Neoadjuvant chemotherapy in patients with stage IVB uterine serous carcinoma: a Turkish multicentric study

Abstract The aim of this study was to evaluate the prognostic factors for and determine the effect of neoadjuvant chemotherapy (NACT) on oncologic outcome in stage IVB pure serous endometrial carcinoma patients who received taxane and platinum. Forty-two patients with 2009 International Federation of Gynecology and Obstetrics (FIGO) stage IVB uterine serous carcinoma were enrolled from six gynecologic oncology centers and a study group was created. The study group had a 2-year disease-free survival (DFS) of 32% and 2-year disease-specific survival (DSS) of 73%. On univariate analysis; lymphadenectomy (not performed vs. performed), paraaortic lymph node metastasis (positive vs. negative) and number of metastatic lymph node count (≤5 vs. >5) were found to have statistical significance for DFS (p < 0.001, p = 0.026 and p = 0.044, respectively). Adnexal metastasis (positive vs. negative) and type of cytoreductive surgery (maximal vs. optimal and suboptimal) had statistical significance for DSS (p = 0.041 and p = 0.015, respectively). Receiving NACT did not affect DFS and DSS in stage IVB uterine serous carcinoma patients. As our sample size was small, precise conclusions could not be made for suggesting the use of NACT in advanced stage uterine serous carcinoma. For more accurate results, more randomized controlled studies are needed in this patient group. IMPACT STATEMENT What is already known on this subject? Endometrial carcinoma is the most common type of gynecologic tract malignancies and usually it is diagnosed at early stages. Although the favorable prognosis, uterine serous carcinoma (USC), one of the rarest subtypes, has a poorer prognosis when compared to other histological subtypes. USC has a propensity to spread beyond pelvis. Due to this aggressive behavior, surgical intervention could not be feasible in advanced stage disease. What do the results of this study add? Our study evaluated the prognostic factors that affect survival in advanced stage USC patients. Also we investigated that neoadjuvant chemotherapy (NACT) could improve oncologic outcomes. Performing lymphadenectomy, presence of paraaortic lymph node and adnexal metastasis, number of metastatic lymph nodes and type of cytoreductive surgery improved survival in advanced stage USC patients. However, NACT did not have a statistical significance as a predictor for disease-specific survival (DSS) and disease-free survival (DFS). What are the implications of these findings for clinical practice and/or further research? Maximal surgical effort should be performed in advanced stage USC according to our results. On the other hand, NACT had no impact on DSS and DFS rates. For this reason, we could not be able to suggest the routine use of NACT in advanced stage USC. But more randomized controlled trials are warranted for confirmation of our results.


Introduction
Endometrial carcinoma (EC) is the most frequent gynecological tract malignancy in developed countries (Siegel et al. 2020).Endometrial carcinomas were divided into two subgroups according to molecular pathogenesis, clinical behavior and histopathology (Bokhman 1983).According to this classification; type I EC, caused by unopposed estrogen stimulation, was related to low-grade endometrioid carcinoma and favorable prognosis.Type II EC, not associated with estrogen stimulation, was reported to have a poorer prognosis.
Uterine serous carcinoma (USC) was first described by Hendrickson in 1982 as a subtype of endometrial carcinomas with a propensity to spread beyond uterine corpus (Hendrickson et al. 1982).USC constitutes approximately 10% of all endometrial carcinomas and 39% of endometrial carcinoma related deaths (Boruta et al. 2009).At the time of surgery, extrauterine metastasis is detected in 23-40% of cases with no myometrial invasion (Slomovitz et al. 2003, Benito et al. 2009).Also in the early-stage disease, lymphovascular space invasion, lymph node metastasis and microscopic spread to intraperitoneal structures are common in USC patients (Fader et al. 2013).Comprehensive surgical staging (total hysterectomy ± bilateral salpingo-oophorectomy þ retroperitoneal lymphadenectomy þ omentectomy þ peritoneal washings and biopsies) with no residual tumor followed by platinum and taxane containing adjuvant chemotherapy regimens are recommended for patients with USC (Boruta et al. 2009).Given the high rates of extrauterine involvement and recurrence; 5-year survival rate for stage IV disease is 19.9% (Slomovitz et al. 2003).In patients with stage IVB USC, 5-year survival rate is reported as 0-5% (Goldberg et al. 2008, Patsavas et al. 2011).Because of low number of patients and lack of randomized controlled studies, the optimal treatment modality for advanced stage USC is controversial.There has been limited data offering the implementation of neoadjuvant chemotherapy (NACT) in these treatment modalities in advanced stage USC patients.
The aim of this study is to evaluate prognostic factors and to determine the effect of NACT to oncologic outcome in stage IVB pure serous endometrial carcinoma patients who received taxane and platinum.Also, we investigated that NACT could improve oncologic outcomes in this patient group.

