As an example, the axial dose distribution of both ncVMAT and coVMAT are illustrated in Fig. 2. It is obvious in Fig. 2 that a better conformity of doselines from 500 cGy to 4000 cGy is shown with ncVMAT on this layer. This can be also seen in the dose volume histogram of PTV and selected OARs in Fig. 3. For this case, ncVMAT provided better sparing for LAD, heart, bilateral lung, right breast and better homogeneity for PTVimn.
A. Dosimetric evaluation of PTV
All plans were normalized to cover 95% volume of PTVall with the prescription dose 50 Gy. The dosimetric parameters of PTVbreast, PTVscn, PTVimn and PTVall are shown in Table. 1. The mean coverage of PTVbreast decreased slightly from 95.98 ± 0.39% to 95.36 ± 0.48% with ncVMAT (p = 0.005). The average coverage of PTVscn was 94.62 ± 0.92%, and there was no significant difference between ncVMAT and coVMAT (p > 0.05). It is interesting that the mean coverage of PTVimn increased significantly from 88.77 ± 3.09% to 91.67 ± 3.84%. Meanwhile, The 110% prescribed dose volume V55 in PTVscn and PTVimn decreased significantly from 5.76 ± 6..03%, 13.73 ± 7.77% to 2.94 ± 3.93%, 3.75 ± 3.36%, separately (p < 0.01). separately. Furthermore, the mean conformity index of the PTVall increased from 0.82 ± 0.02 to 0.86 ± 0.01 (p = 0.005) which means a better conformity for ncVMAT, and there was no significant difference for the homogeneity index.
Table 1
Comparison of dosimetric parameters between coVMAT and ncVMAT for PTV
| Parameters | coVMAT | ncVMAT | p |
PTVall | CI | 0.82 ± 0.02 | 0.86 ± 0.01 | 0.005 |
| HI | 1.11 ± 0.02 | 1.11 ± 0.01 | 0.093 |
| V55 (%) | 11.40 ± 7.88 | 8.28 ± 5.63 | 0.028 |
PTVimn | V45 (%) | 98.40 ± 1.30 | 99.30 ± 0.71 | 0.074 |
| V50 (%) | 88.77 ± 3.09 | 91.67 ± 3.84 | 0.013 |
| V55 (%) | 13.73 ± 7.77 | 3.75 ± 3.36 | 0.005 |
PTVscn | V50 (%) | 94.51 ± 1.76 | 94.62 ± 0.92 | 0.878 |
| V55 (%) | 5.76 ± 6.03 | 2.94 ± 3.93 | 0.007 |
PTVbreast | V50 (%) | 95.98 ± 0.39 | 95.36 ± 0.48 | 0.005 |
| V55 (%) | 13.28 ± 8.95 | 10.21 ± 6.35 | 0.059 |
B. Heart
In breast cancer radiotherapy, a preferential sparing should be provided to heart. It is illustrated in Table 2, the average V30, V20, V10, V5 of heart decline significantly from 5.13 ± 3.22%, 12.18 ± 7.29%, 33.98 ± 15.73% and 72.03 ± 16.79% to 4.86 ± 2.54%, 9.35 ± 5.43%, 25.16 ± 12.42% and 62.27 ± 13.18% (p < 0.05), separately. The mean dose of heart was reduced from 10.47 ± 2.97 Gy to 9.08 ± 2.34 Gy (p = 0.005), and there was a decrease of 13.3% in Dmean of heart.
It can be seen in Table 2, the mean V20, V10, V5 of left lung decline significantly from 24.95 ± 2.61%, 45.44 ± 6.45%, 72.82 ± 13.00% to 22.94 ± 3.94% (p = 0.007), 40.86 ± 6.51% (p = 0.007), 66.24 ± 11.68% (p = 0.007), separately. The mean V10 and V5 of contralateral lung also decreased from 1.53 ± 0.85% and 14.67 ± 6.21% to 0.87 ± 0.77% (p = 0.005) and 8.02 ± 4.37% (p = 0.005), separately. The mean lung dose (MLD) of left lung and right lung were both reduced from 14.80 ± 1.61 Gy and 3.19 ± 0.48 Gy to 13.82 ± 1.79 Gy (p = 0.005) and 2.78 ± 0.39 Gy (p = 0.005). Apparently, it could be concluded that ncVMAT provides a better sparing for bilateral lungs than coVMAT plans.
