Detection of lung tumor movement in real-time tumor-tracking radiotherapy

Purpose: External radiotherapy for lung tumors requires reducing the uncert ainty due to setup error and organ motion. We investigated the three-dimens ional movement of lung tumors through an inserted internal marker using a r eal-time tumor-tracking system and evaluated the efficacy of this system at reducing the internal margin. Methods and Materials: Four patients with lung cancer were analyzed. A 2.0- mm gold marker was inserted into the tumor. The real-time tumor-tracking sy stem calculates and stores three-dimensional coordinates of the marker 30 t imes/s. The system can trigger the linear accelerator to irradiate the tumo r only when the marker is located within the predetermined "permitted dislo cation." The value was set at :+/-1 to +/-3 mm according to the patient's c haracteristics. We analyzed 10,413-14,893 data sets for each of the 4 patie nts. The range of marker movement during normal breathing (beam-off period) was compared with that during gated irradiation (beam-on period) by Studen t's t test. Results: The range of marker movement during the beam-off period was 5.5-10 .0 mm in the lateral direction (x), 6.8-15.9 mm in the craniocaudal directi on (y) and 8.1-14.6 mm in the ventrodorsal direction (z). The range during the beam-on period was reduced to within 5.3 mm in all directions in all 4 patients. A significant difference was found between the mean of the range during the beam-off period and the mean of the range during the beam-on per iod in the x (p = 0.007), y (p = 0.025), and z (p = 0.002) coordinates, res pectively. Conclusion: The real-time tumor-tracking radiotherapy system was useful to analyze the movement of an internal marker. Treatment with megavoltage X-ra ys was properly given when the tumor marker moved into the "permitted dislo cation" zone from the planned position. (C) 2001 Elsevier Science Inc.