Abstract
IVD manufacturers are heavily reliant on novel IVD assays to fuel their growth and drive innovation within the industry. They represent a key part of the IVD industry's value proposition to customers and the healthcare industry in general, driving product differentiation, helping to create demand for new systems and generating incremental revenue. However, the discovery of novel biomarkers and their qualification for a specific clinical purpose is a high risk undertaking and the large, risky investments associated with doing this on a large scale are incompatible with IVD manufacturer's business models. This article describes the sources of novel IVD assays, the processes for discovering and qualifying novel assays and the reliance of IVD manufacturers on collaborations and in-licensing to source new IVD assays for their platforms.
Introduction
The last twenty years or so have seen the emergence of powerful new technologies such as genomics, proteomics, metabonomics and systems biology in biomedical research. These technologies, coupled with advances in computing, biostatistics and bioinformatics have lead to major advances in our understanding of the complex biological mechanisms underlying human diseases. In addition, progress in genetics and genetic profiling have given us new insights into the biological foundation of inter-individual differences and their impact on how different patients respond differently to the same medicines. These technologies are still being refined, but are already highly robust and reproducible and have enabled us, in clinical research settings, to measure hundreds, and in the case of SNPs, hundreds of thousands of biomarkers quickly and, in the context of drug development programs, at an affordable cost.
Novel biomarkers and IVD assays essential for drug discovery, development and personalized healthcare
One significant result of these advances is the identification of numerous biomarkers that are posited to have the potential to improve patient healthcare. Biomarkers are essential to the development of Personalised Healthcare (PHC). The aim of PHC is to identify those patients most likely to benefit from a particular therapy or to exclude those who are unlikely to respond to a medicine or therapy and therefore avoid unnecessary exposure to potential side effects, thereby increasing treatment effectiveness overall. It is our ability to correlate biomarkers to both desired (efficacy) and undesired (toxicity or adverse effects) responses to medicines that forms the basis for advances in PHC.
Novel biomarkers, many of which are “soluble” biomarkers measured using in-vitro assays, are very important tools in the drug development process, enabling earlier and in some cases better decisions about which molecules to progress, informing dose finding and dose scheduling and aiding in the design of more effective and efficient clinical trials. A small subset of these biomarkers will likely prove useful as tools for patient selection, in the way that the Her-2 assay is used to select which breast cancer patients should be treated with trastuzumab, or, in a more recent example, the way that the k-ras mutation assay is used to select which colorectal cancer patients should be treated with cetuximab and panitumumab. If these biomarkers are a required element of the treatment algorithm and are included in the labelling of the medicine, they will per-se become companion diagnostics.
Companion diagnostics are an important consideration for pharmaceutical companies
Pharmaceutical companies are naturally concerned that access to therapies not be inhibited by availability of the companion diagnostic and will therefore generally strive to ensure that the companion diagnostic is broadly available in all markets where the medicine is licensed. The best way to ensure this broad availability is to develop the companion diagnostic as an IVD in collaboration with one or more IVD manufacturers, preferably one with broad market presence.
Novel IVD assays are key for the IVD industry
Novel IVD Assays are also essential to drive growth and fuel innovation for in-vitro (IVD) manufacturers. They are a key part of the IVD industry's value proposition to customers and the healthcare industry in general and drive product differentiation, helping to create demand for new systems with new features and generating incremental revenue. They are also very important for the image of IVD manufacturers, as they evidence the positive impact that the IVD industry has on healthcare as a whole and provide benefits that go well beyond the IVD industry's core base of customers, clinical testing laboratories.
The discovery and qualification of novel IVD assays
The discovery, qualification and development of novel IVD assays is an extremely laborious, long-term and risky process. There are a number of companies and organizations for which this undertaking is a core part of their mission, but to date no company, academic group or organization has been able to create a strong pipeline of novel IVD assays from discovery through to commercialization. This is, in part, due to the wide range of skills and capabilities which it is necessary to master in order to do this and partly due to costs involved in constructing a portfolio of activities and projects necessary to be able to manage the risks inherent in this undertaking.
