Biomarkers in Clinical Medicine Research: A Literature Survey in the PubMed Database and a Critical Evaluation

Biomarker, the short form of "biological marker", appeared in the scientific literature in the 1940s. Since then, many different definitions have been suggested, but a generally applicable explanation of the term biomarker in science is extremely challenging. The word biomarker is found in 1.3 million articles in the scientific database PubMed^® that currently comprises more than 39 million citations for biomedical literature. Biomarkers are closely associated with human health and disease. The present article attempts to approach and evaluate the multifaceted term "biomarker" from a clinical perspective by searching the PubMed database. The search term biomarker was combined with other search terms related to medicine, physiology, biochemistry, and chemistry. Currently generally accepted clinical biomarkers, such as the high-molecular-mass N-terminal prohormone of brain natriuretic peptide (NT-proBNP, 60%), prostate-specific antigen (PSA, 67%), and troponin (37%), serve as a kind of positive control. The combination of the search term biomarker with selected low-molecular substances of clinically non-validated and hence rather experimental character yielded surprisingly high fractions of 41% for 8-iso-prostaglandin F_2α, 39% for symmetric dimethylarginine (SDMA), and 28% for asymmetric dimethylarginine (ADMA). The results of our survey are presented and discussed in detail for a wide spectrum of diseases. We focused on mechanisms that are assumed to underlie the biological activity and specificity of biomarkers. We also considered potential roles of the analytical chemistry of biomarkers including the emerging metabolomics and proteomics. Reliable analytical methods have been used for the quantification of the isomeric low-molecular-mass ADMA and SDMA in human biological samples. ADMA, but not SDMA, is considered an endogenous inhibitor of the endothelium-derived nitric oxide (NO) synthesis, one of the most potent endogenous vasodilators. Paradoxically, the utility of ADMA and SDMA as biomarkers in the renal and cardiovascular systems seems to contradict their main biological activity. This prominent pair is representative of many biomarkers and reveals that the supposed biomarker utility is likely to be predicated on not yet considered biological activity. The majority of human diseases are heterogenic, affect many organs and seem to include different and overlapping biochemical pathways. In recent years, especially proteomic studies provided a series of new potential candidate biomarkers. However, such biomarkers must still be validated in the clinic before they can be introduced into clinical practice. This is perhaps the most critical phase in the discovery of disease biomarkers. Our analysis reveals that the area of biomarker research is highly challenging. With minor exceptions, there is no specific biomarker for a single disease. In addition to clinical examinations, a combination of several biomarkers seems to be needed for reliable diagnosis and therapy. Analytical chemistry, especially proteomics, delivers a huge amount of data, which may complicate and even hinder progress in this area. Specific quantitative analysis of candidate biomarkers observed by proteomics (and metabolomics) is highly recommended to proceed with the same biological samples from studies in which the biomarkers were discovered.