**1. Introduction**

Despite the anticipated boom stemming from proteomic investigations, the rate at which novel protein biomarkers are introduced into clinical practice has remained static over the past 20 years. The reality is that approaches to both discover and validate protein biomarkers remain inadequate, and consequently, many areas of medicine, including the broad field of autoimmune disorders, remain deprived of the tools essential for the optimal management of patients. Most importantly, there is a huge backlog of candidate biomarkers that are yet to undergo thorough investigation and validation to assess their clinical utility. A recent assessment of the situation has estimated that although many tens of thousands of publications claim biomarker discoveries, there are roughly only 100 routinely used in clinical practice (Poste, 2011).

This chapter reviews the potential applications of protein biomarkers to manage autoimmune diseases with a special focus on the transition from the biomarker discovery through to validation phases using proteomic strategies. We emphasize the importance of careful review of the discovery data, the critical roles of protein isoform verification, and the essential features of targeted and thorough validation. Ultimately, when these factors are appropriately considered and implemented, we are optimistic that autoimmune disorders can be transformed by omics technologies and personalized practice can become a reality.

#### **1.1 Biochemical markers and their potential role in autoimmune disease**

Biological markers are widely used in medicine and can provide an objective measure of normal and pathogenic processes or pharmacologic responses to a therapeutic intervention. By the term '*biological markers*' (or biomarkers) we mean an objective molecular indicator or surrogate of pathological processes which possess diagnostic, prognostic or predictive

Validation of Protein Biomarkers to Advance the Management of Autoimmune Disorders 137

validation data have been reported. The significance of major barriers to their widespread clinical adoption is also briefly discussed. These challenges to new marker development and commercialization include factors such as study design, pre-analytical variables, data interpretation, bioinformatics, validation, ethics and commerce. While our focus is on protein biomarkers and proteomic methods in rheumatology, the principles we discuss are

The biomarkers currently used in the practice of rheumatology can be subdivided into three distinct molecular classes: nucleic acid, proteins, and metabolites. Some of these tests and their applications are included in Table 1. This is not an exhaustive list and while an invaluable armamentarium, it is limited in number, clinical utility and performance. With the exception of anti-CCP autoantibody and S100 proteins, the list has remained essentially static for the last 5-10 years. Additional potential biomarkers have been reported, but few, if any, have been adopted in routine practice. There is, however, no dispute that better tools are urgently required for both objective diagnosis and optimal management of rheumatoid

Several areas of medicine lack objective, quantitative measures of a disease and its response to therapy and the inflammatory forms of arthritis are no exception. Below, we define three distinct stages in the clinical progression to chronic inflammatory disease exemplified by some autoimmune disorders and outline where and how biomarkers could aid in managing

i. *The pre-symptomatic stage*: Here asymptomatic individuals, especially those who are genetically susceptible to rheumatoid arthritis (RA), need to be screened for early indications of disease onset. Biochemical markers could remove uncertainty in detecting the disease in its early stages and would allow early and appropriate intervention. Such early detection could help clinicians minimize or even halt disease progression, reduce morbidity and mortality, and markedly lower the costs of health

ii. *The early clinical stage*: In the early stages of rheumatic diseases patients may be symptomatic but there are insufficient clinical or laboratory findings to confirm the diagnosis. A few biomarkers exist which are helpful in predicting outcome at a population level, such as rheumatoid factor and, more recently, anti-cyclic citrullinated peptide antibodies (anti-CCP) for the diagnosis of rheumatoid arthritis (See Table 1). Clinicians believe that early diagnosis and treatment are essential for the best outcome. In the past, the presence of the biomarker, IgM rheumatoid factor (RF), has helped to identify patients likely to have more aggressive disease (Vocovsky et al., 2003). In the last decade, anti-CCP antibodies have been shown to predict a more aggressive disease course in very early disease (van Venrooij, et al., 2008). However, the relatively low sensitivity and specificity of their assay offers relatively low diagnostic sensitivity and this means that a significant minority of patients with aggressive disease cannot be identified (Kastbom et al., 2004; Lindqvist et al., 2005). Furthermore, neither RF nor anti-CCP antibodies are of any use in the other inflammatory arthritides of adulthood, such as psoriatic arthritis and ankylosing spondylitis, or for the vast majority of children

generally applicable to other analytes and areas.

arthritis.

the RA patient.

care delivery in rheumatology.

with juvenile idiopathic arthritis.

