Objective It is unclear how to best sequence adjuvant chemotherapy and radiotherapy for advanced endometrial cancer. We studied the outcomes for women treated with chemotherapy before radiotherapy in a chemotherapy-first (chemotherapy for 6 cycles followed radiotherapy) or ‘sandwich’ approach (chemotherapy for 3 cycles followed by radiotherapy and subsequently chemotherapy for 3 cycles).
Methods Women with stage IIIC endometrial cancer and no gross residual disease treated with chemotherapy before radiotherapy between April 2003 and April 2016 were included. The Kaplan-Meier method was used to estimate recurrence and survival. We performed a meta-analysis of endometrial cancer trials comparing chemotherapy and radiotherapy versus radiotherapy alone.
Results A total of 102 patients were included. The mean (SD) age was 63.8 (10.6) years; 84 patients received the chemotherapy-first approach and 18 patients received the ‘sandwich’ approach. Pelvic and para-aortic nodes were removed in 99% and 88.2%, respectively. Among all the patients, we observed 1 pelvic (1%), 1 para-aortic (1%), and 5 vaginal (4.9%) recurrences. At 3 years, for the ‘sandwich’ and chemotherapy-first approaches, the vaginal recurrence was 11.8% and 4.2%, pelvic recurrence was 0% and 1.5%, para-aortic recurrence was 0% and 1.2%, distant recurrence was 42.9% and 24.4%, and overall survival was 70.3% and 81.7%, respectively. With ‘chemotherapy before radiotherapy’ 94.9% completed 4+ chemotherapy cycles (vs 71–90% reported in the literature for ‘radiotherapy before chemotherapy’). In a meta-analysis of endometrial cancer trials, distant recurrence rates were reduced with 4+ chemotherapy cycles but not with 3 cycles (p=0.01).
Conclusion Chemotherapy before radiation sequencing for stage IIIC endometrial cancer was associated with a high proportion of patients completing 4+ chemotherapy cycles and low locoregional lymphatic recurrence rate, despite delaying radiotherapy until after 3–6 cycles of chemotherapy and not administering concurrent cisplatin.
- uterine cancer
Data availability statement
De-identified data are available upon reasonable request.
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.
Combined adjuvant treatment for advanced endometrial cancer can start with chemotherapy or with radiation.
Women receiving chemotherapy before radiation are more likely to complete 4+ cycles.
Sequencing chemotherapy first improves chemotherapy delivery without compromising locoregional control.
The optimal adjuvant therapy for women with advanced endometrial cancer remains controversial. An updated analysis of PORTEC-3 reported improved progression-free survival and overall survival with chemotherapy and radiotherapy compared with radiotherapy alone.1 Most patients with stage III endometrial cancer are treated with a combination of chemotherapy and radiotherapy.2 To date, no trials have directly compared sequencing approaches for chemotherapy and radiotherapy. Recent endometrial cancer trials3 4 have sequenced radiotherapy first with concurrent cisplatin followed by combination chemotherapy.
Chemotherapy before radiotherapy may be delivered in a chemotherapy-first (chemotherapy for six cycles followed radiotherapy) or ‘sandwich’ approach (chemotherapy for three cycles followed by radiotherapy and subsequently chemotherapy for three cycles).5 Chemotherapy before radiotherapy approaches delay radiation beyond the immediate postoperative period, which could theoretically increase locoregional recurrences (Figure 1). On the other hand, radiotherapy before chemotherapy approaches delay the treatment of micrometastatic disease outside the radiation field and compromise the delivery of chemotherapy due to radiation-related bone marrow suppression. This could increase distant recurrences.5 6
At our institution, we historically used a chemotherapy-first approach; we switched to the ‘sandwich’ approach in 2013. The goal of this study is to summarize the outcomes of patients treated with chemotherapy before radiotherapy in a chemotherapy-first or ‘sandwich‘ approach and compare them to outcomes reported for radiotherapy before chemotherapy in the literature.