Materials and methods
Forty-two patients with 2009 International Federation of Gynecology and Obstetrics (FIGO) stage IVB pure uterine serous endometrial carcinoma were enrolled from six gynaecologic oncology centres and a study group was obtained.A total of 88 patients from six different centres, 31 patients were stage III-IVA, 7 patients received radiotherapy for adjuvant therapy and 8 patients were lost to follow-up.As a result, 42 patients were included in study group (Figure 1).The entire cohort included patients who received platinumbased chemotherapy as neoadjuvant or adjuvant therapy regimens.Patients followed-up for at least 3 months in participating gynecologic oncology clinics were evaluated in this study.Patients with non-serous tumor type, except stage IVB disease and mixed tumors were excluded.Also, patients who received nonplatinum-based chemotherapy, who had secondary primary tumor and patients lost in follow-up period were excluded.Presences of extraabdominal tumor metastasis and receiving radiotherapy and multimodal therapy (chemotherapy followed by radiotherapy, radiotherapy followed by chemotherapy, sandwich therapy) were admitted as exclusion criteria.Additionally, patients in whom interval cytoreductive surgery could not be performed after NACT cycles were excluded from the study group.Staging procedures were performed according to FIGO 2009 criteria.Local ethics committee approval was obtained before the study (14/05/ 2020-07).
Surgical outcome was described as maximal, optimal and suboptimal; patients with maximal cytoreductive surgery (CRS) (no residual tumor), optimal CRS (residual tumor �1 cm) and suboptimal CRS (residual tumor >1 cm) were included in the study group.Tumor size was defined as the largest tumor diameter located within the uterine corpus in final pathology report.
Standard staging surgery was completed by evaluation of intra-abdominal regions, obtaining peritoneal cytologic sampling, performing total abdominal hysterectomy, bilateral salpingo-oophorectomy, total omentectomy and systematic retroperitoneal lymphadenectomy.Cytoreductive surgical techniques were used as part of the staging surgery when a macroscopic pathological finding was detected on intraoperative evaluation.However, surgical radicality was designated by the senior surgeon in all attending centers depending on patient's morbidities.All surgical procedures were performed by expert gynecological oncologists.
When preoperative imaging findings and intraoperative evaluation confirmed that optimal CRS was not feasible, patients were referred to NACT.Number of cycles and type of chemotherapeutic agents were decided in all attending institutions' gynecologic oncology councils.Duration of cycles was determined by respond to NACT.
Failure distal to the pelvic inlet was defined as pelvic failure and between pelvic inlet and diaphragm was defined as upper-abdominal failure.Remaining types of failure were admitted as extraabdominal failure.Also, disease failure in the liver parenchyma, skin, and bone was accepted as extraabdominal failure, too.
After completion of therapy, all patients were followed-up with 3 months intervals in first 2 years, with 6 months intervals in next 3 years and with 1-year intervals after 5 years.Chest X-ray was performed in case of a clinical suspicion or annually.Thoracic and/or abdominal computerized tomography was obtained when necessary.Pap-smear test and serum CA 125 level were also used in follow-up period although not obligatory.

Statistical analysis
Disease-free survival (DFS) was defined as the period from initial surgery to proven disease failure (recurrence or refractory disease) with clinical examination and/or radiological imaging.Also, the period from initial surgery to last contact in patients with no evidence of disease was defined as DFS.The period from initial surgery to death because of disease or last contact was admitted as disease-specific survival (DSS).
Statistical analysis was performed using Statistical Package for Social Sciences (IBM SPSS Inc, Chicago, IL, USA) version 20.0.Descriptive statistics were expressed as mean ± SD or median (min-max) for continuous variables and number/percentage for categorical variables.Categorical parameters were compared with chi-square test and continuous parameters were compared with Annova Table test.DFS and DSS estimates were determined by using the Kaplan-Meier method.Survival curves were compared using the log-rank test.We calculated the power analysis and sample size with 80% power at a two-sided 5% significance level for a difference in relative hazard of 16% during the follow-up (Schoenfeld, 1983).A p < 0.05 was considered as statistically significant.Multivariate analysis could not be carried out because of low number of patients.
The patients who received and did not receive NACT were compared for clinical, surgical and pathological factors.There was not a statistical significance regarding prognostic factors except for omental metastasis between two patient groups.Omental metastasis was 70% in patients who received NACT and 96.8% in patients who did not receive NACT (p ¼ 0.013) (Table 2).Maximal CRS was achieved in eight (80%) patients in NACT group.Similarly, in 26 (81.2%) patients who did not receive NACT, maximal CRS was feasible.The risk of residual tumor was not superior in NACT group (p ¼ 0.930).

