Medicine gets personalised
Matching patients to treatments by screening their genetic makeup is the goal of some drugs companies. Andrew Scott explores the political, economic and scientific issues
- Pharmaceutical companies are turning to genetics to try to streamline healthcare and drug discovery processes by tailoring treatments to individual patients
- Ethical concerns have been raised about whether this approach will only benefit those diseases that drug companies can profit from
- 'Orphan diseases' that aren't seen as worth exploring by drug companies might increase in number if this approach is followed
We are all different. Our genes generate a variety of metabolic profiles and illnesses affect us in diverse ways. Yet traditional pharmacology has adopted a 'one disease, one drug' approach, with little attention paid to complex individual variations. The biggest challenge, and possibly also the biggest opportunity, facing the pharmaceutical industry today is to customise treatments more specifically to suit different categories of patient. The insights from the human genome project, combined with ever-increasing understanding of how genetic variability and metabolic individuality are linked, is opening the doors to a new era of personalised medicine (PM).
The other common labels for this emerging field are pharmacogenomics and pharamacogenetics - combining genetic insights with pharmaceutical knowledge. The dream is that each patient will be analysed to detect which variant of a disease they are suffering from and which therapies would work best for them. They will then be treated by carefully selecting the appropriate drugs from a wide battery of options, each targeted at a very specific subset of disease. The efficacy of treatment will be dramatically improved, and the risk of undesirable side effects greatly reduced.
Undesirable consequences of the traditional unfocused approach to pharmacology, ranging from minor irritation to death, are easy to find. Throughout the world, hundreds of thousands of patients die each year due to adverse drug reactions, and many more are hospitalised but eventually recover. This reveals one clear economic benefit, in health care costs alone, that could be gained from a more personalised approach to medicine.
Anti-depressant drugs are notorious in that they achieve successful therapy in only 30 per cent of patients, so the majority of users are exposed to the risk of side effects that, in their case, may be the only effects they experience. That is just one example of the generality admitted by Allen Roses, a vice-president of GlaxoSmithKline, with his well-publicised remark that 'the vast majority of drugs - more than 90 per cent - only work in 30 or 50 per cent of the people'.
Investigating economics and politics
The drive towards PM seems inevitable, but economics will play a crucial role in how it develops. And with economics comes politics. The economic and political possibilities are complex, because PM can appear to bring both increased costs in some respects, and considerable savings in others.
Making more drugs for smaller more specific populations might be expected to increase costs, losing the economic advantages of scale and meaning each disease must be attacked on multiple research fronts rather than just one or a few. But a medical strategy that is genuinely more effective, helping more people faster, and with fewer side effects, could bring big savings.
The key political and ethical issues are not whether PM might be good for drug companies, but whether it will be good for people suffering from disease and for society as a whole. Some prominent scientists have already voiced concerns.
William Haseltine, former head of Human Genome Sciences in Rockville, Maryland, US, has warned of the dangers of drug companies prioritising patients with a genetic makeup that matches the drugs they want to sell. Will treatment be increasingly focused on only those diseases that happen to be profitable for the drug industry?
An investigation into the potential for PM is currently being undertaken by the UK's Royal Society, and the results should be available in summer 2005. 'This study will look at whether pharmacogenetics, the designing of drug treatments based on a person's genetic makeup, is a scientifically achievable aim,' says David Weatherall, chair of the working group that is conducting the study. 'Equally importantly it will look at whether healthcare systems in the UK and elsewhere have the resources to implement such technologies and what the pharmaceutical industry's assessment is of the significant investment needed to try to develop them in the first place.'
Clearly, drug companies are only going to invest in PM if it makes good business sense. Analyses performed by several prominent consultancy firms suggest there might be a strong business case. One of the key advantages of PM could be bringing down drug development costs and ensuring that a higher proportion of drugs actually make it through all the phases of development and onto the market.
Cut-price drug development
The development cost of each drug is unique, but the total cost of taking a new drug from concept to market is generally estimated at around £500 million, spread over 15 years. That cost could be reduced by more than 30 per cent by advances in PM, according a report from the Boston Consulting Group.
A separate analysis by PriceWaterhouseCoopers (PWC) reaches a similar conclusion, stating: 'Genomics-related technology facilitates the elimination of unfavourable products at earlier stages of development than is currently possible.'
