Abstract
Background/Aim: Traditionally, the radiotherapy of oesophageal cancer has been conformal radiotherapy (CRT). We sought to compare dosimetric parameters of conformal radiotherapy (CRT) with those of two treatment planning systems for hybrid-volumetric modulated arc therapy (h-VMAT) for the treatment of oesophageal cancer. Patients and Methods: In 11 patients, we compared: i) planning target volume coverage, ii) dose to organs at risk, and iii) the dose rate (DR) of the three techniques. We evaluated two treatment planning systems: i) Eclipse and ii) RayStation. Results: The Conformity Index of the CRT plan was significantly higher for the h-VMAT plans, compared to all other parameters. Normal lung tissue volumes receiving >5, 13, or 20 Gy were lower with the RayStation plan compared to Eclipse. The volume of cardiac tissue receiving >40 Gy was highest with the CRT plan. The minimum DR in VMAT was lowest for the RayStation plan (49.5 MU/min). Conclusion: The h-VMAT plan using RayStation is the appropriate choice for reducing lung dose.
Oesophageal cancer is a common cause of cancer death around the world (1). Chemoradiotherapy for oesophageal cancer has better local control and overall survival compared to chemotherapy (2). A cisplatin-based combination is the standard regimen of chemoradiotherapy for oesophageal cancer (3).
The radiotherapy (RT) technique for oesophageal cancer has commonly been conformal radiation therapy (CRT). This technique is often limited by the dose to the organs at risk (OARs). The National Cancer Institute (NCI) does not permit intensity-modulated radiotherapy (IMRT) for treatments in the thorax of patients in NCI-sponsored trials (4, 5). IMRT generally produces widely distributed but lower doses of radiation to normal tissues surrounding the planning target volume (PTV) compared to CRT. The normal tissue volumes involved during RT for oesophageal cancer are larger compared to those in lung cancer treatment because of the longer cranio-caudal length of the PTV.
Volumetric modulated arc therapy (VMAT) has recently been used for the treatment of oesophageal cancer (6). To reduce the volume of normal tissue that receives these extraneous doses, Mayo et al., (7) have developed a new technique called ‘Hybrid-IMRT’. This technique combines static and IMRT beams used concurrently, and has produced results with a better dose conformity and sparing of OARs using VMAT with a shorter treatment time compared to IMRT (8). For this reason, we used volumetric modulated arc therapy (VMAT) instead of IMRT (9). Multiple commercial treatment planning systems (TPSs) have become available. Lafond et al., (10) have compared the dosimetric parameters of two VMAT treatment planning systems for prostate cancer, whereas Langner et al., (11) have compared point doses between Eclipse and Raystation (Eclipse Version 13.7.29; Varian Medical Systems, Palo Alto, CA, USA and RayStation version 4.7.4.4, RaySearch Medical Laboratories AB, Stockholm, Sweden) for proton therapy. The result of calculation is difficult between TPSs, because the calculation algorithm was different. Comparison of hybrid-VMAT (h-VMAT) techniques using two treatment planning systems has not been done before.
The aim of this study was to compare dosimetric parameters of CRT with those of two h-VMAT plans from two different treatment planning systems for the treatment of oesophageal cancer.
Patients and Methods
Patients. Subjects were 11 cases of oesophageal cancer patients treated with h-VMAT technique at our institute between 2017 and 2018. Patient characteristics are summarized in Table I. This study was approved by our ethics committee (No. 42718).
Treatment planning. For the computed tomography (CT) simulation, patients were immobilized with a vacuum pillow (Vac-Lock, Civco Medical Solutions, Iowa, USA) in the supine position with the arms raised. CT images were acquired using the Revolution HD (GE Medical Systems, Milwaukee, WI, USA). The parameters for CT acquisitions were: i) 2 mm slices, ii) 512×512 matrix, and iii) 50 cm field of view. These images were transferred to the Eclipse TPS (Eclipse Version 13.7.29; Varian Medical Systems).
The CRT plan used only Eclipse. The h-VMAT plans used both Eclipse and RayStation. The dose calculation used the analytical anisotropic algorithm (AAA), while the h-VMAT plan from RayStation was recalculated by the AAA of Eclipse.
Target volumes and OARs (e.g., the lung, heart, and spinal cord) were contoured by radiation oncologists. The gross tumour volume (GTV) consisted of identified masses and clinical target volumes (CTV) for boosting and was created by adding potential volumes of tumour extension to GTV. CTV for elective treatment included lymph node regions at potential risk of occult metastasis. The PTV was created by adding an isotropic margin of 5 mm to the CTV. In all cases, high-risk planning target volume (PTVboost) was based on the primary and clinical lymph node metastases, while PTVelective included elective dose areas.
