Background
Adjuvant radiation therapy following breast conserving surgery allows improving local control and overall survival in early stage breast cancer patients [
1]. Treatment techniques for radiation therapy have been established ranging from conventional tangential fields over intensity modulated radiation therapy (IMRT) to volumetric modulated arc therapy (VMAT), and been compared in various planning studies [
2‐
10]. Cardiac sparing in patients with left sided breast cancer has been considered the main advantage of IMRT and VMAT as compared to tangential fields [
2‐
7,
11]. A few planning studies include right sided as well as left sided breast cancer [
8‐
10], but dedicated planning studies for right sided breast cancer seemed not to be of interest up to now. During the last years, however, simultaneous integrated boost radiation therapy superseded sequential boost therapy as the standard fractionation scheme [
12,
13]. In this case, IMRT and VMAT have shown advantages in target coverage and sparing of the organs at risk (OAR) as compared to 3D-CRT techniques [
10,
14], thus becoming more important also for right sided breast cancer.
The latest development in the technology of linear accelerators is the opportunity to irradiate patients without a flattening filter in the beam path to increase dose rate and reduce beam-on times as well as out-of-field doses [
15]. Two planning studies have already been published comparing the two irradiation modes for breast cancer, both of them dealing with left sided breast cancer [
5,
7]. Since dose volume restrictions to the heart might, however, be limiting the reduction of dose to the other organs at risk to a larger extent in the therapy of left sided breast cancer, results cannot simply be transferred to right sided breast cancer cases. Dedicated planning studies for radiation therapy of right sided breast cancer are therefore necessary to identify the optimal treatment technique also for this group of patients.
The aim of the study presented here was to compare the two irradiation modes with (FF) and without flattening filter (FFF) for three different treatment techniques for simultaneous integrated boost radiation therapy of patients with right sided breast cancer.
Discussion
The aim of this study was to investigate the potential of the flattening filter free (FFF) mode of a linear accelerator for three different treatment techniques for patients with right sided breast cancer. The results of the study show, that FFF led to significantly superior results with regard to plan quality, when VMAT or tVMAT were used as planning technique, whereas the opposite was the case for IMRT. Differences in plan quality were, however, in general small and the clinical relevance of the differences remains to be shown. The constraints used for optimization were derived from iterative treatment planning in FF mode and used for optimization in FFF mode without adaptation. This approach was chosen for reasons of comparability in order to not bias the results by the use of different constraints. It might be possible to further improve plan quality in FFF mode by adjusting the constraints for the optimization in FFF mode. This is, however, beyond the scope of this study.
Comparison of all treatment techniques and irradiation modes showed significantly best target coverage and homogeneity for VMAT FFF and lowest doses to the contralateral OAR and normal tissue for tVMAT FFF. This has been confirmed by measurements which showed high agreement between measurements and calculations in the low dose region for VMAT and tVMAT as well as in the high dose area corresponding to the target volumes for all plans. No significant differences could be observed in dose calculation accuracy between the two irradiation modes FF and FFF neither in the high nor in the low dose region. Gamma passing rates were high (above 97 %) for each individual plan independently of the treatment technique and irradiation mode. Dose deviations in the low dose region were significantly higher for IMRT than for VMAT and tVMAT in both irradiation modes, but still in the range of 1 % of the prescription dose, which is excellent considering the recommendations of ESTRO for the verification of simple open fields with tolerances of 3 % of the central axis dose or 30 % of the local dose in the low dose region [
32]. For verification of complete IMRT plans, a confidence limit of 4 % is recommended [
33]. With respect to dose calculation accuracy VMAT and tVMAT are therefore considered the preferred treatment techniques independently on the irradiation mode.
The number of MU and control points was significantly higher in FFF irradiation mode, with an increase by 17 % for tVMAT, 19 % for VMAT and 26 % for IMRT plans. The significantly highest number of MU was observed for VMAT FFF. The increase in MU can be explained by the fact that the linac is calibrated such that 100 MU correspond to 1 Gy under reference conditions on the central beam axis for both FF and FFF. Due to the shape of the dose profile, the dose is lower outside the central axis for FFF beams. Therefore additional MU delivered in smaller off axis segments are required to achieve the same dose away from the central beam in FFF mode. Concerns had been raised that an increase of MU might mitigate the potential advantages of FFF with respect to treatment time and MLC transmission leading to potentially higher OAR and normal tissue doses. The results of our study showed that for VMAT and tVMAT treatment times were significantly reduced by 7 % and 32 % and doses to the OAR were comparable for VMAT and even significantly reduced for tVMAT despite of significantly increased MU in FFF mode. Due to the excellent agreement between dose calculations and measurements for VMAT and tVMAT in the low as well as in the high dose region, the results of the dose volume analysis are considered reliable. For IMRT, longer treatment times and a significant increase in OAR doses were observed in the DVH analysis, normal tissue dose was at the same time significantly reduced. The results of the DVH analysis are, however, somewhat less reliable for IMRT than for VMAT and tVMAT due to the observed uncertainty in dose calculation in the low dose region.
