Cyclooxygenase-2 (COX-2) catalyzes the rate-limiting step in prostaglandin (PG) synthesis and is overexpressed in 70% to 90% of non-small cell lung cancers (NSCLC). Preclinical studies suggest inhibition of COX-2 can enhance the cytotoxic effect of docetaxel. To test this concept clinically, we administered celecoxib (400 mg p.o. twice daily) plus docetaxel (75 mg/m(2) every 3 weeks) to a cohort of patients with recurrent, previously treated NSCLC. Patients first received single agent celecoxib for 5 to 10 days to ascertain the effectiveness of COX-2 inhibition, which was determined by measuring pre- and post-celecoxib levels of urinary 11alpha-hydroxy-9,15-dioxo-2,3,4,5-tetranor-prostane-1,20-dioic acid (PGE-M), the major metabolite of prostaglandin E(2) (PGE(2)). We enrolled 56 patients (35 men, 21 women; median age, 61 years). All patients had received at least one prior chemotherapy regimen. The overall response rate was 11% and median survival was 6 months, similar to that observed with docetaxel alone. Pre-celecoxib urinary PGE-M decreased from a mean level of 27.2 to 12.2 ng/mg Cr after 5 to 10 days of celecoxib (P = 0.001). When grouped by quartile, patients with the greatest proportional decline in urinary PGE-M levels experienced a longer survival compared to those with no change or an increase in PGE-M (14.8 versus 6.3 versus 5.0 months). Our data suggest that combining celecoxib with docetaxel using the doses and schedule employed does not improve survival in unselected patients with recurrent, previously treated NSCLC. However, in light of the apparent survival prolongation in the subset with a marked decline in urinary PGE-M levels, further investigation of strategies designed to decrease PGE(2) synthesis in NSCLC seems warranted.