Article Text
PDF
Abstract
Introduction The purposes of this study were to compare adjuvant treatment modalities and to determine prognostic factors in stage III endometrioid endometrial cancer (EC).
Methods SATEN III was a retrospective study involving 13 centers from 10 countries. Patients who had been operated on between 1998 and 2018 and diagnosed with stage III endometrioid EC were analyzed.
Results A total of 990 women were identified; 317 with stage IIIA, 18 with stage IIIB, and 655 with stage IIIC diseases. The median follow-up was 42 months. The 5-year disease-free survival (DFS) of patients with stage III EC by adjuvant treatment modality was 68.5% for radiotherapy (RT), 54.6% for chemotherapy (CT), and 69.4% for chemoradiation (CRT) (p=0.11). The 5-year overall survival (OS) for those patients was 75.6% for RT, 75% for CT, and 80.7% for CRT (p=0.48). For patients with stage IIIA disease treated by RT versus CT versus CRT, the 5-year OS rates were 75.6%, 75.0%, and 80.7%, respectively (p=0.48). Negative peritoneal cytology (HR: 0.45, 95% CI: 0.23 to 0.86; p=0.02) and performance of lymphadenectomy (HR: 0.33, 95% CI: 0.16 to 0.77, p=0.001) were independent predictors for improved OS for stage IIIA EC. For women with stage IIIC EC treated by RT, CT, and CRT, the 5-year OS rates were 78.9%, 67.0%, and 69.8%, respectively (p=0.08). Independent prognostic factors for better OS for stage IIIC disease were age <60 (HR: 0.50, 95%CI: 0.36 to 0.69, p<0.001), grade 1 or 2 disease (HR: 0.59, 95% CI: 0.37 to 0.94, p=0.014; and HR: 0.65, 95%CI: 0.46 to 0.91, p=0.014, respectively), absence of cervical stromal involvement (HR: 063, 95% CI: 0.46 to 0.86, p=0.004) and performance of para-aortic lymphadenectomy (HR: 0.52, 95% CI: 0.35 to 0.72, p<0.001).
Discussion Although not statistically significant, CRT seemed to be a better adjuvant treatment option for stage IIIA endometrioid EC. Systematic lymphadenectomy seemed to improve survival outcomes in stage III endometrioid EC.
- endometrial neoplasms
- radiotherapy
- lymphatic metastasis
Statistics from Altmetric.com
HIGHLIGHTS
Chemoradiotherapy seems as the best option for stage IIIA endometrial cancer.
Chemotherapy alone increases the risk of recurrence in stage IIIC endometrial cancer.
Pelvic and para-aortic lymphadenectomy improved survival in stage III endometrial cancer.
Introduction
More than 10% of women with endometrial cancer are diagnosed with stage III disease which indicates that the cancer has spread to the uterine serosa, adnexa, vagina or lymph nodes.1 Because of increased risk of relapse and death after surgery alone, patients with stage III disease need adjuvant therapy. However, debate continues regarding the optimal adjuvant treatment.2–5
Patients with stage III endometrial cancer may be treated with adjuvant chemoradiation or adjuvant radiotherapy.6 A recent randomized trial, PORTEC-3, showed a survival advantage of chemoradiation compared with radiotherapy only.7 Women with stage III endometrial cancer with any histological subtypes and any sub-stages have been grouped into the same cluster in randomized clinical trials.2 3 6–9 Similarly, retrospective studies on adjuvant therapy in stage III disease included all histological types. Therefore, the impact of adjuvant treatment modalities in stage III endometrial cancer has not been delineated due to tumor and sub-stage heterogeneities.10–16 However, stage III disease is inherently heterogenous, affecting women at different metastatic sites: women with pelvic and/or para-aortic nodal metastases or patients with ovarian/adnexal/serosal/vaginal metastases or harboring combined metastatic tumor burden.17 Considering all sub-stages together in the same class and applying the same adjuvant treatment modality may lead to over- or under-treatment.
The primary aim of this study was to determine whether different adjuvant treatment modalities (chemotherapy, radiotherapy, and chemoradiation) improve survival outcomes in stage III endometrioid endometrial cancer patients based on the specific sub-stage of the disease. The secondary aim was to assess the prognostic factors in stage III disease.
