IMPORTANCE Treatment of locally advanced non-small cell lung cancer (NSCLC) remains challenging. The rationale of combining a cyclooxygenase 2 (COX-2) inhibitor with concurrent chemoradiation (CCRT) was based on results of preclinical research and prospective clinical studies; however, no randomized clinical trial has provided evidence of a direct comparison with CCRT alone. OBJECTIVE To determine the effect of combined selective COX-2 inhibition with standard CCRT on survival among patients with unresectable stage III NSCLC. DESIGN, SETTING, AND PARTICIPANTS A single-center, open-label, randomized phase 2 clinical trial was performed among 96 patients who had histologically and cytologically confirmed unresectable stage III NSCLC. Participants were enrolled from November 2011 to August 2015. Data were analyzed from February to October 2018. INTERVENTION Patients were randomized to receive thoracic radiation, 60 Gy, for 6 weeks concurrent with etoposide and cisplatin or the same regimen of CCRT combined with 200 mg of celecoxib, taken twice daily. MAIN OUTCOMES AND MEASURES The primary end point was overall survival. The secondary end points were the proportion of patients with treatment-related toxic effects, progression-free survival, and overall survival in subgroups with and without the COX-2 genotype. RESULTS A total of 100 patients were randomized. Following the exclusion of 4 outliers, 96 participants (96.0%) were analyzed (51 randomized to CCRT alone and 45 randomized to CCRT with celecoxib; mean [SD] age, 60.0 [8.3] years; 73.0 [76.0%] male). The median overall survival time was 32.8 (95% CI, 17.0-48.5) months in the group that received CCRT with celecoxib and 35.5 (95% CI, 25.8-45.2) months in the group that received CCRT alone (P = .88). Celecoxib with CCRT was well tolerated; the incidence of symptomatic radiation pneumonitis was 6.6% (95% CI, 1.4%-18.0%) in the group that received CCRT with celecoxib and 11.8% (95% CI, 4.4%-23.9%) in the group that received CCRT alone (P = .49). Among patients with the high-risk genotype, celecoxib plus CCRT was not associated with higher progression-free survival (hazard ratio, 0.36; 95% CI, 0.13-1.04; P = .05) or overall survival (hazard ratio, 0.50; 95% CI, 0.15-1.72; P = .26) compared with CCRT alone. CONCLUSIONS AND RELEVANCE In unresectable stage III NSCLC, adding celecoxib to concurrent chemoradiation did not improve survival. A smaller, not statistically significant proportion of patients in the CCRT with celecoxib group compared with the CCRT alone group developed symptomatic (continued) Key Points Question Could selective cyclooxygenase 2 inhibition combined with standard concurrent chemoradiation therapy improve survival among patients with unresectable stage III non-small cell lung cancer? Findings In this phase 2 clinical trial of 96 patients, adding celecoxib to concurrent chemoradiation did not improve survival. Meaning Adding celecoxib to concurrent chemoradiation did not improve survival.