Article Text

other Versions

Download PDFPDF

Role of adjuvant therapy in stage IIIC2 endometrial cancer
  1. Giorgio Bogani1,
  2. Serena Cappuccio2,
  3. Jvan Casarin3,
  4. Deepa Maheswari M Narasimhulu4,
  5. William A Cilby4,
  6. Gretchen E Glaser4,
  7. Amy L Weaver5,
  8. Michaela E McGree5,
  9. Gary L Keeney6,
  10. John Weroha7,
  11. Ivy A Petersen8 and
  12. Andrea Mariani4
  1. 1Gynecologic Oncology, Fondazione IRCCS Istituto Nazionale dei Tumori di Milano, Milan, Italy
  2. 2Department of Woman's, Child's and Public Health, Fondazione Policlinico Universitario A. Gemelli IRCCS, Rome, Italy
  3. 3Department of Obstetrics and Gynecology, “Filippo Del Ponte” Hospital, University of Insubria, Varese, Italy
  4. 4Division of Gynecologic Oncology, Mayo Clinic Rochester, Rochester, Minnesota, USA
  5. 5Division of Biomedical Statistics and Informatics, Mayo Clinic Rochester, Rochester, Minnesota, USA
  6. 6Division of Anatomic Pathology, Mayo Clinic Rochester, Rochester, Minnesota, USA
  7. 7Division of Medical Oncology, Mayo Clinic Rochester, Rochester, Minnesota, USA
  8. 8Department of Radiation Oncology, Mayo Clinic Rochester, Rochester, Minnesota, USA
  1. Correspondence to Dr Andrea Mariani, Division of Gynecologic Oncology, Mayo Clinic, 200 First St SW, Rochester MN 55905, Minnesota, USA; mariani.andrea{at}mayo.edu

Abstract

Objective The role of the different types of adjuvant treatments in endometrial cancer with para-aortic node metastases is unclear. The aim of this study was to report oncologic outcomes after adjuvant therapy in patients with stage IIIC2 endometrial cancer.

Methods This retrospective single-institution study assessed patients with stage IIIC2 endometrial cancer who underwent primary surgery from January 1984 to December 2014. All patients had hysterectomy (±salpingo-oophorectomy) plus lymphadenectomy (para-aortic nodes, ±pelvic nodes). We included all patients with stage III endometrial cancer and documented para-aortic lymph node metastases (International Federation of Obstetrics and Gynecologists stage IIIC2). We excluded patients who did not provide consent, who had synchronous cancer, or who underwent neoadjuvant chemotherapy. Follow-up was restricted to the first 5 years post-operatively. Cox proportional hazards models, with age as the time scale, was used to evaluate associations of risk factors with disease-free survival and overall survival.

Results Among 105 patients with documented adjuvant therapy, external beam radiotherapy was administered to 25 patients (24%), chemotherapy to 24 (23%), and a combination (chemotherapy and external beam radiotherapy) to 56 (53%) patients. Most patients receiving chemotherapy and external beam radiotherapy (80%) had chemotherapy first. The majority of relapses had a distant component (31/46, 67%) and only one patient had an isolated para-aortic recurrence. Non-endometrioid subtypes had poorer disease-free survival (HR 2.57; 95% CI 1.38 to 4.78) and poorer overall survival (HR 2.00; 95% CI 1.09 to 3.65) compared with endometrioid. Among patients with endometrioid histology (n=60), chemotherapy and external beam radiotherapy improved disease-free survival (HR 0.22; 95% CI 0.07 to 0.71) and overall survival (HR 0.28; 95% CI 0.09 to 0.89) compared with chemotherapy or external beam radiotherapy alone. Combination therapy did not improve prognosis for patients with non-endometrioid histology (n=45).

Conclusions In our cohort of patients with stage IIIC2 endometrioid endometrial cancer, those receiving chemotherapy and external beam radiotherapy had improved survival compared with patients receiving chemotherapy or external beam radiotherapy alone. However, the prognosis of patients with non-endometrioid endometrial cancer remained poor, regardless of the adjuvant therapy administered. Distant recurrences were the most common sites of failure.

  • radiotherapy

Statistics from Altmetric.com

Request Permissions

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.

