The global pharmaceutical industry is twice Australia's annual budget. Australians burn through their fair share of drugs, consuming nine billion dollars of allopathic and natural remedies annually. So it defies belief that to this day, over a third of prescriptions written simply don't work.
Efforts to genetically map health and disease have demonstrated just how complex the picture really is. Disease can be any number of genetically influenced processes which differ between individuals. It helps explain why most drugs work in some and not others. For the first time, tests can help to tell these patient groups apart, before we embark on treatment.
It is said doctors start their careers knowing a little about many drugs and end their careers knowing much about just a handful. In between, enormous sums are spent on ineffectual treatments. Now, thanks to genomics, all that could change.
This transformation is shifting effort from 'what works,' to 'upon whom.' Welcome to the era of personalised medicine. Not just a personal doctor, but where treatments work specifically on you.
Just ten years ago, we were heading in precisely the opposite direction. Blockbuster drugs offered population-wide benefits to reduce blood pressure, lipids and obesity. These drugs were so promising, some joked they could be added to the water supply. Breaking the bank was collateral damage for governments to sort out.
Ultimately it got too expensive to measure these minor health improvements across broad populations. Since 2000, bringing a new treatment to market has blown out from $800 million to a staggering $1.3 billion. New drug applications to the US drug approving body (FDA) steadily fell as result, to around 25 annually.
Worst of all, many new drugs were falling at the last hurdle, with an overall 90% failure rate at the end of phase III human trials. That was often after huge sums had already been invested. The blockbusters appeared promising in animal studies but couldn't always establish their cost-efficacy in humans.
Government is shifting ground too. Tired of paying for drugs unleashed on whole populations, it's in their interest to identify precisely which patients are in the 40% for which prescriptions will fail due to personal variation. Approvers now demand evidence up front on who responds and why. In cancer treatment for instance, government is sick of paying for tumour shrinkage alone. It's now all about quality and duration of extra life. The new term is companion diagnosis; using accompanying biomarker tests to stratify patients into who responds, how well and who won't. The potential to eliminate inefficiency and waste in health systems is enormous.
Medical diagnosis will also be transformed. In the past broad disease categories like multiple sclerosis had complex clinical courses, with subsets of patients having vastly different outcomes. We are uncovering new connections between molecular level processes occurring across a range of diseases.
A deliciously named new treatment called canakinumab exemplifies this shift to molecular pathways. Initially used to treat a few thousand worldwide living with a rare autoimmune disease called CAPS, it has just been revealed that the same molecular process is at play in gout. Suddenly this rare treatment offers hope to three million worldwide struggling with a far more common affliction.
New prognostic and risk information can help us find the right drugs for the right patient. That allows subgroups to be monitored and prevented differently, saving our health systems millions of dollars. As companion diagnosis expands, Government will increasingly seek to remunerate not the drug, but the clinical result. Moving from rewarding transaction to outcomes will ultimately spread from pharmacology across the rest of health care. Genomics already assists with identifying which lymph node negative breast cancer patients will recover without additional chemotherapy. In this case genomics also tells us how to save $100 million a year by not treating. It's a test which more than pays for itself. It isn't much of a leap beyond here to adjust surgical payments according to their clinical outcomes.
Pharmacogenomics is still emerging. Major pharmaceutical companies have identified just a handful of biological pathways for intensive analysis. In 2011, there are around twenty FDA-listed products where prescribers are advised to gather genetic information prior to prescribing. An example is the third of heart attack or stroke patients who lack an enzyme and are therefore unable to respond to Lopidogril. The right test can save us 30% of total drug costs. On top of that we avoid the adverse reactions, the treatment delays, the extra patient consultations and inconvenience.
The days of fiddling with drug structure to extend its life in the market may be numbered. Genomics identifies drug targets, which in turn offers a host of new applications for molecules. Virtual and computational drug design can predict human toxicity long before we need to put compounds into animals or people. PET scans now allow us to image pharmaco-anatomy and witness drugs hitting their target. The immediate challenge is to identify the most likely candidates for intensive development from countless opportunities and bring them to market without breaking the bank.
Public demand in Australia for genetic testing is already on the rise. The US has established a voluntary genetic testing registry which provides standardised information about tests, their purposes, the likelihood of clinically reliable results and probable actions. Australia too needs a one-stop shop on clinical validity and the utility of genetic testing. Government, medical and nursing organisations need to provide the public as much information as they seek in this rapidly evolving frontier.