PWC also points out that many of the 79 per cent of candidate drugs that currently fail in clinical development might actually succeed if they were tested only on sub-populations known to be most susceptible to their beneficial effects and least susceptible to side effects. A targeted approach will stop failures caused by testing candidate drugs on inappropriate patients.
PWC claims that the huge costs of clinical trials could also be significantly reduced by acquiring more specific results with smaller numbers of carefully selected patients. It also expects that PM might change the dynamics of the drug industry in favour of smaller biotech and biopharmaceutical companies, which it says are 'likely to benefit most through expanded opportunities, attainable markets, and increased power in their negotiations with the pharmaceutical giants'. PM also brings a growing market for the diagnostic technologies used to identify which subset of an illness each individual is suffering from. Many of these techniques are currently developed by small-scale biotechnology companies.
Hopes and fears
If it seems clear that PM could be good news for drug companies, does that necessarily mean it will be good for patients and for the government agencies that must fund healthcare costs? It could be, if the increased efficiency predicted by industry analysts allows some of the cost benefits to be passed on to consumers.
There are dangers, however. One of the most prominent is a consequence of the trend that, as advances in genomics reveal the detailed molecular basis of diseases, medicines will become ever more closely targeted on specific subsets of diseases that are currently lumped together. This can be expected to increase the number of so-called 'orphan diseases' which drug companies regard as unviable targets for commercial success, since so few people suffer from them.
A recent commentary in Nature estimated there are currently 8000 orphan diseases, each affecting maybe just 2000 people but together affecting millions. One paradox of progress in PM is that it may reveal how to treat the increasing number of orphan diseases ever more effectively, while at the same time making it increasingly unlikely that drug companies will be prepared to undertake the effort.
Another huge issue is the effect
of the PM revolution on healthcare in developing countries, which cannot even deliver the cheap treatments that are already available to their relatively unhealthy populations.
The World Health Organization (WHO) points out that '90 per cent of all health research expenditure is targeted at problems that affect only 10 per cent of the world's population'. The WHO emphasises that global inequalities in the availability and use of genetic technologies are currently
Its key concern therefore, is that the drive towards PM will deepen the divide between the rich and poor, developed and underdeveloped. In the third world, the key health priority is delivering existing cheap and effective treatments for major diseases, rather than discovering ever more sophisticated approaches.
Despite that concern, the WHO does see opportunities for developing countries, if scientific and political effort can be directed towards their needs. The WHO has acknowledged that 'genetic research has the potential to lead to major medical advances. against such killer diseases as malaria, tuberculosis and HIV/Aids, potentially saving millions of lives, especially in the developing world'.
Miracle or myth?
In the midst of so much publicity for PM's potential, it is important to beware the danger of over-selling the idea. 'Policy makers need to keep a very open mind about whether pharmacogenetics is going to really take off or not,' says Adam Hedgecoe of Sussex University, UK, who has recently published a book on the politics of PM. He cautions against the dangers of politicians making the currently unjustified assumption that PM will automatically produce a healthcare revolution.
Paul Nightingale and Paul Martin, also at Sussex University, recently published a cautious analysis of the biotechnology sector entitled The myth of the biotech revolution. They emphasise that biotech companies, dependent on inward investment, have a vested interest in building exaggerated expectations.
Their research leads them to conclude that 'rather than producing revolutionary changes, medicinal biotechnology is following a well-established pattern of slow and incremental technology diffusion. Unrealistic expectations are dangerous as they lead to poor investment decisions, misplaced hope and distorted priorities'.
At present, the actual penetration of PM products into real practice is rather limited. Adam Hedgecoe points out that up to 70 per cent of adverse drug reactions could probably be avoided by the very low-tech solution of improving prescribing practice. But the trend towards increasing personalisation of medicine, in one form or another, seems certain to continue.
The extent to which the wildest dreams for PM are realised will depend on both science and politics - on what really can be done, who can really benefit from it, and who can afford it. 'The jury is still out on whether pharmacogenetics will deliver widespread changes to healthcare,' says Hedgecoe.
