Prediction of Drug Sensitivity and Resistance of Cancer by Protein Expression Profiling

Although the statistical probability of therapeutic success is known for larger groups of cancer patients, the clinical response to chemotherapy of the individual patient remains uncertain. It would be of great value to know whether or not an individual tumor responds to the proposed therapy. The concept of sensitivity testing of tumors for individualized therapy traces back to the 1970s. Currently, an astonishing revival has taken place due to the thriving development of genomic and proteomic technologies. This review discusses our own results on protein expression profiles of non-small cell lung cancer, kidney carcinoma and acute lymphoblastic leukemia regarding the prediction of drug sensitivity or resistance. A great diversity of drug resistance mechanisms are operative in the clinical drug resistance of cancer e.g., resistance proteins, proliferative, apoptotic, angiogenic factors, proto-oncogenes and tumor suppressor-genes. Hierarchical cluster analyses and cluster image maps reveal different resistance profiles even within cancer types of homogeneous histology. Protein arrays may be appropriate to perform sensitivity or resistance tests for individual patients because thousands of proteins may be detected in a single experiment. On the other hand, results suggest that already a set of a limited number of factors may be sufficient to detect the sensitivity or resistance of a cancer.