This is a single-institution cohort study of women with stage IIIC endometrial cancer who underwent surgery between April 2003 and April 2016 and received adjuvant treatment with chemotherapy and radiotherapy in a chemotherapy before radiotherapy approach. Women with gross residual disease and/or who did not consent to the use of their medical records for research were excluded. The study was approved by the Institutional Review Board.
All included patients had undergone total hysterectomy and bilateral salpingo-oophorectomy. We routinely performed pelvic lymphadenectomy with or without para-aortic lymphadenectomy following our Mayo Clinic surgical algorithm7 until October 2013, when sentinel lymph node assessment was implemented. The planned chemotherapy regimen was carboplatin (AUC 6) and paclitaxel (175 mg/m2) for six cycles. Prior to starting radiotherapy, patients underwent CT scans of the chest, abdomen, and pelvis. External beam radiotherapy was administered using three-dimensional conformal or intensity modulated radiotherapy to standard pelvic doses of 45–50.4 Gy. The field was extended to the para-aortic area as clinically indicated at doses of 45 Gy. Patients in this study did not have gross disease and therefore dose escalation was not routinely performed. Cisplatin was not administered concurrently. Vaginal brachytherapy (10 Gy in 2 fractions) was administered near the completion of external beam radiotherapy to patients with any of the following characteristics: grade 3 histology, non-endometrioid histology, or lymphovascular space invasion.
Comparisons between groups were evaluated using the two-sample t-test or Kruskal-Wallis test for continuous variables and χ2 test or Fisher’s exact test for categorical variables. Duration of follow-up was calculated from the date of surgery. The Kaplan-Meier method was used to estimate recurrence rate (any site), locoregional recurrence rate, distant recurrence rate, and overall survival. The primary statistical analyses were performed using the SAS version 9.4 software package (SAS Institute, Inc, Cary, NC).
A meta-analysis of endometrial cancer trials comparing chemotherapy and radiotherapy versus radiotherapy alone using a random effects model was performed using Stata Statistical Software: release 16 (StataCorp LLC, College Station, TX). The OR was calculated using the event rate, and the effect size was expressed as the 95% CI for each study. The I2 statistic was calculated to measure heterogeneity among the studies.
A total of 102 women were included with a mean (SD) age of 63.8 (10.6) years; of these, 84 patients had the chemotherapy-first approach and 18 patients had the ‘sandwich’ approach. Overall, 89 patients underwent a lymphadenectomy, 12 patients had sentinel node removal followed by lymphadenectomy, and one patient had sentinel node removal only. Pelvic lymphadenectomy was performed in 99% and para-aortic lymphadenectomy in 88.2% (Table 1). The ‘sandwich’ group had higher rates of lymphovascular space invasion (66.7% vs 34.5%, p=0.01) when compared to the chemotherapy-first group.
Radiotherapy: Of the 102 patients, six patients were planned for chemotherapy-first followed by radiotherapy but did not receive radiotherapy. Three of these six patients had distant recurrence prior to starting radiotherapy; the other three discontinued treatment during chemotherapy. Radiotherapy details were unknown for four patients. Of the remaining 92 patients, 91 (98.9%) completed radiotherapy as planned. A total of 39 (42.9%) patients received pelvic and para-aortic radiation, 51 (56.0%) patients received pelvic radiation only, and one patient received para-aortic radiation only. A total of 55 (60.4%) patients received vaginal brachytherapy. Median time from chemotherapy to initiation of radiotherapy was 24 days (IQR 20–27) for the ‘sandwich’ group and 33 days for the chemotherapy-first group (IQR 26–39).
Chemotherapy: The number of cycles was known for 98 patients, of whom 83 (84.7%) completed six cycles of chemotherapy and 93 (94.9%) completed at least four cycles of chemotherapy. Hematological toxicity information was complete for 46 patients; 37%, 8.7%, and 10.9% had grade 3+ neutropenia, thrombocytopenia, and anemia, respectively.
