WHAT IS ALREADY KNOWN ON THIS TOPICWHAT THIS STUDY ADDS

Our international patient cohort demonstrates distinct survival outcomes depending on either full-thickness uterine serosal tumor invasion or adnexal involvement. Patients with full-thickness uterine disease, with or without adnexal involvement, were found to have worse outcomes, even after adjusting for histologic subtype and lymphovascular space invasion.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICYINTRODUCTION

Endometrial cancer rates continue to increase each year, with an estimated 66,200 patients diagnosed in 2023.1 Patients are surgically staged using the guidelines set forth by the International Federation of Gynecology and Obstetrics (FIGO). The most recent iteration of these guidelines was released in 2023 and incorporates pathologic and molecular features, including histology and molecular subtype.2

Approximately 2.6% of patients diagnosed with endometrial cancer will have disease spread through the myometrium to the serosa or the adnexa, comprising a very small subset of surgically staged endometrial tumors.3 The 2009 FIGO staging system designates tumors as stage IIIA if they exhibit either adnexal involvement or full-thickness myoinvasion to the uterine serosa.4 Historically, the 5-year overall survival rate among patients with stage IIIA disease has ranged between 55% and 92%.5–8 Other retrospective studies on this specific stage were largely based on the 1988 FIGO staging system, which could assign stage IIIA based on cytology, involvement of the serosa, or involvement of the adnexa.9–11 The current National Comprehensive Cancer Network (NCCN) treatment recommendation for stage IIIA disease is systemic therapy with or without radiation; although, according to a recent survey assessing practice patterns in advanced endometrial cancer, there are wide variations in treatment.12

The 2023 FIGO staging system further distinguishes stage IIIA disease, designating adnexal involvement (ovary or fallopian tube) as stage IIIA1 and uterine serosal involvement as stage IIIA2.2 There are few data, however, to determine if these distinctions are prognostically relevant, and treatment guidelines do not yet incorporate the 2023 FIGO staging classification. In addition, tumors meeting criteria for stage IIIA disease according to the 2009 FIGO system would now be designated as stage IA3 disease by the 2023 FIGO staging system; criteria include endometrioid histology, FIGO grade 1 or grade 2 disease, <50% myoinvasion, absence of extensive lymphovascular space invasion (defined as involvement of more than five vessels), absence of additional metastases, and adnexal involvement limited to a unilateral ovary (without capsule invasion or rupture).

We aimed to assess clinicopathologic features and oncologic outcomes of patients with full-thickness uterine and/or adnexal spread of endometrial cancer. We also sought to understand the treatment patterns of patients with surgically staged 2009 FIGO stage IIIA endometrial cancer.

METHODS

This was a retrospective study including patient data from five international institutions. After approval was obtained from the institutional review boards at Memorial Sloan Kettering Cancer Center, Mayo Clinic Rochester, MD Anderson Cancer Center, University of Oslo, and Kliniken Essen-Mitte, patient charts and institutional databases were reviewed to identify patients who had undergone complete surgical staging for endometrial carcinoma from January 1, 2000 through December 31, 2019. Surgical staging was required, including hysterectomy, bilateral salpingo-oophorectomy, and regional lymph node assessment. Lymph nodes were assessed via sentinel lymph node dissection, pelvic lymphadenectomy with or without para-aortic lymph node dissection, or sentinel lymph node dissection with additional pelvic lymphadenectomy.

Patients were included if they met 2009 FIGO stage IIIA criteria and had either full-thickness uterine serosal, adnexal, or combined adnexal/serosal involvement. Disease spread to the cervix was stratified between none, mucosa, or stroma with/without mucosa. Adjuvant treatment modalities were stratified into surgery alone, chemotherapy with or without radiation, and radiation alone. Radiation modalities included external beam radiation therapy and intravaginal radiation therapy. To determine the effect of evolving practice patterns during the study period, treatment was also evaluated between two decades—2000 to 2009 and 2010 to 2019. Patients were also analyzed by stratifying between adnexal-only and serosal with/without adnexal involvement. Detailed data regarding degree of lymphovascular space invasion, depth of myometrial invasion in adnexal-only tumors, and more granular pathologic data were not captured across all study institutions; therefore, 2023 FIGO stage was not retrospectively assigned. Patients with sarcoma histology, concurrent endometrial/ovarian malignancy, neoadjuvant treatment, positive lymph nodes, or peritoneal disease were excluded.

