Background Adjuvant therapy in early-stage endometrial cancer has not shown a clear overall survival benefit, and hence, patient selection remains crucial.
Objective To determine whether women with high-intermediate risk, early-stage endometrial cancer with lymphovascular space invasion particularly benefit from adjuvant treatment in improving oncologic outcomes.
Methods A multi-center retrospective study was conducted in women with stage IA, IB, and II endometrial cancer with lymphovascular space invasion who met criteria for high-intermediate risk by Gynecologic Oncology Group (GOG) 99. Patients were stratified by the type of adjuvant treatment received. Clinical and pathologic features were abstracted. Progression-free and overall survival were evaluated using multivariable analysis.
Results 405 patients were included with the median age of 67 years (range 27–92, IQR 59–73). 75.0% of the patients had full staging with lymphadenectomy, and 8.6% had sentinel lymph node biopsy (total 83.6%). After surgery, 24.9% of the patients underwent observation and 75.1% received adjuvant therapy, which included external beam radiation therapy (15.1%), vaginal brachytherapy (45.4%), and combined brachytherapy + chemotherapy (19.1%). Overall, adjuvant treatment resulted in improved oncologic outcomes for both 5-year progression-free survival (77.2% vs 69.6%, HR 0.55, p=0.01) and overall survival (81.5% vs 60.2%, HR 0.42, p<0.001). After adjusting for stage, grade 2/3, and age, improved progression-free survival and overall survival were observed for the following adjuvant subgroups compared with observation: external beam radiation (overall survival HR 0.47, p=0.047, progression-free survival not significant), vaginal brachytherapy (overall survival HR 0.35, p<0.001; progression-free survival HR 0.42, p=0.003), and brachytherapy + chemotherapy (overall survival HR 0.30 p=0.002; progression-free survival HR 0.35, p=0.006). Compared with vaginal brachytherapy alone, external beam radiation or the addition of chemotherapy did not further improve progression-free survival (p=0.80, p=0.65, respectively) or overall survival (p=0.47, p=0.74, respectively).
Conclusion Adjuvant therapy improves both progression-free survival and overall survival in women with early-stage endometrial cancer meeting high-intermediate risk criteria with lymphovascular space invasion. External beam radiation or adding chemotherapy did not confer additional survival advantage compared with vaginal brachytherapy alone.
- uterine cancer
- lymphatic vessels
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Optimal adjuvant plan for high-intermediate risk endometrial cancer with lymphovascular space invasion is unknown.
In this retrospective cohort study, adjuvant therapy, irrespective of the type, improved progression-free survival and overall survival in this group.
No survival advantage favored a particular adjuvant therapy approach.
Endometrial cancer is the most common gynecologic malignancy in the United States with over 60 000 cases diagnosed annually. Despite the decrease in the rate of many common cancers in women, the incidence of uterine cancer is steadily increasing.1 At diagnosis, 67% of patients have disease confined to the uterus. Most women diagnosed with early-stage endometrial cancer have a good prognosis, with 5-year survival rates approaching 90%.2 However, among patients who experience recurrence, outcomes are poor.
According to National Comprehensive Cancer Network guidelines, endometrial cancer confined to the uterus is generally managed surgically with hysterectomy, salpingo-oophorectomy, and staging.3 Adjuvant therapy in this setting is dependent on pathologic and age-based risk factors. While several landmark clinical trials have attempted to identify women at risk for recurrence in order to determine the optimal treatment paradigm, the superior strategy remains elusive. In Gynecologic Oncology Group (GOG) 99 by Keys et al, 392 women with early-stage endometrial cancer were randomized to either adjuvant external beam radiation therapy or observation following staging surgery.4 With a median follow-up of 69 months, the 2 year recurrence rate was 3% vs 12% (p=0.007); this difference was particularly evident within a subset of patients designated as 'high-intermediate risk'. Patients meeting high-intermediate risk criteria based on their age and uterine factors were found to benefit significantly from adjuvant external beam radiation compared with observation (recurrence rate 6% vs 26%, HR 0.42). However, there was no overall survival benefit. These findings were confirmed by Post-Operative Radiation Therapy in Endometrial Carcinoma (PORTEC) 1, a large European randomized trial.5 Similarly, PORTEC 2 showed improvement in only the progression-free survival in both the external beam radiation and the vaginal brachytherapy arms.6 A wide range of treatment options used remain without clear overall survival benefit.
