Abstract
Aim: To report preliminary results of a prospective study using pelvic volumetric-modulated arc therapy and simultaneous integrated boost (SIB-VMAT) on vaginal cuff postoperatively in patients with endometrial cancer (EC). Patients and Methods: Fifty consecutive patients, submitted surgery for EC, were recruited to SIB-VMAT prescribing a dose of 54 Gy to the pelvis and 66 Gy to the vaginal cuff in 30 fractions. A 2 mm transvaginal probe and magnetic resonance imaging were used to define the vaginal cuff. Toxicity data were collected according to Common Terminology Criteria for Adverse Events v4.0; clinical outcomes were analyzed. Results: The median follow-up was 26 (range=12 to 39) months. According to International Federation of Gynecology and Obstetrics 2009, the stages were: IB1 in 20%, IB2 in 28%, IIA2 in 16%, IIB in 6%, IIIA in 2%, and IIIC in 28%. The 2-year Overall Survival and Local Control were 96% and 100%, respectively. Two pelvic node failures were registered. Acute gastrointestinal toxicity was: G0 in 12%, G1 in 52%, G2 in 36%; no case of toxicity G3 or more was observed. Acute genitourinary toxicity was: G0 in 10%, G1 in 42%, G2 in 48%; no case of toxicity G3 or more was observed. No late severe gastrointestinal or genitourinary toxicities were reported. A statistical correlation was found between acute G2 gastrointestinal toxicity with bowel V20 Gy ≥30%, V20 ≥40%, V30 ≥30%, Dmax ≥45 Gy. Acute G2 genitourinary toxicity was threefold higher with chemotherapy. Conclusion: In patients with EC, SIB-VMAT is feasible, and well tolerated. Preliminary data of clinical outcome are promising. Further prospective studies are advocated.
Postoperatively in patients with endometrial cancer (EC), locoregional control (LC) is improved by radiation therapy (RT), without advantage in overall survival (OS). Based on clinicopathological features, the Gynecologic Oncology Group (GOG)-99 trial identified an intermediate-high-risk group of patients in which the improvement in locoregional control was primarily due to a reduction in the relapse to the vaginal cuff, which accounted for approximately 75% of recurrences. A higher gastrointestinal (GI) toxicity was observed in the pelvic radiation therapy (PRT) groups (1, 2). Based on this recurrence pattern and the side-effect profile associated with PRT, vaginal brachytherapy (VB) alone was evaluated by PORTEC-2 trial that enrolled patients with intermediate-risk EC comparing VB with PRT. Regarding recurrence rates, no significant difference was observed in the two arms, with decreased GI toxicity (13% vs. 54%) in favor of the VB group (3).
Thus, VB has become a widely used option for postoperative early-stage EC (4). The vaginal failure rate following VB has been shown in multiple studies to be less than 3%, with an acceptable associated toxicity profile (5). However, in patients unable to receive VB or in cases of patients refusing to receive this treatment, alternative therapeutic options should be offered.
Volumetric-modulated arc therapy RapidArc® (VMAT) has been shown to be able to maintain a low-moderate toxicity profile in PRT. Additionally, it showed a great potential for highly conformal dose to target volume (6, 7). VMAT to the pelvis with simultaneous integrated boost (SIB) to the vaginal cuff could potentially be a viable alternative to intracavitary brachytherapy boost (BRT) combined with external beam PRT. In our Institution, a trial was designed to try to avoid brachytherapy time in adjuvant radiation treatment prescribing a dose of 54 Gy to the pelvis and 66 Gy as concomitant boost to the vaginal cuff in 30 fractions with SIB-VMAT technique. The aim of the current analysis is to report feasibility and preliminary clinical outcomes of the SIB-VMAT approach.
Patients' (n=50) baseline characteristics and demographics.
Patients and Methods
Fifty consecutive patients, older than 18 years old, submitted to hysterectomy with bilateral salpingo-oophorectomy and pelvic lymphadenectomy for EC, were considered eligible for adjuvant radiotherapy with/without chemotherapy, according to risk category and were recruited in a prospective study approved by the Internal Review Board of the Institute (RT-01/2011).
