February 2009: Volume 35, Number 2
Is There a Role for Clinical Labs Now?
By James Ritchie, PhD, and Henry Rodriguez, PhD, MBA
Proteomics, one of the many “omics” buzzwords today, is loosely defined as the systematic analysis of the proteins in a biological sample, especially proteins expressed during a disease state. As such, clinical proteomics is nothing new to clinical chemists who have been measuring proteins in biological samples for years. What has changed in the last 10 years, however, is the ability to simultaneously measure multiple proteins in a biological sample. This change has been fueled by the rapid advances in mass spectrometry that began in the mid-1980s and continues today.
The discovery of disease biomarkers using proteomic techniques holds enormous potential to revolutionize clinical practice and improve patient care. It seems, however, that new disease-associated protein biomarkers are described in the scientific literature every day, yet few of these candidate biomarkers have advanced past the discovery phase to become clinical diagnostic biomarkers. This lack of progress indicates an obstruction somewhere in the pipeline between the discovery and clinical validation phases (Figure 1).
The Pipeline of New Protein Biomarkers from Proteomics
Click for figure
Numbers of analytes is total possible proteins which can be measured. Numbers of samples is the numbers of patient samples which need to be tested to validate the new biomarker. LC-MS/MS: liquid chromatography tandem mass spectrometry; SID: stable isotope dilution; MRM: multiple reaction monitoring.
Reprinted by permission from MacMillan Publishers Ltd: Nature Biotechnology, 24:971-983, 2006.
The National Cancer Institute recognized the promise of clinical proteomics for the early detection and treatment of cancer and launched a program in 2006 to tackle the challenges faced by the proteomics research community. The Clinical Proteomic Technologies for Cancer (CPTC) initiative aspires to bring together institutions interested in advancing the promise of proteomics. One of the goals of the initiative is to make available well-characterized reagents and processes for the verification phase of proteomic assay development, which is critical for transferring proteomic assays and analytes from research labs to clinical labs.
This article describes some of the prominent technology in the proteomics field and the CPTC’s initiatives, as well as the important role clinical chemists are poised to play in the development of these new clinical analyses.
Overview of Proteomic Technologies
Proteomic research laboratories collect, store, and study proteins in different ways, using a variety of platforms and practice standards. This myriad of technologies and practices introduces an additional layer of variability on top of the immense biological complexity of the human proteome. Furthermore, lack of standardization, well-characterized reagents, and inter/intra-assay variability has made it extraordinarily difficult to compare results among labs and has greatly hindered clinical assay validation for new biomarkers.
Recently, proteomics labs have begun to focus on two mass spectrometry technology platforms: matrix assisted laser desorption/ionization-time off flight (MALDI-TOF) and linear trap quadrapole orbitrap (LTQ orbitrap). Both of the technologies offer the exceedingly high mass accuracy and resolution needed for accurate protein measurements of minute quantities of protein analytes.
Additionally, technologies are evolving for verifying biomarkers. Stable-isotope dilution, multiple reaction monitoring (SID-MRM) has been used previously for measuring small molecules, such as drugs and metabolites, and is already a standard technique in some clinical labs. In proteomic applications, the technique involves enzymatically digesting the sample, typically with trypsin, and then identifying three-to-five proteotypic peptides from the protein of interest. For analysis, synthetic stable isotope-labeled versions of each peptide are added to the sample as internal standards. Comparison of the mass spectrometry signals from the endogenous tryptic-peptides to those of the stable, isotope-labeled internal standard peptides allows accurate concentration determinations (Figure 2). Currently, this approach is limited to proteins in the mid-to-low ng/mL range.
Verification of Candidate Protein Biomarkers from Proteomics
Click for figure
The diagram shows the process of multiple reaction monitoring as it applies to proteomics. Once candidate biomarkers are identified, signature tryptic peptides must be elucidated and stable, isotope-labeled peptides must be synthesized (left side). A plasma sample can then be quantitated by the digestion of the endogenous proteins of interest. For the mass spectrometer analysis, labeled, internal standard peptides are used for quantitation of the daughter ions generated from each peptide (right side).
Reprinted by permission from MacMillan Publishers Ltd: Nature Biotechnology, 24:971-983, 2006.
