Advances in clinical applications of tissue proteomics: opportunities and challenges

Abstract

Mass spectrometry (MS)-based proteomics assays provide unprecedented opportunities for determination of protein expression in tissues for clinical applications. Within the last few years a number of clinical assays have been established. These include multiplexed selected reaction monitoring based approaches to quantify protein targets of therapeutic agents in cancer tissues and bottom-up shotgun MS-based proteomics approaches to diagnose and classify protein deposition disorders such as amyloidosis. These examples highlight the vast potential of MS-based proteomics as analytical and descriptive tools for clinical diagnosis of human disease. However, a number of major challenges remain for broader application and acceptance of such technologies in routine clinical care.

Keywords::

• amyloidosis
• cancer
• immunoglobulin
• mass spectrometry
• proteomics
• transthyretin

Mass spectrometry (MS)-based proteomics assays provide unprecedented opportunities for determination of protein expression in tissues for clinical applications. The strengths of such assays include the ability to identify a comprehensive compendium of proteins expressed in human tissues, outstanding analytical specificity and accuracy, and depending on the preanalytical methods used, sensitivity similar to that achieved by immunoassays. Unlike immunoassays, MS-based proteomics does not require the development of high affinity antibodies specific for each protein epitope of interest and can be easily applied to a wide variety of matrices including fresh cells, frozen tissues and most importantly, formalin-fixed paraffin-embedded (FFPE) tissues, which represent the bulk of specimens collected for clinical diagnosis. For this reason, within the last 10 years, often supported and funded by national initiatives, MS-based proteomics has been extensively used for biomarker discovery in human liquid specimens, cell lines and tissues [1–5]. More recently, such methods have been applied successfully to FFPE clinical specimens [5,6]. Mann and colleagues using a combination of sophisticated preanalytical digestion and separation methods and post-analytical informatics tools have reported identification of over 10,000 proteins from FFPE colon cancer specimens, suggesting that virtually all proteins expressed in cancer cells, including those shown to be expressed at very low levels, can be analyzed and quantified in routine clinical specimens [6].

However, translation of such efforts into clinical practice has been problematic due to a number of reasons discussed below. Two best established clinical applications of MS-based tissue proteomics have taken two different approaches. Burrows and colleagues, following their preliminary elegant work on FFPE specimens [7], have taken the approach of selected reaction monitoring (SRM) assays to identify and quantify a number of therapeutic targets important in the management of breast cancer [8]. In this approach, laser microdissection of FFPE tumor tissue was used to ensure that the proteins analyzed are from the tumor but not from the surrounding benign tissue. SRM assays were developed in a stepwise approach starting from recombinant proteins to identify the tryptic peptides most suitable for MS analysis and validating these on cell lines and FFPE tumor specimens. The SRM assays showed excellent correlation for quantitation of the selected targets when compared to conventional immunoassays such as ELISA or immunohistochemistry. Although the clinical utility of such sophisticated quantitative protein assays for tissue proteomics has not been entirely established, the approach offers unparalleled opportunities for clinical applications as many SRM targets can be multiplexed in a single test, which could be performed on a single 10 micron section of the tumor. Integration of SRM assays to stratify patients into clinical trials targeting specific protein abnormalities will no doubt accelerate the adaptation of such strategies to clinical laboratory testing.

The other successful clinical application of MS-based proteomics has been in the field of amyloidosis. Amyloidosis is the clinical term used to define a number protein deposition disorders caused by abnormal deposition of proteins in extracellular tissues. These deposits cause tissue damage by a variety of mechanisms and organ dysfunction, in particular in the heart, kidneys and nerves. There are at least 30 different proteins that can cause amyloidosis, and by microscopic examination, they are indistinguishable from one another. The clinical management entirely depends on the type of the protein deposited, thus on the underlying pathogenesis [9]. For example, so-called AL amyloidosis is caused by deposition of clonal immunoglobulin light chains produced by an underlying systemic plasma cell neoplasm. The treatment targets the plasma cell neoplasm and often involves high-risk approaches such as high-dose chemotherapy and peripheral blood stem cell transplantation. In contrast, hereditary ATTR amyloidosis is caused by deposition of a mutated transthyretin protein and is treated by liver transplantation. Thus the stakes are high and accurate typing of amyloid deposits is critical for major management decisions. To address this important clinical problem, the author’s group utilized a semi-quantitative bottom-up shotgun MS-based proteomics approach to develop clinical tests to diagnose and classify systemic and localized amyloidosis [10]. The first test was developed to address the cases where a diagnosis of amyloidosis was established on FFPE biopsy material but the etiology was unclear. The method involved microdissection of minute fragments (60,000 µ² or the size of single glomeruli) of amyloid deposits from FFPE sections [11]. The fragments were disrupted by heat and sonication and digested into peptides using trypsin and the peptides were fingerprinted using nano-flow tandem MS and a customized informatics pipeline. The clinical cut-offs were established by spectral counting and the test was validated according to federal and state regulations and implemented for clinical diagnosis. In the validation cohort, the test showed 100% specificity and sensitivity, outperforming all other assays available (usually 40–80% specificity). The second test the author’s group developed was a screening assay for diagnosis and classification of amyloidosis using blindly obtained fresh abdominal fat aspirate specimens from patients suspected to have amyloidosis. The assay classified amyloidosis with 88% sensitivity and 96% specificity, well-beyond the analytical requirements for a screening clinical test [12]. Since the implementation of the tests as clinical platforms in 2008–2009, both assays have been extensively used for clinical management of thousands of patients with amyloidosis and performed within the analytical parameters established by the validation studies. Unlike the immunoassays where the diagnosis depends on the availability of specific reagents and knowledge of possible clinical targets, MS-based proteomics assays do not require prior knowledge or hypothesis and simply establish the protein content of the amyloid plaque. Using these tests, we were not only able to provide accurate diagnostic information, but were also able to highlight the presence of pathogenic proteins variants, including hereditary mutations responsible for the clinical phenotypes [13,14]. The author’s group was also able to identify new causes of amyloidosis including an iatrogenic amyloidosis caused by deposition of a peptide drug [15], a new hereditary amyloidosis caused by deposition of serum protein beta-2-microglobulin [16] and a type of systemic amyloidosis caused by deposition of a chemokine, LECT2 [17,18].

The above examples highlight the vast potential of MS-based proteomics as analytical and descriptive tools in clinical diagnosis of human disease. Although the technology has been widely used in the research setting to identify biomarkers of human disease, clinical adaptation has significantly lagged. The reasons for this are multiple and include the lack of understanding of clinical test development requirements by researchers, the lack of know-how, expertise and infrastructure in clinical laboratories, significant upfront costs to establish the technology in the clinical setting, the lack of robust instrumentation and systems engineered and designed for clinical use, the lack of standardized informatics pipelines for data analysis, the complexity of regulatory requirements for implementation of clinical MS-based proteomics assays and the lack of financial structures for reimbursement of such tests. For MS-based proteomics assays to fulfill the clinical promise beyond a few niche applications described above, close cooperation with researchers, clinical laboratories, industry, professional organizations, regulatory agencies and health care service providers will be required.

Financial & competing interests disclosure

The author has no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.