Among the 102 patients, 42 deaths were documented and 31 patients had a recurrence within 3 years following surgery (median follow-up among those alive at last follow-up was 5.9 years). Site of first recurrence was locoregional only in four patients, distant only in 24 patients, and both locoregional and distant recurrences in three patients (Online supplemental table 1). At 3 years, for the ‘sandwich’ and chemotherapy-first approach, the vaginal recurrence was 11.8% and 4.2%, pelvic recurrence was 0% and 1.5%, para-aortic recurrence was 0% and 1.2%, distant recurrence was 42.9% and 24.4%, and overall survival was 70.3% and 81.7%, respectively (Table 2, Online supplemental figure 1).
The majority of the locoregional recurrences (5/7) were in the vagina (Figure 2, Table 2). Of the five patients with vaginal recurrence, one did not receive radiotherapy; the remaining four patients received radiotherapy (two external beam radiotherapy only, two external beam radiotherapy and vaginal brachytherapy) and had one of these risk factors—grade 3 tumor, lymphovascular space invasion, or cervical stromal invasion. A pelvic recurrence was seen in one patient and para-aortic recurrence in one patient, both of whom had endometrioid histology and were treated with chemotherapy-first.
Patients with endometrioid histology had a lower 3 year distant recurrence rate and a higher 3 year overall survival when compared with patients with non-endometrioid histology (Table 2). Among the 34 patients with grade 1 or 2 endometrioid endometrial cancer treated with chemotherapy-first, the 3 year distant recurrence rate was only 3% (95% CI 0.0% to 8.7%). In a univariate analysis looking at predictors of distant recurrence, non-endometrioid histology was associated with increased risk (p<0.001), whereas age, stage (IIIC1 vs IIIC2), lymphovascular space invasion, and depth of myometrial invasion were not significant.
Literature Review: Impact of Sequencing of Adjuvant Chemotherapy and Radiotherapy on Chemotherapy Delivery
A literature review was performed to identify endometrial cancer trials that administered a combination of chemotherapy and external beam radiotherapy using the search terms ‘endometrial cancer’, ‘chemotherapy’, and ‘radiotherapy’ in PubMed. Chemotherapy before radiotherapy approaches are able to deliver more cycles of chemotherapy when compared with radiotherapy before chemotherapy (Online supplemental table 2 and Online supplemental figure 2). A range of 92–100% of women were able to complete at least four chemotherapy cycles with the chemotherapy before radiotherapy approach3 8 9 as compared with 71–90% with the radiotherapy before chemotherapy approach.3 4 8 10–13 Endometrial cancer trials that planned for six cycles of chemotherapy in a chemotherapy before radiotherapy approach were able to administer 73–85% of the planned six cycles.14 15
Literature Review and Meta-Analysis: Impact of Number of Cycles of Planned Chemotherapy on Distant Recurrence Rate
We updated the meta-analysis published by Yi et al16 on adjuvant chemotherapy and radiotherapy (various sequences) versus radiotherapy alone in high-risk endometrial cancer by adding published data from PORTEC-3 (Figure 3). There was no significant difference in the distant recurrence rate between chemotherapy and radiotherapy versus radiotherapy alone when three cycles of chemotherapy were planned (three studies, 886 participants, OR 1.37, 95% CI 0.73 to 2.58, I2=49.2%).8 17 18 However, there was a significant reduction in distant recurrences with chemotherapy and radiotherapy versus radiotherapy alone when four or more cycles of chemotherapy were planned (three studies, 1219 participants, OR 0.70, 95% CI 0.53 to 0.92, I2=0%).4 8 19 This alludes to a potential dose-response relationship between chemotherapy and distant recurrence rate in endometrial cancer (P for interaction=0.01).
Summary of Main Results
Our current approach of chemotherapy before radiotherapy was associated with low lymphatic locoregional recurrence rates despite delaying radiotherapy until after 3–6 cycles of chemotherapy and not administering concurrent cisplatin. Most of the locoregional recurrences in our series were in the vagina. A total of 60% of women in our series received vaginal brachytherapy. All patients with vaginal recurrences following pelvic radiotherapy had more than one risk factor: grade 3 tumor, lymphovascular space invasion, or cervical stromal invasion.
