Para-aortic and pelvic extended-field radiotherapy for advanced-stage uterine cancer : dosimetric and toxicity comparison between the four-field box and intensity-modulated techniques

Background In patients with advanced-stage endometrial carcinoma (eca), extended-field radiotherapy (efrt) is traditionally delivered by the 3-dimensional conformal (3d-crt) 4-field box technique. In recent years, the use of intensity-modulated radiotherapy (imrt) in gynecologic cancers has increased. We compared the delivery of efrt by the 3d-crt and contemporary imrt techniques. Methods After surgical staging and adjuvant chemotherapy in 38 eca patients, efrt was delivered by either imrt or 3d-crt. Doses to the organs at risk, side effects, and outcomes were compared between the techniques. Results Of the 38 eca patients, 33 were stage iiic, and 5 were stage ivb. In the imrt group, maximal doses to rectum, small intestine, and bladder were significantly higher, and mean dose to bladder was lower (p < 0.0001). Most acute gastrointestinal, genitourinary, and hematologic side effects were grade i or ii and were comparable between the groups. In long-term follow-up, only grade 1 cystitis at 3 months was statistically higher in the imrt patients. No grade iii or iv gastrointestinal or genitourinary toxicities were observed. No statistically significant differences in overall and disease-free survival or recurrence rates were observed between the techniques. Conclusions In advanced eca patients, imrt is a safe and effective technique for delivering efrt to the pelvis and para-aortic region, and it is comparable to the 3d-crt 4-field box technique in both side effects and efficacy. For centres in which imrt is not readily available, 3d-crt is a valid alternative.


INTRODUCTION
Endometrial cancer (eca) is the 4th most common cancer in women and the 7th leading cause of cancer death in women in developed countries 1 .Patients with locally advanced disease or distant metastasis account for 15%-30% of cases, but for more than half the cancer-related deaths 2 .Treating para-aortic lymph nodes in high-risk patients is imperative because those nodes appear to be involved in more than 50% of patients with positive pelvic lymph nodes.Moreover, most para-aortic lymph node metastases (88%) are located between the inferior mesenteric artery and the renal veins 3 .
Extended-field radiotherapy (efrt) including the pelvis and the para-aortic area has traditionally been delivered as 3-dimensional conformal radiotherapy (3d-crt) using the parallel opposed fields or 4-field box techniques.In recent years, the use of intensity-modulated radiotherapy (imrt) in gynecologic cancers has increased significantly 4 .Intensity-modulated radiotherapy allows for the highly conformal delivery of therapeutic radiation doses to oncologic targets, while the dose to healthy tissue is limited, thus reducing acute and late gastrointestinal (gi) and genitourinary toxicity 5 .Dose-sparing around the para-aortic nodes is particularly significant, given their close anatomic relationship with the small bowel and kidneys 6 .Possible disadvantages of imrt include smaller treatment volumes.A greater emphasis on accuracy is consequently needed in target delineation.Furthermore, the risk of a second cancer is likely increased because of a larger integral dose and up-surging mean costs 7,8 .Yet despite the increasing trend in utilization, the breadth of the supporting literature remains limited.
Here, we present our experience treating a cohort of patients with advanced-stage eca.The purpose of our study was to examine doses to targets and to organs at risk, the associated toxicity profile, and disease-free (dfs) and overall survival (os) in patients treated with imrt compared with the 3d-crt 4-field box technique.