**1.3 Biomarkers currently used in the practice of rheumatology** 

**1.4 The future: additional markers & rheumatology practice** 

utility. These are distinct from '*clinical markers*' which rely on physical variables or symptoms such as joint count, pain assessment or radiological findings. Biomarker examples include: cardiovascular risk assessment through cholesterol checks; pituitary and target gland hormone determinations to assess endocrine function and dysfunction; hemoglobin A1c (HbA1c) evaluations to monitor blood glucose levels in diabetic patients; liver function tests (LFT) in liver disease; and, prostate-specific antigen (PSA) determinations to assess prostate cancer risk. Not surprisingly, there is considerable interest in developing additional clinical biomarkers in medicine; however, the path from their discovery to routine adoption is painstakingly complex and slow.

#### **1.2 Why target protein biomarkers**

Although genomic and transcriptomic methods are powerful, they cannot predict downstream events. Specifically, they can't predict what protein forms will be expressed in a particular tissue or biofluid, nor can they reliably estimate expressed protein levels. Because it is the gene products that contribute directly to physiological or pathological change, they alone provide the best clues to function in health and dysfunction in disease. Just as importantly, protein modifications including a plethora of post-translational changes are not evident upstream. As discussed later, proteins may require cleavage of a specific sequence to become biologically active, additional sidegroups may be added at specific amino acids including phosphorylation to propagate signal transduction or glycosylation to transport the molecule to a specific site. The analysis of hundreds if not thousands of proteins – proteomics – is therefore potentially the most illuminating of all multiplexed strategies.

Proteomics, however, is an imperfect science, and although its methods are rapidly evolving, it is important to acknowledge that all existing approaches have serious limitations. Notably, the available discovery tools offer poor precision, sensitivity, specificity and low throughput. An example of a low throughput proteomic platform is classic liquid chromatography separation of a single complex biological sample, coupled with electrospray ionization to generate of mass spectra. These limitations place severe constrains on study design (*e.g.,* small sample numbers) and they can lead investigators to place a disproportionate confidence in the data generated in a discovery setting. Regardless of these shortcomings, proteomic methods have been widely adopted and have generated many potential candidate markers. As we will illustrate, candidates identified using these techniques should be considered a set of ideas or leads from which the investigator can generate testable hypotheses. The study design and process of testing or validating these candidates is especially important if the test is to be applied in a clinical setting.

This chapter focuses on several key areas of the biomarker development process and uses rheumatoid arthritis as a case model of autoimmune disease for discussion. We first discuss the issue of clinical need, *i.e.,* how biomarkers *are currently* used in the practice of rheumatology what biomarkers might offer in *the future* in a clinical setting, then highlight some clinical scenarios describing considerations for study design. Because discovery methods have been reviewed elsewhere (Gibson & Rooney, 2007; Hu et al., 2006; Tilleman et al., 2005) our emphasis here is on describing and evaluating the targeted proteomic methods that are essential to candidate validation. It is the authors' belief that the task of validation has not received adequate attention. We are overwhelmed with discovery studies and data arising from them; validation approaches are infrequently reviewed and relatively little validation data have been reported. The significance of major barriers to their widespread clinical adoption is also briefly discussed. These challenges to new marker development and commercialization include factors such as study design, pre-analytical variables, data interpretation, bioinformatics, validation, ethics and commerce. While our focus is on protein biomarkers and proteomic methods in rheumatology, the principles we discuss are generally applicable to other analytes and areas.
