Objective To investigate the role of forehead dot marking in immobilization during radiotherapy for brain tumors. Methods A total of 84 patients with brain tumors were immobilized using the full-head mask from 2013 to 2014; among these patients, 36 received dot marking at the center of the forehead (group A), 27 received routine immobilization (group B), and 21 received dot marking at one side of the forehead (group C). Kilovoltage cone-beam computed tomography (CBCT) was used to compare the deviation in spatial position. The t-test was used to compare the deviation of the first CBCT in group A with the errors of subsequent CBCT. Results The passing rates of 212 times of CBCT in group A and 158 times of CBCT in group B were 95.2% and 93.8%. The errors in x, y, and z directions in groups A and B were 0.868±0.618 mm vs. 1.319±0.935 mm (P=0.005), 1.118±0.692 mm vs. 1.149±0.807 mm (P=0.824), and 0.868±0.790 mm vs. 0.936±0.919 mm (P=0.665). In group C, the error in the x direction was 1.162±0.866 mm, similar to those in groups A and B (P=0.070 and 0.430). In group A, the error in the x direction showed a difference between the first time and the sixth time (0.790±0.656 mm vs. 1.280±0.724 mm, P=0.030). Conclusions Compared with the routine technique, dot marking at the center of the forehead can improve the accuracy of positioning in the x direction and reduce rotation error, and provides a new method for immobilization in radiotherapy for brain tumors. Key words: Brain neoplasms/radiotherapy; Forehead dot marking; Set-up error