METHODS
This retrospective cohort study evaluated 38 patients with International Federation of Gynecology and Obstetrics (2009) stages iiic1, iiic2, and ivb eca.The patients were treated with surgery, followed by adjuvant chemotherapy, efrt, and vaginal brachytherapy (vbt).
After approval was received from the institutional review board, eligible patients were identified in the hospital's database.Demographic, surgical, and pathology data, and information about chemotherapy, radiotherapy, and clinical outcomes were extracted.Total hysterectomy, bilateral salpingo-oophorectomy, peritoneal washings, pelvic and para-aortic lymph node dissection, omentectomy, and maximal tumour debulking were performed by either laparotomy or the robotically assisted minimally invasive technique.Paclitaxel-carboplatin chemotherapy (6 cycles) was administered every 21 days.After chemotherapy, patients were restaged either by computed tomography (ct) imaging of the thorax, abdomen, and pelvis or by wholebody positron-emission tomography imaging.All patients received prophylactic antiemetics.
Extended-field radiotherapy, including to the pelvis and para-aortic areas, was delivered by either the 3d-crt 4-field box technique or the imrt technique to a total dose of 45 Gy (1.8 Gy daily administered in 25 fractions, 5 days per week, for 5 consecutive weeks).In patients with macroscopic disease after surgery and chemotherapy or with an area of significant risk of recurrence, a concomitant boost of 5-10 Gy was given (2-2.2Gy fractions).Patients were given oral contrast, and a vaginal marker was used for simulation.
Patients were simulated on a ct simulator in the supine position with a knee rest.The superior border of all fields was placed to cover the contoured para-aortic clinical target volume up to the L1 level.It was extended to include adjacent suspicious visible lymph nodes or lymph nodes that were positive on pathology.The inferior border was placed to cover the upper two thirds of the vagina, usually at the inferior aspect of the obturator foramina.For 3d-crt, the targets-including the upper two thirds of the vagina; the internal, external, and common iliac vessels; and the aorta to L1-were contoured on ct slices.A 1.5-cm margin was given for field edges.For imrt, target volumes for the pelvis were drawn according to the consensus guidelines published by Small et al. 9 .In both treatment techniques, critical structures including bladder, rectum, small bowel, and kidneys were contoured on full and empty bladder scans.
The 3d-crt and imrt plans were generated using the Eclipse treatment planning system (Varian Medical Systems, Palo Alto, CA, U.S.A.).Beams for the 3d-crt and imrt plans were 18 MV and 6 MV respectively.High-dose-rate vbt using either a Nucletron ct-and magnetic resonance imaging-compatible cylinder (Elekta, Stockholm, Sweden) or a Miami applicator was given in all patients immediately after efrt.Patients received a total of 6-18 Gy in 1-3 fractions, depending on the involvement of the lower uterine segment or cervix and the status of the resection margins.The dose was prescribed to a depth of 5 mm from the vaginal vault and at the surface of the upper 4 cm of the vagina.The dose to organs at risk was kept within the dose limits set by Emami et al. 10 .
Toxicity grading was based on the U.S. National Cancer Institute's Common Terminology Criteria for Adverse Events, version 4.03, and the highest grade of any observed toxicity was reported for each patient.All patients were monitored for side effects at each visit, once weekly during treatment, and every 3 months after completion of treatment.After 2 years, patients were followed every 6 months for an additional 3 years, and annually thereafter.The treating radiation oncologist alternated with the referring gynecologist oncologist in providing follow-up.
Disease-free survival was defined as the duration from the date of surgery to the date of recurrence.Overall survival was defined as the duration from the date of diagnosis to the date of death or the date of last follow-up.Descriptive statistics were calculated for demographic and clinical variables of interest.Patient characteristics were compared using chi-square and t-tests, depending on the normality of the distribution.
Probability estimates of dfs and os were determined using the Kaplan-Meier method.Hazard ratios were computed using a Cox proportional hazards model.Survival curves for categorical covariates were compared using the log-rank test.Statistical tests were two-sided and considered significant at p values less than 0.05.All analyses were performed using the SAS 9.3 software application (SAS Institute, Cary, NC, U.S.A.).

