Personalised medicine and companion diagnostic device development
Personalised medicine is a new field that has often been defined as ‘the right treatment for the right person at the right time’. The essence of the concept is accounting for the differences in an individual patient’s genetic profile in a way that anticipates and overcomes potential complications and adverse drug reactions. Companion diagnostics is a key attribute of personalised medicine and is the development of diagnostic tests whereby molecular assays that measure levels of proteins, genes or specific mutations are used to provide information for a therapy specific to an individual’s condition.
This article in the International Pharmaceutical Industry Journal by Sagentia's Nick Rollings discusses this topic in some detail.
Healthcare in the developed world is on the edge of a precipice; populations are aging and healthcare costs are rising against the backdrop of uncertain economic times. Consulting the latest OECD figures shows the cost of healthcare as a percentage of GDP rising from 13.6% to 17.4% for the period 1999 to 2009 in the USA alone. This has placed increasing pressure on healthcare systems that already appeared unsustainable in the long run. Whilst proposed measures to deal with the situation continue to polarise political viewpoints, no one denies that there is a move underway from payment for procedure — the traditional reimbursement paradigm — toward outcome-based reimbursement or “payment by performance”.
In parallel to this, the pharma industry itself is undergoing a period of intense change. The ‘patent cliff’ is a major industry issue and is coupled with the fact that most firms do not have enough new products in their clinical pipelines to replenish the revenues they are poised to lose and the everincreasing cost of drug development.
There is also the issue of increased regulatory scrutiny over the safety of marketed drugs that contributes to rising costs. Cancer is a case in point. On average, the top 15 selling oncology drugs have a 35% positive response in patients, which means that 65% of cancer patients suffered from toxicity and wasted time on futile therapy. Moreover, given the cost of these drugs was $26.4 billion, the 65% failure rate implies that $17 billion of previous healthcare dollars were wasted.
These changes in the pharma industry combined with the pressures and demands on healthcare providers have led many observers to suggest that we are now heading towards a new era referred to as ‘personalised medicine’.
This new field has often been defined as ‘the right treatment for the right person at the right time’. The essence of the concept is accounting for the differences in an individual patient’s genetic profile in a way that anticipates and overcomes potential complications and adverse drug reactions.
That all of this is possible is down to the tremendous progress made in the field of molecular biology since the discovery of the DNA double helix structure by Watson and Crick in 1953. The human genome project is a case in point; large collaborative projects such as this have laid the groundwork for understanding the role of genes in the development of human physiology and the existence of single nucleotide polymorphisms (SNPs) that account for some of the genetic variability between individuals. Understanding the DNA genome as a genetic information archive allows researchers to understand the protein signalling pathways that result from each of the genes. This is important given that the functional aspects of cells are controlled through proteins, not genes.
This has subsequently led to the rise of the field of proteomics; the comprehensive analysis and characterisation of all of the proteins and protein isoforms encoded by the human genome. Studying proteins allows for the development and application of protein biomarkers across a wide variety of disease areas.
Biomarkers can be considered molecular sensors of health status, providing clinically relevant information about the health status of each individual. This allows the determination of those patients that will safely and effectively respond to a given drug, thus effectively allowing personalised treatment. Breast cancer is a relevant example of this; the HER2 test identifies the 15-20% of breast cancer patients who may benefit from treatment with Herceptin. When it is considered that the cost of treatment can vary from $50,000 - $80,000, not only does the patient benefit but it also facilitates the efficient outlay of healthcare budgets.
As well as indicating the state of a patient’s health, proteins from diseased tissue are also found in the bloodstream, which allows clinicians to measure protein biomarkers via the use of a diagnostic test with a blood sample. This is highly preferable to the alternative of obtaining a biopsy of the disease tissue itself.
As demonstrated above, a key attribute of personalised medicine is the development of diagnostic tests — so called ‘companion diagnostics’— whereby molecular assays that measure levels of proteins, genes or specific mutations are used to provide information for a therapy specific to an individual’s condition.