The dose prescriptions were 40 and 20 Gy for the PTVelective and PTVboost (2 Gy/fraction), respectively. For the CRT plan, doses were delivered to the isocentre. For the h-VMAT plans, doses were prescribed to the mean dose to the PTV. The CRT plan consisted of opposite anterior-posterior (AP) fields to deliver PTVelective and opposite oblique fields to deliver PTVboost to the isocentre. The h-VMAT plans used two full arcs in VMAT and opposite AP fields. The prescription dose of the h-VMAT plan was delivered at 50% with VMAT, and 50% in the AP direction. The following dose constraints were used for OARs: i) maximal dose (Dmax) to the spinal cord at ≤45 Gy, ii) V5 for the lung at ≤65%, and iii) V20 for the lung at ≤35%. While maintaining dose constraints of PTV coverage and spinal cord, the three plans had added constraints to minimize lung dose.
The typical field arrangements and isodose distribution from 5 Gy to max dose of the three treatment techniques are illustrated in Figure 1.
The following dosimetric parameters were analysed: i) D98, ii) D95, iii) D2, iv) conformity index (CI) for PTV, v) homogeneity index (HI) for PTV, vi) Dmax to the spinal cord, vii) V5, viii) V13, ix) V20, x) V30 and xi) mean of lung-PTVelective (MLD), xii) V40 and xiii) mean to heart. Dx% indicates the dose that includes x% of the target and Vx% indicates the volume% of the target receiving x% of the prescribed dose. CI was determined by dividing the volume receiving the prescribed dose by the target volume. HI was evaluated as the difference between D2 and D98 of PTV divided by D50 of PTV. The Dx of PTVboost was calculated adding plans of PTVelective and PTVboost. The dose rates (DR:MU/minute) of the two arc beams of two h-VMAT plans were compared.
Patient characteristics.
Data analysis. Results are expressed as mean ± standard deviation. Paired Student's t-tests were used to compare the dosimetric parameters. A value of p<0.017 was defined as having statistical significance using the Bonferroni's method.
Results
Target coverage. Table II shows the dosimetric parameters of PTV for CRT and the two h-VMAT plans. D95 of PTVelective and PTVboost were not significantly different and >95% of the prescribed dose. The D50 of PTVelective for the Eclipse plan was not significantly different from that of the RayStation plan. The D50 for PTVboost of the Eclipse plan was significantly different from that of the RayStation plan (p<0.01). The CI for PTVelective of the CRT plan was significantly higher compared to that of the Eclipse and RayStation plans (p<0.001). The CI for PTVboost of the Eclipse plan was also significantly lower compared to that of the RayStation plan (p<0.001).
Organs at risk. Table II shows OAR doses. The V5 for lung-PTVelective of the Eclipse plan was significantly higher compared to those of the CRT and RayStation plans (p<0.001 and 0.01, respectively). The V13 for lung-PTVelective of the RayStation plan was significantly lower compared to that of the CRT plan (p=0.005). The V20 for lung-PTVelective of the RayStation plan was significantly lower compared to that of the CRT plan (p<0.001). The V30 for lung-PTVelective of the CRT plan was not significantly different from those of the Eclipse and RayStation plans. The mean dose for lung-PTVelective of the CRT plan was significantly different from those of the Eclipse and RayStation plans. The maximum spinal cord dose of the CRT plan was significantly higher than that of the Eclipse and RayStation plans (p=0.026 and 0.035, respectively).
Axial (upper) and coronal (lower) dose distribution of >5 Gy for the three planning techniques for one patient. (A) CRT plan, (B) h-VMAT plan using Eclipse, (C) h-VMAT plan using RayStation. The color of dose distribution is >5 Gy (blue), >20 Gy (light blue), >30 Gy (green), >45 Gy (yellow), >55 Gy (orange).
Dose rate. The mean DR of the RayStation plan was significantly higher compared to that of the Eclipse plan (p<0.001). The minimum DR of the RayStation plan was lower compared to that of the Eclipse plan (49.53 MU/min vs. 62.06 MU/min). The beams of the CRT plan and AP beams of the h-VMAT plan were 600 MU/min stably. Figure 2 shows an example of DR for h-VMAT plans between Eclipse and RayStation.
Discussion
In this study, we compared the dosimetric parameters of CRT and of two h-VMAT plans using two treatment planning systems in 11 oesophageal cancer patients. The two h-VMAT plans showed a significant improvement in CI compared to CRT. The h-VMAT plan using RayStation in CI showed good coverage compared to Eclipse plans. In normal lung, the RayStation plan provided lower low-dose volumes except for the V5 of the CRT plan. The heart V40 was highest in CRT compared the other plans. The Dmax to spinal cord was not significantly different among all methods. The minimum of DR was lowest for the RayStation plan (49.5 MU/min), though the RayStation plan had a wider range of DR compared to the Eclipse plan.