RESULTS
Doses to organs at risk, side effects, and outcomes were evaluated in the study cohort.Most of the patients had stage iiic disease; 5 patients had stage ivb eca without liver or lung metastasis.More than 60% of the patients had deeply invading non-endometrioid tumours, with associated high-risk factors such as lymphovascular space Current Oncology, Vol.22, No. 6, December 2015 © 2015 Multimed Inc.
invasion and lower uterine segment involvement.In 17 patients (45%), pelvic and para-aortic lymph nodes alike were involved with metastatic tumour.
A comparison of the clinical and pathologic characteristics of the patients treated using the two efrt techniques showed that the imrt patients significantly more often underwent robot-assisted surgery and had more para-aortic lymph nodes extracted; however, they had a significantly shorter follow-up (Table i).After surgery, all except one 86-year-old patient received chemotherapy.Radiotherapy was initiated within 2 months after the last chemotherapy.All patients except one received 25 radiotherapy fractions for an average total dose of 46 Gy (range: 32-55 Gy).One patient received only 32 Gy in 16 fractions without additional vbt because of non-recovering grade 4 thrombocytopenia.
Immediately after efrt, vbt was given in 1-3 fractions for a total dose of 6-18 Gy.Compared with patients treated using 3d-crt, those treated using imrt experienced significantly higher maximal doses to rectum, small bowel, and bladder, and a lower mean dose to bladder (p < 0.0001, Table ii).Radiotherapy treatments were completed within 36 days, without delays or dose reduction.During radiotherapy, regardless of the technique used, side effects were mild to moderate and easy to manage.Grade 1 diarrhea and nausea (79% and 55.3% respectively) were the most significant immediate side effects, common to all patients.Hematologic toxicities were encountered in 15 patients (39.5%), with only 2 (13.3%) having grade 3 and 4 toxicity.
At first follow-up, 3 months after treatment, 63% of the patients complained of an increase in urination frequency, urgency, dysuria, nocturia, or new onset of incontinence.One year after treatment, 37.8% of the patients still had symptoms compatible with grade 1 cystitis.The only statistically significant difference in side effects between the treatment groups was a higher rate of grade 1 cystitis at 3 months in imrt-treated patients (93.7% vs. 40.9%,p = 0.0016).At last follow-up (an average of 40.2 months after completion of treatment), no patient complained of radiotherapy-related side effects.
In imrt patients, diarrhea was the most significant side effect during treatment, encountered in 75% of patients.At 3 and 12 months after treatment completion, cystitis was the most prevalent side effect, affecting 93.7% and 53.3% of patients respectively.Grades 1-2 anemia and neutropenia were recorded in 12.5% and 25% of patients respectively.Two patients (12.5%) experienced thrombocytopenia, one having a grade 3 occurrence, and the other having an idiopathic non-recovering grade 4 occurrence that required termination of treatment.
The average follow-up for all patients was 48 months; it was significantly longer for the 3d-crt group (63.4 months vs. 26.2months, p < 0.0001).Recurrent disease affected 9 patients (23.7%: 2 treated with imrt and 7 treated with 3d-crt) at an average of 18.5 months after the last radiotherapy treatment.No statistically significant differences in the recurrence rate between the two groups were evident (p = 0.42).All patients with recurrence experienced widespread distant metastasis.In 2 patients treated using the 3d-crt technique, recurrences were noticed in a previously irradiated field (vaginal vault and para-aortic lymph node).