Continuous variables were analyzed with the Kruskal-Wallis test. Categorical variables were assessed using the Χ2 test. Progression-free survival was measured from date of surgery to date of recurrence or last follow-up. Overall survival was measured from date of surgery to date of death or last follow-up. The Kaplan-Meier method was used to estimate survival distributions, which were compared with the log-rank test. Cox proportional hazards models were used for univariate and multivariate analyses. P values <0.05 were considered statistically significant. All statistical tests were performed on SPSS version 28. In accordance with the journal’s guidelines, we will provide our data for independent analysis by a selected team by the Editorial Team for the purposes of additional data analysis or for the reproducibility of this study in other centers if such is requested.

RESULTS

A total of 185 patients met the study inclusion criteria; 139 patients had tumors with adnexal involvement only, 40 with serosal involvement only, and six with combined adnexal/serosal involvement. For patients with tumors with adnexal, serosal, and combined adnexal/serosal involvement, median age at diagnosis was 58 years (range 23–83), 66 years (range 51–89), and 71 years (range 58–76), respectively, (p<0.001; Table 1); median body mass index was 29.2 kg/m2 (range 19.0–55.2), 28.0 kg/m2 (range 18.0–65.9), and 32.8 kg/m2 (range 24.8–37.0), respectively (p=0.3).

Table 1

Clinicopathologic characteristics of patients with full-thickness serosal, adnexal, or combined full-thickness serosal with adnexal involvement of endometrial carcinoma.

Sixty-seven patients (48.2%) with adnexal involvement, 23 (57.5%) with serosal involvement, and 2 (33.3%) with combined adnexal/serosal involvement underwent minimally invasive surgery (p=0.4). Pelvic lymph node dissection (with or without para-aortic lymph node dissection) was performed in 97 patients (69.8%), 25 patients (62.5%), and 4 patients (66.7%) with tumors with adnexal, serosal, and combined adnexal/serosal involvement, respectively; sentinel lymph node dissection with additional lymphadenectomy was performed in 14 patients (10.1%), 3 patients (7.5%), and 1 patient (16.7%), respectively (p=0.7).

There was a significant difference in histology among the three groups (p=0.015). The six patients with tumors with combined adnexal/serosal involvement had a higher proportional representation of clear cell (n=1, 16.7%), carcinosarcoma (n=2, 33.3%), and mixed/other (n=2, 33.3%) histology compared with patients with serosal-only (p=0.003) and adnexal-only (p=0.009) involvement (Online Supplemental Table 1). Among 115 patients with endometrioid histology, FIGO grade 3 was present in 17 (19.1%) of 89 tumors with adnexal involvement, 12 (48.0%) of 25 with serosal involvement, and 0 of 1 with combined adnexal/serosal involvement. This difference was significant when comparing tumors with adnexal-only and serosal-only involvement (p=0.007). Lymphovascular space invasion was documented for 175 patients, and present in 50 (37.3%) of 134 tumors with adnexal involvement, 23 (63.9%) of 36 tumors with serosal involvement, and 3 (60.0%) of 5 tumors with combined adnexal/serosal involvement. This difference was significant only when comparing tumors with adnexa-only and serosa-only involvement (p=0.004) (Online Supplemental Table 2).