A subset analysis of PORTEC 1 and 2 data identified the presence of substantial lymphovascular space invasion as a risk factor for distant metastasis, pelvic recurrence, and overall survival.7 Lymphovascular space invasion refers to the presence of cancer cells within the lymphatic or vascular space of the uterus. Lymphovascular space invasion is present in 15% of stage I and II endometrial cancer8 and has been shown to be an independent predictor of nodal metastasis (OR 11.04, p<0.00001)9 as well as distant metastasis.9–11 The rate of recurrence is significantly higher with lymphovascular space invasion,9 12 13 even after adjusting for negative lymph node status (p=0.02),10 and overall survival is worse.11 14 Some authors advocate systemic therapy in this subgroup of patients15; however, data are limited.16 Further investigation is needed for patient selection.
In part due to the lack of effect on overall survival, adjuvant therapy for high-intermediate risk endometrial cancers remains controversial; moreover, how lymphovascular space invasion modifies this risk has not been investigated. The primary objective of this study was to examine whether women meeting high-intermediate risk criteria in early-stage endometrial cancer with lymphovascular space invasion particularly benefit from adjuvant therapy. The secondary objective was to compare the effect on outcome of various modes of adjuvant therapy.
A multi-institutional retrospective study was conducted in collaboration with eight academic centers across the United States including the Cleveland Clinic (Cleveland, Ohio), Johns Hopkins University School of Medicine (Baltimore, Maryland), University of Oklahoma Health Sciences Center (Oklahoma City, Oklahoma), University of North Carolina (Chapel Hill, North Carolina), Mayo Clinic (Rochester, Minnesota), University of Virginia Health System (Charlottesville, Virginia), Duke University Medical Center (Durham, North Carolina), and the University of Cincinnati Academic Health Center (Cincinnati, Ohio). Institutional review board approval was obtained at all sites.
Inclusion criteria were women with stage IA, IB, and II endometrial cancer with lymphovascular space invasion who underwent hysterectomy with or without lymphadenectomy from January 2005 to December 2015. All staging was adjusted according to the International Federation of Gynecology and Obstetrics (FIGO) 2009 categorization based on review of the pathology report. All sampled lymph nodes were negative for malignancy. High-risk histology, including clear cell, serous, and carcinosarcoma, were excluded. Patients meeting criteria for high-intermediate risk based on GOG 99 were selected: at least 70 years of age with one of the other risk factors, at least 50 years of age with two of the other risk factors, or any age with all three of the other risk factors. The risk factors were increasing age, moderate to poorly differentiated tumor grade, presence of lymphovascular space invasion, and outer third myometrial invasion. These patients were then stratified by the type of adjuvant treatment received: (1) observation, (2) external beam radiation, (3) vaginal brachytherapy, (4) brachytherapy + chemotherapy, (5) external beam radiation + brachytherapy, (6) chemotherapy, (7) external beam radiation + chemotherapy, (8) external beam radiation + brachytherapy + chemotherapy, and (9) hormonal therapy. The four largest groups were selected for statistical analysis: (1) observation, (2) external beam radiation alone, (3) vaginal brachytherapy alone, and (4) combined brachytherapy + chemotherapy. External beam radiation dosing included 4500 cGy in 25 fractions to 5040 cGy in 28 fractions. Vaginal brachytherapy dosing included 700 cGy to 3000 cGy, with the majority of the patients receiving 2100 cGy in three fractions. Further information regarding radiotherapy techniques and equipment was not collected. Patients who had persistent disease after surgery were excluded. Clinical and pathologic features were abstracted and compared. Demographic data were collected from respective electronic medical records, including age, year of surgery, race, parity, menarche, menopause, and body mass index. Operative reports were reviewed for details regarding the type and route of surgery, pelvic washings, and lymph node assessment. Pathology reports were reviewed for the histopathology of tumor characteristics. Postoperative adjuvant treatment and recurrence data were abstracted.