DISCUSSION
The adjuvant management of advanced-stage eca is not yet established.In the absence of prospective randomized controlled trials, patients are managed according to contemporary clinical practice patterns and institutional protocols 11 .Founded on the metastasis and recurrence patterns of lymph-node positive eca, the current collective    evidence suggests that adjuvant systemic chemotherapy with concurrent or sequential radiotherapy is a good alternative for patients with advanced-stage disease 3 .External-beam radiotherapy alone or with vbt has been shown to reduce pelvic relapses, but treatment failure at distant sites, including the para-aortic area, is frequently observed 12 .Similarly, the impact of efrt on lymphatic recurrences beyond the pelvis remains uncertain 13 .
Studies of clinical outcomes with use of imrt in eca are limited in number; most results are confined to small heterogeneous analyses grouping endometrial and cervical cancers with high-risk stage i and stage iiic eca (Table iii).
We present a cohort including only patients with advanced-stage eca treated using tri-modal therapy (surgery, chemotherapy, and efrt plus vbt), with the application of two different external-beam radiotherapy techniques.The clinical characteristics of the patients in the two study groups were similar except for shorter follow-up and more robotic surgery in the imrt-treated patients.
The dosimetric advantage of imrt over traditional radiation techniques has been demonstrated in numerous studies 21 .In our cohort, the mean doses to rectum and small bowel, as well as the volume of both kidneys receiving 20 Gy or more, were slightly but not statistically significantly lower in the imrt group.Mean dose to bladder was significantly lower, and the maximal doses to rectum, small intestine, and bladder were significantly highermost probably reflecting "hot-spots" in the imrt fluence.
Early retrospective studies with short follow-up and a heterogeneous mixture of gynecologic cancers, postoperative combined treatments, and disease stages have reported that the dosimetric advantages of imrt result in a reduction of both acute and chronic genitourinary and gi toxicity (to 60% from 91% and to 3% from 20% respectively [22][23][24] ).
More recently, two phase ii prospective multi-institutional trials-rtog 0418 and rtcmiendometredemonstrated the benefits of imrt in patients treated postoperatively to the pelvis for eca.In rtog 0418, the incidence of acute grade 2 or higher gi toxicities was lower at 28%, and rtcmiendometre showed a lower incidence of acute grade 2 gi and genitourinary toxicities at 27% and 20% respectively 17,20 .In combined retrospective studies of patients with endometrial and cervical carcinoma, extended-field imrt was associated with 6.5%-11.7%acute and chronic grade 3 or greater gi toxicities 6 .Shirvani et al. 25 reported an even higher grade 3 or greater toxicity rate (19%) in 27 patients with recurrent eca.In their clinical guidelines on imrt in the treatment of gynecologic cancers, D'Souza et al. 26 state that "imrt should be considered if acute and chronic toxicities are the main outcomes of interest." In our cohort of patients with advanced-stage eca treated using efrt and followed for almost 4 years, only 1 patient (treated using 3d-crt) was experiencing grade 2 cystitis at 3 months after completion of treatment.Furthermore, none of the patients reported radiotherapy-related toxicities at their last follow-up.Except for a statistically higher grade 1 cystitis rate at 3 months after treatment in imrt patients, 3d-crt-and imrt-treated patients had similar rates of side effects.
It has been suggested that imrt can reduce hematologic toxicity, but given the large volume of hematopoietically active marrow in the pelvis and lower lumbar spine, specific planning constraints are required 19 .Tierney et al. 15 used imrt and chemotherapy to treat 14 patients with eca.Grade 3 leucopenia occurred in 6 patients, grade 3 anemia in 3, and grade 3 thrombocytopenia in 3.Only 1 patient in our cohort experienced grade 3 thrombocytopenia.Another developed grade 4 non-recovering thrombocytopenia that required termination of treatment.The latter occurrence was investigated by the Hematology service and was deemed idiosyncratic to the patient's reaction to chemotherapy and radiotherapy.No significant difference between the two treatment techniques was noticed.
To date, only a few heterogeneous single-institution studies have evaluated the clinical outcomes of imrt in eca (Table iii).Shih et al. 19 examined 46 patients with stages iiii eca (stage iiic: 26 patients).The 5-year dfs and os rates were 88% and 97% respectively after a median follow-up of 52 months.In our cohort of patients staged iiic and ivb and treated by imrt, the estimated 5-year dfs and os rates were 76.1% and 92.3% respectively.Intensity-modulated radiotherapy compares well to conventional treatments, and overall, it demonstrates at least equivalent survival.But care must be taken in interpreting retrospective institutional series 8 .Taken together, the evidence is insufficient to recommend imrt over 3d-crt for women receiving treatment for advanced eca 26 .
Intensity-modulated radiotherapy is associated with distinct disadvantages, such as increased cost secondary to an increase in planning and technology requirements 8 , a potentially higher risk of second cancers because of an increased number of radiation fields, and an increased risk of fracture, particularly spinal fracture, after efrt 27 .Considering life expectancy in this group of older patients, a 1%-1.75% increase in the risk of second cancers for patients surviving more than 10 years represents an acceptable risk 28 .Precautionary measure such as bone densitometry could prevent spinal fractures.
The results of our study are subject to the limitations inherent to any single-institution experience.Nonetheless, every attempt was made to select a homogenous patient population for whom detailed toxicity and outcome information were available.The grouping of stages iiic and ivb, and particularly the inclusion of 5 stage ivb patients, did not affect os or dfs rates.Compared with other studies, we present a detailed dosimetric, toxicity, and survival analysis of patients with advanced-stage eca.We found no significant differences in clinical characteristics or survival rates between patients treated with 3d-crt and with imrt techniques.Yet, our study, like others, might have been underpowered to detect modest differences between the groups.