Methods
The Institutional Review Board of the Kazakh Institute of Oncology and Radiology, Almaty, Kazakhstan, approved the study protocol (Date: 01.02.2019. No: 0202/1129). A retrospective, multicenter, international study of women with stage III endometrioid endometrial cancer was performed: SATEN III—Splitting Adjuvant Treatment of stage III ENdometrial cancers. Data were collected from 13 institutions participating in the Central Asia and East Europe Trial (CentEAST) Group.
Patients with endometrial cancer who were surgically staged and diagnosed with stage III disease from January 1, 1998 to August 31, 2018 were included. Exclusion criteria were non-endometrioid tumors, neoadjuvant chemotherapy, and incomplete medical records. We also excluded patients who died within 30 days of surgery due to operative complications or non-endometrial cancer-related causes as well as those with a follow-up <3 months.
The data, including age, menopausal status, pre-operative serum CA 125 level, surgical procedures, adjuvant treatment modalities, and follow-up were extracted from the medical files of the patients. The diagnosis was made via conventional microscopy by an experienced gynecological pathologist for each case. Histopathological data were obtained from the primary pathologist’s report and were not reviewed centrally. The histologic subtype was determined according to the World Health Organization classification.18 Tumor grade and surgical stage were defined in accordance with the International Federation for Gynecology and Obstetrics (FIGO) criteria.19 In patients treated before 2009, stage was determined retrospectively based on surgical and pathologic assessment. Lymphovascular space invasion was defined as the presence of tumor cells in endothelium lined channels of uterine specimens extracted at the time of surgery as described by GOG-99.20
Peritoneal, hematogenous, and nodal recurrences outside the retroperitoneal area (ie, inguinal, axillary, mediastinal, and supraclavicular) were considered as distant failures. Recurrences located in pelvic and/or para-aortic nodes were considered as retroperitoneal failures whereas relapses at vaginal vault, vagina, and/or central pelvis were considered as loco-regional recurrences. Lymph node ratio was defined as the percentage of metastatic lymph nodes to total lymph nodes removed.
Adjuvant radiotherapy included external beam radiation therapy with or without vaginal brachytherapy. The external beam radiation therapy had been delivered with image-guidance or intensity-modulated radiation therapy. Adjuvant chemotherapy usually consisted of multi-agent therapy. Carboplatin plus paclitaxel was the most frequently used regimen. Decisions regarding the choice of adjuvant treatment were not standard and differed among centers.
Disease-free survival was defined as the duration in months between the date of surgery and the date of first recurrence or the date of death from any cause, whichever occurred first, or the date of the last visit for patients alive without recurrence. Patients alive with no evidence of disease were censored at the date they were last known to be alive in disease-free survival analyses. The duration in months between the date of surgery and the date of death from any cause or the date of last contact was defined as overall survival. Patients alive at the last known follow-up were censored in overall survival analyses.
Statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS) software version 21 (IBM, USA). The one-way analysis of variance (ANOVA) test and Kruskal-Wallis test were used for continuous variables to compare patient characteristics among the three subgroups. In subgroup analyses of categorical variables, we used χ2 test or Fisher’s exact test, as appropriate. We investigated the association between recurrence site and adjuvant treatments within each subgroup using χ2 test or Fisher’s exact test, as appropriate. The Kaplan-Meier method was used to estimate the survival distribution. Survival curves were compared by using the log rank test. A Cox proportional hazards model, including all the parameters statistically significant in univariate analysis, was used to account for the influence of multiple variables. Values of p<0.05 were considered to be statistically significant.
Results
We identified 1229 patients. We excluded 214 patients with ineligible histology, 15 patients with incomplete medical records, eight patients who died within the first 3 months after primary treatment due to non-endometrial cancer causes, and two patients who received neoadjuvant chemotherapy. Finally, 990 women with stage III disease were included. The distribution of variables between stage IIIA, IIIB, and IIIC disease is presented in Table 1. Patients with stage IIIC disease were more likely than patients with stage IIIA disease to undergo pelvic and para-aortic lymphadenectomy (75% vs 47%, p<0.001), to have lymphovascular space invasion (81% vs 53%, p<0.0001), to have ≥50% myometrial invasion (76% vs 54%, p=0.001), and to receive adjuvant chemoradiation (58.2% vs 49.5%, p<0.0001); 5.7% and 5.5% of patients with stage IIIA and IIIB disease received no adjuvant therapy, respectively; and only 0.7% of patients with stage IIIC disease received no adjuvant therapy. The number of pelvic nodes removed was similar among patients who underwent pelvic lymphandectomy in all sub-stages, though more para-aortic nodes were removed in patients with stage IIIC disease compared with those in other groups.