Normally distributed continuous measures were summarized using means and SD, and compared using t-tests or analysis of variance. Continuous measures that showed departure from normality as well as ordinal measures were summarized using medians/quartiles or frequencies/percentages, and compared using Kruskal-Wallis tests. Categorical factors were summarized using frequencies/percentages and were compared using Pearson’s chi-square tests or Fisher’s exact tests. Survival analyses were done from day of surgery, with a month being defined as 30 days. Cox proportional hazards right-censored univariate models were built for progression-free survival and overall survival. Pre-specified multivariable Cox models were built to examine the effect of adjuvant treatments. Progression-free survival was adjusted by stage (IA, IB, II) and tumor grade 2 or 3. Overall survival was adjusted by stage, tumor grade, and age of diagnosis.
All p values were two-sided, with 0.05 as level of statistical significance and 95% CI. Statistical analyses were performed using Statistical Analysis Software version 9.4 (SAS Institute Inc., Cary NC). Data were collected and managed using REDCap electronic data capture tool hosted at the Cleveland Clinic.17 The de-identified data were shared for the final database as specified by the institutional review board.
In all, 728 patients were identified in the original database with surgically staged IA, IB, and II endometrial cancer with positive lymphovascular space invasion, of which 76% of the cases were contributed by half of the participating institutions. From this database, 405 patients met criteria for high-intermediate risk as defined by GOG 99; 101 patients (24.9%) underwent observation and 304 patients (75.1%) received any form of adjuvant therapy. Specifically, 46 patients underwent external beam radiation alone (15.1%), 138 vaginal brachytherapy alone (45.4%), 58 combined brachytherapy +chemotherapy (19.1%), 21 combined external beam radiation +brachytherapy (6.9%), and eight chemotherapy alone (2.6%). The remaining 33 patients (10.9%) underwent external beam radiation +chemotherapy, external beam radiation +brachytherapy + chemotherapy, or hormonal therapy (Online supplementary file 1).
The total adjuvant therapy group and the four largest subgroups were analyzed (Tables 1 and 2). The mean age of patients undergoing observation (67.8±11.9 years) was comparable to that of the total adjuvant therapy group (66.2±8.8 years, p=0.22) as well as to that of the individual analysis subgroups: external beam radiation (67.7±8.8 years), vaginal brachytherapy (67.9±9.5 years) and combined brachytherapy + chemotherapy (64.3±7.2 years) groups (p=0.11). Body mass index, race, parity, age of menarche, and menopausal status were not statistically different.
Surgical and pathologic factors are shown in Tables 1 and 2. The majority of patients underwent minimally invasive hysterectomy: 72.3%, 91.3%, 86.2%, and 93.1% in the observation, external beam radiation, vaginal brachytherapy, and brachytherapy +chemotherapy groups, respectively (p<0.001). Lymph node assessment was performed in 83.6% of patients (n=336), including 75.0% lymphadenectomy (n=303) and 8.6% sentinel lymph node biopsy (n=33). Women who received adjuvant therapy were more likely to have undergone lymphadenectomy (63.0% in observation, 76.1%, 82.6%, and 86.2% in external beam radiation, vaginal brachytherapy, brachytherapy + chemotherapy, respectively, p=0.001).
As expected, the adjuvant therapy groups had a higher stage distribution than the observation group (p=0.003, Table 2). Grade and histology were not significant. Lower uterine segment involvement was the lowest in the vaginal brachytherapy group and the highest in the brachytherapy + chemotherapy group (p=0.019). Myometrial invasion and tumor size were not different among the groups.
There was a significant difference in the 5-year progression-free survival: 69.6% in the observation group and 77.2% in the total adjuvant therapy group, HR 0.55 (p=0.011, Table 3, Figure 1). Moreover, the 5-year overall survival was significantly different: 60.2% in the observation group and 81.5% in the adjuvant treatment group, HR 0.42 (p<0.001, Figure 1). Recurrence occurred in 27% of the observation group and 18% of the adjuvant treatment group. The length of follow-up ranged from 35.9 months in the observation group to 38.4 months in the external beam radiation group, 42.8 months in the vaginal brachytherapy group, and 54.1 months in the brachytherapy + chemotherapy group (p=0.009). Site of recurrence was not statistically significant among the groups, with distant metastasis being the most common site (45.6% of all recurrences, overall p=0.15). Interestingly, the lowest rate of combined distant and multi-site recurrence was in the brachytherapy + chemotherapy group (8.6% vs 13.9% observation, 19.5% external beam radiation, 12.3% vaginal brachytherapy).
