Lung cancer has the highest mortality rate of any cancer, often due to the fact that it is detected at a late stage in its progression when it has already metastasised. The levels of certain biomarkers, such as proteins, metabolites and chemokines, in biological fluid or tissue could potentially detect cancer at an early stage, determine cancer subtype, or monitor the sensitivity to cancer treatment. Currently available lung cancer markers lack the sensitivity and specificity to be of great benefit and there is room for improvement. The research in this thesis aims to discover new biomarkers, using proteomic techniques, with the potential to improve or supersede those used at present for diagnosis and prognosis of lung cancer. Discovery phase was performed on conditioned media of lung cancer cell lines using 2D-DIGE in the hope it might mimic the serum/plasma environment of lung cancer patients. Further discovery phase was performed on serum using immunodepletion, proteominer, 2D-DIGE, label-free mass spectrometry followed by pathway analysis, metabolomic analysis, and multiplex assay analysis of cancer panels and a matrix-metalloproteinase panel. Validation in serum and plasma was performed using ELISAs, biochemical assays, and multiplex platforms, and in tissue using immunohistochemistry. Lung cancer subtypes examined in validation phase were squamous cell carcinoma, adenocarcinoma, and small cell lung cancer. There were insufficient serum/plasma samples to include a large cell carcinoma group. Potential biomarkers discovered include hnRNPA2B1, pyruvate kinase M2 (PKM2), HSC70-interacting protein (Hip), tenascin C, vascular endothelial growth factor alpha (VEGF-α), MMP-1, MMP-8, MMP-9, 12-HETE, and phenylalanine. In serum hnRNPA2B1, PKM2, Hip, tenascin C, VEGF-α, MMP-1, -8, and -9, and 12-HETE were increased in cancer compared to normal. Phenylalanine had similar levels in normal and small cell lung cancer but was decreased in non-small cell lung cancer. Of those examined in plasma hnRNPA2B1, VEGF-α, MMP-1 and -9, and 12-HETE were decreased in cancer compared to normal whereas PKM2 and tenascin C were increased. Where benign lung disease controls were used the markers were present at levels similar to lung cancer except for MMP-9 in plasma, VEGF-α in serum, and PKM2 in both serum and plasma. Immunohistochemistry of tissue showed an overall increase of expression of hnRNPA2B1 and HSC70-interacting protein in cancer tissue compared to normal. Functional assays were performed on hnRNPA2B1 showing its potential role in invasion and migration of lung cancer; its knockdown in the DLKP-M lung cancer cell line with two siRNA’s showed a 57% and 44% decrease in invasion and a 32% and 39% decrease in migration respectively. The research provided in this thesis demonstrates the importance for intensive validation before conclusions can be drawn on the overall usefulness of candidate lung cancer markers; a good serum marker does not necessarily make a good plasma marker and a marker that differentiates normal conditions from cancer conditions does not necessarily differentiate lung diseases/benign tumour from cancer.