Bridging the Biomarker Gap
How to bring better diagnostic tests to the market to benefit patients.
Biomarker discovery and validation relies heavily on reproducible and robust analytical methodology. Separation science and notably high performance liquid chromatography (HPLC) are essential and efforts spent in developing more efficient and robust HPLC stationary phases, together with advanced mass spectrometers, have made modern biomarker research possible. Given that many clinical biomarkers are proteins, advances in protein bioanalysis and proteomics have also been critical in driving the biomarker field forward. Many new biomarker candidates are proposed for various diseases every week – but often from small-scale studies lacking statistical power.
Sorting through this mountain of information and prioritizing biomarkers for further validation is a challenge. And rather disappointingly, only a few biomarker candidates survive the validation phase in large clinical studies – even fewer enter commercial development and clinical application. This “biomarker gap” is recognized and major efforts are being deployed to professionalize biomarker discovery and validation. The recently founded Dutch Biomarker Development Center, a public-private partnership consortium, is a good example of the actions taken (http://biomarkerdevelopmentcenter.nl/).
I will talk about the challenges inherent to any biomarker discovery and development program with a focus on analytical science at HPLC 2015 in Geneva. I shall highlight pre-analytical factors that may bias biomarker studies, leading to discoveries that cannot be validated later on. I will exemplify this with studies on cervical cancer (1) and multiple sclerosis (2), and refer to other published studies where appropriate. While HPLC coupled to mass spectrometry holds great promise to gain a better understanding of the intricate changes that occur in protein and metabolite profiles in body fluids or tissue, it is vital that researchers are aware of the need for equally powerful data processing and statistical analysis approaches.
I’ll also highlight some examples showing that data processing and statistical analysis alone may influence the final result considerably (3). I have no doubt that the trio comprising well-designed comparative clinical studies addressing relevant disease-related questions, validated and robust analytical techniques, and reliable data processing and analysis forms the basis for successful biomarker research. Notwithstanding some setbacks, the field is alive and still holds great promise notably in the field of personalized medicine.
See you at HPLC 2015 in Geneva (www.hplc2015-geneva.org).
- A. P. Boichenko et al., “A Panel of Regulated Proteins in Serum from Patients with Cervical Intraepithelial Neoplasia and Cervical Cancer”, J. of Proteome Res. 13, 4995-5007 (2014).
- T. Rosenling et al., “The Impact of Delayed Storage on the Measured Proteome and Metabolome of Human Cerebrospinal Fluid”, Clinical Chemistry 57, 1703-1711 (2011).
- T. Rosenling et al., “The Effect of Pre-Analytical Factors on Stability of the Proteome and Selected Metabolites in Cerebrospinal Fluid (CSF)”, J. Proteome Res. 8, 5511-5522 (2009).
In 1987, Rainer Bischoff joined Transgene (Strasbourg, France) where he began protein-related research. He pioneered the application of mass spectrometry to characterizing recombinant proteins and published one of the first papers on this technique in 1990. Rainer continued his industrial career at AstraZeneca R&D (Lund, Sweden) before joining the Faculty of Mathematics and Natural Sciences at the University of Groningen (The Netherlands) in 2001, where he is full professor of Analytical Biochemistry. His research interests focus on biomarker discovery and validation, bioinformatics, the bioanalysis of biopharmaceutical proteins and the development of novel instrumental analytical techniques. He is the author of more than 160 peer-reviewed publications and book chapters and holds 12 patents.
