Improvement of three-dimensional treatment planning models of small lung targets using high-speed multi-slice computed tomographic imaging.

Abstract

To improve the reliability of the patient model and reduce treatment volume by acquiring multi-slice computed tomographic (CT) data with the patients' single holding of breath at normal inhalation and exhalation. Seven patients with nine small peripheral lung cancer tumors underwent CT scanning under three respiration conditions using multi-slice CT: free breathing (FB), shallow inspiration (SI), and shallow expiration (SE). To compare the treatment plan created using the two-respiratory-phase images (SI + SE) with the plans created using only SE images or using only FB images, we attempted to calculate the true dosimetric characteristics for three-dimensional treatment planning taking respiratory movement into consideration. Minimum dose to the gross tumor volume (GTV) and ipsilateral lung dose-volume histogram (DVH) were calculated for the inhalation and exhalation positions of shallow breathing. There was no significant difference between minimum doses of the GTV in the three treatment plans when using anteroposterior/posteroanterior parallel-opposed fields. However, there was a significant difference between the minimum doses of the GTV in the two-phase treatment plan and the minimum dose in the other treatment plans when using the four-field technique, consisting of shaped anterior, posterior, right and left lateral fields (p = 0.03, 0.04). Comparison of the percent volume of ipsilateral lung receiving a dose exceeding 20 Gy (V(20)) based on inhalation and exhalation CT data revealed that the V(20) of the two-phase plan was the smallest of the three treatment planning fields (p < 0.001). Two-phase planning using multi-slice CT provides an immediate reduction in the amount of normal tissue treated and improved reliability of patient data for DVH modeling.