OBJECTIVES: A bronchial genomic classifier had been prospectively validated to improve diagnostic accuracy of bronchoscopy for suspected lung cancer in order to identify patients at low probability who may be more suitable for active surveillance as opposed to a biopsy via an invasive procedure. Our objectives were to assess the frequency of invasive procedures and the cost-effectiveness of bronchoscopy plus genomic classifier versus bronchoscopy alone in diagnostic work-up of patients at intermediate risk for lung cancer in the United States. METHODS: A decision-analytic Markov model was developed to project the costs and effects of the two competing strategies, using test performance from the AEGIS-1 and AEGIS-2 studies. Diagnostic accuracy of non-invasive and invasive follow-up, as well as associated adverse event rates were derived from published literature. Procedure costs were based on claims data and 2016 inpatient and outpatient reimbursement amounts. The model projected the number of invasive follow-up procedures, two-year costs and quality-adjusted life years, by strategy, and the resulting incremental cost-effectiveness ratio (ICER), discounted at 3% per annum. RESULTS: Use of invasive procedures were reduced by 28% at one month and 18% at two years, respectively. Total costs and QALY gain were similar with classifier use ($27,221 vs. $27,183, and 1·512 vs. 1·509 QALYs, respectively), resulting in an ICER of $15,052 per QALY. CONCLUSIONS: The use of a genomic classifier for inconclusive bronchoscopies is associated with meaningful reductions in invasive follow-up procedures at about equal costs and is therefore a high-value strategy in the diagnostic work-up of patients at intermediate risk of lung cancer.