18040 Background: The clinical benefit of non-cisplatin doublets vs a single agent in elderly or unfit p is still controversial. The present study focuses on clinical outcome with VRB/GEM in elderly p and the role of functional status and comorbidities. Genetic predictive markers of response to VRB/GEM will also be examined in genomic and cDNA from tumor and circulating tumor DNA. Methods: 145 chemonaive p with stage IIIB (pleural effusion or supraclavicular lymph nodes)-IV or recurrent NSCLC and age > 70 years were accrued at 32 sites between April 2004 and January 2006. Treatment consisted of VRB 25 mg/m2 IV or 60–80 mg/m2 oral plus GEM 1200 mg/m2, days 1, 8 every 21 days. Activities of daily living (ADL), instrumental activities of daily living (IADL) and comorbidities were evaluated. DNA samples were collected from primary tumors for the assessment of microtubule associated protein 4 (MAP4) and from serum for the checkpoint with forkhead-associated and ring finger (CHFR) methylation. Results: Data on 130 p is available. Median age 76 years (69–83); males: 86.8%; smokers: 70.5%; PS 0–1: 83.9%; adenocarcinoma: 34.4% / squamous: 48%; stage IIIB: 22.7%, IV: 77.3%. Self-sufficiency in ADL and IADL was 77.4% and 45.2% of the p analyzed. 68% of the p had comorbidities. Median cycles: 3 (1–8). Hematological toxicities (%p): grade 3/4 neutropenia, 7.8%/4.7%; grade 3/4 thrombocytopenia, 2.3%/0.8%; grade 3 anemia, 3.1%. Efficacy in evaluable population: PR, 23.2% (95% CI, 15.1% to 32.9%); SD, 41.1%. 24 p died during the treatment period (non toxicity related) and 21 p were not evaluable. With a median follow up of 5.8 months, median survival for the whole population was 4.97 months (m), progression free survival 4.53 m, event free survival 3.43 m, 1-year survival 26.6%. Statistically significant differences in median survival were observed among subgroups: PS 0–1/2, 6.5/2.3 m (p<0.001); sex male/female, 4.5/9.7 m (p 0.027); ADL <6/=6, 3.4/7.1 m (p 0.023). Conclusions: This trial confirms that VRB/GEM is effective, presenting a favorable toxicity profile in elderly p with advanced NSCLC. Complete data on genetic markers will be presented. No significant financial relationships to disclose.