LBA7727 Introduction: The mild toxicity profile of pemetrexed (P), seen in a phase III trial of P 500 mg/m2 vs docetaxel 75 mg/m2 in vitamin-supplemented patients (pts), suggests higher doses of P can be given in pts with previously treated advanced non-small cell lung cancer (NSCLC) without significant toxicity. We conducted a phase III trial to determine if a higher dose of P (900 mg/m2) could improve survival vs the standard P dose in pts with NSCLC. Methods: Pts with stage III/IV NSCLC, previously treated with platinum-based chemotherapy, were randomized to P 500 or 900 mg/m2 iv q3week. After the second planned interim analysis, the Data Safety Monitoring Board recommended discontinuation of enrollment due to low probability of demonstrating a survival advantage and a greater incidence of some toxicities on the P 900 arm. Patients were allowed to continue treatment at the P 500 dose. Results: 588 randomized pts were evaluated for efficacy and 581 pts, who received =1 dose, for safety. Safety data for pts who transitioned from P 900 to P 500 were analyzed separately. The treatment arms were balanced regarding baseline characteristics and prior treatment. Both arms had: ∼67% males, median age 62 yrs, 87% pts with an ECOG PS of 0 or 1, and 77% pts with stage IV disease. Key results are shown in the table . There was no statistical difference between arms for any efficacy measure. In general, the incidence of adverse events (AEs) was comparable or numerically higher in the P 900 arm. Some AEs (all grades) reported a >5% difference between the P 900 (N=240) and P 500 arms: fatigue (41.7% vs 32.8%), anemia (32.9% vs 22.1%), vomiting (20.0% vs 13.1%), stomatitis (17.9% vs 10.0%), diarrhea (15.4% vs 10.0%), and thrombocytopenia (11.3% vs 5.5%). Conclusion: P 900 offers no advantage over P 500 mg/m2 as second-line therapy for pts with advanced NSCLC. Certain toxicities were somewhat more pronounced in the P 900 group. No significant financial relationships to disclose. [Table: see text]