The difference in the influence of irradiation mode on treatment time can be explained by the fact that for step and shoot IMRT delivery times increase with the number of control points due to the interruption of irradiation during movement of the MLC. For VMAT techniques the number of control points does not affect the treatment time, since the beam stays on during MLC movement. The number of MU required in FFF plans increased by a factor of 1.2 whereas dose rates delivered in FFF mode increased by up to a factor of 3.4, leading to shorter treatment times for FFF in VMAT and tVMAT techniques. The maximum dose rate is, however, not applied throughout the whole treatment, due to limitations in the speed of mechanical movements therefore the ratio of treatment time in FFF mode to treatment time in FF mode is in general larger than 1.2: 3.4. In simultaneous integrated boost irradiation using IMRT, VMAT or tVMAT dose conformity is higher as compared to conventional tangential field techniques for whole breast treatment. Therefore reduction of intrafractional movement becomes more important, to avoid shifts of the dose distribution relative to the target due to systematic drifts in patient position [
34]. Wiant et al. showed in their study of intrafraction motion of breast cancer patients, “that the patients tend to drift further away from their initial position and they tend to have more short-term random motion as time in the treatment position increases” [
35]. The authors observed a linear increase of the mean shift with time during the first 5 min of a fraction. The reduction of the total treatment time observed for tVMAT FFF is therefore corresponding to a reduction of the mean shift by 32 % as compared to tVMAT FF and 46 % as compared to VMAT FFF. With respect to treatment time and reduction of shifts of the dose distributions relative to the target, tVMAT FFF is therefore considered the preferable treatment option. An investigation of plan robustness against intrafractional breathing movement as a function of treatment technique was, however, beyond the scope of the study. It is therefore advised to use any form of motion control in combination with volume imaging to reduce uncertainties due to breathing.
Balancing target coverage, OAR sparing, agreement of dose calculations and measurements, and delivery time, VMAT FFF and tVMAT FFF are considered the preferred treatment options for SIB radiation therapy of right sided breast cancer in this study. Due to the relatively large standard deviations especially in the target coverage for tVMAT plans and in the V20Gy of the ispilateral lung, the clinical decision is depending on the individual patient’s plan.
Two planning studies have been published comparing the two irradiation modes for left sided breast cancer, one of them dealing with whole breast irradiation [
5], the other with SIB treatment [
7]. Spruijt et al. [
7] performed a planning study for SIB treatment of left sided breast cancer comparing FF and FFF for different static field and IMRT techniques planned with Eclipse treatment planning system for a Varian True Beam Linac. They found comparable plan quality and lower delivery times when FFF beams were used. For verification of the dose calculation one phantom case was created for out-of-field dose measurements 0.3 to 3.1 cm from the field edge. They found an average reduction in out-of-field dose of 10 %. Comparison to dose calculations in this region showed, however, an underestimation by the treatment planning system Eclipse of 26 % to 85 % as compared to measurements. Because of the uncertainties in the dose calculation the authors abandoned evaluation of dose volume parameters of the contralateral breast and lung in their planning study. The agreement between calculations and measurements is substantially higher in our study with (0.1 ± 2.1) % to (−9.3 ± 3.0) % dose deviation in the low dose region of around 20 cGy depending on the treatment technique. A comparison of the plan quality achieved in their study to our results is therefore not possible.
Koivumaki et al. [
5] compared FF and FFF irradiation for tVMAT and tangential IMRT for whole breast irradiation of left sided breast cancer in a hypofractionated scheme of 15 × 2.67 Gy. Treatment planning was conducted in an earlier version of Monaco v3.0 for an Elekta Infinity linac. The authors found a significant reduction in beam-on time when FFF was used at comparable plan quality for tVMAT and degraded plan quality for the tangential IMRT plans. The IMRT and VMAT techniques used in their study differ from the techniques presented here, but the effect that FFF seems beneficial for VMAT but not for IMRT treatments can be observed in both studies. Comparing the results of both studies in plan quality, the differences in total dose of 40 Gy versus 50.4 Gy and 63 Gy have to be taken into account. Whereas mean doses to the contralateral breast and lung are similar for the tVMAT plans in both studies, the mean and maximum dose to the heart and V
20Gy and the D
mean of ipsilateral lung are lower in the study presented here. The same trend could be observed in comparison to the results of Pasler et al. [
4] who compared tVMAT to VMAT for left sided breast cancer. For the dose to the heart, differences are expected due to the location of the heart. The differences in the ipsilateral lung might be explained by the fact that dose volume restrictions to the heart are limiting the reduction of dose to the ipsilateral lung to a larger extent in the therapy of left sided breast cancer than for right sided breast cancer.