Neoadjuvant chemotherapy and survival analysis
Ten patients received platinum-based NACT between three and eight cycles.Paclitaxel þ carboplatin combination was applied to all patients during NACT cycles.Six patients received three cycles of NACT, one patient got four cycles of NACT, two patients got six cycles of NACT and one patient got 8 cycles of NACT.In our study group, NACT was not applied to 32 patients.Also, platinum-based chemotherapeutic agents were used as adjuvant chemotherapy in this subgroup.Paclitaxel þ carboplatin combination was used in 24 (75%) patients, paclitaxel þ cisplatin in four (12.5%) patients, docetaxel þ carboplatin in two (6.2%) patients, paclitaxel þ carboplatin þ gemcitabine in one (3.1%)patient and paclitaxel þ cisplatin þ adriamycin in one (3.1%)patient.There was not a statistical significance for total chemotherapy cycles between NACT and adjuvant chemotherapy subgroups (median 6; range 3-10 vs. median 6; range 2-12, p ¼ 0.738, respectively).
Median follow-up time was 18.5 months (range 2-152).Twenty (47.6%) patients developed recurrence in follow-up period.Two (4.8%) of these recurrences were located in only pelvic region and 13 (30%) were located in extraabdominal regions.Eleven (26.2%) patients died because of disease.The power analysis determined that a total of 11 event would be sufficient.
Study group had a 2-year DFS of 32% and 2-year DSS of 73%.5-year DFS was 25% and 5-year DSS was 32%.On univariate analysis; lymphadenectomy (not performed vs. performed), paraaortic lymph node metastasis (positive vs. negative) and number of metastatic lymph node count (�5 vs. >5) were found to have statistical significance for DFS (p < 0.001, p ¼ 0.026 and p ¼ 0.044, respectively).Adnexal metastasis (positive vs. negative) and type of CRS (maximal vs. optimal and suboptimal) had statistical significance for DSS (p ¼ 0.041 and p ¼ 0.015, respectively).Receiving NACT did not affect DFS and DSS (Table 3 and Figures 2 and 3).Multivariate analysis was not possible because of low number of patients in entire cohort as well as low number of relapsed patients.

