Technologies Behind Biomarker Discovery and Development

Evidence-Based Oncology, February, Volume 18, Issue SP1

Metabolomics, Proteomics, and Epigenetics

For many years, researchers have recognized the promise that biomarkers hold for advancing the way we think about and treat cancer. Biomarkers are molecules found in biological specimens (eg, blood, bodily fluids, or tissues) that can serve as indicators of normal biological processes, pathogenic processes, or response to medical therapy.1 Clinical applications of biomarkers in oncology are expected to improve diagnosis, prognosis, prediction of response or recurrence, and disease monitoring. Biomarker discovery and development is moving forward through novel uses of complex technologies.

An educational session at the 2011 San Antonio Breast Cancer Symposium focused on 3 important technologies used in biomarker research: metabolomics, proteomics, and epigenetics. The first speaker was Arun Sreekumar, PhD, Baylor College of Medicine, who talked about metabolomics as a vehicle for biomarker identification and development. Gordon Mills, MD, PhD, University of Texas MD Anderson Cancer Center, followed with a discussion of proteomics. The final speaker, Stephen Baylin, MD, of Johns Hopkins University, provided an overview of epigenetics.

The American Association of Cancer Research, Food and Drug Administration, and National Cancer Institute (AACRFDA- NCI) Cancer Biomarkers Collaborative defines biomarker discovery and development as a pipeline of linked processes: hypothesis generation, research study design, sample collection, data collection, data analysis, assay development, assay validation, clinical qualification, regulatory approval, and clinical use.2 Collectively, the speakers touched upon many of these processes. While each speaker concentrated on a particular approach toward biomarker discovery and evaluation, they all adhered to a central theme: the successful application of biomarkers toward the design of combined rational therapy will require integrative efforts that exploit the information provided by these and other technologies.3,4 Dr Mills summarized the overarching goal of this type of research: “The key goal as we move forward is to have a sufficient understanding of what is happening in our patients’ tumors to develop rational combinatorial therapy and link biomarkers of patients who are likely to benefit to the best targeted therapy.”

The availability of high-throughput technologies has vastly broadened the potential for untargeted biomarker discovery, making it possible to scan hundreds of molecules at once. All 3 speakers emphasized the importance of establishing validity and maintaining quality control throughout each phase of biomarker discovery and development.


Metabolomics is the study of metabolites and the metabolome (sidebar). In his talk, Dr Sreekumar defined it as “the comprehensive and simultaneous systematic determination of metabolite levels in the metabolome and their changes over time as a consequence of stimuli, for example, disease, environmental changes, and drugs.” He explained that metabolites, as the end products of protein action, serve as cellular physiology indicators.

Dr Sreekumar began his talk by putting genomic, transcriptomic, proteomic, and metabolomic technologies into context with one another. The metabolome comprises about 3000 metabolites, making it less complex than the other “—omic” approaches. 5 According to Dr Sreekumar, they are also more approachable: “Metabolites are much more stable than genes or proteins and can be easily identified in biofluids.” Cancer-focused metabolic research is ongoing to identify metabolic markers of disease and tumor-specific biochemical pathways that can serve as druggable targets. During an interview conducted after his presentation, Dr Sreekumar remarked, “There are metabolic tests already available, like in the case of diabetes, you can look at glucose. Testing of metabolites is clinically done for metabolic syndrome, diabetes, and so on, but similar tests in the field of cancer have not come about yet. Our goal is to see whether we can provide such a test.”


Proteomics refers to studies of the proteome— the proteins of an organism or cellular system along with any modifications to those proteins—to understand cellular biology. In general, targeted therapies exploit a specific protein product within a perturbed cellular signaling pathway. The benefits of targeted therapies are often short-lived because cells are capable of adaptively responding after exposure to these drugs. These adaptive responses are revealed at the protein level through changes in homeostatic networks, feedbacks, and crosstalks. It is hoped that the identification of the mechanisms behind adaptive responses will inform the design of rational combined therapies. Dr Mills spoke about why it is important to move beyond RNA and DNA studies. “Although we have some incredible targeted therapeutics, as we look at what is going on with patients, most of the time only a subpopulation of patients respond. We obviously do not know why certain subpopulations respond and, in most of the cases, those responses are short-lived.”

Protein biomarkers can be assayed using relatively inexpensive standard assays, such as fluorescent in situ hybridization (FISH), enzyme-linked immunosorbent assay (ELISA), and Western blot, already in practice in CLIA-certified laboratories. While protein biomarkers are ripe with potential, their identification has been hampered by the inherent complexities of protein biology and the limitations that current technologies present. Dr Mills reviewed the efforts of the National Cancer Institute’s Clinical Proteomic Technologies Tumor Analysis Consortium, established to advance proteomics research.6 It is hoped that information gained from this program’s integrative approaches will produce a deeper understanding of cancer biology, with high-quality data sets, reagents, and analytically validated quantitative assays to be made publicly available.