In the adjuvant radiation treatment, a dose of 54 Gy to the pelvis and 66 Gy as concomitant boost to the vaginal cuff in 30 fractions with SIB-VMAT technique were prescribed.
Before the treatment, all patients were generally staged with: gynecological examination, chest X-ray and pelvic magnetic resonance imaging (MRI). The stage system used was the International Federation of Gynecology and Obstetrics (FIGO) 2009 criteria (8).
Adjuvant treatment started not later than 4-6 weeks after surgery. We prospectively collected clinical outcomes and toxicity data. Patients' baseline characteristics and histopathological findings (grading, myometrial infiltration, nodal status etc.) are shown in Table I.
Treatment procedure. After patient immobilization inthesupine position by Combifix™ frame (Civco Inc.®), computed tomographic (CT) planning of the abdomen and pelvis (3 mm-thick slide) without intravenous contrast media was performed. A 2 mm soft radiopaque transvaginal probe was used to define the vaginal cuff for all patients (Figure 1). The lower limit of the scanned area was set to 2 cm below the lower limit of the lesser trochanter, the upper limit was the L2-L3 interspace. Patients were simulated and treated with full bladder: they were asked to drink 500 cc of water 30 min before CT planning and before each treatment session. If the rectum was abnormally full or distended, CTs were carried out following enema. The transvaginal probe was re-introduced before each RT session to visualize the vaginal cuff on cone beam CB, CT).
Pelvic MRI staging was merged with the CT planning to better define target volumes and organs at risk (OARs), including the bladder, the bowel defined as intestinal cavity, rectum, and femoral heads according to the Radiation Therapy Oncology Group (RTOG) consensus panel atlas (9).
A clinical target volume (CTV)PELVIS was defined to include the upper two thirds of the vagina, the para-vaginal tissues, pelvic lymph nodes (LNs) and pre-sacral LNs. A 7 mm isotropic expansion was used to obtain a pelvis-planning target volume (PTVPELVIS). A total dose of 54 Gy in 30 fractions was prescribed to the PTVPELVIS.
A CTVSIB including the vaginal cuff and the upper two-thirds of the vagina was defined matching MRI to CT planning and with the aid of an endovaginal probe. The PTVSIB was generated using a 5 mm isotropic expansion on CTVSIB. A total dose of 66 Gy in 30 fractions was prescribed to the PTVSIB.
RT was delivered as 1 fraction/day for five days per week. Before each daily session, patients were submitted to image guided radiotherapy (IGRT) procedure by means of a daily kV CBCT to check and to correct set-up errors. IGRT was also utilized to correct the positioning oft the vaginal cuff by means of superposition of the transvaginal probe between images of daily CBCT with planning CT.
In cases of stage 2 or greater disease, RT was delivered sequentially not later than 4-6 weeks after platinum-based chemotherapy (10, 11).
Planning. All dose distributions were calculated with the anisotropic analytical algorithm on an Eclipse treatment planning system (Varian Medical Systems, Palo Alto, CA, USA), setting a dose grid size of 2.5 mm. RA treatments were delivered with 6 MV beams from either a Varian Clinac Trilogy® or a Varian TrueBeam™ device, both equipped with a Millennium 120-MLC (multileaf collimators with 5 mm leaves in the central 20 cm). RapidArc plans consisted of one or two full arcs using 6-MV photon beams. Plans were optimized with the Progressive Resolution Optimizer in Eclipse (Varian) treatment planning system, version 10, using two full arcs.
For all PTVs, the planning objectives for targets were 95% of the PTVs receiving >95% of each prescribed dose. Maximum dose <107% was requested for PTVSIB only.