Patient characteristics
The median follow-up was 42 months (IQR 4–253 months), and the 5 year disease-free survival and overall survival rates of the entire cohort were 61.1% and 72.6%, respectively. The 5 year disease-free survival rates were 66.0%, 59.0%, and 58.7% for patients with stage IIIA, IIIB, and IIIC disease, respectively (p=0.37). The 5 year overall survival rates were 78.9%, 66.3%, and 70.5% for those with stage IIIA, IIIB, and IIIC, respectively (p=0.17).
Kaplan-Meier survival plots of disease-free survival and overall survival in stage IIIA, IIIB and IIIC patients for the three different adjuvant treatment modalities are shown in Figure 1. Women with stage IIIA disease who received adjuvant radiotherapy, adjuvant chemotherapy, or adjuvant chemoradiation had 5 year disease-free survival rates of 65.8%, 54.6%, and 69.4%, respectively (p=0.11), whereas the 5 year overall survival rates were 75.6%, 75.0%, and 80.7%, respectively (p=0.48).
Disease-free survival (DFS) by adjuvant treatment modality of the patients with: stage IIIA (68.5%, 54.6%, and 69.4% for radiotherapy (RT), chemotherapy (CT) and chemoradiation CRT), respectively; p=0.11) (A); stage IIIB (66.7%, 68.6%, and 53.6% for RT, CT, and CRT, respectively; p=0.96) (B); and stage IIIC (65%, 52.7%, and 59.1% for RT, CT, and CRT, respectively; p=0.23) (C). Overall survival (OS) by adjuvant treatment modality of the patients with: stage IIIA (75.6%, 75%, and 80.7% for RT, CT, and CRT, respectively; p=0.48) (D); stage IIIB (66.7%, 85.7%, and 53.6% for RT, CT, and CRT, respectively; p=0.7) (E); and stage IIIC (78.9%, 67%, and 69.8% for RT, CT, and CRT, respectively; p=0.08) (F).
Regarding stage IIIB disease, the 5 year disease-free survival rates were 66.7%, 68.6%, and 53.6%, after adjuvant radiotherapy, chemotherapy, and chemoradiation, respectively (p=0.96), whereas the 5 year overall survival rates were 66.7%, 85.7%, and 53.6%, respectively (p=0.7).
Among patients with stage IIIC disease, the 5 year disease-free survival rates were 65.0%, 52.7%, and 59.1% for those who received adjuvant radiotherapy, adjuvant chemotherapy, and adjuvant chemoradiation, respectively (p=0.23). The corresponding figures for 5 year overall survival were 78.9%, 67.0%, and 69.8%, respectively (p=0.08).
A total of 243 patients (24.5%) had a recurrence. Recurrence site was not significantly associated with the type of adjuvant treatment for stage IIIA and IIIC (Table 2). Although not statistically significant, distant metastases were less common in the adjuvant radiotherapy group when compared with the other adjuvant treatment groups. Clinicopathological characteristics of the adjuvant treatment groups are presented in online supplementary Table 1. Among patients with stage IIIA disease, those who received adjuvant chemotherapy were significantly more likely to undergo pelvic and para-aortic lymphadenectomy (81%) compared with those who received adjuvant radiotherapy (35%) or chemoradiation (43%) (p<0.0001). In stage IIIC disease, grade 2 or 3 tumors were more common in the chemotherapy (80.6%) and chemoradiation (86.1%) groups compared with the radiotherapy (72.7%) group (p=0.005). Similarly, lymphovascular space invasion was more common in women who received chemotherapy (79.4%) or chemoradiation (85.0%) compared with those who received radiotherapy (67.7%) (p<0.0001). A significantly increased lymph node ratio was found in the chemoradiation group (median 0.15), compared with the radiotherapy (median 0.013) and chemotherapy (median 0.034) groups (p=0.02).
