In the year 2003 only, 157,200 people are expected to die from lung cancer, exceeding the mortality from breast, prostate, and colorectal cancers combined. Identification of biomarkers for early lung cancer detection is essential to reduce the mortality of this disease. We have developed an antibody-based approach to biomarker discovery in cancer, focusing on differential protein expression rather than RNA/gene expression. Previously, we have selected a subset of lung cancer specific antibodies from our large collection of polyclonal antibodies (∼100,000) raised against individual human proteins and their isoforms. This subset was generated using matched normal and diseased tissues from lung cancer patients. We have applied the lung cancer specific antibody subset to over 200 human plasma specimens from normal, high-risk, benign, and early adenocarcinoma stage I lung cancer patients. Screening was performed using our Matrix Protein Array Technology (MPAT), a multiplex protein array immunoassay. Total plasma proteome from normal and diseased samples was printed on a membrane, then allowed to interact with individual antibodies. Antibody-sample reaction was detected by chemiluminescence, and computer analysis of a CCD-acquired (charge-coupled device) image was applied to quantify each spot. Statistical analysis of the data points thus generated, enabled us to select stage-specific antibody panels that recognize differentially expressed targets in early stages of lung cancer. Further validation of such antibodies using Western blot and larger cohort of patients in a longitudinal study is presented. In summary, we have discovered a panel of plasma secreted proteins correlating to early lung cancer. These biomarkers offer potential applications in the early diagnosis, prognosis and therapy response of lung cancer.

[Proc Amer Assoc Cancer Res, Volume 45, 2004]