Biomarker Discovery - Mechanism of Action

Mechanism of Action

The way that these tests have been found can be viewed as biomarker discovery; however, their identification has primarily been made one at a time. Many well-known tests have been identified based on biological insight from the fields of physiology or biochemistry; therefore, only a few markers at a time have been considered. An example of biomarker discovery is the use of insulin to assess kidney function. From this process a naturally occurring molecule (creatinine) was discovered, enabling the same measurements to be made without insulin injections.

The recent interest in biomarker discovery is spurred by new molecular biologic techniques, which promise to find relevant markers rapidly without detailed insight into the mechanisms of a disease. By screening many possible biomolecules at a time, a parallel approach can be attempted; genomics and proteomics are some technologies used in this process. Secretomics has also emerged as an important technology in the high-throughput search for biomarkers; however, significant technical difficulties remain.

The identification of clinically significant protein biomarkers of phenotype and biological function is an expanding area of research which will extend diagnostic capabilities. Biomarkers for a number of diseases have recently emerged, including prostate specific antigen (PSA) for prostate cancer and C-reactive protein (CRP) for heart disease. Using biomarkers from easily assessable biofluids (e.g. blood and urine) is beneficial in evaluating the state of harder-to-reach tissues and organs. Biofluids are more readily accessible, unlike more invasive or unfeasible techniques (such as tissue biopsy).

Biofluids contain proteins from tissues and serve as effective hormonal communicators. The tissue acts as a transmitter of information, and the biofluid (sampled by physician) acts as a receiver. The informativeness of the biofluid relies on the fidelity of the channel. Sources of noise which decrease fidelity include the addition of proteins derived from other tissues (or from the biofluid itself); proteins may also be lost through glomerular filtration. These factors can significantly influence the protein composition of a biofluid. In addition, simply looking at protein overlap would miss information transmission occurring through classes of proteins and protein-protein interactions.

Instead, the proteins' projection onto functional, drug, and disease spaces allow measurement of the functional distance between tissue and biofluids. Proximity in these abstract spaces signifies a low level of distortion across the information channel (and, hence, high performance by the biofluid). However, current approaches to biomarker prediction have analyzed tissues and biofluids separately.