Personalised Medicine: great for patients, bad for business?

The concept of personalised medicine has always been at the heart of every physician's decision making process but the "one-treatment-fits-all-disease-phenotypes" is very often not satisfactory. The regularly used "trial-and-error" approach in treating many patients, often determined on cost factors, delays appropriate treatment.

Finding alternative ways of diagnosing patients by using biomarkers, however, can dramatically improve this. For some patients the use of molecular diagnostics can also mean increased safety because pharmaceutical drugs can be harmful for some patients but not for others. Rather than rejecting the treatment option altogether, these patients can be identified and excluded from the treatment population. It can also lead to the resurrection of drugs. For example, Vectibix from Amgen didn’t show improved efficacy until non-responders with a mutation in the K-RAS gene were identified and excluded from the treatment. This is a prime example of personalised medicine but uncovers a new issue. Amgen would most probably not have gone ahead with developing the drug for a market that is 60% smaller than expected. Vectibix sales are below expectations and far from a blockbuster drug. So who should pay for these in the future?

There is almost a moral obligation for pharmaceutical companies to bring these drugs to market but this is not a business model and not sustainable for future products. To ensure that innovative products like Vectibix get to market in the future, the high development costs of around €1.5 billion will have to be reimbursed.

In the UK a new program has been launched that will foster development of personalised medicine with a much needed cash injection. The program is called the Stratified Medicines Innovation Platform and will oversee an investment of over £50 million of government funding in innovative research and development in areas such as tumour profiling, drug biomarker discovery and testing new business models. The first competitions, worth up to £11 million, open in January 2011.

Collaboration is needed in personalised medicine and is a prime requirement for these competitions. Details on the 3 competitions can be found here:

• Strand A: Developing Business Models and Value Systems
• Strand B: Tumour Profiling and Data Capture to Improve Cancer Care Flyer
• Strand C: Developing Biomarkers for More Effective Drugs

I hope that all other European countries will follow suit, however, cash injections and grants are not the only thing that is missing to make personalised medicine a reality.


We need:

• Clear guidelines from the regulatory authorities and payers which will help companies to develop their business plans
• Increase collaboration and involvement of physicians to make sure that these pioneering developments will reach the front line and the patients
• Government funding for sound projects that are not commercially valuable but with a great potential to improve patients' lives
• Access to national biobanks and patient data to researchers to ensure that the samples that are used and data that is generated is of the highest standard. Bad samples in - bad data out
• Agreements on standards in bio-sampling to ensure long time usability of biological samples and biobanks

Personalized medicine comes one step closer

Researchers from the Stanford University School of Medicine have given us a glimpse of what the future of healthcare could look like.

We have had human genome sequences for a while now, but it has taken some time to overcome the information analysis challenges. Euan Ashley and colleagues published a study in The Lancet last week assessing the potential of a patient’s genome in a clinical setting to find out what a physician could do for a patient with a specific genetic profile. This was only the fifth human genome ever sequenced, and the first time that this type of study was done.

The field of genomics and genetics has generated a vast amount of information to date, and the researchers tested the integration of these results with the genome and medical history of a specific patient. They were able to test disease risk and link it to the patient’s medical history. They also tested pharmacogenomics, the genetic predisposition of an individual to react to a drug treatment, information that can tell a physician if a patient will benefit from a particular drug and if the patient is likely to experience adverse effects.

The patient, who had a family history of vascular disease and early sudden death, fell just short of the regular criteria for a recommendation of a cholesterol-lowering drug. But his genome revealed that he had increased genetic risk for coronary artery disease. The pharmacogenomic study revealed his treatment options based on his genetic make-up: He would probably respond well to lipid-lowering therapy and to low initial dosing requirements for Warfarin, a popular and effective drug for preventing thrombosis and embolism. He had a low risk of side effects. This could be suggested as a very good treatment option, if prescription of drugs became necessary. Under the current system, this could have taken a month to determine.

So, is this what the future of healthcare will look like?

Probably, but there are still a few challenges ahead, not only from a technological point of view but also from an ethical point of view. The technical challenges focus on automated annotation with extremely high accuracy and ways to bringing down the cost below $1000. Scientists are working around the clock to achieve this and it is only a question of time until we get there. But the ethical question is far greater and something that each and everyone of us will face! How would you cope with 2.5 million reference points about your health, which could tell you about hundreds of disease risks and medications that will have no effect? Something like genetic psychological counseling will have to be introduced!

Undoubtedly, the transition to personalized medicine will change the way healthcare is delivered.