The discovery and qualification challenge
Organizations setting out to pursue the systematic discovery and qualification of novel clinical biomarkers generally assemble several sets of resources and capabilities. A systematic approach may not be essential, but will likely ameliorate many of the risks involved in this process. The steps below are listed in a logical sequence, but individual groups or organizations have been able to discover and qualify novel biomarkers without having followed all of these steps or undertaken them in the order listed.
Clear identification of an unmet need
If one is to set out to systematically try to discover relevant in-vitro biomarkers for a specific clinical purpose, it makes sense to have a clear understanding of the unmet clinical need. This allows for a structured survey of current clinical algorithms, estimations of the value that a novel marker could create, the creation of a business case, if required, development of target specifications for the desired performance of the marker and the construction of an appropriate “discovery platform”. In some cases, the unmet need may be an opportunity to improve the safety/benefit ratio of a particular therapy through stratification (the basic value proposition of PHC) and the value generated will likely be due to a mix of improved healthcare outcomes and improved use of healthcare resources. In other cases, with markers for screening for diseases such as colorectal cancer, the value generated will mostly be due to improved healthcare outcomes. Lastly there will also be cases where a marker is able to e.g. replace a more expensive technology, such as in-vivo imaging, and the value generated will be largely based on cost savings to the healthcare system.
Development of a discovery hypothesis
Once a need has been identified it is important to develop a scientific and technical approach to find biomarkers which address that need. This will generally take the form of a series of experiments which set out to define and test a hypothesis e.g. that a novel marker for the differential diagnosis of acute myocardial infarction can be found by proteomic approaches.
Collection of appropriate samples for both the discovery and the qualification phase
An essential part of any biomarker discovery effort is the collection of appropriately clinically annotated samples. Usually the samples will be collected from the clinical setting in which the unmet need has been identified, such that the discovery and qualification procedure produces biomarkers which will prove useful in this clinical setting. A great deal of care should be taken in creating the right collection of discovery and qualification samples, using standard operating procedures (SOPs) to ensure that these samples are collected and stored in such a way as not to introduce confounding factors which will interfere with biomarker measurements.
The qualification collection should consist of samples which numerically “recreate” the clinical setting toward which the biomarker is targeted i.e. if the biomarker is going to be used to screen for cancer X, the qualification collection should mirror the screening population in this setting. It is particularly important that the number of samples from patients who are known positives for the cancer in question are correctly proportionally represented within the population otherwise the positive predictive value of the biomarker will be overestimated as the prior probability of encountering the analyte in question in a patient sample will be incorrect. Similarly, the qualification collection should also numerically recreate the prevalence of other diseases and conditions in the screening population, as these diseases and conditions are likely to contribute to the presence of confounding factors, or may indeed contribute to changed levels of the biomarker being investigated.
Development and validation of an appropriate discovery platform
One of the key tools in any marker discovery effort is a discovery platform. The nature of the platform will vary depending on whether a genomics, proteomics, metabonomics or other approach is used, but will generally be capable of measuring many hundreds, and possibly thousands of analytes in parallel, allowing researchers to cast a very wide net. However, these discovery platforms are currently not suitable for use as IVD platforms. In clinical laboratories a single patient specific determination generally yields a clinical diagnostic result and the platform must be robust, reliable and relatively simple to operate.
The discovery platform must be adequately validated, with measures in place to minimize the effects of artefacts, false discovery rates etc. and yield sufficiently promising candidates for the qualification step.
Development of an appropriate assay prototype
Once a biomarker has been adequately qualified for its posited role as an IVD, a prototype assay which is compatible with the future clinical platform needs to be developed. If the biomarker in question is a protein species or a species against which antibodies can effectively be raised this will likely be an immunochemistry assay. If the biomarker is a nucleic acid species the prototype assay will likely be real-time PCR assay (this assuming that the biomarker is a soluble species to be assayed in serum, plasma or another body fluid). This prototype assay can then be tested against the qualification panel which was used to qualify the biomarker in the previous step. Alternatively, biomarker qualification can be performed using the prototype assay.
