Paclitaxel and Carboplatin With or Without Celecoxib Before Surgery in Treating Patients With Stage IIIA Non-Small Cell Lung Cancer

OBJECTIVES: * Compare the complete pathological response rate and/or minimal residual microscopic disease in patients with stage IIIA non-small cell lung cancer treated with preoperative paclitaxel and carboplatin with vs without celecoxib. * Compare the clinical response rate in patients treated with these regimens. * Compare chemotherapy-related toxicity in patients treated with these regimens. * Compare the time to progression, disease-free survival, and overall survival of patients treated with these regimens. OUTLINE: This is a randomized, double-blind, multicenter study. Patients are stratified according to use of aspirin for prior cardiovascular disease (yes vs no). Patients are randomized to 1 of 2 treatment arms. * Arm I: Patients receive paclitaxel IV over 3 hours and carboplatin IV over 1 hour on days 1, 22, and 43. Patients also receive oral celecoxib twice daily beginning on day 1 and continuing until the morning of surgical resection. * Arm II: Patients receive paclitaxel and carboplatin as in arm I and an oral placebo twice daily beginning on day 1 and continuing until the morning of surgical resection. In both arms, patients undergo surgical resection and complete mediastinal lymph node dissection within 3-6 weeks after completion of chemotherapy. Patients resume oral celecoxib or placebo twice daily within 28-42 days after surgery and continue until 3 years from the date of randomization in the absence of disease progression or unacceptable toxicity. Patients are followed every 3-6 months. PROJECTED ACCRUAL: A total of 110 patients (55 per treatment arm) will be accrued for this study.