Background: We recently showed that IGFBP2 is overexpressed in primary lung cancer tissues. This study aims to determine whether IGFBP2 is elevated in blood samples of lung cancer patients and whether its level is associated with clinical outcomes. Methodology/Principal Findings: Plasma IGFBP2 levels were determined blindly by enzyme-linked immunosorbent assay in 80 lung cancer patients and 80 case-matched healthy controls for comparison. We analyzed blood samples for IGFBP2 levels from an additional 84 patients with lung cancer and then tested for associations between blood IGFBP2 levels and clinical parameters in all 164 lung cancer patients. All statistical tests were two-sided and differences with p<0.05 were considered significant. The mean plasma concentration of IGFBP2 in lung cancer patients was significantly higher than that in healthy controls (388.12 +/- 261.00 ng/ml vs 219.30 +/- 172.84 ng/ml, p<0.001). IGFBP2 was increased in all types of lung cancer, including adenocarcinoma, squamous cell cancer, and small-cell cancer, regardless of patients' age, sex, or smoking status. IGFBP2 levels were mildly but significantly associated with tumor size and were significantly higher in stage IV than stage I or III disease. A multivariate analysis showed that lung cancer patients whose blood IGFBP2 was higher than 160.9 ng/ml had a poor survival outcome, with a hazard ratio of 8.76 (95% CI 1.12-68.34, p=0.038 after adjustment for tumor size, pathology, and stage). The median survival time for patients with blood IGFBP2 > 160.9 ng/ml is 15.1 months; whereas median survival time was 128.2 months for the patients whose blood IGFBP2 was <= 160.9 ng/ml (p =0.0002). Conclusions/Significance: Blood IGFBP2 is significantly increased in lung cancer patients. A high circulating level of IGFBP2 is significantly associated with poor survival, suggesting that blood IGFBP2 levels could be a prognostic biomarker for lung cancer.