The chemotherapy before radiotherapy approach had favorable chemotherapy delivery, with 94.9% completing 4+ cycles of chemotherapy. In a meta-analysis of endometrial cancer trials, the distant recurrence rate was reduced with 4+ chemotherapy cycles but not with three cycles (p=0.01); this should be considered hypothesis-generating. Women with endometrioid endometrial cancer had a favorable 3 year distant recurrence rate of 13.2% (95% CI 2.8% to 22.5%) with the chemotherapy-first approach. We still observed a high rate of distant failures in women with non-endometrioid endometrial cancer.
Results in the Context of Published Literature
Among 102 included patients, we observed seven (6.9%) locoregional recurrences. This is comparable to GOG-258 (5 year pelvic and/or para-aortic recurrence rate in the chemotherapy and radiotherapy arm was 11%) and PORTEC-3 (15% of women with stage IIIC disease developed a pelvic recurrence).1 3 The safety of delaying radiotherapy while administering chemotherapy is also supported by data from breast cancer studies.20
In GOG-258, the improved locoregional control in the chemotherapy and radiotherapy arm was offset by the higher distant recurrence rate.3 Improving chemotherapy delivery by sequencing chemotherapy before radiotherapy could potentially improve the distant recurrence rate without compromising locoregional control. A recent National Cancer Database (NCDB) analysis found that sequencing radiotherapy before chemotherapy was associated with a 25% higher mortality (HR 1.25, 95% CI 1.13 to 1.39) compared with chemotherapy before radiotherapy for stage IIIC disease.5 An older NCDB analysis found similar results with a survival benefit for sequential chemotherapy before radiotherapy when compared with concurrent chemotherapy and radiotherapy.21
Compared with the chemotherapy-first approach, the ‘sandwich’ approach has the theoretical benefit of initiating radiotherapy sooner after surgery, the downside being the unknown impact of interruption in chemotherapy mid-course. Some retrospective studies support the superiority of the ‘sandwich’ approach over chemotherapy-first,22 23 but others studies show no difference.6 We did not formally compare outcomes between these approaches in our series.
Strengths and Weaknesses
The strengths of our study include our institution’s uniform approach to patient care, the standardized pathology review, radiotherapy planning, and the chemotherapy protocol. We included a homogenous patient population with stage IIIC disease and no gross residual disease. Limitations include a retrospective design and a potential for referral bias. Most of our patients had complete lymphadenectomy; caution must be exercised when extrapolating to patients who had sentinel node removal only. The low false negative rate associated with the sentinel node assessment technique is reassuring; however, the risk of para-aortic recurrence may be higher.24 Nevertheless, ours is a large cohort reporting outcomes after chemotherapy before radiotherapy sequencing for stage IIIC endometrial cancer.
Implications for Practice and Future Research
Chemotherapy before radiotherapy sequencing for stage IIIC endometrial cancer was associated with a high proportion of patients completing 4+ chemotherapy cycles and low locoregional lymphatic recurrence rate, despite delaying radiotherapy until after 3–6 cycles of chemotherapy (9–18 weeks) and not administering concurrent cisplatin. Uniform application of vaginal brachytherapy for women with risk factors may improve vaginal disease control. Chemotherapy before radiotherapy sequencing for stage IIIC endometrial cancer needs to be studied in prospective trials.
Chemotherapy before radiotherapy sequencing for stage IIIC endometrial cancer was associated with a high proportion of patients completing 4+ chemotherapy cycles and low locoregional lymphatic recurrence rate.
Data availability statement
De-identified data are available upon reasonable request.
Mayo Clinic IRB # 07-002816.
Contributors Study conception and design: MSB, GEG, AM, DN, IAP and ALW. Acquisition of data: DN. Analysis and interpretation of data: MSB, GEG, AM, MMM, DN, IAP and ALW. Drafting of manuscript: DN. Critical revision: MSB, GEG, AK, CLL, AM, MEM, DN, IAP and ALW.
Funding This work was supported by a grant from the National Center for Advancing Translational Sciences (CTSA Grant Number UL1 TR002377), a component of the National Institutes of Health (NIH).
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.