It was aimed to investigate the dosimetric differences among linac-based and robotic stereotactic radio-therapy (SRT) techniques for the treatment of uveal melanoma and to evaluate secondary malignancy risks for these different SRT techniques. Ten patients who received robotic SRT with CyberKnife were ret-rospectively included in this study. A total dose of 54 Gy in three fractions was prescribed to the planning target volume (PTV). For each patient, non-coplanar micro-multileaf collimator based dynamic conformal arc (DCA), intensity-modulated radiotherapy (IMRT) and circular cone based DCA (cDCA) plans were gen-erated. During the analysis dose-volume histogram (DVH) parameters, homogeneity index, new confor-mity index, the volume received more than or equal to 30% and 50% of the prescribed dose were com-pared. Additionally, secondary malignancy risk for each technique was estimated using the risk factors recommended by The International Commission on Radiological Protection. Robotic SRT plans provided a high degree of conformity within the PTV and better normal tissue sparing compared to linac-based treatment plans. However, dose distribution was more heterogeneous in robotic SRT plans than that in linac-based techniques. Estimated secondary malignancy risk was also found as 3.4%, 1.4%, 1.4% and 1.6% for robotic SRT and linac-based IMRT, DCA, cDCA plans, respectively. Treatment parameters of uveal melanoma patients planned with robotic SRT had superior conformity and organ-at-risk (OAR) sparing compared with those planned with the linac-based system. However, estimated secondary malignancy risk was almost two-times higher in robotic SRT than that in linac-based techniques. (c) 2021 American Association of Medical Dosimetrists. Published by Elsevier Inc. All rights reserved.