Year: 2016
Type: article
Abstract: Clinical proteomics has come a long way and is at a point where the expertise, technology, assays, and knowledge have matured sufficiently to begin enabling personalized and precision medicine. Although traditional laboratory medicine seems to be stuck in a bit of a rut, the field is now starting to take advantage of some of the approaches needed to monitor a person's dynamic state of health or disease. Unfortunately, these quantitative approaches have often been based on a “static” diagnostic model that focuses on a few basic protein changes describing what has happened, or genomic markers that might describe what possibly could happen in the future. Static markers are biomarkers that define a single disease entity or are used for a single clinical diagnosis are often limited; we can do better. Proteomics has the potential to determine the “dynamics” of an individual's health and this is the level of understanding we need to move toward. This requires an intimate knowledge and precise quantification of the important modified forms of proteins (proteoforms) within a proteome, and the recognition that proteoforms are dynamic and important to an individual's phenotype and crucial in determining the immediate present and potential future response to therapies. If we don't fully characterize and understand what we are measuring we could misdiagnose or lose an opportunity to have a biomarker(s) with a better selectivity toward a particular disease. This is also true for companion diagnostics. Unless the assays can capture the mechanism of the drug, which occurs most often at the proteome level, the assays could fail to guide good drug development and cause late stage failures of promising therapeutics, an increasingly acute problem for big pharmaceutical companies. We need to more fully map the proteins we are measuring and take advantage of their modified forms in our clinical assays. Mass spectrometry (MS)3 can detect an intact protein or a subset of composite or surrogate peptides, which provides more coverage of isoforms or posttranslational modifications than could be observed or quantified by using traditional immunoassays. The sheer number of different protein forms, many of which are disease state or individual-specific, provides an opportunity for more specific molecular phenotyping, or a form of “proteotyping.” Proteotyping was coined in 2003 (1), but here we use it to encompass the concept of determining the proteome that is unique to an individual's phenotype. We propose that the proteotyping of individuals should provide the granularity and specificity required for precision medicine. It is clear that to precisely target therapies, we also need to have a greater understanding of the downstream disease reporter proteins and the modified enzymatic processing reporters to avoid impacting false targets or missing important ones. This was illustrated in a Nature publication (2) that showed that brain natriuretic peptide (BNP)/protein kinase G signaling was regulated differently than nitric oxide/protein kinase G pathways and demonstrated that failure to understand the significance of personalized pathways around BNP and protein kinase G as drug-able targets had led to late-stage issues in expensive drug-development programs. Circulating BNP signaling is complicated. In the heart, preproBNP undergoes intracellular cleavage and then can undergo a chain of cleavage products in the blood (3, 4). It is important to develop assays that are “true” to the biology and quantify either the fragments directly or the specific proteolytic activity responsible for their formation. MS is able to do this quantification in a way that traditional immunoassays find extremely challenging or impossible. This ability to quantitatively understand the components that constitute an individual's proteotype has other benefits. It also offers us the ability to move away from “single snap shot” measurements against what is “normal” for a population. It is well understood that the granularity offered by traditional diagnostics is a blunt tool with an inability to recognize small changes in a protein that might get lost in the morass of “normal ranges,” but for that individual are a vital clue to their health. Having the ability to look at multiple “small” biological/pathological changes becomes a multiplier of diagnostic power. The ability to understand, at a genetic level, an individual's potential for disease, in addition to understanding what is actually happening with a person's proteotype in real time, will change how we look at disease, wellness, and health. This is especially true if the analysis is continuous and the comparison is tracked against that individual's own profile rather than some population-based reference interval (5). This should, hopefully, move biomarkers from being disease-centric (Diseasecare) to encompassing wellness and health (Healthcare). MS is an ideal tool to open up this new paradigm, but for this to happen, instrument manufacturers must develop a strategy, both in terms of instrumentation and analyte content, to become an important detection technology in the clinical laboratory. Improved remote and self-directed sampling technologies, for those diagnostics based on peptides and enrichment approaches, and improved QC methods that are MS centric, will drive the uptake of protein-based biomarkers, which are most effective when multiplexed or need to measure more than just a single proteoform. In addition, improved sample processing, automation, and postaccessioning will be required to manage this process. Liquid chromatography and MS manufacturers should focus on improvement of instrument sensitivity, total speed to result, robustness, ease of use, reduction of “downtime,” software, and their own manufacturing costs, but also invest in biomarkers, which are best quantified by MS. The additional challenge for anyone introducing a new platform is analyte content. Fortunately, MS is overflowing with content, however much of it is restricted at the research-use-only level or has not had its clinical utility defined. MS is ideal to take advantage of customer/manufacturer partnerships that can deliver content quickly. Fortunately, MS offers, within the same instrument, the ability to capture deep proteomic detail including posttranslational modifications. MS can also be used for both targeted methods and broader discovery-level analysis offered by data-dependent or independent acquisition methods. This opens up the concept of proteomics creating “big data” on large numbers of individuals. The final piece in this puzzle is how to make sampling easier, and ensure that the information delivered is useful, actionable, and valuable for physicians, patients, and consumers alike. MS-based protein measurement has been perceived to be difficult to do well and owned by experts in academics or in MS-based reference laboratories (6). There are real challenges in MS, for example, in ensuring consistent interlaboratory variation and quality of results. The procedures to control these issues have already been laid out for immunoassays, where the challenges are much greater, by adopting quality assurance and accreditation schemes. Another challenge is how regulatory authorities will safely and efficiently manage the introduction of innovation that encompasses such a depth of information and the opportunity to truly multiplex tests. The good news is that MS is a more transparent technology since the measurement of analytes is related to their empirical quantification. On the other hand, development of traditional immunoassay tests from discovery requires action earlier by diagnostic companies. High-quality antibody pairs are required. In some cases it takes multiple generations of antibody pair development to accurately test the analyte in complex clinical settings. A diagnostic company most often does the evolution of an antibody pair. As well, the initial immunoassays are often tested in a research-grade 96-well ELISA format and they must be adapted by one of the diagnostic companies to fit on commercial automated instruments. These often mean that the initial discoverers of the biomarker(s) need to interact with industry much earlier to a product with immunoassays than with MS. The route to developing clinical MS tests is more direct, linear, and transparent once the instrumentation is in place. This is an exciting time for clinical diagnostics. We now have the ability to quantify an individual's proteotype on a large scale. This should translate to more specific and granular biomarkers. The challenge for MS manufacturers is: who is going to figure out the ease of use, robustness, and automation that will unlock the opportunity? However, the game changer may not even be the analytical process but something as simple as an easy-to-use, mailable, and quality-controlled self-sampling device that allows remote monitoring that can easily be coupled to a proteomic-based sample preparation and MS platform. In this scenario, an MS pipeline will be needed in clinical laboratories to meet an unmet need in healthcare rather than diagnosis of a specific disease state. mass spectrometry brain natriuretic peptide. Thanks to Beckman Coulter, Sciex, and Thermo Fisher for instrumentation and Koen Raedschelders and Shenyan Zhang for know-how and work.
Source: Clinical Chemistry
Authors: Ian Wright, Jennifer E Van Eyk
Language: English
Subfield: Spectroscopy
Field: Chemistry
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2550136454