The benefits of molecular diagnostic tests are well understood. These include the ability to carry out risk assessments on a patient to identify if a condition has the potential to deteriorate further. Early detection of a condition is also possible, thus allowing earlier intervention, especially relevant for cancers and other highly progressive diseases. Definitive diagnosis is another distinct advantage of diagnostic tests, as the human body is limited in the number of symptoms it can display, making definitive diagnosis challenging. Once a condition has been accurately diagnosed this information can be used to guide treatment decisions and create prevention strategies. This can help a physician decide whether to use a particular drug on a patient-bypatient basis.
It is true that many diagnostic tests come with a high price-tag, however if that high initial cost can offset the higher treatment costs as exemplified with cancer above then this cost can be justified.
Most molecular diagnostics tests are clinical laboratory tests that provide useful information to the physician regarding the use of a therapeutic product, but they are not a determining factor in the safe and effective use of the therapeutic product. Companion diagnostic tests differ as they are required before the administration of a drug and the information they provide is essential to the safe and effective use of the corresponding therapeutic.
This scenario now raises significant concerns about the safety and effectiveness of the test and any device used to perform that test. These safety concerns also have an influence on the development of any therapeutic product, as the companion diagnostic test will require regulatory clearance and will need to be marketed at the same time the therapeutic is available.
To address this issue and the relevant safety concerns, the FDA published a guidance document in July of this year to bring together current advice and best practice in this new and developing field. One of the key messages behind the document is that the therapeutic product depends on the use of the respective diagnostic test and that a lack of test approval effectively equals lack of therapeutic product approval. This implies that the diagnostic test must be developed in parallel with the drug and ideally validated at the same Phase III clinical trial. Combined validation is critical in establishing the safe use of the drug based on the companion diagnostic test result.
This is clearly a fundamental shift away from conventional therapeutic product development with the diagnostic test now equally as important as the drug. This new approach provides challenges for the pharma industry, where a competency in companion diagnostics, including understanding diagnostic technologies, launching diagnostic products and submitting regulatory 510k and PMA applications, all become contributory factors in the success of certain therapeutic products.
This is especially relevant as the FDA is now actively encouraging the development of therapeutic products that depend on the use of approved or cleared IVD companion diagnostic devices when an appropriate scientific rationale supports such an approach.
Whilst this is non-trivial, it is possible to turn these challenges into opportunities with the correct approach.
Diagnostic system development
Due to the increased interdependence of the diagnostic test and the therapeutic product, it becomes immediately clear that each one will heavily influence the other during the course of a development process. So just how would one go about approaching the daunting task of integrating these different aspects?
The starting point of any therapeutic product is the medical condition that is to be dealt with. This then drives the mode of intervention and therefore the relevant diagnostic technology and the associated device that the tests will be carried out on. The test scenario will also require a level of consideration. For example; is this a diagnostic test carried out in a central hospital lab or next to the patient at the point of care? Whilst a similar process is carried out in both scenarios, the manner in which the patient sample is collected, prepared, analysed and the information presented is very different. Having a clear definition of not only the test requirements, but also the information required from the test, the usage scenario and the potential user profile, are all prerequisites before moving forward.
In the development of any diagnostic test and associated device there may be a number of collaborators involved, often from different disciplines and with different drivers that can sometimes conflict. These drivers can be from both a commercial and technical perspective. Understanding what drives each stakeholder and their expectations right at the beginning of a project is essential.
Once these are understood it is possible to put together a top level programme work plan. At this stage of development it is helpful to consider the therapeutic and the diagnostic test as an inter-related system. This allows the interactions that occur between each aspect of the system to be mapped out and a managed development path established that pays due regard to the key variables in that system. This is particularly important as the key variables may change during the course of development, particularly as biochemistry processes mature or as separate work streams operated by other development teams progress, generating further data.
Another major aspect to be addressed early on in the development of any companion diagnostic is the detection technology that will be required. This can depend on a great number of factors, including the sample collection and preparation steps required and the sensitivity and degree of multiplexing required. For example, this would be the case if a pattern of biomarkers is to be analysed for certain conditions.