Axial computed tomographic slice with trans-vaginal probe to define the vaginal cuff
OAR planning objectives were as follows: for rectum V50 Gy <45%, V60 Gy <30%, V65 Gy <20%, Dmax <70 Gy; for bladder V60 Gy <35%; for femoral heads D1cm3<50 Gy; for bowel V20 Gy <40%, Dmean <20 Gy, Dmax <48 Gy were the dose constraints used.
Various Vx (structure volume receiving at least dose x) and Dy (an amount y of the structure volume receiving at least the dose D) values were recorded for all OARs and targets to be included in the overall statistical analysis.
A final additional dosimetric end-point was to determine new dose–volume objectives for such cases.
Follow-up. The median follow-up was 26 (range=12 to 39) months. During RT, physical examination and complete blood count were performed weekly. After RT, patients were evaluated for disease status every three months until two years and every six months thereafter.
Before RT, a baseline assessment of quality of life was performed using the European Organization for Research and Treatment of Cancer (EORTC)- QLQ-CX24 Scoring manual (12). During RT and follow-up, genitourinary (GU) and GI toxicities were evaluated using the Common Terminology Criteria for Adverse Events (CTCAE) v4.0 (13). Clinical data were collected and evaluated for statistical analyses.
Statistical analyses. LC and overall survival (OS) were estimated using the Kaplan-Meier method. OS was calculated from the date of diagnosis to the death or last follow-up date. Time to progression (TTP) was considered from the date of surgery to the time of recurrence or distant metastasis. LC was defined from the date of surgery to locoregional relapse date. We considered local recurrence as any relapse in the pelvis or the vaginal cuff.
Treatment-related toxicity was considered as the primary end-point. LC and OS were chosen as the secondary end-points.
Correlation between clinical and dosimetric parameters with GI and GU toxicities were analyzed with contingence tables and Fisher's exact test. A p-value less than 0.05 was considered significant. Statistical analyses were performed using R-software® (The R Foundation for Statistical Computing, University of California, Los Angeles).
Transvaginal probe in sagittal CT scan. Definition of pelvis planning target volume and simultaneous integrated boost planning target volume
Results
The actuarial 2 year-OS and LC rates were 96% and 100%, respectively (Figures 2 and 3). The median TTP was 25 months (range=12-30 months). No recurrence at the vaginal cuff was registered. The two local failures were pelvic nodes metastases in stage IIIC disease in which chemotherapy was reduced due to G3 neutropenia.
Five patients had systemic progression of disease (two cases of peritoneal carcinosis, one of liver metastases, one of bone metastases, one of sub-cutaneous metastases). Of these, three patients had unfavorable histopathological findings (serous or clear-cell carcinomas, G3, lymphovascular invasion, myometrial invasion ≥50%), the others had disease stage of II or more. Only two cases of cancer-related death were registered.
According to FIGO 2009 criteria, the frequencies of stages were: IB1: 10/50, 20%; IB2: 14/50, 28%; IIA2: 8/50, 16%; IIB: 3/50, 6%; IIIA: 1/50, 2%; IIIC: 14/50, 28%. No patient had positive surgical margins.
Twenty-eight patients (56%) were candidates for adjuvant chemotherapy. In four patients, chemotherapy was stopped after two cycles due to hematological toxicity; in three patients, a 25% reduction of total dose was adopted due to G3 neutropenia.
Concerning RT, all patients received their treatment as planned. The median duration was 42 (range=41-44) days. In most cases, dose constraints were respected. Greater attention was given to target coverage or sparing of OARs when the anatomical characteristics or the co-morbidities of the patient did not allow the internal protocol constraints to be followed. In the clinical evaluation, particular attention was given to GI and GU toxicities.
Toxicity was recorded in all cases. During treatment, the median week of onset of toxicity was week 4 (range=2-6) for GI toxicity and 3 (range=2-6) for GU toxicity.
Acute GI toxicity was registered as follow: G0 in six patients (12%), G1 in 26 (52%), and G2 in 18 (36%). No case of toxicity of G3 or more was observed. Acute GU toxicity was registered as follows: G0 in five patients (10%), G1 in 21 (42%), and G2 in 24 (48%). No case of toxicity G3 or more was observed. At the end of RT, gynecological examination revealed a G1 vaginal erythema in most cases. Vulval pruritus was referred by seven patients.