Highlights

  • Chemotherapy and radiotherapy improved survival in patients with stage IIIC2 endometrioid endometrial cancer.

  • Patients with non-endometrioid endometrial cancer had a poor prognosis, regardless of adjuvant therapy.

  • Most stage IIIC2 endometrial cancer cases had distant recurrences; isolated para-aortic failure was rare.

Introduction

For patients with endometrial cancer clinically confined to the uterus, lymphadenectomy became more common after the Gynecologic Oncology Group published study GOG33, which showed the prognostic value of occult lymph node metastases.1 In 1988, the International Federation of Obstetrics and Gynecologists (FIGO) introduced the concept of surgical staging.2 According to this staging system, the presence of pelvic or para-aortic lymph node metastases in patients with tumors confined to the uterine corpus restages the disease from stage I to stage IIIC.2 The 2009 revised FIGO staging system recognized the different prognoses associated with pelvic and para-aortic node metastases and stratified stage IIIC into stages IIIC1 and IIIC2, depending on whether neoplastic cells were present in the para-aortic nodes.3 4

Although the medical literature consistently recognizes the prognostic significance of aortic node metastases,4–13 no clear and mature evidence is currently available about the efficacy of adjuvant therapy in endometrial cancer with para-aortic node metastases. Owing to the relatively small number of patients with para-aortic nodal disease (±pelvic nodal disease), prior studies generally grouped all patients with advanced (stage III and IV)14–16 or all stage IIIC endometrial cancer (stage IIIC1 and IIIC2) for analysis,17–19 making it difficult to determine whether the findings of the overall stage IIIC population can be applied to patients with stage IIIC2 disease.

In our study, we aimed to evaluate prognostic factors influencing the survival of patients with stage IIIC2 endometrial cancer. Additionally, we sought to identify the effectiveness of adjuvant therapy for different histologic subtypes and to determine how the different therapeutic strategies might influence patterns of recurrence.

Methods

This retrospective analysis was approved by the Mayo Clinic Institutional Review Board. We identified all patients who underwent primary surgery for endometrial cancer at Mayo Clinic (Rochester, Minnesota, USA) from January 2, 1984, through December 31, 2014. All patients underwent hysterectomy (±salpingo-oophorectomy) plus lymphadenectomy (para-aortic nodes,±pelvic nodes). Other staging and adjunctive procedures were performed according to patient and disease characteristics. All patients had negative chest imaging. Demographic characteristics and disease- and treatment-specific details were abstracted from the health records. We excluded patients who did not give consent for the use of their clinical information for research purposes, those who had synchronous cancer, or those who underwent neoadjuvant chemotherapy. For this study, we selected all patients with stage III endometrial cancer and documented para-aortic lymph node metastases (FIGO stage IIIC2).

Details of surgical treatment at our institution have been reported previously.10 12 20–22 In 1999, Mayo Clinic gradually implemented pelvic and para-aortic lymphadenectomy, and in 2004, a surgical algorithm was developed that considered tumor diameter and histologic features determined during intra-operative frozen section analysis. According to this algorithm, in the absence of extra-uterine macroscopic disease, systematic pelvic and para-aortic lymphadenectomy was performed in patients with at least one of the following findings: (1) deep myometrial invasion (>50%), (2) tumor diameter >2 cm, (3) grade 3 histology, (4) tumor extending outside the uterine corpus, and (5) non-endometrioid endometrial cancer. Systematic lymphadenectomy was defined as pelvic plus para-aortic lymphadenectomy resulting in resection of ≥10 pelvic nodes and ≥5 para-aortic nodes. In the presence of macroscopic extra-uterine disease, lymphadenectomy was performed when optimal cytoreduction was anticipated.12 20–22 Histologic subtypes and architectural grade, as well as surgical stage, were consistent with the taxonomy proposed by the WHO and FIGO criteria, respectively.23

Adjuvant therapy included radiotherapy and chemotherapy. External beam radiotherapy was administered by using three-dimensional conformal or intensity-modulated radiotherapy to deliver standard pelvic doses of 45–50.4 Gy and para-aortic doses of 45 Gy. Platinum-based combination chemotherapy, usually with paclitaxel or doxorubicin (or both), was the treatment of choice. When chemotherapy was the sole adjuvant modality, 4–6 cycles (more commonly 6) were delivered in standard doses. Chemotherapy and radiotherapy were always delivered in sequence (never concomitantly), in accordance with the radiation and medical oncologist recommendations.

