Stereotactic body radiotherapy (SBRT) is increasingly utilized in the management of ultra-central thoracic tumors, although concerns regarding significant toxicity remain. We sought to evaluate the toxicity and efficacy of SBRT to these tumors at our institution. Patients with ultra-central lung tumors or nodes treated at our institution with SBRT between 2009 and 2019 were retrospectively reviewed. Ultra-central was defined as having the planning target volume (PTV) overlapping or abutting the central bronchial tree and/or esophagus. All SBRT plans were generated with homogenous dose distributions using target coverage objectives of ITV V100% >99%, PTV V95% >99%, and an ideal PTV Dmax <105% (strict <120%). All plans were reviewed in quality assurance rounds by a team of dosimetrists, physicists, and radiation oncologists. The primary endpoint was incidence of severe toxicity (ST), defined as SBRT-related grade ≥3 toxicities, graded using the Common Terminology Criteria for Adverse Events V5.0. Secondary endpoints included local failure (LF), progression-free survival (PFS) and overall survival (OS). Competing risk analysis was used to estimate incidence and predictors of ST and LF, with death as a competing risk. Kaplan-Meier method was used to estimate PFS and OS. A total of 154 patients who received 162 ultra-central courses of SBRT were included, with a median follow-up of 21.5 months. Treatment intent was most commonly for oligoprogression (46%), oligometastasis (30%), followed by curative (20%). The most frequent tumor histologies were NSCLC (41%) and RCC (26%). SBRT prescription doses ranged from 30-55 Gy in 5 fractions (BED10 range 48-115 Gy). The most common prescription was 50 Gy in 5 fractions (42%). The cumulative incidence of ST was 8.9% at 3-years. The most common ST was pneumonitis (n = 4). Notable toxicities included bronchopleural fistula (n = 2, grade 3 and 4), bronchial stricture (n = 1, grade 3), and esophagitis leading to bleeding (n = 1, grade 4). There were no esophageal strictures or perforations, and no bronchial bleeds. There was 1 possible treatment related death from pneumonitis/pneumonia. Predictors of any ST included increased lung V5 Gy, decreased PTV V95%, and not having prior radiation therapy to the chest. The cumulative incidence of LF was 4.8%, 11% and 14% at 1-, 2-, and 3-years respectively. Predictors of LF included younger age, and greater volume of overlap between the PTV and esophagus. Median PFS was 8.4 months, while median OS was 3.7 years. In one of the largest case series of ultra-central thoracic SBRT reported to date, homogenously prescribed SBRT plans were associated with relatively low rates of ST and LF across a variety of treatment indications. Predictors of ST should be interpreted recognizing the heterogeneity in toxicities observed. Identified predictors of both ST and LF can contribute to future work to optimize the therapeutic ratio in treatment of ultra-central thoracic tumors.