During the follow-up, no moderate-severe GI or GU toxicities were reported. At the time of the analysis, late GI toxicity was registered as follow: G0 in 34 patients (68%), and G1 in 16 (32%). No case of toxicity of G2 or more was observed. In one case, a pelvic abscess was documented at one year from RT. For this case, it is difficult to relate the toxicity event to RT. No late GU toxicity was registered. No cases of vaginal stenosis or mucosal atrophy were reported.
For statistical correlations, we calculated median bowel and bladder volumes. The median bowel volume was 2250 cc (range=1800-2500 cc), the median bladder volume was 200 cc (range=150-350 cc).
At multivariate analyses, a significant correlation was found between acute G2 GI toxicity with bowel dose constraints V20 Gy ≥30% (p=0.02; 95%CI=0.23-1.5), V20 ≥ 40% (p=0.02; 95% CI=0.25-1.6), V30 ≥ 30% (p=0.004; 95% CI=0.27-2) and with 1 cc bowel Dmax ≥45 Gy (p=0.001; 95% CI=0.15- 0.5). The odds ratio (OR) for developing G2 GI toxicity was found to be 4.5 if bowel V20 ≥30%, 4 if bowel V20 ≥40%, ≈7 if bowel V30 ≥30%. No correlation was found with other dose–volume constraints, prior abdominal surgery (p=0.25) or chemotherapy (p=0.5).
Regarding GU assessment, the risk of developing G2 acute toxicity was found to increase with adjuvant chemotherapy by ≈3 fold (p=0.07). No correlation was found between G2 toxicity and bladder dose–volume parameters, bladder volume or diabetes (p=0.5).
Discussion
Adjuvant treatment of EC is controversial. In PRT using 3D-conformal radiation therapy (3DCRT), acute G2 GI toxicity occurred in 50-60% of cases and late GI toxicity was reported in 26%. In patients who experienced acute toxicity of G2 or more, a severe complication rate was around 3%. In the GOG-99 trial, six patients had acute GI toxicity G3 or more, including two deaths due to small bowel injury (1, 2, 14). In recent years, intensity-modulated radiation treatment (IMRT) has been assuming an important role in cancer therapy, minimizing the dose to healthy tissue and probably reducing the risk of acute and long-term effects. In other settings, for example in PRT for prostate cancer, the risk of acute toxicity was reduced by IMRT compared to 3DCRT due to better bowel sparing with IMRT (15, 16). Anatomical and clinical predictors of acute bowel toxicity after PRT using image-guided SIB-IMRT techniques have been also assessed in the same setting (17, 18).
In gynecological cancer, encouraging preclinical studies of PRT-IMRT demonstrated that the percentage of bowel irradiated may be reduced by 30-50% (19, 20). Postoperatively in patients with EC, a phase II multi-institutional trial (RTCMIENDOMETRE) showed that acute GI toxicity (grade 2 or higher) was around 30% with pelvic-IMRT (21). Similar toxicity was reported in the RTOG 0418 trial, a phase II trial of postoperative pelvic IMRT, involving cervical carcinoma and EC (22).
VMAT, an IMRT technique continuously delivered during gantry rotation, potentially offers a more conformal dose to the PTV with improved homogeneity and conformity in several settings, including treatment of the pelvis (7, 23, 24). Treatment planning studies showed a better of OARs sparing and shorter delivery time, with an average 12% integral dose to healthy tissues reduced (25).
