The utilization of stereotactic body radiotherapy (SBRT) in the management of ultra-central thoracic tumors is increasing. We sought to evaluate toxicity and efficacy of ultra-central thoracic tumors treated with SBRT in our institution.In this institutional research ethics board approved study, patients with primary or metastatic ultra-central thoracic tumors treated with SBRT from 2015 - 2019 were retrospectively reviewed. Ultra-central was defined where the planning target volume (PTV) of a primary tumor and/or thoracic lymph node overlapped or abutted the proximal main bronchus (PMB) and/or esophagus. The primary endpoint was SBRT-related grade 3 toxicities using common terminology criteria for adverse events (CTCAE v.5). Multivariable stepwise regression was performed to determine relevant organs at risk (OAR) dosimetric parameters as predictors of grade 3 toxicity. Overall survival (OS) and progression-free survival (PFS) were estimated using the Kaplan-Meier method. Competing risks analysis was used to estimate the cumulative incidence of local recurrence (LR) using death as a competing risk factor. Predictors of OS/PFS, as well as LR were evaluated using univariable Cox regression and Fine & Gray competing risks regression, respectively. Predictors with P < 0.05 on univariable regression were entered into a multivariable regression model.A total of 81 patients were included, and the median follow-up was 16.4 (range, 0.7 - 69.0) months. SBRT doses ranged from 30-60 Gy in 4-5 fractions, and mean biological equivalent dose (BED10) was 84.1Gy. Tumors were most commonly metastatic (n = 62) followed by NSCLC (n = 15) and recurrent NSCLC (n = 4). In 20 patients, SBRT was also prescribed to thoracic lymph nodes. There were 14 (17.3%) episodes of grade 3 toxicity (n = 8 pulmonary, n = 4 PBT, n = 2 esophagus). There were 3 potentially treatment-related deaths (n = 2 pneumonitis, n = 1 bronchus-pleural fistula). On multivariable modelling, there were no significant dosimetric predictors of grade 3 toxicity. The 1-and 2-year LC for the cohort were 81.5% and 71.5%, respectively. Higher BED10 and higher PTV coverage with the prescribed dose were significantly associated with improved LC. The median PFS was 9.9 (range, 7.6 - 12.4) months; and 1-and 2-year PFS were 39.1% and 23.3%, respectively. Larger PTV volume and patients treated for indications other than primary NSCLC were associated with significantly lower PFS. The median OS was 25.3 (range, 17.6 - 31.7) months. The 1-and 2-year OS were 77.2% and 51.4%, respectively.While SBRT for ultra-central thoracic tumors is associated with encouraging rates of LC, attempts to limit major toxicity should be prioritized to optimize the therapeutic ratio.