Overexpression of tumor-associated antigens (TAAs) has been reported in many types of cancer and may trigger secretion of their autoantibodies. The present work was thus designed to test whether circulating antibody to p16 protein-derived antigens was altered in lung cancer. Two hundred seventy-one patients with non-small cell lung cancer (NSCLC) and 226 control subjects matched in age, gender, and smoking history were recruited in this study. The levels of circulating anti-p16 IgA and IgG antibodies were tested using an enzyme-linked immunosorbent assay (ELISA) developed in-house with linear peptide antigens derived from p16 protein. Student’s t test showed that patients with NSCLC had a significant higher level of anti-p16 IgG antibody than control subjects (t = 2.74, P = 0.0063) but did not have a significant increase in IgA antibody levels (t = 1.92, P = 0.056). The sensitivity against >90% specificity was 19.7% for the IgG assay with an inter-assay deviation of 11.6%, and 10.3% for the IgA assay with an inter-assay deviation of 14.7%. Based on a cut-off value determined by the 99th percentile of control IgG levels, the anti-p16 IgG positivity was 6.7% in patients with NSCLC compared to 0.88% in control subjects (χ (2) = 10.58, P = 0.001, OR = 7.97, 95% CI 1.84–34.85). Circulating anti-p16 IgG levels were increased with stages of NSCLC, and patients with stage IV NSCLC had the highest IgG level among all four stages (t = 2.42, P = 0.016, compared with the control group). Pearson correlation analysis showed a significant correlation between circulating levels of IgA and IgG in the patient group (r = −0.2, df = 236, P = 0.0021) but not in the control group (r = −0.1, df = 205, P = 0.146). Circulating IgG antibody to p16 protein may be a potential biomarker with prognostic values for lung cancer.