In our study, postoperative SIB-VMAT for EC was proven to be feasible and safe. An acceptable GI toxicity profile was found, with approximately 75% acute G0-G1 GI toxicity and no case of G3 or more. Most patients complained of GI discomfort during the last weeks of RT. At a median follow-up of 26 months, no moderate/severe late toxicity was reported. Although the mixed population evaluated represents an evident limitation, our analyses showed important bowel dose constraints to be useful for improving the potential of VMAT in this clinical setting. Bowel V20 ≥30% (p=0.02), V20 ≥40% (p=0.02), V30 ≥30% (p=0.004) and Dmax ≥45 Gy (p=0.001) are related to acute G2 GI toxicity.
To our knowledge, only a single experience has dealt with the SIB-VMAT approach in high-intermediate-risk EC (26). The authors compared dosimetric parameters and clinical findings with those of a similar series treated with concomitant boost 3DCRT in the ADA-I trial (27). Their results, with a smaller population (n=32 patients) compared to our analysis and analyzing only acute toxicity, confirmed the advantage of the SIB-VMAT approach in reducing moderate GI adverse events compared to concomitant boost 3DCRT. Doses prescribed were 45 Gy to the pelvis plus 10 Gy boost to the vaginal vault in 25 fractions.
In the current analyses, we found a slightly higher acute G2 GI toxicity (around 35%), but this phenomenon could be related to the higher prescription dose in our series. The rationale of this choice was to compensate the gap in terms of equivalent dose with VB. In a dose–volume analysis for inoperable patients with stage I-II EC treated with PRT with BRT, using an α/β of 10 for tumour and an α/β of 3 for late-responding tissues, a total dose of 62 Gy equivalent dose in 2 Gy fractions (EQD2) was calculated. The 5-year rates of LC and OS were 100% and 90%, respectively (28).
Kaplan–Meier curve for overall survival
In our study, using a similar α/β, a total dose of 67.10 Gy EQD2 was calculated. The actuarial 2 year-OS and LC rates were 96% and 100%, respectively. A longer follow-up is necessary to validate these encouraging preliminary rates. No vaginal relapse and two regional recurrences, in the presence of unfavorable prognostic factors, were recorded.
Although our analysis presents several limitations due to the heterogeneity of the population, results in terms of tolerability and LC appear promising.
Obviously, the current treatment approach represents, for selected patients, a potential alternative to the standard of care. The general consensus is that IMRT techniques will not replace the role of BRT in EC because the latter clearly offers some advantages concerning organ immobilization, with steep dose gradients and highly conformal dose distribution. Despite these characteristics, some experiences reported a reduced vaginal recurrence rate using PRT alone and excluding VB (vaginal failure around 1-2%) (29). On the other hand, a subset of patients with unknown prognostic factors are still amenable to the administration of postoperative external beam RT. In the PORTEC-2 trial, for example, the estimated 5-year rate of pelvic recurrence was significantly higher in the VB arm compared to the arm treated with pelvic external beam RT (4% versus 0.5%, p=0.02) (3). A similar pelvic control rate was reported by Sorbe et al. in intermediate risk EC (1.5% in the PRT arm versus 5% in the BRT arm, p=0.013) (30). In addition, BRT is not free from toxicity. In a randomized trial, 9% of patients receiving BRT experienced G1/G2 GU toxicity, especially vaginal, compared to 1.5% in the observation arm. A significant reduction in vaginal length was noted when 30 Gy/6 fractions was delivered, with an increased rate of vaginal fibrosis (31).
Kaplan–Meier curve for local control
In our series, acute GU effects would seem to be chemotherapy-related, although statistical significance was not reached due to the sample size (p=0.07). In the current approach, no case of vaginal fibrosis was noted and this aspect could be considered non negligible in sexually active women.
Conclusion
In the present analysis, good tolerability and feasibility of the treatment were preliminarily reported, confirming that pelvic VMAT with SIB to the vaginal cuff could potentially be a viable alternative for VB, when BRT is not logistically available or the patient is not suitable for such therapy. Prospective randomized trials are needed to confirm the role of VMAT with SIB approach in the adjuvant setting for selected patients with EC.
- Received December 22, 2014.
- Revision received January 16, 2015.
- Accepted January 20, 2015.
- Copyright© 2015 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.