On multivariable analysis, progression-free survival was adjusted by stage (IA vs IB, HR 1.92, p=0.023; II, HR 2.81, p=0.014) and grade 2/3 (HR 1.92, p=0.1). Overall survival was adjusted by stage (IA vs IB HR 2.20, p=0.006; II HR 3.47, p=0.003), grade 2/3 (HR 3.03, p=0.006), and age (HR 1.06, p<0.001). Improved progression-free survival was demonstrated for vaginal brachytherapy and brachytherapy + chemotherapy compared with observation (HR 0.42, p=0.003; HR 0.35, p=0.006 respectively, Table 4). Progression-free survival in the external beam radiation group compared with observation was not significant (p=0.054). However, all groups including external beam radiation, vaginal brachytherapy, and brachytherapy + chemotherapy were associated with improved overall survival compared with observation (HR 0.47, p=0.047; HR 0.35, p<0.001; HR 0.30, p=0.002, respectively). Pairwise comparison for external beam radiation, vaginal brachytherapy, brachytherapy + chemotherapy groups did not reveal a statistically significant difference in progression-free survival and overall survival when each group was compared with the other (Table 5).
Patients with early-stage endometrial cancer with lymphovascular space invasion are at high risk for metastasis, recurrence, and poor survival.7 9–11 While randomized data demonstrate improved loco-regional control with adjuvant pelvic radiation in women meeting high-intermediate risk criteria, current literature does not address how treatment should be tailored in patients with lymphovascular space invasion. In this multi-institutional analysis of 405 patients meeting high-intermediate risk criteria with positive lymphovascular space invasion, we showed that adjuvant therapy improves both progression-free and overall survival. Survival advantage did not favor a particular adjuvant therapy approach.
Several studies have shown improved progression-free survival in patients receiving adjuvant therapy for high-risk uterine factors. PORTEC 1 was an intention-to-treat analysis of 714 unstaged patients with unfavorable uterine factors. Similarly to GOG 99, this study showed improved loco-regional recurrence in the external beam radiation group compared with control (4% vs 14%, p<0.001).5 In contrast, our study did not find a progression-free survival benefit for women undergoing external beam radiation (p=0.054). This finding may be due to limitations of the small sample size, differing treatment patterns secondary to lymph node status, or potential selection bias towards healthier patients receiving external beam radiation. One of the main criticisms of PORTEC 1 and 2 is the lack of full staging, including lymph node assessment. Our study included more than 83.6% of patients who were surgically staged and confirmed these trial findings.
Moreover, neither of the PORTEC studies included lymphovascular space invasion as a stratification criterion, which is a unique feature of our study. In a later subset analysis of PORTEC 1 and 2, Bosse and colleagues demonstrated that substantial lymphovascular space invasion is a strong prognostic factor for pelvic recurrence (HR 6.2, 95% CI 2.4 to 16), distant metastasis (HR 3.6, 95% CI 1.9 to 6.8), and overall survival (HR 2.0, 95% CI 1.3 to 3.1).7 In their cohort of 129 patients with lymphovascular space invasion, treatment with brachytherapy significantly decreased vaginal recurrence, and external beam radiation significantly reduced both vaginal and pelvic recurrences. Neither modality affected distant recurrence or survival; however, this study was not powered to compare treatment modalities. Similarly, a 10-year analysis of PORTEC 2 confirmed substantial lymphovascular space invasion as a strong risk factor for metastasis and also cancer-related survival,18 which was consistent with prior smaller retrospective studies.11 14 Both these secondary analyses of the PORTEC trials showed improvement in only the progression-free survival with adjuvant radiation.7 18 In our study, which contained a higher number of lymphovascular space invasion positive women in each group, we were able to detect an improvement in overall survival. The present study builds on the important findings of the PORTEC sub-analyses but also demonstrates that vaginal brachytherapy alone may be sufficient adjuvant treatment in patients with high-intermediate risk endometrial cancer and lymphovascular space invasion who have undergone surgical staging with lymphadenectomy.