Previous publications on IMRT/VMAT have recognized some advantages compared to CRT (12-14). However, the IMRT/VMAT plans give a higher lung dose compared to the CRT plans (V5, V10, V20, MLD) (15). To reduce the volume of normal lung for V5, V13, V20, V30, MLD, the h-VMAT plan offered better conformity and lower low-dose volumes except for the V5 of the lung (9). It is generally understood that V20 and mean lung dose (MLD) are major predictors for radiation pneumonitis (RP). A critical review of the dose-volume effect in the lung recommends limiting V20 to ≤30-35% and MLD to 20-23 Gy (16). Schallenkamp et al., (17) have reported results from a study that examined MLD, V10, V13, V15, V20, V30, and effective lung dose for the treatment of a series of 99 lung cancer patients. They concluded that larger volumes of lung exposed to lower doses (V13) may be more predictive of complications compared to V20 or V30. V5 may also be important (18). Wu et al., (19) have compared CRT, IMRT, and VMAT in oesophageal cancer patients, and have demonstrated that all plans were able to meet the prescription and there was no clear distinction on PTV coverage. VMAT can decrease the high dose area but delivers more volume of the low dose area. The h-VMAT plan using RayStation compared to CRT and Eclipse successfully lowered the V20 and mean lung dose, however, the low dose was similar to that of the CRT plan, together with an improvement in CI. Regarding the cardiac dosimetry, the RTOG 0617 clinical trial reported that the heart V5 and V30 are important predictors of patient survival (20). In this study, the heart V40 was extremely high with the CRT plan. This was due to the presence of oblique beams used to reduce the lung dose. The heart V40 remains significantly associated with OS in the multivariable analysis (21).
Dose parameters of PTV coverage, OARs.
CIs to PTVelective were significantly higher for the CRT plan. For this reason, a lung volume receiving >20 Gy of the CRT plan was significantly higher than that of the Eclipse and RayStation Plans. Chan et al. have compared the CIs of three treatment techniques: i) CRT, ii) VMAT, and iii) h-VMAT for locally-advanced non-small cell lung cancer. They have described that the CIs of VMAT (1.13±0.07) and h-VMAT (1.13±0.05) plans were lower than that of CRT (1.42±0.17) (9). Similarly, in our study, the CIs of PTVelective and PTVboost for the CRT plan were higher compared to the other plans. A wide distribution of low dose to the surrounding normal tissues can be harmful for the patients (18). The differences in DRs for the h-VMAT plan between the two treatment systems have not been compared yet. Lafond et al., (10) have compared the DR between two treatment planning systems [Monaco (Elekta, Crawley, UK) and Pinnacle (Philips Medical Systems, Best, the Netherlands)] for VMAT in prostate cancer. The average DR was higher with Monaco (230 MUs/min) compared to Pinnacle (160 MUs/min).
Example of DR for h-VMAT plan between Eclipse (blue) and RayStation (red).
In that study, the VMAT treatment plans obtained with Monaco and Pinnacle offered clinically acceptable dose distributions for prostate cancer (10). The DR of the RayStation plan compared to that of the Eclipse plan was wider and had a lower mean in our study. Due to a wide range of DR, the RayStation plan was a steep distribution compared to the Eclipse plan.
Some limitations exist in this study. First, the V5 of lung for the CRT and h-VMAT plan using RayStation were not significantly different. The lung dose of the h-VMAT plan may be reduced by changing the ratio of the A-P beam and VMAT. Second, this study was calculated using AAA. AAA can be suboptimal in low density tissues, such lung, where it may overestimate the dose. Inaccuracy of dose calculation may be reduced by using Monte Carlo (22) or a commercial software (Acuros XB, Varian Medical Systems) (23).
The CI of the h-VMAT plan was significantly lower compared to that of the CRT plan for PTVelective. For this reason, lung doses except for V5 were significantly higher with CRT compared to the h-VMAT plan. The V5 mean of Lung-PTVelective of the h-VMAT plan using RayStation was significantly lower compared to that of the Eclipse plan. The h-VMAT plan using RayStation is an approach for reducing the lung volume receiving a low dose volume during the treatment of oesophageal cancer.
Acknowledegments
The Authors would like to thank Libby Cone, MD, MA, from DMC Corp. (www.dmed.co.jp<http://www.dmed.co.jp/>) for editing a draft of this manuscript.
Footnotes
Authors' Contributions
All Authors contributed to data collection, and participated in the writing and final approval of the manuscript.
This article is freely accessible online.
Conflicts of Interest
No actual or potential conflicts of interest exist.
- Received October 12, 2019.
- Revision received October 30, 2019.
- Accepted November 5, 2019.
- Copyright © 2020 The Author(s). Published by the International Institute of Anticancer Research.
