The present study is a multi-institutional study; this implies that there are some "subjective" factors depending on planning philosophy of the respective hospital e.g. number of beam directions, number of segments and arcs, limitations of the MLCs, weighting of the importance of PTV and OAR. Another role plays the level of experience of the planners in the different centres that's why we selected for every technology and TPS combination experienced users. But in the last consequence the results of this multi-institutional study show that all used IMRT technologies together with their TPSs have the power to provide treatment plans with a satisfying target coverage while at the same time respecting the defined OAR criteria. At least there is no best technology with respect to all evaluation parameters, i.e. all techniques are connected with some advantages and with some disadvantages. As far as treatment planning is concerned, there were substantial differences in terms of usability to specify the planning goals for the different volumes. It would be of great help for treatment planning if functions where available in TPS that excluded intersections automatically or where priorities to different PTVs with intersections could be assigned.
The results are in good agreement with published data [26–29] regarding the volumatric arc therapy. Only the results of our study getting with sliding window are much better than in . A differentiation of the patients in the two groups (post-operative patients and primary RT) did not show significant differences in the results.
All treatment plans offer a very good coverage of the PTV1 and a good coverage of the PTV2. The lowest dose to the PTV2 with clearly inferior results compared to the other techniques was achieved with the Pinnacle step&shoot combination. The median doses for the PTV2 and the PTV1 were in a range between 100% and 106%. This implies that the planners of the participating institutes improved the coverage of the PTVs with the help of an increase of the median dose. The requirements demanded by the HR0022 protocol are more or less fulfilled. ICRU recommendations for prescribing, reporting and recording IMRT have just been which will be helpful in the future to harmonize IMRT practice .
Sliding window, RapidArc and Tomo techniques resulted in better target dose homogeneity for the PTV1 compared to VMAT and step&shoot with Panther DAO, Pinnacle and KonRad.
All technologies TPS combinations fulfill the OAR constrains. Only the high myelon maximal dose receiving with sliding window is demonstrative (but with a margin of 7 mm clinically acceptable). The highest median dose to the spared parotid while using the RapiArc is peculiar too.
The volume which receives equal or more than 5Gy is lowest with the sliding window technique (3800 ccm), followed by the VMAT and KonRad step&shoot (about 4500 ccm). Pinnacle step&shoot, Tomo, Panther DAO and RapidArc deliver doses of equal or more than 5Gy to volumes of 5000 ccm or bigger. It is of interest that neither the "classic IMRT" with fixed gantry angles nor the rotation based IMRT is clearly the superior solution. It seems that rotational IMRT techniques do not automatically generate more volume that receives dose of equal or more than 5Gy. The volume could probably be even further reduced using higher photon beam energies.
The treatment delivery times obtained in the present study were shortest for the RapidArc solution. The delivery times for Tomo and Panther DAO were in the medium range while VMAT, step&shoot with Konrad or Pinnacle and with sliding window were characterised by the longest ones. As far as the VMAT results on delivery efficiency are concerned, it needs to be emphasized that Monaco Version 2.01 was used in the present study, which was improved recently with a new sequencer available in successive versions of this TPS.
The MUs are significantly reduced for the DAO step&shoot (408MU), RapidArc (437MU) and VMAT (501MU). The MUs needed for a step&shoot KonRad plan is situated in the centre (about 800MU). Pinnacle step&shoot needs 1060MU and sliding window takes the highest number of 1140MU. It is known that the number of MU is one factor which influences the peripheral dose, but there are some other factors like the linac head shielding and collimation system (shape, thickness, material), the focus body distance and the spectrum of the beam. The peripheral dose is of importance without any doubt but in the particular case subordinated relativ to the treatment plan quality.