Peritoneal washings were collected for 147 patients; washings were negative for 89 patients (60.5%), positive for 50 patients (34.0%), and suspicious for 8 patients (5.4%). Positive peritoneal washings were identified in 42 (36.8%) of 114 tumors with adnexal involvement, 7 (24.1%) of 29 tumors with serosal involvement, and 1 (25.0%) of 4 tumors with combined adnexal/serosal involvement. Cervical stromal involvement was found in 15 (10.8%) of 139 tumors with adnexal involvement, 1 (2.6%) of 39 with serosal involvement, and 1 (16.7%) of 6 with combined adnexal/serosal involvement.

Adjuvant treatment was prescribed to 130 (94%) of 139 patients with adnexal-only, 36 (90%) of 40 with serosal-only, and 6 (100%) of 6 with combined adnexal/serosal involvement (Table 1). No patients during the study time period received immunotherapy in the upfront setting. Adjuvant radiation therapy alone was prescribed to 57 patients (41.0%) with adnexal-only involvement, 18 (45.0%) with serosal-only involvement, and 3 (50.0%) with combined adnexal/serosal involvement. The most common radiation regimen in each group was external beam radiation therapy; 46 (80.7%) of 57 patients with adnexal-only involvement, 11 (61.1%) of 18 with serosal-only involvement, and 2 (66.7%) of 3 with combined adnexal/serosal involvement received external beam radiation therapy (Online Supplemental Table 3). Additional differences in adjuvant treatment by institution were also noted (Online Supplemental Table 4). The entire patient cohort was stratified by decade of diagnosis (2000–2009 vs 2010–2019), and adjuvant treatment was assessed between surgery alone (n=13), chemotherapy alone (n=52), radiation alone (n=78), and combined chemoradiation (n=42). The rate of radiation therapy alone decreased from 54.3% (n=44) in 2000 to 2009, to 32.7% (n=34) in 2010 to 2019, while the rate of combined chemoradiation increased from 12.3% (n=10) to 30.8% (n=32) (Table 2).

Table 2

Adjuvant treatment modalities prescribed to entire patient cohort (n=185) during the entire study period (2000–2019), stratified by decade.

Median follow-up for the entire cohort was 77 months (range 0.6–254); 81 months (range 0.6–254) for patients with adnexal involvement, 59 months (range 3–171) for patients with serosal involvement, and 73 months (range 4–79) for patients with combined adnexal/serosal involvement. The 5-year progression-free survival and overall survival rates for the entire cohort of patients with stage IIIA disease included in the study was 73.8% (SE 3.5%) and 81.0% (SE 3.1%), respectively. Among patients with tumors with adnexal, serosal, and combined adnexal/serosal involvement, 5-year progression-free survival rates were 80% (SE 3.8%), 61% (SE 8.3%), and 33% (SE 19.2%), respectively; (p<0.01; Figure 1); 5-year overall survival rates were 85% (SE 3.3%), 70% (SE 7.8%), and 60% (SE 21.9%), respectively (p=0.09; Figure 2). The progression-free survival remained statistically significant when comparing the adnexal-only and serosal-only tumors (p=0.03), whereas overall survival did not (p=0.06).

Figure 1

Progression-free survival of patients with adnexal, serosal, or combined serosal with adnexal involvement of endometrial carcinoma.

Figure 2

Overall survival of patients with adnexal, serosal, or combined serosal with adnexal involvement of endometrial carcinoma.

Patients were then dichotomized into two cohorts—adnexal-only involvement and serosal with/without adnexal involvement. On univariate Cox regression analysis, tumors with serosal with/without adnexal involvement, non-endometrioid histology, and lymphovascular space invasion were significantly associated with a higher risk of recurrence (Table 3). Tumors having serosal with/without adnexal involvement retained their statistically significant association with recurrence even after adjusting for histology and lymphovascular space invasion (adjusted HR=2.2, 95% CI 1.2 to 4.3; p=0.01). Peritoneal washings, cervical involvement, and adjuvant treatment were not significantly associated with recurrence on univariate analysis. These findings remained consistent when excluding combined serosal with adnexal involvement (n=6) and comparing adnexal-only (n=139) to serosal-only tumors (n=40; Online Supplemental Table 5).