Vaginal brachytherapy was administered to some patients, regardless of the primary adjuvant therapy. Recipients were mostly patients with at least one of the following characteristics: grade 3 disease, non-endometrioid endometrial cancer, or lymphovascular space invasion.

The date and type of each patient’s first recurrence were used for the analysis of disease-free survival. The type of recurrence was classified as either local (vaginal), locoregional (lymphatic pelvic and para-aortic), or distant (hematogenous, peritoneal, and distant lymphatic [other than pelvic and para-aortic nodes]). Data were summarized using standard descriptive statistics. Duration of follow-up was calculated from the date of surgical treatment through the date of death or last follow-up. However, considering the 31-year span of the study period, the duration of follow-up was restricted to the first 5 years after surgery so that fair comparisons could be made for the various adjuvant strategies throughout the whole study period. Disease-free survival and overall survival were estimated using the Kaplan-Meier method. Various potential risk factors were evaluated for an association with disease-free survival and overall survival based on fitting univariable Cox proportional hazards models, with age as the time scale (instead of time since surgery) to allow for complete age adjustment. Associations were summarized by calculating HRs and corresponding 95% confidence intervals (CIs). All calculated p values were two-sided and p values <0.05 were considered statistically significant. Analyses were performed with the SAS software package, Version 9.4 (SAS Institute Inc).

Results

Characteristics and type of post-operative therapy

Among the 1222 patients with endometrial cancer undergoing para-aortic lymphadenectomy, 123 (10.1%) had documented stage IIIC2 disease. Patient-, disease-, and treatment-specific characteristics are shown in Table 1. Nine patients (7%) declined adjuvant treatment; of these, eight (89%) died of disease at a median of 5 months (IQR 3–11 months) after surgery and one (11%) was alive at last follow-up (88 months). Another nine (7%) patients had insufficient details on the type of adjuvant therapy performed, mainly because they received treatment at other institutions.

Table 1

Characteristics of patients with stage IIIC2 endometrial cancer stratified by type of adjuvant therapy*

Of the 105 patients with well-documented adjuvant therapy, 81 patients underwent radiotherapy (±chemotherapy) and 24 patients underwent chemotherapy without radiotherapy. Among the 81 patients undergoing radiotherapy (±chemotherapy), radiotherapy was delivered to the pelvic and para-aortic area for 65 patients (80%), to the pelvic area only for one (1%), and to the para-aortic area for two (2%); details about the area receiving radiotherapy were unclear for 13 patients (16%). Fifty-six of the 81 patients received a combination regimen (chemotherapy and external beam radiotherapy); the most common sequence was chemotherapy followed by external beam radiotherapy (n=36, 64%), followed by a 'sandwich' regimen of chemotherapy, external beam radiotherapy and subsequent chemotherapy (n=9, 16%), and less commonly external beam radiotherapy followed by chemotherapy (n=7, 13%). Four patients (7%) had incomplete information on the sequence of treatments. Of the 80 patients who received chemotherapy, six patients (8%) had <4 cycles of chemotherapy administered, 57 patients (71%) had ≥4 cycles, and 17 patients (21%) had incomplete information on the number of cycles administered. Among those with available information, 12 of 16 patients (75%) who had chemotherapy without external beam radiotherapy (±vaginal brachytherapy) had ≥4 chemotherapy cycles compared with 45 of 47 patients (96%) who had chemotherapy and external beam radiotherapy.

Sites of recurrence, stratified by treatment type

We focused our analyses on the 105 patients with well-documented adjuvant therapy and aimed to determine whether the addition of chemotherapy to radiotherapy improved disease-free survival, overall survival, or both. In this group, 36 patients (34%) died of disease, 9 (9%) died of other causes, and 18 (17%) had an unknown cause of death at the time of last follow-up. The remaining 42 patients (40%) were alive at a median of 7.5 years (IQR 4.6–12.2 years) after surgery. In addition, 46 (44%) patients had a recurrence within 5 years, and 46 (44%) patients died within the first 5 years (34 deaths were due to disease). Most relapses (31/46, 67%) had a hematogenous or peritoneal component (or both) and accounted for 47% (7/15), 75% (9/12), and 79% (15/19) of all recurrences in the chemotherapy, external beam radiotherapy, and chemotherapy and external beam radiotherapy treatment subgroups, respectively.