There are three drivers here; firstly the analytical ability of the technology, i.e. its ability to measure the required target accurately; secondly the clinical validity, i.e. the ability to distinguish between patients that will respond to a drug and those that will not; and thirdly the functionality - determining if every feature of a technology is required in a particular application. These characteristics of a technology should be established well in advance of a clinical trial. These aspects can also be influenced by the handling of the sample at every stage from collection through to analysis, and so developing proof of principle prototypes and technology evaluation hardware can be a useful tool to understand the performance of proposed technologies with the relevant tests. Quantifying these characteristics correctly is especially important when inadequate performance of the device could have severe therapeutic consequences.
Clearly there are many different technologies that can be used for different scenarios in different ways. Assessing the available options and their suitability for a certain application requires an understanding of the core science behind the therapeutic and the companion diagnostic test whilst still being able to envisage the final device.
Intellectual property landscape
It is not possible to discuss diagnostic technologies without paying due regard to the intellectual property landscape. Entering into a new area can be a potential minefield for anyone not particularly familiar with the major players, the key technologies and their various embodiments.
Specialist knowledge in this area is required when one considers the mature molecular diagnostics sector and the established patent portfolios of the major players in this space, compared to the new and emerging challenges and opportunities in the companion diagnostics field.
In addition to technical requirements and understanding the patent landscape, the other substantial part of developing the diagnostic test relates to the regulatory pathway that will be followed. There are numerous options available at this stage and each needs to be fully considered.
The first option is the pre-market approval (PMA) route for an entirely new test and device. This is particularly suitable for a new area with high unmet clinical needs, but carries an increased degree of risk due to concerns the diagnostic test might not get initial clearance.
The other option for the diagnostic device may be the 510(k) route where the companion diagnostic device demonstrates ‘substantial equivalence’ to an existing device. This implies a lesser degree of regulatory risk, but may impose constraints on the diagnostic test that could cause problems during the development cycle, particularly relevant if a new application for the therapeutic product is being considered. In both cases the creation of a design history file (DHF) documenting the development of both the test and the device is an important part of any regulatory submission. The correct path will differ for each application and depends on the risk involved and the existing situation, therefore careful consideration of the different factors is required before deciding on the best route.
Another area that should not be overlooked is the scenario in which a test or indeed any medical device is used. Although medical staff are highly trained, they may not all be experts in the area of diagnostic testing, and so paying attention to human factors and the workflow required of the user from sample collection to display of the test result is imperative. This is especially relevant when the test could dictate the application of a powerful treatment that could cause harmful side-effects. Data visualisation and interpretation are of particular relevance here to ensure that the test result is not open to misinterpretation.
Pulling together all of the factors described above is no small challenge, and requires the application of many different specialists during concept generation, initial research, feasibility testing, detailed design and prototyping, before transfer to manufacture of both the therapeutic and the diagnostic device. This should also be underpinned by a rigorous quality management system such as ISO 13485 to satisfy regulatory requirements.
Despite this, going forward it is clear that companion diagnostics are set to become an increasingly important part of the future. As research drives a greater understanding of disease and the underlying molecular pathways there will be a need to exploit this knowledge, particularly if there are additional benefits such as reduced patient suffering and economic advantages in the targeted applications of drugs.
In this new future it is clear that very few, if any, current organisations have all the required expertise and resources in-house, and that the creation of multidisciplinary partnerships will almost certainly be required. Pharmaceutical companies may need to engage with numerous diagnostic companies, as it is likely that no single technology will be suitable for all of a particular company’s varied applications. They will also need to be flexible and identify the most suitable diagnostic for a particular application at a particular time. This need for flexibility in a fast moving environment is especially important as parallel developments may influence each other in unforeseen ways.
Identifying and establishing competency in relevant diagnostic technologies is therefore an important first step. Ideally this can then be followed by the rapid development of a diagnostic device to maximise the patent protection period a therapeutic product has, and to get the drug and test onto the market as soon as possible. Before this can all happen, however, the right resources and/or partners need to be established. We believe pharma companies must get this right to achieve success in this new and exciting area of personalised medicine.Download the complete article