Evaluation of positioning accuracy of four different immobilizations using cone-beam CT in radiotherapy of non-small-cell lung cancer.

Abstract

To evaluate the positioning accuracy of four different immobilizations by use of cone-beam computed tomography guidance for radiotherapy of non-small-cell lung cancer (NSCLC). Sixty-seven patients with NSCLC received conventional or stereotactic body radiotherapy. Of these, 30 were immobilized with a thermoplastic frame (TF), 16 with a thermoplastic frame and active breathing control (TF-ABC), 7 with a stereotactic body frame (SBF), and 14 with a stereotactic body frame and active breathing control (SBF-ABC). Cone-beam computed tomography scans at initial setup and after correction were registered to planning computed tomography. The positional errors in the left-to-right, superior-inferior, and anterior-posterior directions were analyzed. The planning target volume margins were calculated. The precorrection systematic and random errors ranged from 1.9 to 4.2 mm for TF, 1.9 to 4.3 mm for SBF, 1.2 to 5.8 mm for TF-ABC, and 2.3 to 3.9 mm for SBF-ABC. The postcorrection systematic and random errors ranged from 0.3 to 1.9 mm for the four immobilizations. The planning target volume margins (conventional vs. stereotactic body radiotherapy) were 15.6 vs. 13.9 mm (TF), 14.9 vs. 14.8 mm (TF-ABC), 14.4 vs. 13.4 mm (SBF), and 9.9 vs. 9.4 mm (SBF-ABC) before correction and 7.3 vs. 6.9 mm (TF), 4.0 vs. 3.8 mm (TF-ABC), 7.5 vs. 7.1 mm (SBF), and 4.5 vs. 4.2 mm (SBF-ABC) after correction. The positioning accuracies of SBF and TF were similar. Active breathing control increased positioning error but reduced internal margin. Cone-beam computed tomography online correction improved the positioning accuracy of NSCLC patients.