DNA is subject to a number of modifications that do not involve a change in the actual DNA sequence, for example, methylation of DNA (the addition of a methyl group to a DNA strand) and histone deacetylation (removal of an acetyl group from a histone protein). Epigenetics is the study of functionally relevant changes of this nature (Figure 1).7

Dr Baylin focused most of his talk on the deregulation of DNA methylation, just 1 type of epigenetic abnormality (Figure 2).8 It is becoming apparent that methylation losses and gains happen simultaneously across the cancer genome in a nonrandom manner. In effect, what happens is that some epigenetic changes turn off (repress) critical tumor suppressor genes while other changes maintain genes that drive the cancer in an active configuration.

Next-generation sequencing projects such as the Cancer Genome Atlas Project have revealed that mutations in epigenome- modifying proteins occur at high frequency in cancer cells.8,9 What these mutations actually mean to the phenotype of the epigenome is not clearly understood. According to Dr Baylin, “This is going to be a big exercise that you are going to see, I predict, over the next few years—linking the change in the hard drive (DNA) to what happens to make the change in packaging of the DNA. I think this is going to be extremely instructive not only in providing targets of therapy all built on this growing understanding but also in our understanding of these genetic and epigenetic linkages.”

Earlier in the session, Dr Sreekumar presented published data from a metabolomic analysis of bladder cancer that affirmed the notion that a true understanding of cancer biology is going to come from studies that link all of these fields together. The data demonstrated that deregulation of an important enzyme pathway is linked to epigenetic modifications in the form of methylation.10

Therapeutics that target epigenetic aberrations may be able to reverse the changes that shut down tumor suppressors. This differs from the mechanism of many anticancer interventions that aim to kill cancerous cells. In fact, in an ongoing study presented by Dr Baylin, researchers are taking old drugs like the demethylating agent azacitidine and the histone deacetylation inhibitor entinostat and using them at nanogram doses that are much lower than historical cell-killing doses. Early results from a small study involving patients with recurrent non-small-cell lung cancer are promising. Preclinical studies of low dose azacitidine for the treatment of breast cancer are also ongoing, with encouraging preliminary results.

Looking Forward

Cancer treatment decisions are based, for the most part, on cancer type and stage, and not molecular characteristics of the cancer cells. The hope is that biomarkers will enable physicians to base treatment selection on the molecular profile of each patient’s cancer.

Most biomarkers identified to date have been stalled in the research setting and have not been carried over to clinical use. Clearing the hurdles that are preventing crossover will require a collaborative effort that brings together scientists and policy makers from a variety of disciplines and who represent academic, private, and public settings.4

Dr Mills emphasized this in his introductory statements by saying, “As we move forward to helping patients and translating our observations to patients we have to learn how to work with industry.…Many of the things we are doing are moving forward and we are working with industry in a way that is going to help our patients.”Author Disclosures: The author (KF) reports no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (KF); drafting of the manuscript (KF); and critical revision of the manuscript for important intellectual content (KF).1. Biomarkers Definitions Working Group. Biomarkers and surrogate endpoints: preferred definitions and conceptual framework. Clin Pharmacol Ther. 2001;69:89-95.

2. Khleif SN, Doroshow JH, Hait WN; for the AACRFDA-NCI Cancer Biomarkers Collaborative. AACR-FDANCI Cancer Biomarkers Collaborative Consensus Report: advancing the use of biomarkers in cancer drug development. Clin Cancer Res. 2010;16:3299-3318.

3. Mills G. Proteomics. 34th Annual CTRC-AACR San Antonio Breast Cancer Symposium; 2011; San Antonio, TX.

4. Nass SJ, Moses HL, eds. Cancer Biomarkers: The Promises and Challenges of Improving Detection and Treatment. Washington, DC: National Academies Press; 2006.

5. Beecher C. The human metabolome. In: Harrigan GG, Goodacre R, eds. Metabolic Profiling: Its Role in Biomarker Discovery and Gene Function Analysis. Boston, MA: Kluwer Academic Publishers; 2003.

6. Clinical Proteomic Tumor Analysis Consortium. Office of Cancer Clinical Proteomics Research Web site. Accessed December 19, 2011.

7. Baylin SB, Schuebel KE. Genomic biology: the epigenomic era opens. Nature. 2007;448:548-549.

8. Baylin SB, Jones PA. A decade of exploring the cancer epigenome-biological and translational implications. Nat Rev Cancer. 2011;11:726-734.

9. The Cancer Genome Atlas. Accessed December 19, 2011.

10. Putluri N, Shojaie A, Vasu VT, et al. Metabolic profiling reveals potential markers and bioprocesses altered in bladder cancer progression [published online ahead of print October 11, 2011]. Cancer Res. 2011;71:7376-7386.