Challenges to personalised medicine
Rebecca Aris interviews Duncan McHale
Duncan McHale, Vice President of Global Exploratory Development at UCB Pharma, tells us about personalised medicines, the issues and challenges that their development faces, and what the future holds.
Duncan McHale has a long history in personalised medicines. He helped to set up the pharmacogenomics programme at Pfizer, where it orchestrated biomarker technologies to predict drug responses. He then headed up the clinical personalised medicines department at AstraZeneca for four years before moving to UCB. He is also the ABPI stratified medicines lead and was previously the EFPIA pharmacogenomics lead.
He speaks with pharmaphorum on the current issues and challenges surrounding personalised medicine.
RA: What are personalised medicines?
DM: ‘Personalised medicines’ (also known as stratified medicines) is a very broad descriptor for a range of medicines targeting sub-populations of patients with the same disease. It is generally used to describe the application of clinical or diagnostic tools to identify patients who are likely to respond to a particular treatment and therefore achieve superior outcomes in the management of a patient’s disease (maximise the benefit or minimise the risk).
It’s important to point out that it encompasses much more than the application of genetic testing, which it is commonly thought to describe. In fact, one has to look at a whole range of tools including clinical phenotypes, histopathology, biochemistry, molecular tests and how one might use all of these results in an integrated fashion to understand its full scope and potential.
RA: What do you see as the key challenges facing personalised medicines?
DM: For many disease areas the key challenge to the advancement of personalised medicines is the need for a greater understanding of the basic science of a disease, this will help us to more accurately predict drug response. Disease classifications will need to be deeper, more diverse and based around molecular aetiology rather than phenotypic features (currently the normal classification of diseases does not represent its aetiology or pathology).
“…the key challenge to the advancement of personalised medicines is the need for a greater understanding of the basic science of a disease…”
Another key challenge that personalised medicines face is the regulatory process, which currently does not facilitate their development. The evidence base needed for diagnostic tests is considerable and is required to be available at the same time as the treatment evidence base. When there is a clear stratification biomarker hypothesis to be tested, based on preclinical testing, then this is possible. However, for most programs this is not available and the stratification has to be identified during the phase II program. This makes it very difficult to have a prototype diagnostic test available for the start of phase III. The current regulatory paradigm in the US requires prospective stratification as part of the pivotal trial so a failure to have this at the start of phase III means that additional pivotal large scale studies would be needed to get the diagnostic test approved. This is a disincentive to the development of companion diagnostic tests after the phase III program has been run.
Funding for research for personalised medicines can also be a challenge, let me explain further. For a pharmaceutical company the price of a new medicine is negotiated at the time of launch. By their nature, companion diagnostics exclude a proportion of patients from receiving the medicine as they would get no benefit. So through personalised medicine the overall value the drug brings to the population remains the same or increases as the excluded subjects were getting little benefit and potentially some harm. However, the revenue to the company decreases as fewer subjects are taking the new medicine. As the price cannot be increased the company loses revenue despite increasing value. This is a disincentive for research into personalised medicines after a new medicine is launched and the price already set. Whilst there is still an incentive for the diagnostic company, as they would get increased revenue, it’s worth noting that the cost of having to run prospective clinical trials and the current reimbursement levels for diagnostics also create a disincentive for a diagnostic company to invest in this research.
A final challenge is getting the diagnostic test(s) incorporated into everyday clinical practice. Implementation in clinics of diagnostic tests that are sometimes different and more complex than existing practice will be difficult as some of these tests will be outside much of traditional clinical training and practice. So there is a significant challenge to be addressed in changing medical practice and training to take on board what a new and more detailed approach to classifying disease and mapping personalised treatment.
“Another key challenge that personalised medicines face is the regulatory process…”
RA: Which disease areas will benefit from personalised medicines in the foreseeable future?
DM: The one area that we can see starting to benefit from personalised medicines is oncology. Implementation is also quick in this area, due to the fact that the evidence base is quite strong and to the fact oncologists are well versed with complex testing and prescribing regimens.
Another area that looks set to benefit is neuro-degeneration. With the next wave of treatments for neurodegenerative disease (e.g. Alzheimer’s or Parkinson’s disease) focussing on disease modification rather than symptomatic treatment, it is likely that effective medication will need to start early and ideally before symptoms arise. The implication of this is that we may need to look at conditions such as mild cognitive impairment and stratify this “condition” into those who will go on to develop Alzheimer’s disease and those who will not. This will drive the development of diagnostic tools in a way that has not yet happened for other chronic degenerative conditions.
In general, the trend in the development of personalised medicines is around diseases where we can look to prevent or modulate rather than staying with the approach of symptom management you see in diseases such as diabetes or asthma.
RA: What is the role of diagnostics within personalised medicine?
DM: There is a concern around the current model for the development, implementation and reimbursement of diagnostic tests.
Currently, mature diagnostic companies and companies looking at the emerging technologies for personalised medicines are often separate. Companion diagnostics are generally a small part of most established diagnostic companies. This split means there are few companies that have the scientific, regulatory and commercial experience to develop companion diagnostic tests on novel platforms e.g. next generation sequencing.
However, we can see a trend of ‘biomarker companies’ increasingly looking to create diagnostics, and ‘diagnostic companies’ buying up early technology companies. This is a trend we can only hope to see increase and with that the momentum that will drive increased personalisation, as it will bring together the technology platforms, which may be required for more complex diagnostics with companies with experience in the development and commercialisation of diagnostics tests.
“This will drive the development of diagnostic tools in a way that has not yet happened for other chronic degenerative conditions.”
RA: What is the future long-term outlook for personalised medicines?
DM: In ten years’ time, I think we will lose the concept of the companion diagnostic test and see more tests delineating the molecular pathology of diseases. Disease classifications will be deeper and more diverse and based around molecular aetiology rather than phenotypic features.
The clinical sector will not be able to cope with the large numbers of biomarker platforms becoming diagnostic platforms, and so we are likely to see a small number of platforms being implemented and standardised to limit capital demands on reference laboratories.
Decisions will need to be made regarding reimbursement of diagnostic tests. This is currently not based on the value the test brings to the patient but on the cost of consumables. A value-based pricing approach would provide a strong stimulus to more diagnostic research being performed. The discussion around funding should not focus on the division of the current drug budgets between the biopharmaceutical and diagnostic companies but about the funding of healthcare generally and the value the diagnostic tests and therapies bring to patients and society.
About the interviewee:
Duncan McHale is Vice President, Global Exploratory Development, UCB.
UCB, Brussels, Belgium (www.ucb.com) is a global biopharmaceutical company focused on the discovery and development of innovative medicines and solutions to transform the lives of people living with severe diseases of the immune system or of the central nervous system. With more than 8,000 people in about 40 countries, the company generated revenue of EUR 3.2 billion in 2011. UCB is listed on Euronext Brussels (symbol: UCB).
How can we develop disease understanding to drive personalised medicine?