Table 2
Comparison of dosimetric parameters between coVMAT and ncVMAT for heart and lung
| Parameters | coVMAT | ncVMAT | p |
Heart | V5 (%) | 72.03 ± 16.79 | 62.27 ± 13.18 | 0.007 |
| V10 (%) | 33.98 ± 15.73 | 25.16 ± 12.42 | 0.005 |
| V20 (%) | 12.18 ± 7.29 | 9.35 ± 5.43 | 0.005 |
| V30 (%) | 5.13 ± 3.22 | 4.86 ± 2.54 | 0.047 |
| Dmean (Gy) | 10.47 ± 2.97 | 9.08 ± 2.34 | 0.005 |
Left lung | V5 (%) | 72.82 ± 13.00 | 66.24 ± 11.68 | 0.007 |
| V10 (%) | 45.44 ± 6.45 | 40.86 ± 6.51 | 0.007 |
| V20 (%) | 24.95 ± 2.61 | 22.94 ± 3.94 | 0.007 |
| Dmean (Gy) | 14.80 ± 1.61 | 13.82 ± 1.79 | 0.005 |
Contralateral lung | V5 (%) | 14.67 ± 6.21 | 8.02 ± 4.37 | 0.005 |
| V10 (%) | 1.53 ± 0.85 | 0.87 ± 0.77 | 0.005 |
| Dmean (Gy) | 3.19 ± 0.48 | 2.78 ± 0.39 | 0.005 |
Dosimetric parameters of right breast and LAD are evaluated in Table 3. The V10, V5 and Dmean of contralateral breast declined from 0.50 ± 0.70%, 8.90 ± 5.64% and 2.45 ± 0.64 Gy to 0.15 ± 0.30% (p = 0.005), 5.87 ± 2.70% (p = 0.008) and 2.26 ± 0.35 Gy (p = 0.005), separately. Accordingly, the V10 in right breast was negligible for ncVMAT plans. The average maximum does (Dmax) and Dmean in LAD descending coronary artery were 43.61 ± 10.65 Gy and 22.88 ± 7.40 Gy for ncVMAT plans comparing to 45.41 ± 8.74 Gy and 23.77 ± 6.73 Gy for coVMAT plans. The reduction of Dmax and Dmean for LAD were both statistically significant (p < 0.01).
The evaluation of some other OARs is shown in Table 3. It is illustrated that the Dmean of esophagus increased slightly from 30.85 ± 3.96 Gy to 31.90 ± 4.43 Gy (p < 0.05), which was acceptable in clinical practice. Furthermore, reduction of Dmax for both esophagus and Thyroid were realized with ncVMAT. In addition, there were no significant differences for V30 and Dmean of left humeral head, Dmax and Dmean of left brachial plexus, V30, Dmean of Thyroid, and Dmax of spinal cord (p > 0.05). Moreover, all the dosimetric parameters of these OARs were clinically acceptable.
Table 3
Comparison dosimetric parameters between coVMAT and ncVMAT for other OARs
| Parameters | coVMAT | ncVMAT | p |
Contralateral breast | V5 (%) | 8.90 ± 5.64 | 5.87 ± 2.70 | 0.005 |
| V10 (%) | 0.50 ± 0.70 | 0.15 ± 0.30 | 0.008 |
| Dmean (Gy) | 2.45 ± 0.64 | 2.26 ± 0.35 | 0.005 |
LAD | Dmax (Gy) | 45.41 ± 8.74 | 43.61 ± 10.65 | 0.005 |
| Dmean (Gy) | 23.77 ± 6.73 | 22.88 ± 7.40 | 0.005 |
Left humeral head | V30 (%) | 6.59 ± 14.27 | 15.54 ± 15.21 | 0.093 |
| Dmean (Gy) | 17.74 ± 6.69 | 21.91 ± 4.48 | 0.059 |
Esophagus | Dmax (Gy) | 53.09 ± 0.38 | 52.54 ± 0.88 | 0.037 |
| Dmean (Gy) | 30.85 ± 3.96 | 31.90 ± 4.43 | 0.037 |
Left brachial plexus | Dmax (Gy) | 54.30 ± 0.62 | 54.43 ± 1.24 | 0.646 |
| Dmean (Gy) | 50.42 ± 1.33 | 49.96 ± 0.97 | 0.646 |
Thyroid | V30 (%) | 49.42 ± 8.80 | 47.25 ± 5.66 | 0.878 |
| Dmax (Gy) | 54.58 ± 0.95 | 53.99 ± 0.92 | 0.028 |
| Dmean (Gy) | 31.30 ± 4.03 | 31.28 ± 2.83 | 0.646 |
Spinal cord | Dmax (Gy) | 30.26 ± 4.58 | 31.22 ± 5.97 | 0.646 |
The average MU for increased from 833 ± 126 MU for coVMAT to 932 ± 172 MU for ncVMAT, but the increase is statistically insignificant. The average treatment time increased from 136 ± 12 s for coVMAT to 280 ± 15 s for ncVMAT (p = 0.005). The time increase is significant, and caused by manually rotating the couch from 0˚ to 90˚.