The ultimate qualification of a biomarker for clinical use may require testing in a clinical trial, with its clinical utility being demonstrated in an intervention trial in comparison to a current clinical algorithm. This clinical trial will usually require use of an assay in its final development form (once it has been applied to the IVD platform) though it may be possible to run the trial using a late-stage prototype with a cross-over study validating the performance of the assay used in the trial with the final commercial IVD assay.
Application of the assay to an IVD platform
Once the biomarker in question has been adequately de-risked through the qualification process it can then be developed on a commercial IVD platform. The costs per marker of this step are usually high in relation to the development of the prototype and this development work will not be undertaken unless there is a high likelihood that the resulting novel IVD assay will attain regulatory approval and demonstrate sufficient clinical utility to warrant broad commercial uptake.
IVD manufacturer's business models incompatible with large, risky investments in biomarker discovery and qualification
The business model that has been adopted by the major companies making up the IVD industry (that is, companies engaged in the research, development, manufacture and marketing of IVD products) is one that is centred around a systems model. It encompasses the development and manufacture of highly automated, fully integrated systems comprising an instrument, software and on-board reagents, requiring limited user intervention.
Most of these companies currently market a broad portfolio of these systems, with the biggest companies being active in clinical chemistry, immunochemistry, molecular diagnostics, urinalysis and haematology. For each of these systems to compete effectively in the market they need to be regularly updated with additional features and capabilities, such as reagent and sample sparing capabilities, new quality assurance features and updated software packages. These systems also generally need to offer very broad test menus. Whenever a new or updated system is introduced, all of the tests in the reagent menu need to be adapted to the new platform. This may involve significant rework of testing parameters in order to ensure that the test runs optimally on the new platform.
The result of this is that large IVD manufacturers have to invest significant resources in maintaining their current product portfolios and most are able to invest only a small proportion of their total R&D budgets on the discovery and validation of novel IVD markers. In general, the steps of the discovery, qualification and development process described above which are compatible with the IVD manufacturer's business model are steps 5 and 6, those concerned with developing the prototype and final clinical reagents on a commercial IVD platform. The IVD manufacturer will also generally be best suited to create the necessary documents for an IVD regulatory submission, should that be required.
Sources of novel IVD assays
If IVD manufacturer's are only about to invest relatively small proportions of the R&D budgets into the discovery and validation of novel IVD assays, then where are these assays to come from? There are a number of possible sources:
Academic groups, particularly major groups such as the National Institutes of Health, the National Cancer Institute or the German Cancer Research Centre
Consortia, such as The Biomarker Consortium (www.biomarkersconsortium.org), The Predictive Safety Testing Consortium (www. c-path.org/pstc.cfm), the High Risk Plaque Initiative (www.hrpinitiative.com) and other public private partnerships.
Biotech companies focused on biomarker discovery and validation e.g. Celera, Genomic Health, Epigenomics, XDx, Agendia, Predictive Biosciences and others (a number of which were founded as spin outs from academic groups)
Pharma companies' efforts, related to drug discovery and development, particularly collaborations with pharma companies on companion diagnostics assays. As was explained earlier, pharma companies are investing increasing amounts in the discovery and validation of novel biomarkers and some of this investment will doubtless result in the development of novel IVD assays
IVD companies own marker discovery and validation efforts
Collaboration essential for IVD manufacturers to develop a pipeline of novel IVD assays
Given the broad range of possible sources of novel IVD assays and IVD manufacturers own limited investments in the discovery and development of novel assays, it is essential that IVD manufacturers can effectively interface with those who are discovering and developing these novel assays, be this through collaborations, licensing activities or other methods, if they are to secure access to sufficient novel assays. IVD manufacturers are also ideal collaboration partners for the members of the pharmaceutical industry, biotech companies and academic groups who are discovering and developing biomarkers with potential utility in the clinical diagnostics marketplace as IVD manufacturers have all of the capabilities necessary to apply these assays to existing IVD platforms and also control access to these platforms. It seems therefore very likely that the path to clinical use of novel IVD assays will involve collaboration between IVD manufacturers and those discovering and qualifying novel biomarkers with IVD potential.