The new model of the Leksell Gamma Knife Icon is equipped with additional technical functionalities for fractionated treatment, namely a cone beam computed tomography (CBCT). Performing CBCT before treatment enables to define the stereotactic space, and to correct the radiation plan in case the patient’s position changes. The purpose of this work was to assess differences between frame-based and CBCT defined stereotactic space, to identify predictors of the observed findings and to assess their effects on the clinical outcome. The study included 122 patients treated on the LGK Icon unit from July 2018 to December 2019. Both the information about the differences (rotational and translational shifts) between frame-based and CBCTdefined centers of the stereotactic and maximum shot displacement (MSD) were reported by the registration module of treatment planning system. We also collected the potential predictors of the differences. To identify the cause of the observed discrepancies between traditional and CBCT localizations, 19 parameters were investigated. Multiple linear regressions were performed to evaluate associations between parameters. Discrepancies between the coordinate systems were revealed depending on the localization method. 2.4% out of 122 cases exceeded 1 mm and 1 degree in translational and rotational shifts, respectively. Tumor coverage decreased more than 5% in 4.9 percent of cases. As a result of linear regression analysis, we found that the fiducial errors, weight of the patient, diagnosis, KPS were predictors of the increased rotational and translational shifts, as well as the MSD. Fifty-one patients diagnosed with multiple brain metastases were analyzed to determine the effect of stereotactic coordinate shifts on clinical outcome. The follow-up brain MRI scan showed an increase of 10 (5.3%) and 3 (5.8%) irradiated targets after 3 and 6 months, respectively. The increases in the remaining (7 of 188) targets were differentiated as post-radiation changes. However, the resulting sample of patients is insufficient to assess the effect of target coverage differences on clinical outcome. Using the methods of mathematical statistics based on the data on the onset of local recurrence in patients with metastases, it was found that the minimum required sample size should be at least 425 patients for a complete analysis of the effect of coordinate systems shifts on the clinical outcome. As the next step of the study, it is planned to increase the patient sample under consideration to assess the effect of the existing difference between the two methods of localization on the clinical outcome of treatment. [ABSTRACT FROM AUTHOR]