OBJECTIVES In this study, we sought to evaluate the performance of the Xpert MTB/RIF (Cepheid) assay for the detection of Mycobacterium tuberculosis (MTB) complex DNA on fresh and formalin-fixed, paraffin-embedded (FFPE) tissue specimens from oncology patients in an area with a low prevalence of tuberculosis. We also aimed to retrospectively assess the potential impact of Xpert MTB/RIF on the duration of airborne infection isolation (AII). SETTING A 473-bed, tertiary-care cancer center in New York City. DESIGN A total of 203 tissue samples (101 FFPE and 102 fresh) were tested using Xpert MTB/RIF, including 133 pulmonary tissue samples (65.5%) and 70 extrapulmonary tissue samples (34.5%). Acid-fast bacilli (AFB) culture was used as the diagnostic gold standard. The limit of detection (LOD) and reproducibility were also evaluated for both samples types using contrived specimens. The potential impact of the Xpert MTB PCR assay on tissue samples from AII patients on AII duration was retrospectively assessed. RESULTS Using the Xpert MTB/RIF for fresh tissue specimens, the sensitivity was 50% (95% CI, 1.3%-98.7%) and the specificity was 99% (95% CI, 94.5%-99.9%). For FFPE tissue specimens, the sensitivity was 100% (95% CI, 63.1%-100%) and the specificity was 98.3% (95% CI, 95.5%-100%. The LOD was 103 colony-forming units (CFU)/mL for both fresh and FFPE tissue specimens, and the Xpert MTB/RIF was 100% reproducible at concentrations 10 times that of the LOD. With an expected turnaround time of 24 hours, the Xpert MTB PCR could decrease the duration of AII from a median of 8 days to a median of 1 day. CONCLUSIONS The Xpert MTB/RIF assay offers a valid option for ruling out Mycobacterium tuberculosis complex (MTBC) on tissue samples from oncology patients and for minimizing AII resource utilization. Infect Control Hosp Epidemiol 2018;39:462-466.