Objective To evaluate the dosimetric effect of three-dimensional(3D)printed headrests made of different materials in radiotherapy for head and neck tumor,and to evaluate whether the existence of customized 3D-printed headrest can be neglected during planning phase by comparing differences in target area doses,homogeneity index(HI),conformity index(CI),monitor units(MU),and organ-at-risk(OAR)dose.Methods Ten patients with head and neck tumors,including 5 cases with nasopharyngeal carcinoma and 5 with other head and neck tumors,were enrolled.The headrest contours were outlined using Monaco treatment planning system,and with the same calculation parameters,treatment plans were generated for scenarios without a headrest,with a standard headrest,and with 3D-printed headrests made of 10%filled polylactic acid and thermoplastic polyurethane.The target area doses,OAR dose,MU,and other results were recorded and subjected to statistical analysis.Results No significant differences were observed among the 4 groups(ignoring headrest,standard headrest,3D-printed headrests with 10%filled polylactic acid and thermoplastic polyurethane)in parameters such as D95,D5,Dmean,HI,CI,MU,and OAR dose(the maximum dose to the spinal cord)(P>0.05).Considering the presence of the headrest,significant differences were found in CI for nasopharyngeal carcinoma and other head and neck tumors(P<0.05).Conclusion During the planning phase,the existence of customized 3D-printed headrest can be ignored,but it should be noted that the presence of a headrest may reduce the MU in radiotherapy plans.When considering the headrest,the average CI of patients with other head and neck tumors is significantly higher than that with nasopharyngeal carcinoma.