As such, the adjuvant treatment paradigm based on uterine factors alone has not been shown to improve overall survival. On the other hand, Beavis et al analyzed 478 all-comers with stage I and II endometrial cancer with positive lymphovascular space invasion. In that study, adjuvant therapy improved progression-free survival compared with observation (chemotherapy±radiation HR 0.18, radiation only HR 0.31, both p<0.05) but, again, not overall survival.15 Combined, these data suggest that patients meeting both high-intermediate risk criteria and positive lymphovascular space invasion may uniquely benefit from adjuvant therapy in improving overall survival. Prospective studies examining this particular patient population are warranted, and this criterion may be a useful factor for stratification in clinical trials.
Finally, GOG 249 was a randomized phase III trial of 601 patients with high-intermediate and high-risk endometrial cancer which found equivalent outcomes in those who received external beam radiation and those who received combined brachytherapy + chemotherapy.19 PORTEC 3 demonstrated similar results in a much more expanded patient population.20 According to the authors of PORTEC 3, chemotherapy showed the greatest benefit in stage III endometrial cancer because of the higher risk of recurrence. This finding was reflected in the study by Beavis et al, which found superior progression-free survival in patients receiving chemotherapy in grade 3 early-stage endometrial cancer with lymphovascular space invasion compared with radiation alone (HR 0.25, p<0.05).15 In our study, the combined brachytherapy + chemotherapy group did not show improved survival relative to other adjuvant modalities; however, distant and multi-site recurrence was the lowest in this group. The role of chemotherapy in high-intermediate risk patients with lymphovascular space invasion remains to be confirmed.
Limitations of this study include its retrospective nature, as well as differing practice patterns at multiple institutions. Due to the small sample size, other subgroups such as chemotherapy alone, other combined radiation therapies, and hormonal therapy could not be statistically analyzed. In addition, we were unable to control for factors which may affect patient outcomes, such as radiation techniques and provider bias in treating higher stage disease, and this was perhaps suggested by the higher stage distribution of the adjuvant therapy groups compared with observation. Despite this, we were able to find a survival advantage, which further supports our conclusion. Because part of our data predates the 2009 staging change, it is unclear whether pelvic washing status influenced treatment planning. Patient medical co-morbidities could not be elicited from the available data, which may particularly affect the observation group. Likewise, the various institution and provider-based criteria for electing observation over adjuvant therapy could not be elicited, which is a major limitation. Finally, the extent of lymphovascular space invasion was not sub-typed as this was not yet routinely reported at every institution.
Despite these limitations, our study represents a multi-institutional cohort of surgically staged patients with relatively homogeneous pathology undergoing a full spectrum of adjuvant treatment, and strengthens the present data supporting adjuvant treatment to improve oncologic outcomes. The fact that any adjuvant therapy improved overall survival in this cohort is a novel finding. This is counter to results from several prospective trials of patients with high-intermediate risk endometrial cancer; however, our study is the only one to look specifically at the sub-group of higher-risk patients with positive lymphovascular space invasion.
In conclusion, our retrospective analysis showed that the subset of patients with high-intermediate risk endometrial cancer with positive lymphovascular space invasion may uniquely benefit from adjuvant treatment in improving both progression-free and overall survival. The morbidity of each adjuvant modality should be carefully considered when selecting a treatment regimen. Specifically, the survival advantage of brachytherapy was comparable to external beam radiation and combined brachytherapy + chemotherapy and may be considered for adjuvant treatment alone in this patient population.
Twitter @laurajmoulton, @amandanfader, @wcburkett
Contributors Conception, data analysis and interpretation, manuscript drafting: JS, LMC, SR. Additional interpretation and analytical edits: CC, CMM, MY, PAG, DT. Statistical support: MY. Conception and management of the original full database: ALB, T-TY, ANF. Data collection, management: JS, LMC, CC, ALB, T-TY, RLS, SLW, ANF, WCB, DLR, AS-S, SA, PAG, DT, SCD, MWS, SCM, CW, AVe, JE, LH, AAS, AL, KG, AJ, AVi. All authors critically revised the manuscript and approved the manuscript in its final version prior to submission.
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 All authors reported their potential conflict of interest and there were none directly relating to this study.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement All data relevant to the study are included in the article or uploaded as supplementary information.