsen E. A new aggressive approach to high-grade endometrial cancer of possible benefit to patients with stage I uterine papillary cancer. Gynecol Oncol 1993;48:32 – 7. [13] Boronow R, Morrow C, Creasman W, et al. Surgical staging in endometrial cancer: clinical – pathologic findings of a prospective study. Obstet Gynecol 1984;63:825 – 32. [14] Cirisano F, Robboy S, Dodge R, et al. Epidemiologic and surgicopathologic findings of papillary serous and clear cell endometrial cancers when compared to endometrioid carcinoma. Gynecol Oncol 1999;74:385 – 94. [15] Creasman W, Morrow P, Bundy B, et al. Surgical pathologic spread patterns of endometrial cancer. Cancer 1987;60:2035 – 41. [16] Gallup D, Gable D, Talledo E, et al. A clinical – pathologic study of mixed Mu¨llerian tumors of the uterus over a 16-year period—The

[17] [18] [19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29] [30]

[31]

Medical College of Georgia experience. Am J Obstet Gynecol 1989;161:533 – 9. Jeffrey J, Krepart G, Lotocki R. Papillary serous adenocarcinoma of the endometrium. Obstet Gynecol 1986;67:670 – 4. Major F, Blessing J, Silverberg, et al. Prognostic factors in early-stage uterine sarcoma. Cancer 1993;71:1702 – 9. Ramirez-Gonzales C, Adamsons K, Mangual-Vazquez T, et al. Papillary adenocarcinoma in the endometrium. Obstet Gynecol 1987;70:212 – 5. Sartori E, Bazzurini L, Gadducci A, et al. Carcinosarcoma of the uterus: a clinicopathological multicenter CTF study. Gynecol Oncol 1997;67:70 – 5. Shaw R, Lynch P, Wade-Evans T. Mullerian mixed tumour of the uterine corpus: a clinical histopathological review of 28 patients. Br J Obstet Gynaecol 1983;90:562 – 9. Sreenan JJ, Hart WR. Carcinosarcomas of the female genital tract. A pathologic study of 29 metastatic tumors: further evidence for the dominant role of the epithelial component and the conversion theory of histogenesis. Am J Surg Pathol 1995;19:666 – 74. Trimbos B, Vergote I, Bolis G, et al. Impact of adjuvant chemotherapy and surgical staging in early-stage ovarian carcinoma: European Organisation for Research and Treatment of CancerAdjuvant ChemoTherapy in Ovarian Neoplasm Trial. J Natl Cancer Inst 2003;95:113 – 25. Huh W, Powell M, Leath C, et al. Uterine papillary serous carcinoma: comparisons of outcomes in surgical stage I patients with and without adjuvant therapy. Gynecol Oncol 2003;91:470 – 5. Ramondetta L, Burke T, Levenback C, et al. Treatment of uterine papillary serous carcinoma with paclitaxel. Gynecol Oncol 2001;82:156 – 61. Zanotti K, Belinson J, Kennedy A, et al. The use of paclitaxel and platinum-based chemotherapy in uterine papillary serous carcinoma. Gynecol Oncol 1999;74:272 – 7. Le TD, Yamada SD, Rutgers JL, et al. Case report: complete response of a stage IV uterine papillary serous carcinoma to neoadjuvant chemotherapy with taxol and carboplatin. Gynecol Oncol 1999;73: 461 – 3. Price FV, Amin RM, Sumkin J. Case report: complete clinical responses to neo-adjuvant chemotherapy for uterine serous carcinoma. Gynecol Oncol 1999;73:140 – 4. Resnik E, Taxy JB. Case report: neoadjuvant chemotherapy in uterine papillary serous carcinoma. Gynecol Oncol 1996;62:123 – 7. Bitterman P, Chun B, Kurman RJ. The significance of the epithelial differentiation in mixed mesodermal tumors of the uterus. Am J Surg Pathol 1990;14:317 – 28. Silverberg SG, Major FJ, Blessing JA, et al. Carcinosarcoma (malignant mixed mesodermal tumor) of the uterus. A Gynecologic Oncology Group pathologic study of 203 cases. Int J Gynecol Pathol 1990;9:1 – 19.