PURPOSE/OBJECTIVES: There are limited data on outcomes in patients with ultracentral pulmonary oligometastases treated with SABR. The purpose of this study was to determine whether ultracentral location was prognostic for toxicity and survival. MATERIAL AND METHODS: Oligometastatic lung lesions treated on the single-arm phase 2 SABR-5 trial were retrospectively stratified into 2 cohorts: ultracentral tumors (UC), defined as planning target volume overlap or direct tumor abutment to the proximal bronchial tree, esophagus, great vessels, or heart, and nonultracentral tumors. Cohorts were compared with respect to grade ≥ 2 toxicity, progression-free survival (PFS), and overall survival (OS). RESULTS: In total, 41 patients with 45 ultracentral metastases and 93 patients with 172 nonultracentral metastases underwent SABR. The most common primary histologies were colorectal (30%), lung (13%), and renal (13%), and these did not differ between groups. Patients with UC had a lower median PFS of 5.8 months compared with 15.8 months in patients with non ultracentral tumors (P <
.001). OS was also worse in the UC cohort: median 29.0 months versus not yet reached (P <
.001). On multivariable regression, UC remained prognostic for worse PFS (hazard ratio 2.18, P = .004) and OS (hazard ratio 3.45, P <
.001). Groups had similar rates of local tumor control. Patients with UC had higher 2-year cumulative incidence of polymetastatic progression: 69.2% versus 31.4% (P <
.001). The 2-year cumulative incidence of grade ≥ 2 toxicity was 14.6% for patients with UC and 9.8% for patients with nonultracentral tumors (P = .74). There were no grade 4 or 5 toxicities. CONCLUSIONS: In this prospective patient cohort, SABR for ultracentral tumor had low toxicity rates and good local control. However, ultracentral location was an adverse prognostic feature for survival. This finding should be validated with larger studies and may be a factor when weighing the benefit versus risk of SABR in patients with pulmonary oligometastases.