CONCLUSIONS
Intensity-modulated radiotherapy is a safe and effective technique to deliver efrt to the pelvis and para-aortic region in patients with advanced eca, even in the setting of a high-intensity tri-modality approach.Based on our data, imrt is comparable to 3d-crt by the 4-field box technique in both side effects and efficacy.For centres in which imrt is not readily available, the 3d-crt technique appears to be a valid alternative.The impact of imrt on survival rates and toxicity in gynecologic cancers requires further investigation in large trials.Currently, the Radiation Therapy Oncology Group is conducting a phase iii randomized trial-the rtog/gog 1203 (Time-C trial)-that is comparing 3d-crt with imrt in post-hysterectomy endometrial and cervical carcinoma patients.

ACKNOWLEDGMENTS
The Israel Cancer Research Fund, the American Physicians Fellowship for Medicine in Israel, the Israel Cancer Association, and Ben-Gurion University of the Negev supported the fellowship of Alex Rabinovich.This work was presented at the 28th Canadian Association of Radiation Oncology Annual Scientific Meeting, St. John's, NL, 25-28 August 2014.

CONFLICT OF INTEREST DISCLOSURES
We have read and understood Current Oncology's policy on disclosing conflicts of interest, and we declare that we have none.

a
Significant results shown in boldface type.IMRT = extended-field intensity-modulated radiotherapy; 3D CRT = 3-dimensional conformal radiotherapy, 4-field box technique; V 20 = volume expected to receive 20 Gy or more.

FIGURE 1
FIGURE 1 Disease-free survival with radiotherapy in patients with stages IIIC and IVB endometrial carcinoma.Solid line = 3D conformal radiotherapy using the 4-field box technique; dotted line = extendedfield intensity-modulated radiotherapy (IMRT).

FIGURE 2
FIGURE 2 Overall survival with radiotherapy in patients with stages IIIC and IVB endometrial carcinoma.Solid line = 3D conformal radiotherapy using the 4-field box technique; dotted line = extended-field intensitymodulated radiotherapy (IMRT).

TABLE II
Doses to organs at risk in patients with stages IIIC and IVB endometrial carcinoma treated using two different radiotherapy (RT) techniques

TABLE III
Toxicity and survival in patients with stages IIIC and IVB endometrial carcinoma treated using extended-field intensity-modulated radiotherapy (IMRT) a Occurring at 3 months or less after IMRT.b Occurring at more than 3 months after IMRT.Current Oncology, Vol.22, No. 6, December 2015 © 2015 Multimed Inc.
*Division of Gynecologic Oncology, Segal Cancer Centre, Jewish General Hospital, McGill University, Montreal, QC; † Division of Radiation Oncology, Segal Cancer Centre, Jewish General Hospital, McGill University, Montreal, QC; ‡ Department of Oncology, Division of Cancer Epidemiology, McGill University, Montreal, QC.