Supplemental material
Recurrence patterns in regards to adjuvant treatment modalities
Multivariate analysis revealed age, grade, myometrial invasion, cervical involvement, and execution of lymphadenectomy as independent prognostic factors for both disease-free survival and overall survival in all women with stage III disease (online supplementary Table 2). Only age independently impacted disease-free survival in stage IIIA disease. Negative peritoneal cytology and execution of lymphadenectomy were determined as independent prognostic factors for improved 5 year overall survival in stage IIIA disease (Table 3). Multivariate analysis revealed age, grade, cervical involvement, and para-aortic lymphadenectomy as independent prognostic factors for both disease-free survival and overall survival in stage IIIC disease (Table 4). Additionally, myometrial invasion appeared to be another independent prognostic factor for disease-free survival in stage IIIC disease (Table 4)
Supplemental material
Multiple regression analysis of disease-free survival and overall survival in patients with stage IIIA endometrial cancer
Multiple regression analysis of disease-free survival and overall survival in patients with stage IIIC endometrial cancer
Discussion
The results of the SATEN III trial, the largest retrospective study to date on stage III endometrial cancer, showed that adjuvant chemoradiation was associated with better survival in women with stage IIIA disease when compared with adjuvant chemotherapy or radiotherapy. Execution of lymphadenectomy was associated with improved overall survival in stage IIIA disease, whereas performance of para-aortic lymphadenectomy improved survival outcomes in stage IIIC disease.
In the present study, the 5 year overall survival of patients with stage III disease was comparable to that reported in the PORTEC-3 trial (72.6% vs 74.0%, respectively).8 However, it should be noted that the latter included non-endometrioid histologies as well. In the current study, age >60 years, grade 3 disease, myometrial invasion ≥50%, cervical involvement, and performing no lymphadenectomy were identified as independent prognostic factors for decreased disease-free survival and overall survival in stage III disease. Better survival in younger patients has been reported previously.7 8 However, grade 3 disease, deep myometrial invasion, and cervical involvement are well-known risk factors for nodal involvement, recurrence, and worse overall survival in clinically early-stage endometrial cancer.4 21 22
Our results showed that no adjuvant treatment modality was superior to any other in terms of improving disease-free survival or overall survival. However, a Cochrane review of four randomized trials demonstrated that overall survival and progression-free survival were significantly improved after adjuvant chemotherapy compared with adjuvant radiotherapy in stage III disease. In subgroup analyses, the positive effect of chemotherapy on survival was also shown for sub-stages IIIA and IIIC.23
The PORTEC-3 study included women with high-risk endometrial cancer.8 An analysis of patients with FIGO stage III disease showed that adjuvant chemoradiation was superior to radiotherapy alone in terms of failure-free survival (5 year failure-free survival 69.3% vs 58%, p=0.03; 5 year overall survival 78.7% vs 69.8%, p=0.07). However, subgroup analyses of GOG-258 that compared adjuvant chemoradiation with chemotherapy did not identify a subgroup of patients who benefited more from one alternative over another.6
In the present study, the 5 year overall survival of patients with stage IIIA disease was 79.0%. Retrospective studies reported the 5 year overall survival rates varying from 55% to 92% for stage IIIA endometrial cancer.24–26 In our study, almost one-fourth of the patients who received any adjuvant treatment developed recurrences. Of those, 60% were distant recurrences. Likewise, a study on stage IIIA endometrial cancer stated that two-thirds of recurrences occurred outside of the pelvis.24 Recurrence patterns by adjuvant treatment modality suggested that adjuvant chemoradiation achieved a lower rate of distant metastasis in women with stage IIIA disease (19.4%, 15.3%, and 12.1% distant metastasis in the radiotherapy, chemotherapy, and chemoradiation group, respectively).
Negative peritoneal cytology and execution of lymphadenectomy appeared as independent prognostic factors for improved overall survival in stage IIIA disease in our study. Similarly, a previous study on stage IIIA disease indicated that lymphadenectomy was associated with improved survival (n=1257; HR 0.48, 95% CI 0.39 to 0.59).27 The presence of undefined metastatic foci left behind might have led to worse survival outcomes in women who did not undergo lymphadenectomy, given that the execution of pelvic and para-aortic lymphadenectomy contributes towards ascertaining that patients are correctly classified as stage IIIA disease and reduces the number of occult stage IIIC1 and IIIC2 tumors.
