Protein Biomarker Patterning: Entering the “Golden Age”.

It is widely described in many scientific circles that the development of “omics” technologies has encouraged a shotgun approach to biological science rather than hypothesis-driven strategies. And why not? These powerful technologies allow us to examine the effects of disease, drugs, genetics or the environment on multiple pathways and systems simultaneously and generate more data points in one experiment than was provided by years of traditional work. Older techniques required that we make precise decisions concerning what analytes to measure. This demanded a workable hypothesis and a focused experimental testing approach. As we all know, biology is not a precise science. Now with the advent of the “omics” one can simply cast a wide net to maximize our success at finding the relevant “biomarkers”.

Biomarkers are broadly defined to include genotypic markers, transcripts, proteins, lipids, and metabolites. With the exception of genomics, we know that perturbations to biological systems result in reproducible ripple effects throughout all of these biochemical systems. Although all of these molecular targets are important and potentially useful, many researchers have chosen to concentrate efforts on proteins and more specifically those proteins found in the plasma proteome.

There are many compelling reasons for this strategy. The vast majority of our therapeutic interventions today are designed to affect a protein. In fact the fastest growing class of therapeutics, biologics, are themselves proteins. The literature contains thousands of references that describe predictable alterations of protein expression in response to various treatments on the systemic level. From a business point of view in the future we will determine the genotype of an individual once in their lifetime but their phenotype (protein expression patterns) will be determined regularly. Our billion dollar diagnostics industry uses many blood-based protein measurements to guide medical decisions. We started fifty years ago with enzymatic tests for protein analytes such as liver enzymes that have become a mainstay of the industry. All of these early tests measured protein activity. With the advent of immunoassays, it became possible to precisely measure the concentration of numerous diagnostic proteins in the blood.

Blood is the standard clinical sample whose collection, treatment and storage is well characterized and understood. Protein measurement from blood is rugged and robust. Plasma and serum can be used directly for protein analysis without the cumbersome and often irreproducible effects of sample preparation or fractionation. Perhaps someday other testing techniques that include such “pre-treatments” will be used routinely in the clinical laboratory. However, today and in the foreseeable future, a simple plasma or serum sample is still our best choice for discovery, validation and use of new diagnostic biomarkers.

This is not to downplay the significance of the other omics such as proteomics, transcriptomics and metabolomics that utilize powerful new technologies. They pattern huge numbers of biomarkers using mass spectrometry and nucleic acid detection arrays that are on the cutting edge of important new findings. They are informative techniques when using samples derived from the cells or tissues of interest but they have limited applications for blood-based testing. Tissue biopsies, although integral to cancer diagnostics, can not be considered routine sample types for the clinical laboratory. However, this information does serve to provide new assay targets as these proteins or their surrogates inevitably find their way into the systemic circulation. For these reasons, we feel that the measurement of multiple members of the plasma proteome, what we call “Protein Biomarker Patterning” provides the greatest near-term potential for clinical application.

Most clinical diagnostic tests are based on single markers and as a result are subject to a high degree of uncertainty due to the inherent phenotypic variation of the general population. Consequently, the use of multiple biomarkers drives a substantial reduction in the uncertainty of drug safety, efficacy, and disease diagnosis. Protein patterning allows us to take advantage of ‘biomarker stacking’ and is a key component of the strategy. Biomarker stacking increases both the specificity and sensitivity for diagnostics, prognostic and theragnostic tests. To perform protein patterning of the plasma proteome, one must have quantitative tools that allow for many determinations to be made simultaneously from a small volume of sample.

Nucleic acid array technology provided our first glimpse of the power of this type of profiling. Protein arrays were late to the party because of the problems inherent with performing multiple immunoassays in a single space. The ability to perform multiple immunoassays simultaneously is termed ‘multiplexing”. There are now four new types of protein immunoassay arrays on the market in both the research and clinical arenas. The first is a copy of the nucleic acid arrays that place a miniature immunoassay on a planar surface that when developed results in a fluorescent signal that is imaged by a CCD camera. The second is a miniaturization of the Enzyme-Linked Immunoassay (ELISA) which has been a standard in the industry for many years. Although both sensitive and rugged, the standard ELISA suffers from the “one-tube/one assay” paradigm which limits the number of proteins measured in a sample. The new “multiplexed” ELISAs perform 2-16 immunoassays in the bottom of a single microtiter well somewhat improving the comprehensive nature of the data stream. The third is the “Lab-on-a-Chip” strategy that uses microfluidics to move assay reagents in repetitive small channels to achieve multiplexing. The fourth and most successful approach has been the use of a “suspension array” consisting of fluorescently-encoded microspheres where each individual immunoassay is performed on one of these addressable sphere sets. Multiple immunoassays can be performed on a small sample with the same precision and ruggedness characteristic of the single ELISA or the traditional immuno-analyzers found in the clinical lab today.

