3D-conformal RT, fixed-field IMRT and RapidArc, which one is better for esophageal carcinoma treated with elective nodal irradiation.

The purpose of this study is to compare the characteristics of 3D-conformal radiotherapy (3D-CRT), fixed-field intensity-modulated radiotherapy (IMRT) and RapidArc for esophageal squamous cell carcinoma (ESCC) treated with elective nodal irradiation (ENI). CT datasets of 20 patients with ESCC were included and plans for single and double arcs of RapidArc (RA1 and RA2), 7-field IMRT and 3D-CRT were created and optimized for each patient. The goal was to deliver 59.6 Gy to ≥95% of the planning target volume (40 Gy to electively irradiated lymph nodal regions) while meeting the same normal-tissue dose constraints. The plans were compared based on dosimetric characteristics of target and organs at risk (OARs), monitor units (MUs), and appraised beam-on time. Both RA2 and IMRT resulted in similar target coverage (V95%, 97.84±1.50% for RA2 versus 96.96±1.15% for IMRT), homogeneity index (HI, 0.11±0.02 for RA2 versus 0.10±0.01 for IMRT) and conformity index (CI, 0.81±0.03 for RA2 versus 0.79±0.04 for IMRT), which displayed slightly better than single arc (V95%=94.55±1.50%, HI=0.12±0.02, CI=0.80±0.02) and much better than 3D-CRT (V95%=91.17±2.89%, HI=0.15±0.03, CI=0.60±0.07). The total lung V20, V30 was reduced approximately from 31%, 16% (3D-CRT) to 22%, 13% (IMRT) and 20%, 12% (RA2); the heart V30, V40 from 29%, 21% (3D-CRT) to 28%, 20% (IMRT) and 27%, 18% (RA2). The maximum dose to the spinal cord was 44.26±2.60 Gy for 3D-CRT, 42.47±2.40 Gy for IMRT, and 42.79±1.81 Gy for RA2. The number of MUs per fraction reduced from 990±165 (IMRT) to 503±70 (3D-CRT) and 502±79 (RA2). Appraised beam-on time of RapidArc was 1.2-2.4 min, which was lower than IMRT with 5.4 min by average. RapidArc, especially for double arcs plan could provide slight improvements in OARs sparing and lower MUs without compromised target qualities compared with IMRT, which was much better than 3D-CRT for ESCC treated with ENI.