Our findings indicated that women with stage IIIC disease treated with adjuvant chemotherapy alone were two to three times more likely to develop vaginal recurrence (4.1%) when compared with those treated with adjuvant radiotherapy alone (2%) or treated with adjuvant chemoradiation (1.5%). Additionally, adding chemotherapy to radiotherapy did not seem to improve survival. Similar results were reported in another retrospective multicenter study.15 Among 265 patients with stage IIIC disease, those treated with adjuvant chemotherapy alone experienced seven-fold more vaginal recurrence and three-fold more pelvic recurrence than patients treated with adjuvant chemoradiation. The authors also reported that disease-free survival and overall survival rates were similar between adjuvant radiotherapy and adjuvant chemoradiation groups.15 However, a more recent analysis of 200 patients with stage IIIC disease showed that adjuvant chemoradiation should be preferred in these patients for its overall survival benefit over other adjuvant treatment modalities.10 In the current study, analyses of the clinicopathological features of the adjuvant treatment groups in stage IIIC disease showed that the radiotherapy group had more favorable features such as low grade disease, negative lymphovascular space invasion status, and decreased lymph node ratio (online supplementary Table 1). These results seem to increase the possibility of confounding, and reduce the reliability of our findings regarding adjuvant treatment in stage IIIC disease.
Our study demonstrated the execution of para-aortic lymphadenectomy along with age <60 years, grade 1–2 disease, and no cervical involvement as independent prognostic factors for overall survival in stage IIIC disease. Of 247 patients with pelvic nodal metastasis in whom pelvic and para-aortic lymphadenectomy was performed, 167 (67.6%) also had para-aortic nodal involvement. Our findings suggest that para-aortic lymphadenectomy should be performed in the presence or suspicion of pelvic nodal metastasis. However, a prospective study involving >400 patients from the Mayo Clinic reported the rate of para-aortic nodal disease as 8% in the ‘at risk’ (other than grade 1–2, <50% myometrial invasion, and tumor diameter ≤2 cm) population of endometrioid endometrial cancer.28 The prevalence of para-aortic nodal metastasis was 45% (for only endometrioid histology) in the presence of pelvic nodal involvement.28
Given the increased recurrence risk, stage III disease is classified as high-risk.4 However, the group of high-risk tumors is heterogenous and includes various histologic types and grades. Both a Gynecologic Oncology Group (GOG) study and the PORTEC-3 trial reported worse outcomes in serous and clear cell histologies compared with the endometrioid type in advanced stages.8 29 Yet, all available randomized clinical trials associated with stage III disease included various histological types. The main strength of our study is the inclusion of women with only pure endometrioid histology. Considering the long-term follow-up data from centers in East Europe and Central Asia, the results of the SATEN-III study may be applicable to the general population. The main limitations of this study are its retrospective design as well as lack of a central pathology review and lack of standard adjuvant treatments. In addition, there were no data audit process to confirm data entered at each site with all source documents from each of these sites.
Our findings revealed chemoradiation as a better adjuvant treatment option for stage IIIA disease when compared with adjuvant radiotherapy or chemotherapy. A definitive conclusion on adjuvant treatment in stage IIIC disease could not be drawn as the chemoradiation group exhibited worse clinicopathological features compared with the radiotherapy and chemotherapy groups. Systematic pelvic and para-aortic lymphadenectomy improved survival in each sub-stage of stage III endometrioid endometrial cancer. Future prospective studies focusing on adjuvant treatment modalities in each sub-stage of stage III endometrioid endometrial cancer, rather than those using risk-classification systems which combine all sub-stages into a group, may help to standardize adjuvant treatment recommendations.
References
Footnotes
Contributors IK: Concept, design, monitored data collection, analysis, draft, writing the manuscript, final approval. MMM: Materials, supervision, interpretation of data, data curation, critical review, final approval. ST: Data collection, materials, final approval. AA: Materials, supervision, critical review, final approval. CS: Materials, data collection, interpretation of data, final approval. MS: Materials, data collection, review, final approval. FD: Materials, data collection, supervision, interpretation of data, final approval. FO: Materials, data collection, review, final approval. DH: Materials, data collection, critical review, analysis, final approval. VS: Materials, data collection, review, final approval. DK: Materials, data collection, interpretation of data, final approval. AS: Materials, data collection, review, final approval. FF: Materials, data collection, final approval. SA: Materials, data collection, review, final approval. EU: Materials, data collection, review, final approval. DK: Data collection, review, final approval. IY: Data collection, interpretation of data, data curation, review, final approval. SM: Data collection, review, final approval. HA: Data collection, review, final approval. MES: Concept, design, literature seach, review, final approval. AP: Data collection, interpretation of data, review, final approval. KA: Data collection, final approval. TB: Materials, review, final approval. MK: Materials, review, final approval. DV: Materials, review, final approval. MG: Materials, concept, design, monitored data collection, analysis, interpretation of data, critical review, final approval.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; internally peer reviewed.
Data availability statement Data may be obtained from a third party and are not publicly available.