Because all patients had disease in the para-aortic area, we then focused our analysis on para-aortic recurrences. Among the 46 patients with recurrent disease, nine (20%) had para-aortic recurrences, alone or in combination with other routes of dissemination (only one of nine recurrences was isolated to the para-aortic area, with no other associated areas of dissemination). In particular, in the chemotherapy group, four of the 15 recurrences (27%) were para-aortic (one was the isolated para-aortic recurrence); in the external beam radiotherapy group, three of the 12 recurrences (25%) were para-aortic; in the chemotherapy and external beam radiotherapy group, two of the 19 recurrences (11%) were para-aortic. However, these numbers were small and the differences between subgroups lacked statistical significance.

Risk factors for recurrence and death

Association of various risk factors with recurrence and death (Table 2) may help counsel patients on treatment options. We adjusted each analysis for age because older age is generally associated with an increased risk of recurrence (HR per 10-year increase in age 1.28; 95% CI 0.96 to 1.70) and death (HR per 10-year increase in age 1.73; 95% CI 1.28 to 2.36). Patients with grade 3 disease were more likely to have poorer disease-free survival (p=0.004) and poorer overall survival (p=0.04) compared with patients with low-grade disease. Likewise, patients with non-endometrioid histology were more likely to have poorer disease-free survival (p=0.003) and poorer overall survival (p=0.03) than patients with endometrioid histology. Furthermore, the disease-free survival and overall survival of patients with grade 3 endometrioid disease appeared to have intermediate clinical behavior compared with that of patients with low-grade endometrioid tumors (which had a better prognosis) and non-endometrioid cancers (worse prognosis) (Figure 1). The type of adjuvant therapy administered was a strong predictor of disease-free survival and overall survival (Figure 2): chemotherapy and external beam radiotherapy favorably affected disease-free survival (p=0.01) and overall survival (p=0.06) compared with chemotherapy or external beam radiotherapy as sole treatment modalities. After adjusting for age, additional factors univariately associated with poorer disease-free survival included lymphovascular space invasion, cervical stromal invasion, and positive peritoneal cytology. Overall survival was also negatively influenced by lymphovascular space invasion, adnexal invasion, serosal invasion, and positive peritoneal cytology.

Figure 1

Disease-free survival and overall survival after surgery for stage IIIC2 endometrial cancer, stratified by histology and International Federation of Gynecology and Obstetrics grade. Values in parentheses indicate the number at risk.

Figure 2

Disease-free survival and overall survival after surgery for stage IIIC2 endometrial cancer, stratified by adjuvant therapy. Values in parentheses indicate the number at risk. CT, chemotherapy; EBRT, external beam radiotherapy; CT+EBRT, combination of the two treatments.

Table 2

Univariate analysis of factors potentially associated with recurrence or death in patients with stage IIIC2 endometrial cancer and known adjuvant therapy (n=105)*

Stratified analysis by histology and grade

Considering non-endometrioid and endometrioid groups separately, we observed that different adjuvant therapy strategies did not significantly influence disease-free survival and overall survival in patients with non-endometrioid disease (Figure 3B and D). For these patients (non-endometrioid histology), when comparing chemotherapy and external beam radiotherapy to single-therapy modalities, the age-adjusted HR for disease-free survival was 0.50 (95% CI 0.19 to 1.33; p=0.16), and for overall survival, the HR was 0.49 (95% CI 0.19 to 1.25; p=0.13). However, for patients with endometrioid histology, chemotherapy and external beam radiotherapy favorably affected disease-free survival (HR 0.22; 95% CI 0.07 to 0.71; p=0.01) and overall survival (HR 0.28; 95% CI 0.09 to 0.89; p=0.03) compared with single-therapy modalities (Figure 3A and C). When focusing on site-specific recurrence among patients with endometrioid histology, chemotherapy and external beam radiotherapy favorably affected locoregional disease-free survival (HR 0.11; 95% CI 0.01 to 0.95; p=0.04) and distant disease-free survival (HR 0.19; 95% CI 0.04 to 0.95; p=0.04) compared with chemotherapy or external beam radiotherapy as sole treatment modalities (Online supplementary figure 1A and C).