Andrew Scott is a writer and lecturer based in Perth, UK
- The politics of personalised medicine - pharmacogenetics in the clinic, Adam Hedgecoe, Cambridge University Press, 2004
- The myth of the biotechnology revolution, Paul Nightingale and Paul Martin, Trends in Biotechnology, 2004, 22, No 11
Personalisation in practice
The anti-cancer agent Herceptin is by far the best publicised of the few drugs already on the market that come with some sort of personalised medicine (PM) claim. Herceptin is a humanised monoclonal antibody that binds to the human epidermal growth factor receptor 2 protein (HER2), that is overproduced in approximately 25 per cent of breast cancer patients. This overproduction of HER2 protein is believed to lead to continuous growth promoting signals being transmitted to affected cells, as a key part of the development of cancer.
Herceptin binds to the HER2 protein and mediates anti-cancer effects through a combination of mechanisms. These apparently include promotion of HER2 protein endocytosis and destruction, promotion of the destruction of diseased cells by the immune system, and enhancement of the effect of cytotoxic drugs generally administered alongside Herceptin.
The PM part of the process comes in the initial testing of cancer patients to see if they fall into the HER2 overproduction category. This preliminary test ensures that the drug is targeted at those patients who can benefit. Those who cannot benefit directly also win, because time and effort is not wasted trying out a therapy that will not work for them.
Another early runner in the PM field is Velcade, an anti-neoplastic small molecule drug used for the treatment of multiple myeloma bone marrow cancers. Velcade inhibits the proteasome enzyme complex within cancer cells and is approved in the US for patients whose cancers have already resisted other treatments. Millennium, the manufacturer of Velcade, is gathering detailed genomic data from patients and is hoping to identify one or more genetic markers to indicate which subsets of patients will respond to the drug most effectively.
In a recent development, in May 2005, Roche Diagnostics announced a collaboration with Eli Lilly and Company to classify biomarkers that may be used to identify patients most likely to respond to specific cancer therapies. The first phase of the agreement targets biomarkers linked to Lilly's Alimta (pemetrexed) and Gemzar (gemcitabine) anti-cancer treatments
In looking for real PM success stories, however, candidates are thin on the ground right now, but many new arrivals on the scene, and new successes are widely anticipated.
Knowing your cytochrome P450s
One of the most significant factors in each person's unique response to pharmaceuticals is the set of cytochrome P450 enzymes produced in the liver. These drug-metabolising enzymes are involved in degrading the majority of prescribed drugs, so our personal P450 complement is crucial to determining what effect drugs have on us, what dosages are appropriate, and what dangers the drugs pose. Eighty per cent of the most serious adverse reactions to medicines appear to involve drugs that are metabolised by the variable P450 enzyme system.
Understanding individual variations in cytochrome P450s has become possibly the main focus of attention for researchers developing personalised therapeutics. So far 57 genes for human P450 enzymes have been identified and variants of some of these enzymes have already been linked with the inability of some commonly prescribed drugs to be effective in some patients. Up to 10 per cent of the population, for example, do not experience any pain relief from codeine due to the specific version of one of the P450 enzymes (CYP2D6) found in their livers. Different forms of the same enzyme are believed to be responsible for the very variable response to antidepressants such as Prozac.
In the latest significant development in this area, Roche Diagnostics recently received clearance in both the US and EU for its AmpliChip CYP450 test. This is the first microarray-based diagnostic test for detecting genetic variations that can influence drug efficacy and adverse drug reactions. It reveals which variants people have in their livers of the genes for two of the most significant cytochrome P450 enzymes, namely CYP2C19 and CYP2D6. Knowing more about a patient's drug-metabolising enzymes will permit more accurate dosage and better avoidance of side effects.
As they look to the long term, personalised medicine researchers envisage a day when a trip to the GP or hospital involves a new and crucial phase before doctors print their prescriptions. Comments like 'Let's look at your genetic makeup' and 'Let's get a genetic profile of these cells' may become very familiar to us as genetics and pharmacology join forces in recognition of how different each of us is from everybody else.
Genetics, genomics and the patenting of DNA - review of potential implications for health in developing countries
World Health Organization, 2005
PriceWaterhouseCoopers, February 2005
Royal Society policy report, September 2005
External links will open in a new browser window