Table 3

Univariate and multivariate analyses for the association with recurrence among patients with adnexal-only or serosal with or without adnexal involvement.

DISCUSSIONSummary of Main Results

Our multi-institutional, international study found distinct survival outcomes for patients with endometrial cancer with full-thickness serosal, adnexal, and combined serosal/adnexal involvement. Progression-free survival was poorest among patients with serosal involvement, regardless of adnexal involvement, even when adjusting for histologic subtype and lymphovascular space invasion. Receipt of adjuvant treatment was not significant on univariate analysis. These results support the notion that serosal involvement with or without adnexal involvement is associated with worse outcomes compared with adnexal involvement alone.

Results in the Context of Published Literature

The FIGO staging system has evolved and been refined over the past four iterations (Online Supplemental Table 6). The 2023 FIGO staging system considers endometrial carcinoma with locoregional spread to the uterine serosa or to the adnexa as separate entities.2 Stage IIIA1 refers to spread to the adnexa, and stage IIIA2 refers to spread through the uterine serosa or involvement of the uterine subserosa. Histology and pathologic features may further modify the staging criteria of adnexal involvement to the newly designated stage IA3; this applies to low-grade endometrioid tumors with superficial myometrial involvement, absence of extensive lymphovascular space invasion, and presence of ovarian tumor that is unilateral and confined to the ovary. Our collaborative multi-institutional data did not collect all parameters necessary to accurately define stage IA3 disease in our patient population, and therefore we were limited in our ability to analyze this disease stage. According to NCCN guidelines, the current treatment paradigm for surgically staged, stage III endometrial carcinoma is systemic therapy with or without radiation. Furthermore, combination therapy is dependent on the risks of locoregional and distant metastates.13 Notably, there is a very small representation of stage IIIA disease in clinical trials that include patients with stage III disease. Recent retrospective studies report mixed results in determining benefit of specific treatment modalities.8 14–19 The historical literature assessing serosa and adnexal involvement of stage IIIA disease are heterogeneous, as they include past iterations of FIGO staging, which included cytology in the IIIA staging designation.9 11 A 2011 study by Jobsen et al. assessed serosal and adnexal tumor involvement with 2009 FIGO staging among 67 patients. This study found worse progression-free survival for serosal-only tumors compared with adnexal tumors (5-year progression-free survival rate of 76.4% vs 59.6%), with no difference in disease-specific survival (5-year disease-specific survival rate of 76.3% vs 75.4%).10

The Post Operative Radiation Therapy in Endometrial Carcinoma (PORTEC)-3 and Gynecologic Oncology Group (GOG)-258 trials are both prospective, randomized controlled studies that included patients with 2009 FIGO stage III endometrial cancer. PORTEC-3 compared radiation alone with chemoradiation and found improved progression-free survival for patients with combination treatment. GOG-258 compared chemotherapy alone with chemoradiation and did not report a significant improvement in progression-free survival with either treatment, although improvement in local disease control was observed in the chemoradiation arm.20 21 PORTEC-3 reported results on patients with stage III disease (n=295); however, lymphadenectomy was not mandatory for inclusion across all surgically treated patients in the trial. Alternatively, all 151 patients with 2009 FIGO stage IIIA disease on the GOG-258 trial underwent full surgical staging, which included pelvic and/or para-aortic lymph nodes. These landmark studies have led to contemporary challenges in treatment management due to the inclusion of patient populations with overlapping pathologic and treatment features and the need to clarify which patients might benefit most from radiation treatment.