To increase sensitivity, researchers have developed another technique: stable isotope standards with capture by antipeptide antibodies (SISCAPA). In this approach, antipeptide antibodies are made against the selected peptides formed by enzymatic digestion of the proteins of interest and used to enrich the digestion mixture before mass spectrometric analysis. In a typical assay, the protein mixture from plasma or urine is first digested with trypsin. The stable, isotope-labeled exogeneous peptides are added to the digest and the mixture is then passed through an affinity column or mixed with antibody affinity beads to which the antibodies specific to the signature peptides for the protein are bound. The contaminants are washed off the matrix and the tryptic peptides plus their internal standards are eluted in a small volume. This fraction is then analyzed by the MRM process described above.
Antibodies used in SISCAPA must have high affinity for the tryptic-peptides, but high selectivity is not required due to the high specificity of mass spectrometry detection. This process has the advantage of enriching the plasma-digest peptides by >1,000 fold, lowering the limits of quanitation into the low ng/mL range, and improving throughput by enabling multiple proteins to be measured simultaneously, the process known as multiplexing. In fact, the first clinical assay using this technique to quantify thyroglobulin in plasma was recently reported in Clinical Chemistry (See Suggested Reading).
The Clinical Proteomic Technologies for Cancer Initiative
Proteomic analysis requires sophisticated technology not widely used in clinical labs today. As described in Figure 1, the clogs in the biomarker pipeline must be fixed in order for new biomarkers resulting from proteomic technologies to become clinically useful. To get to that endpoint, proteomic research labs need standards, protocols, and reagents—in essence a “proteomics toolkit” —so that biomarker discoveries can be translated into real clinical utility.
The goal of the CPTC’s 5-year initiative is to reduce the layers of variability that currently prevent applying proteomics to clinical practice. To help meet this ambitious goal, CPTC has partnered with scientists from nearly 50 federal, academic, and private sector organizations (Figure 3). This collaborative network is working not only to understand the reproducibility and sensitivity of various biomarker discovery platforms, but also to define the best technologies for quantitative verification of biomarkers. Success in these areas will ultimately advance protein biomarkers to clinically validated products faster, cheaper, and more effectively.
The Clinical Proteomic Technology Assessment for Cancer Network
Click for figure
Lead centers are depicted in red, direct partners in blue, and centers with supporting expertise in green. Laboratorians who want to become involved in proteomics should contact researchers in the NCI-CPTAC network.
The proteomics toolkit, which will be provided to the scientific community, will allow proteomic data to be reproducible and comparable among labs. This means that investigators will have the assurance that protein/peptide measurement results are due to changes in the biological sample and not to variability in the instrument, assay performance, reagents, operator, or site. Such efforts should dramatically improve the quality of biomarker candidates that enter clinical practice.
CPTC’s initiatives fall into three major programs: the Clinical Proteomic Technology Assessment for Cancer, the Advanced Platforms and Computational Sciences, and the Proteomic Reagents and Resources Core.
Clinical Proteomic Technology Assessment for Cancer Program. A multidisciplinary network made up of five leading institutions or teams, the CPTAC program includes The Broad Institute of MIT and Harvard, Memorial Sloan-Kettering Cancer Center, Purdue University, University of California at San Francisco, and Vanderbilt University School of Medicine.
However, the CPTAC network actually extends beyond these five centers, bringing in expertise from both the public and private sectors (Figure 3). The teams are focused on identifying and reducing sources of variability in the current biomarker pipeline, with the goal of creating universal reference materials that will be used to characterize performance for all mass spectrometers, separation techniques, and data analysis methods.
The CPTAC teams are also focused on improving the biomarker pipeline by implementing an extra step, verification before clinical validation, which will pre-qualify biomarker candidates before costly clinical trials. Like panning for gold, this extra step sifts through large candidate lists for the gold nuggets or really useful biomarkers. For example, MRM, an established technology but not widely used in proteomics, is being optimized for this purpose. If MRM works as anticipated, it will increase the number of biomarkers entering clinical use and could be transformative to the proteomics field.