Supplemental material

Figure 3

Disease-free survival and overall survival after surgery for stage IIIC2 endometrial cancer, stratified by adjuvant therapy and histology. (A) Disease-free survival, endometrioid. (B) Disease-free survival, non-endometrioid. (C) Overall survival, endometrioid. (D) Overall survival, non-endometrioid. CT, chemotherapy; EBRT, external beam radiotherapy; CT+EBRT, combination of the two treatments.

After stratifying patients by histology and grade, non-endometrioid cancer was confirmed to have the poorest prognosis, regardless of the adjuvant treatment administered. The chemotherapy and external beam radiotherapy combination was associated with the best disease-free survival and overall survival for patients with low-grade and high-grade endometrioid cancer compared with sole treatment modalities; however, the small sample size limited the value of the present analysis.

Discussion

This single-institution retrospective study investigated outcomes after post-operative adjuvant therapy for patients with stage IIIC2 endometrial cancer. We observed that combination therapy (chemotherapy and external beam radiotherapy; mainly chemotherapy followed by external beam radiotherapy) improved disease-free survival and overall survival compared with either chemotherapy or external beam radiotherapy as a sole modality. However, combination adjuvant therapy had a significant effect only for patients with endometrioid histology. For patients with non-endometrioid tumors, the prognosis was poor, regardless of the type of adjuvant therapy received. Unsurprisingly, in endometrioid endometrial cancer, administration of combined therapy was associated with more favorable locoregional and distant disease-free survival than either treatment alone. When stratifying by grade and histology, patients with low-grade tumors had the best prognosis, those with non-endometrioid cancers had the worst prognosis, and high-grade endometrioid endometrial cancer had an intermediate prognosis. Finally, among women who had recurrences, two-thirds had a peritoneal or hematogenous component and only 20% had a para-aortic component (2% had isolated para-aortic recurrence). These findings emphasize the ongoing lack of adequate systemic therapy for patients with stage IIIC2 disease, and it points to an area that requires concentrated research efforts.

Although the prognostic value of aortic node metastases is well accepted, the most appropriate post-operative therapy for patients with positive aortic nodes is unclear. Because external beam radiotherapy and chemotherapy provide locoregional and systemic control, respectively,16 19 20 a combination regimen theoretically would improve post-operative outcomes compared with either single-modality treatment. Only limited evidence in the 'modern' surgical staging era (after 1988) is available for patients with stage IIIC2 endometrial cancer.24–29

The GOG258 trial30 randomized 736 patients with stage III–IVA endometrial cancer to post-operative chemoradiotherapy or chemotherapy alone. They reported no significant differences in disease-free survival or overall survival between the two groups. However, chemotherapy alone was associated with fewer distant recurrences and more local recurrences compared than the combined regimen. Most patients enrolled in that study (73%) had stage IIIC disease and 183 (25%) had stage IIIC2 disease. Overall, 224 patients (30%) had non-endometrioid histology. Although the GOG258 trial seems to draw conclusions that are inconsistent with those of our study, we note that the combined chemoradiotherapy regimen used in GOG258 differed from the sequential regimens used in our series. In fact, it is certainly possible that the sequence of chemotherapy followed by radiotherapy may provide the combined effect of controlling distant disease and providing local control. Thus, our conclusion is not contradictory to the GOG258 findings. Also, our study suggests that the combined regimen (chemotherapy and external beam radiotherapy) provides the best results, even when compared with chemotherapy alone, only for the subgroup of patients with endometrioid stage IIIC2 tumors. In GOG258, in the separate analysis of stage IIIC2 patients (combined with an additional six patients with stage IVA endometrial cancer), patients were not stratified by histology.

Recently, the PORTEC-3 trial prospectively evaluated the role of adjuvant chemoradiotherapy versus radiotherapy alone for patients with high-risk endometrial cancer.31 32 In a subgroup analysis of approximately 300 patients with stage III endometrial cancer, adjuvant chemoradiotherapy improved 5-year disease-free survival (70.9% vs 58.4%; p=0.01) and 5-year overall survival (78.5% vs 68.5%; p=0.04) compared with external beam radiotherapy alone.32 However, no specific data regarding stage IIIC2 endometrial cancer were reported for that trial.