A 2020 survey by the Society of Gynecologic Oncology described the practice patterns in advanced-stage endometrial cancer. Among 187 respondents, 65% favored treating 2009 FIGO stage IIIA disease with combination therapy (either external beam radiation therapy or vaginal brachytherapy), with 7–10% of respondents choosing radiation alone12; choice of treatment did not differ based on adnexal or serosal involvement. In a 2005 survey by the Society of Gynecologic Oncology with 425 respondents, radiation alone as first-line treatment for stage IIIA endometrial cancer was favored by 63% of respondents for treatment of tumors with serosal involvement, and by 69% of respondents for treatment of tumors with adnexal involvement.22The 2005 survey used 1988 FIGO staging criteria, which included cytology in stage IIIA disease, but included details pertaining to the contemporary staging with uterine serosal and/or adnexal involvement.23

The modalities of radiation differed between the 2005 and 2020 Society of Gynecologic Oncology surveys. While potential responses on the 2005 survey included whole pelvic radiation, whole abdominal radiation, intraperitoneal radioactive chromic phosphate, and vaginal vault radiation, responses on the 2020 survey were limited to external beam radiation therapy and vaginal brachytherapy. These surveys highlight the evolution of radiation treatment and the shift away from radiation alone toward a chemotherapy-based regimen. This shift was likely spurred by results of the GOG-122 trial, which demonstrated superior survival outcomes with chemotherapy compared with whole abdominal radiation in advanced-stage endometrial cancer.24

Strengths and Weaknesses

Our study is strengthened by having one of the largest international, multi-institutional, retrospective cohorts of patients with stage IIIA endometrial cancer and either full-thickness uterine serosal, adnexal, or combined adnexal/serosa involvement. All surgical specimens in our study underwent expert pathology review at each institution, and only patients who were surgically staged were included. In addition to limitations inherent to retrospective study design, we recognize the evolution of endometrial cancer care over the study period, which might have affected survival outcomes. Although not an aim of our study, we were unable to retroactively re-stage patients using the updated 2023 FIGO staging system, as collection of information on required pathologic variables was limited. For example, there were 58 patients who could have potentially had a stage change from 2009 FIGO stage IIIA to either 2023 FIGO stage IIIA1 or stage IA3. Our study was designed to study serosal involvement with full-thickness myoinvasion; 2023 FIGO stage IIIA2 disease also includes tumors with subserosal involvement. Our results support the high-risk groupings between full-thickness serosa- and adnexa-containing tumors but cannot evaluate tumors with deep myoinvasion and no serosal involvement.

Implications for Practice and Future Research

Our multi-institutional data provide a real-world assessment of changing treatment patterns and exemplify the challenge in designing standardized regimens. Within our study population, most patients received radiation alone (n=78, 42.2%), followed by chemotherapy (n=52, 28.1%), and combined chemotherapy with radiation (n=42, 22.7%). Our cohort contains international data that were collected over a time period of practice-changing clinical trials, and we have not seen a trend toward combination therapy in the last decade.20 21 25 The distribution between chemotherapy alone, radiation alone, and combination therapy, however, was nearly equal during this time (Table 2). Our data affirm that treatment considerations must account for concerning pathologic features, as patients with high-risk histology or lymphovascular space invasion had poorer survival outcomes.

CONCLUSIONS

The definitions of stage III disease have evolved over the past three iterations of the FIGO staging system for endometrial cancer, first with the inclusion of cytology and now further stratification based on serosal and adnexal involvement. The current study adds to the literature describing the high-risk nature of serosal involvement in stage III endometrial carcinoma. Treatment trends within our patient cohort changed over time (by decade) and provide insight into the management of stage IIIA disease, an uncommon stage of endometrial carcinoma.

Data availability statement

Data are available upon reasonable request. The data that support the findings of this study are available from the corresponding author upon reasonable request.

Ethics statementsPatient consent for publication

Not applicable.

Ethics approval

This study involves human participants and was approved by the institutional review boards at Memorial Sloan Kettering Cancer Center, Mayo Clinic Rochester, MD Anderson Cancer Center, University of Oslo, and Kliniken Essen-Mitte. Participants gave informed consent to participate in the study before taking part.