Objective To study the novel methods of VMAT planning based dose volume histogram ( DVH) optimization, evaluated the dosimetry and planning efficiency in VMAT planning for Esophageal Carcinoma. Methods Twelve Esophageal carcinoma patients were enrolled in this study. The conventional VMAT planning as the reference, using multi?criterion optimization DVH ( MCO?DVH ) and overlapping volume histogram prediction DVH ( OVH?DVH ) two different methods to get ideal objectives function for informing VMAT plans, Then evaluate the dosimetric, planning efficiency for all new VMAT plans. The difference between the paired t?test groups. Results The two VMAT plans based DVH objective function can meet the clinical needs. Compared with the conventional VMAT plan, Conformity index ( CI ) and Homogeneity index ( 0. 77 vs. 0. 72, P=0. 017 and 0. 10 vs. 0. 12, P=0. 047 ) is better in DVH informed plans;lung V5 and spinal cord V50 are better in MCO?DVH informed plan (54. 66 vs.60. 23,P=0. 013 and 0. 98 vs.0. 49,P=0. 037).Furthermore,the DVH informed plans had higher planning efficiency (8. 2 vs. 19. 5,P=0. 023) . Conclusions DVH Objective informed VMAT Planning can achieve clinical needs with much uniform dose to target,lower OAR dose and higher planning efficiency.

Objective To compare the errors of double-center and single-center setup, and to study the role of both on reducing the rotational setup errors for the patients with esophageal carcinoma depend on rigid registration errors between online kV-cone-beam computed tomography (kV-CBCT) images and plans for CT images. Methods 20 patients with middle esophageal carcinoma received image scanning before treatment every week by using double-center setup and CBCT, and single-center setup images of 20 patients were taken from the X volume image (XVI) system. Then the images of both setup types, registration errors of CT image and rotational setup errors were compared respectively. Every patient received kV-CBCT scanning analysis before treatment every week, and 6 times in total. 240 group of kV-CBCT images from all of the patients were off-line matched with plans for CT images to calculate the errors of X-axis, Y-axis, Z-axis. Then the data of linear errors and rotational setup errors from patients were collected, aiming at putting the error data into the patients treatment program and analyzing the significances. Results The standard registration of double-center setup was as follows: T (X) (0.28 ±0.19) cm, T (Y) (0.27 ±0.19) cm, T (Z) (0.33 ±0.12) cm, R (X) (0.40 ±0.19)° , R (Y) (0.30 ±0.18)° , R (Z) (0.30 ±0.19)° . The standard registration of single-center setup was as follows:T(X) (0.32±0.20) cm, T(Y) (0.29±0.25)cm, T(Z) (0.31±0.16) cm, R(X) (2.2±0.68)°, R(Y) (0.5±0.32)°, R(Z) (2.10±0.60)°. There were statistical differences between linear errors in T(X) and rotational setup errors in R(X), R(Y) or R(Z) (P< 0.05). Conclusion Double-center position can reduce the rotational setup errors, especially in X-axis, Y-axis errors, and may provide more help for the radiation oncology departments without on-board CBCT.