Rules-Based Medicine (RBM) commercialized their “suspension array” approach in 2002. Since that time, RBM has built a solid reputation in the industry by further maximizing this technology to deliver reproducible, quantitative, multiplexed immunoassay data. RBM’s Multi-Analyte Profiles (MAPs) measure hundreds of proteins from a small volume of sample such as serum, plasma, urine, cerebrospinal fluid, tissue culture supernatants, tissue extracts and others. These MAPs provide a comprehensive and cost-effective evaluation of protein expression patterns critical to applications such as drug safety and efficacy, disease diagnosis and disease modelling. The proven success of RBM’s multi-analyte profiling has resulted in numerous submissions to the FDA and patents filed based on the company’s data.

Significant Contributions

Perhaps the best indication of this success is RBM’s customer base which includes over 130 biotech, consumer products and pharmaceutical companies and over 100 academic and government institutions. The service performed for these organizations typically begins with basic research efforts that lead to pre-clinical trials in rodents or monkeys. Biomarker patterns discovered with these animal models are then applied in human clinical trials. In other cases, pharma have utilized banked clinical trial samples retrospectively looking for biomarker patterns that might indicate why the drug failed or might suggest other applications. A few examples of RBM’s success stories include:

• A biomarker pattern determined in a pre-clinical rodent trial that indicated deleterious dose-dependent effects that was then used in a Phase I clinical trial to monitor dosage.
• A biomarker pattern seen in a clinical trial that identified those candidates who would become responders to the trial compound.
• A biomarker pattern, first seen in mice and then validated in humans that correlated with drug efficacy.
• A diagnostic biomarker pattern that stratified a clinical trial population by more accurately identifying those candidates with a complex disease indication.
• A diagnostic biomarker pattern that detected sepsis earlier in hospitalized patients leading to earlier, more efficacious treatment.
• RBM and its customers have filed numerous biomarker patents using the MAP approach for both diagnostic and therapeutic applications.

The Strengths of this MAP Approach

• High Quality Data - RBM relies on proven suspension array technology coupled with automated liquid handling systems. Together they provide excellent assay precision and reproducibility while enjoying cost efficiencies and reduced sample volume requirements. RBM is capable of supporting Good Laboratory Practice (GLP) studies which is required by the pharmaceutical industry for certain discovery and preclinical trial activities. In addition, RBM is certified under CLIA which is a clinical laboratory standard required by pharma for support of human clinical trials
• Comprehensive Biomarker Data - The RBM service products measure more analytes quantitatively than any other platform. Both the RodentMAP™ and the HumanMAP® assess levels of hundreds of plasma proteins involved in numerous biochemical pathways. This comprehensive evaluation has been successful in discovering meaningful biomarker expression patterns across numerous species and sample types. This breadth allows investigators to utilize RBM’s services throughout their research and development cycle.
• Small Sample Volume - Often the volume of a biological sample is a limiting factor causing the investigator to make difficult, and sometimes erroneous, choices as to which analytes to measure. RBM’s multiplexing service alleviates this problem by enabling the performance of hundreds of assays on the same sample. For example, in the past an investigator with only 100 uL of sample was restricted to performing only 4-5 different immunoassays. With RBM, the same investigator can now perform hundreds of assays on the same sample.
• Speed to results - RBM delivers high-quality data in about two weeks following receipt of samples. This allows investigators to concentrate on the data thus expediting their research and development programs. Blockbuster drugs generate millions of dollars per day. With RBM services, they do not spend the time and resources necessary to set-up, validate and maintain their own assay systems.
• Cost Effective - The automated, multiplexed format of RBM’s testing services provides efficiencies and economies of scale that are passed on to the customer. For the customer, there are no capital expenditures, no personnel to hire and train, no time consuming and expensive validation procedures and no laboratory facilities to maintain.

The Future

We are at the earliest stages of the “Golden Age for Biomarkers”. As with other periods of rapid scientific discovery and progress, technology is leading the way. As the “omics” technologies mature and enter the commercial realm, their contributions to our understanding of complex biological systems will expand logarithmically. The future includes inexpensive whole genome scanning and eventually sequencing for identification of potentially deleterious genotypes, transcriptional profiling of tumor tissue that will indicate the severity of the malignancy and the most efficacious therapy, and metabolite profiling of urine that could provide sensitive and accurate prognostics for monitoring disease and therapy.

We also know that protein biomarker patterning outlined here has already yielded significant findings. The power of the approach will continue to grow with each new immunoassay added to the menu. This will enhance its ability to characterize the effects of new drug therapies; it will identify more sensitive diagnostics which will allow earlier intervention when therapy is most effective; it will improve our understanding of biological systems and it will lead us into the impending Golden Age.