Discussion
In our study, performing lymphadenectomy, presence of paraaortic lymph node metastasis and number of metastatic lymph node count were found to have statistical significance on DFS; whereas presence of adnexal metastasis and type of CRS had statistical significance on DSS on univariate analysis.In addition, receiving NACT was not a predictor for DFS and DSS in stage IVB USC patients.
Different investigators evaluated the effect of lymphadenectomy in EC patients.In one of these studies, Eggemann et al. (2016) enrolled 1502 patients with EC independent of histologic subtypes.They found that performing pelvic þ paraaortic lymphadenectomy improved overall survival (OS) in EC patients at intermediate and high risk for recurrence on multivariate analysis.In their novel study, Todo et al. (2010) detected OS was longer in EC patients of intermediate or high risk for recurrence with lymphadenectomy.In addition, Zhong et al. (2018) determined that adnexal involvement and lymph node metastasis were independently associated with DFS and recurrence rates.In their multicenter study, 241 USC patients from all stages (42% stages III-IV) were reviewed.They determined that removed lymph node count and metastatic lymph node count affected survival rates in this patient group.
Information focused on the importance of NACT in advanced stage EC is scarce.In one of these studies, Eto et al. (2013) enrolled 426 stage IVB EC patients from different gynecologic oncology centers.In entire cohort, only 47% of the primary chemotherapy group underwent subsequent surgery.Like our study, they found that OS was similar between primary chemotherapy followed by surgery group and primary surgery group.For patients who were not suitable for primary surgery, preoperative chemotherapy was suggested as a treatment option for all histologic subtypes of stage IVB EC patients.In a review article, 106 stage IVB EC patients to date of 2015 were evaluated for the impact of NACT (Rabinovich 2016).Although no precise conclusions were stated, NACT followed by interval debulking was recommended for patients with poor performance status and for whom maximal debulking was not feasible.This study group was not classified according to histological subtypes of EC.Vandenput et al. (2009) investigated the effect of NACT in stage IV serous endometrial cancer patients in a prospective study.They revealed that 24 (80%) of 30 patients underwent optimal cytoreduction after three to four cycles of NACT.In addition, postoperative minor and major complications were better in NACT group.They suggested NACT followed by interval debulking surgery as a treatment option for advanced stage EC.
In a multicenter study, 102 advanced stage (III and IV) EC patients were evaluated for the impact of NACT followed by interval debulking surgery (De Lange et al. 2019).The study cohort included 44 (43%) serous type EC along with other histologic subtypes.The authors demonstrated NACT followed by interval debulking surgery was a valuable treatment modality for advanced stage EC patients regardless of tumor type.Similarly, Khouri et al. (2019) reviewed 39 advanced stage EC patients (44% serous type) and established that less than half of the patients undergoing NACT for EC underwent interval debulking surgery.None of the patients had complete response.But, response to NACT and performing interval debulking surgery improved overall survival.In a large retrospective series, investigators determined the impact of NACT use in 952 stage IV EC patients (Tobias et al. 2020) Serous tumor histology was present in 26.2% of study group.Unlike our study, they found that patients treated with NACT had superior survival in short term; for the first 8 months after diagnosis.This superiority in short-term survival was confirmed in a recent review (Huang et al. 2021).A total of 5844 patients of which 1317 received NACT were evaluated between years 1990 and 2020.According to this review, all studies offered similar OS rates except for one study by Tobias et al.Finally, the reviewers stated that NACT followed by interval debulking surgery reduces perioperative morbidity with similar OS.
In another study, 10 patients with stage IV USC received NACT and 34 patients underwent primary debulking surgery (Wilkinson-Ryan et al. 2015).Investigators reported that there was no difference in progression-free survival and overall survival rates between these patient groups.However, operation time and hospital stay were shorter in NACT group.Thus, NACT could be used as a treatment option in select patients.Similar to our study, maximal CRS rate and optimal CRS rate did not differ between two 18.8 a n:33 patients (Pathologic reports about of lymphovascular space invasion in nine patients was not reported).b Glandular and/or stromal.c n:33 patients (Pathologic reports about of peritoneal in nine patients was not reported).d n:41 patients (Pathologic reports about of omental metastasis in one patient was not reported).e n: 41 (One patient underwent bilateral salpingo-oophorectomy before endometrial cancer).f n:34 (Eight patients did not underwent lymphadenectomy).CRS: cytoreductive surgery.groups.Bogani et al. (2019) assessed 34 patients diagnosed with stage IVB serous endometrial carcinoma.Fifteen patients undergoing NACT plus interval debulking surgery and 19 patients undergoing primary surgery followed by adjuvant treatment were compared.Cytoreduction rate was similar between groups (p ¼ 0.962).Also, DFS and OS rates had no statistical significance (p ¼ 0.663 and p ¼ 0.349, respectively).They stated that NACT might be a proper option for unresectable stage IVB serous endometrial carcinoma.Similarly, all of these studies were conducted with low number of patients.Because of the rarity of advanced stage USC, data extrapolated from ovarian cancer studies was used for interpretation of treatment modalities in advanced stage EC patients.NACT followed by interval debulking surgery was not inferior from initial surgery in terms of progression-free survival and overall survival in two randomized trials (Vergote et al. 2010, Kehoe et al. 2015).USC has similar tumor biology to ovarian cancer; frequently metastasises and requires aggressive cytoreductive surgery (Hamilton et al. 2008).Due to the fact that conducting randomized controlled studies including advanced stage USC patients is not feasible, results obtained from ovarian cancer studies can be used for advanced stage EC.
Retrospective design is the main limitation of our study.However, a homogeneous study cohort was created with inclusion of only stage IVB USC patients.We think that our results are reliable because of the similarity of clinicopathological features between the two groups with received NACT and not received NACT that reduced the possibility of bias.In addition, the entire cohort received platinum-based chemotherapy and none of the patients had extraabdominal metastasis.Also, all patients underwent staging procedures performed by gynecologic oncologists.Pathology results were evaluated by experienced gyneco-pathologists.These are the main advantages of our study.
In conclusion, receiving NACT did not affect DFS and DSS in stage IVB USC patients.Despite the fact that our sample size was small, precise conclusions could not be made for suggesting the use of NACT in advanced stage USC.For more accurate results, more randomized controlled studies are needed in this patient group.

Table 1 .
Clinical, surgical and pathological characteristics of patients.

Table 2 .
Clinical, surgical and pathological factors of patients received neoadjuvant chemotherapy compared the patients who did not receive neoadjuvant chemotherapy.
a Median value.b Patient with paraaortic lymph node metastasis with or without pelvic lymph node metastasis.c Patient with pelvic lymph node metastasis with or without paraaortic lymph node metastasis.d Glandular or stromal.