An interesting aspect of the combination regimen is the sequence of chemotherapy and external beam radiotherapy. In our experience, most patients (88% of those receiving combined treatment) had chemotherapy before external beam radiotherapy or received the sandwich regimen (chemotherapy followed by external beam radiotherapy with subsequent chemotherapy). Data regarding the best sequencing strategy are still scant for stage IIIC endometrial cancer, particularly stage IIIC2. Recently, a multi-institutional retrospective study focused on the role of chemotherapy and external beam radiotherapy and evaluated 179 patients with stage IIIC endometrial cancer (including 59 with stage IIIC2)33 ; they reported that the sandwich regimen was superior to chemotherapy followed by external beam radiotherapy in terms of disease-free survival and overall survival.

The main strengths of the current investigation are the focus on stage IIIC2 endometrial cancer and the stratification of outcomes according to histologic subtype and grade. Also, we report the outcomes of patients who received adjuvant combined chemotherapy and external beam radiotherapy, with most patients receiving chemotherapy before external beam radiotherapy. The main limitations of this study are the inherent biases of a retrospective single-institution cohort of non-randomized patients who were treated during a period that spanned more than three decades. Because of the long study period and the estimated majority of our patients receiving post-operative treatment outside our institution, the type of adjuvant treatment in our population is not homogeneous. In addition, the study design precluded the ability to draw any conclusions about the absolute efficacy of chemotherapy as a single-modality therapy in non-endometrioid endometrial cancer, so only relative efficacy was evaluated.

Another point that deserves attention is the more limited use of para-aortic lymphadenectomy in the sentinel node era. In fact, para-aortic nodal dissection is usually omitted for patients undergoing sentinel node mapping, even in high-risk cases.34 However, it is well known that approximately two-thirds of patients with lymphatic dissemination have para-aortic metastases,35 and this percentage is even higher in the presence of high-grade tumors and large lymphatic metastases in the pelvis.36 Therefore, it is reasonable to assume that most 'high-risk' patients with positive lymph nodes may have occult para-aortic disease and must be treated accordingly. This assumption is less applicable to patients with lymphatic dissemination, low-grade histology, and low-volume nodal disease, in which the risk of para-aortic dissemination may be negligible.35 36

Conclusion

In this large series of patients with stage IIIC2 endometrial cancer, we observed that older age, high grade, and high-risk histology were the strongest predictors of a poor prognosis. Among patients with endometrioid histology, the combination of chemotherapy and external beam radiotherapy favorably affected disease-free survival and overall survival compared with chemotherapy or external beam radiotherapy as sole modalities. In contrast, for patients with non-endometrioid endometrial cancer, the different types of adjuvant treatment did not significantly affect outcomes. In contrast to patients from the GOG258 trial who received chemoradiotherapy followed by chemotherapy as combined therapy, our patients mainly received chemotherapy followed by external beam radiotherapy. This difference in treatment may explain the differences in our findings.

Large prospective studies are needed to evaluate the efficacy of combined chemotherapy and radiotherapy, in terms of dosage and sequencing, for patients with stage IIIC2 endometrioid endometrial cancer. However, randomized studies are unlikely to be feasible because of the relative rarity of patients with stage IIIC2 disease. Also, the fact that the majority of recurrences (67%) had a distant component emphasizes the importance of discovering new systemic therapies for patients with stage IIIC2 endometrial cancer. Future multi-institutional retrospective collaborative studies may allow us to identify the most effective treatment modality for patients with stage IIIC2 endometrial cancer.

References

Footnotes

  • GB and SC contributed equally.

  • Presented at Presented in part as a poster at the 45th Annual Meeting on Women’s Cancer, Tampa, Florida, March 22-25, 2014.

  • Contributors Conceptualization: AM, GB. Methodology: All authors. Statistical analysis: MEM, ALW. Project administration: AM. Supervision: AM, ALW, WAC. Writing, original draft: GB, SC, JC, ALW, AM. Writing, review, and editing: All authors.

  • 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; externally peer reviewed.

  • Data availability statement Data are available upon reasonable request.