Identification of Serum Biomarkers for Lung Cancer

Background: Symptoms of lung cancer (LC) often do not appear until the disease is advanced; only 15% of LC cases are discovered while the tumor is in the early stages of development. Carcinogen exposure, asthma and smoking have been determined to be risk factors for the development of LC. Early detection of LC will likely have a major impact on the natural history of the disease and facilitate curative treatment. The objective of this study was to apply multiplexed immunoassays to identify a panel of biomarkers for early detection of LC.

Methods: Normal (NO) serum controls (n = 30) from healthy volunteers and lung cancer patients (n = 30) were acquired from a commercial vendor. Baseline (pre-treatment) serum specimens from individuals with asthma (AST; n = 28) and lung cancer risk (LCR; n = 73) were available from clinical trials of two novel agents that are being developed by JBNI Inc. for the respective indications. Serum levels of 59 cytokines, growth factors and biologically active peptides were quantified using multiplexed immunoassays using the Luminex platform to identify biomarkers that are expressed in a significantly different manner in individuals with LC, LCR, or AST in comparison with NO subjects. Data were reduced using nearest neighbor cluster analysis with squared Euclidian distance to separate patients into groups across analytes with inter-pathology comparisons determined using Student’s t test.

Results: Multiple analytes showed highly significant differences (p < 0.0002) between LC and healthy controls as single indices of pathologic state. In addition we were able to differentiate AST from LC (p < 0.002) and LCR and LC (p < 0.0001) using a panel of thirteen markers in various combinations. Using multiplexed assays we found significant differences in biomarker levels in sera of LC compared to NO, in NO compared to AST and LC compared to AST samples. Our results support an extended multiplexed immunoassay-based analysis of serum biomarker profiles as supplementary tools for the diagnosis of pathologic and as an aid in the development of novel agents for prevention, early detection and treatment of LC.

Conclusions: We have identified a group of markers having high inter-pathology discrimination power that are capable of reliably differentiating AST and LC from control specimens. This panel remains to be validated in a larger set of specimens but we are confident that these measures will produce clinical assays capable of reliably diagnosing lung pathology.