Advanced Proteomic Platforms and Computational Sciences Program. This program is focused on developing the next generation of quantitative proteomic technologies. NCI has granted 15 awards to academic institutions across the nation: seven are developing innovative proteomic technologies that are rapid, specific, reliable, and inexpensive and eight are working on extremely powerful computational tools that are necessary for accurate analysis of very large proteomic data sets. Advancements from these research programs will also help provide the necessary tools to bring new protein biomarkers into routine clinical use.
The Clinical Proteomic Reagents and Resources Core Program. A significant bottleneck within the field of proteomics is the lack of affordable, high-quality, well-characterized, and validated affinity reagents. While a plethora of reagents are available commercially, few are well-characterized, and many have highly variable quality.
Through collaborations with public, pri-vate, academic, and international institutions, the program is building a system for developing renewable sources of monoclonal antibodies and hybridomas for use by the entire proteomic community. Fully characterized antibodies against several cancer-related antigens are now available on CPTC’s website (http://cptc.abcc.ncifcrf.gov).
The program will also provide researchers with protein mixtures, standard operating procedures, and other resources needed for proteomic analysis. It is expected that this critical resource will be launched before the end of the year and will serve as a central source of reagents for the scientific community.
Clinical Chemists and Proteomics
As the field of proteomics continues to grow and mature, multiplexed proteomic assays eventually will become routine in clinical labs where they will be used to solve old analytical problems, as well as fulfill the promise of new disease-specific biomarkers. Standardization, reproducibility, accuracy, and clinical validation issues are now recognized as critical to proteomic assay development, and groups of diverse scientists are coming together to solve these problems. It is extremely important that the expertise of practicing clinical chemists be incorporated into these groups. The clinical lab community has the expertise in pre-analytical variation, assay performance, diagnostic accuracy, reference ranges, and overall analytical utility as they relate to real-world patient diagnoses.
To ensure that proteomic assays live up to the standards and performance characteristics of clinical labs, AACC’s Proteomics Division is preparing a white paper: Standards for Clinical Proteomic Diagnostics (see Box). It is envisioned that this document will be the first organized set of real-world performance standards for proteomic diagnostic testing. In addition, the document will outline a clear process by which investigators and manufacturers could meet these standards. Topics to be covered include: minimizing and assessing pre-analytical variation; assay validation and verification; appropriate quality control for multiplexed assays; reference ranges and diagnostic utility requirements; diagnostic algorithm validation; and requirements for clinical platform development.
Just as clinical labs embraced molecular diagnostics in the 1990’s, the advancements and initiatives described here are poised to put multiplexed proteomic assays on lab menus in the near future. Clinical chemists have been performing and overseeing protein diagnostic analyses for more than 100 years. As a profession, we must now step up and assist in helping shape the next wave of in vitro protein diagnostics. Indeed, we are the only clinical lab professionals qualified to do so.
Rafai N, Gillette M, Carr SA. Protein biomarker discovery and validation: the long and uncertain path to clinical utility. Nature Biotechnology 2006;24:971–983.
Hoofnagle, AN, Becker, JO, Wener, WH, Heinecke, JW, Quantification of thyroglobulin, a low- abundance serum protein, by immunoaffinity peptide enrchment and tandem mass spectrometry 2008, 54 :11, 1796–1804.
Horton GL, Jortani SA, Ritchie JC, et al. Proteomics: A new diagnostic frontier. Clin Chem 2006;52:1218–1222.
Mischak H, Apweiler R, Rosamonde EB, et al. Clinical proteomics: A need to define the field and to begin to set adequate standards. Proteomics Clin Appl 2007;1:148–156.
Scigelova, M and Makarov, A. Orbitrap mass analyzer—Overview and applications in proteomics, Practical Proteomics (a supplement to Proteomics Clinical Applications) 2006;6(52):2–87.
Online Resources for Proteomics
A good overview of mass spectroscopy
AACC Proteomics Division Website
NCI’s Clinical Proteomics Technologies for Cancer
James Ritchie, PhD, DABCC, FACB is professor of Pathology & Laboratory Medicine at Emory University Hospital, Atlanta, Ga. He is Past Chair of the AACC Proteomics Division.
Henry Rodriguez, PhD, MBA is director, Clinical Proteomic Technologies for Cancer, Office of Technology and Industrial Relations, Center for Strategic Scientific Initiatives, National Cancer Institute, Bethesda, Md.