Molecular and Cellular Endocrinology 230 (2005) 95–106 Proteomic patterns: their potential for disease diagnosis Zhen Xiao, DaRue Prieto, Thomas P. Conrads, Timothy D. Veenstra, Haleem J. Issaq ∗ Laboratory of Proteomics and Analytical Technologies, SAIC-Frederick Inc., National Cancer Institute at Frederick, P.O. Box B, Frederick, MD 21702, USA Received 1 October 2004; received in revised form 6 October 2004; accepted 14 October 2004 Abstract Alterations in proteins abundance, structure, or function, act as useful indicators of pathological abnormalities prior to development of clinical symptoms and as such are often useful diagnostic and prognostic biomarkers. The underlying mechanism of diseases such as cancer are, however, quite complicated in that often multiple dysregulated proteins are involved. It is for this reason that recent hypotheses suggest that detection of panels of biomarkers may provide higher sensitivities and specificities for disease diagnosis than is afforded with single markers. Recently, a novel approach based on the analysis of protein patterns has emerged that may provide a more effective means to diagnose diseases, such as ovarian and prostate cancer. The method is based on the use of surface-enhanced laser desorption/ionization (SELDI) time-of- flight mass spectrometry (TOF-MS) to detect differentially captured proteins from clinical samples, such as serum and plasma. This analysis results in the detection of “proteomic” patterns that have been shown in recent investigations to distinguish diseased and unaffected subjects to varying degrees. This review will discuss the basics of SELDI protein chip technology and highlight its recent applications in disease biomarker discovery with emphasis on cancer diagnosis. © 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Proteomic patterns; SELDI; Cancer; Biomarker discovery 1. Introduction The multi-factorial nature of many diseases necessitates the use of biomarkers for early disease detection, for moni- toring the response to therapy and, in the best-case scenario, to predict the clinical outcome. These biomarkers can be cat- egorized according to their clinical applications. Diagnostic markers are used to initially define the histopathological clas- sification and stage of the disease, while prognostic markers can help forecast the development of disease and the prospect of recovery. Based upon the peculiarities of individual cases, the predictive markers can be applied to choose different ther- apeutic modalities. A biomarker could include patterns of single nucleotide polymorphisms (SNPs) or DNA methylation or changes in mRNA, protein, or metabolite abundances providing these patterns can be shown to correlate with the characteristics of the disease (MacNeil, 2004). It has been demonstrated, ∗ Corresponding author. Tel.: +1 301 846 1226; fax: +1 301 846 1438. E-mail address: issaqh@ncifcrf.gov (H.J. Issaq). however, that there is often no predictive correlation between mRNA abundances and the quantity of the corresponding functional protein present within a cell. Hence, since pro- teins represent the preponderance of the biologically active molecules responsible for most cellular functions, it is be- lieved that the direct measurement of protein expression can more accurately indicate cellular dysfunction underlying the development of disease (MacNeil, 2004). Advances in separations and analysis methods have pro- vided the means to advance our knowledge of proteins, how- ever, these methods do not feasibly lend themselves to routine use in a clinical setting due to technical concerns. There still exists a great need for development of novel methods capable of discovering protein biomarkers that possess high predic- tive and prognostic characteristics that can be translated into clinical practice. This need is greatest in the case of cancer, where the survival rates can be greatly increased with early diagnosis and intervention. Proteomics is a field that seeks to develop methods ca- pable of completely characterizing the entire complement of proteins in a biological sample. Many of the technologies 0303-7207/$ – see front matter © 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.mce.2004.10.010