RATIONALE: Erlotinib hydrochloride and celecoxib may stop the growth of tumor cells by blocking some of the enzymes needed for cell growth. Celecoxib may also stop the growth of lung cancer by blocking blood flow to the tumor. Giving erlotinib hydrochloride together with celecoxib may kill more tumor cells.

PURPOSE: This randomized phase II trial is studying how well giving erlotinib hydrochloride together with celecoxib works compared with erlotinib hydrochloride alone in treating patients with stage IIIB-IV non-small cell lung cancer.

PRIMARY OBJECTIVES:

I. Comparison of progression-free survival (PFS) in patients receiving erlotinib + celecoxib vs. erlotinib + placebo for advanced NSCLC.

SECONDARY OBJECTIVES:

I. Objective tumor response rate as defined by RECIST Criteria for subjects receiving erlotinib/celecoxib treatment arms.

II. Categorize the change in e-cadherin expression from baseline to week 8 in a subset of subjects.

III. Evaluation of overall survival (OS). IV. Measurement of COX-2, EGFR by immunohistochemistry and EGFR amplification by FISH, and EGFR mutation status to correlate with clinical response.

V. Measurement of change in urinary PGE-M and correlation with response.

OUTLINE: Patients are randomized to 1 of 2 treatment arms.

ARM I: Patients receive oral erlotinib hydrochloride once daily and oral placebo twice daily on days 1-28.

ARM II: Patients receive oral erlotinib hydrochloride once daily and oral celecoxib twice daily on days 1-28.

In both arms, treatment repeats every 28 days for 12 courses in the absence of disease progression or unacceptable toxicity.

After completion of study treatment, patients are followed periodically.

Inclusion

• Pathologically proven NSCLC, stage IIIB (defined as: with pleural effusion or recurrence after mediastinal radiation and chemotherapy) or IV
• Available tumor tissue for mutation screening
• Measurable stage IIIb or IV disease by RECIST guidelines
• ECOG performance status of 0 or 1
• Progressive disease despite >= 1 prior chemotherapy regimens as standard of care or subject's refusal or inability to receive standard chemotherapy
• Normal renal function (defined as serum creatinine =< 2mg/dl)
• Normal liver function (defined as serum total bilirubin =< 1.5, and serum transaminases =< 2.5X the upper limits of normal [ULN]); if liver metastases are present, serum transaminases > 5X the ULN
• No evidence of coagulopathy (defined as PT and/or PTT =< 1.5X ULN or platelets >= 100,000)
• No evidence of leukopenia (defined as absolute neutrophil count >= 1,500 mm^3)
• Negative pregnancy test prior to initiation of treatment and adequate contraception throughout treatment

Exclusion

• Cytotoxic chemotherapy agents within 4 weeks of initiating treatment; all toxicities must be recovered to baseline or NCI CTCAE v3.0 Grade 1 from all acute effects of prior cancer treatment, except alopecia or any clinically insignificant effect, prior to study initiation
• Evidence of NYHA class III or greater cardiac disease, history of myocardial infarction, cerebral vascular accident, symptomatic ventricular arrhythmia, or symptomatic conduction abnormality
• Non-cytoxic therapy within 2 weeks of initiating treatment ; all toxicities must be recovered to baseline or NCI CTCAE v3.0 Grade 1 from all acute effects of prior cancer treatment, except alopecia or any clinically insignificant effect, prior to study initiation
• Prior radiotherapy to target lesions is not permitted unless completed more than 4 weeks prior to treatment within the study and that there has been documented progression at these sites (Radiotherapy to non-target lesions is permitted within 2 weeks of study entry provided all acute effects of the radiotherapy have resolved at least grade 1)
• Comorbid disease or a medical condition that would impair the ability of the subject to receive or comply with the study protocol
• Prior malignancy within the last 3 years with the exception of non-melanoma skin cancer or cervical cancer in situ
• Hypersensitivity of erlotinib or celecoxib or to any of the excipients of these products
• Hypersensitivity to sulfonamides, aspirin or other NSAIDS
• Prior history of EGFR inhibitor for the treatment of cancer
• Previous history of gastrointestinal ulceration, bleeding or perforation
• Concurrent use of COX-2 inhibitors or other NSAIDS (For subjects on NSAIDS prior to study initiation, cessation of the drug for 72 hours prior to study entry is required)
• Chronic or concurrent use of steroids (topical steroids are acceptable if medically indicated)
• Subjects who require treatment with fluconazole or lithium
• Any evidence of clinically active interstitial lung disease (patients with chronic stable radiographic changes who are asymptomatic need not be excluded)
• Renal insufficiency (defined as serum creatinine > 2 mg/dl)
• Liver insufficiency (defined as serum total bilirubin > 1.5, or serum transaminases > 2.5C the upper limits of normal [ULN]); if liver metastases are present, serum transaminases > 5X the ULN
• Coagulopathy (defined as PT and/or PTT > 1.5X ULN or platelets < 100,000)
• Leukopenia (defined as absolute neutrophil count < 1,500/mm^3)
• Pregnancy or inadequate contraception
• Lactating females
• Active CNS metastasis (stable, treated CNS metastasis acceptable)