Karen Smith

The Personal Touch - Tailor Made Medicine.

"We have discovered the secret of life!" Francis Crick,

1953

In 1953 Cambridge scientists Watson & Crick discovered the double-helix structure of DNA. 50 years later other Cambridge scientists such as John Sulston were key in decoding the human-genome - another fundamental breakthrough had occurred that might allow us to one day to personalise medicine.

Figure 1: Watson & Crick with their double-helix model of DNA

Imagine it - you visit your GP and the first thing he does is to

painlessly collect a sample of your DNA from your cheek or fingerprint.

A press of the key on his computer brings up your individual genome

and a couple of minutes later your DNA has been analysed and your

doctor can prescribe you the drugs that will suit your individual

genetic make-up. The analysis might predict your response to a particular

medicine, recommend the dose you need and indicate what side-effects

you might have. The need for educated guess work - gone! Although

such a scenario sounds fantastical, pharmacogenetics, pharmacogenomics

and advances in new technologies mean that it could become a reality.

A reality that could dramatically change the delivery of health

care by altering the way diseases are defined, patients are treated,

and new medicines are discovered and developed.

Figure 2: Doctors prescribe treatments based on the patient's symptoms and use their skills to prescribe a regime which is hopefully safe and efficacious

Scientists have known since the 1950's that individuals can react

differently to any single drug, and much of this is due to individual

genetic variation. Personalised Medicine is thus most often associated

with pharmacogenetics which is the study of how individual genetic

variation affects our response to medicines. Most medicines today

are essentially based on a 'one size fits all principle'.

Doctors prescribe treatments based on the patient's symptoms and

use their skills to prescribe a regime which is hopefully safe and

efficacious. However, there are problems with this approach. Doctors

may often try several medicines on a patient before they find one

which works (efficacy). This is because one disease may actually

have a number of underlying causes based on different genetic characteristics.

Common drugs used to treat this 'single' disease may only work well

in a subset of patients - indeed many current treatments only work

in 30-60% of patients. Patients also vary in their ability to metabolise

drugs (safety); this is due to individual differences in key liver

enzymes (such as CYP2D6) which are responsible for processing ingested

drugs.

Figure 3: AmpliChip CYP450 - the world's first pharmacogenomic microarray designed for clinical applications

"The vast majority of drugs - more than 90 per cent - only

work in 30 or 50 per cent of the people" Allen Roses, worldwide

VP Genetics, GlaxoSmithKline, 2003.

Pharmacogenetics and associated techniques have the potential to

address these patient-centred problems, and also alleviate some

commercial drug development issues. As an example, in 2004 Roche

introduced a microarray based diagnostic chip (Amplichip CYP450)

that can determine which liver enzyme variants an individual possesses.

This information gives a physician a more precise idea of how a

patient will tolerate a particular drug and thereby avoid dangerous

side-effects. This represents Personalised Medicine in action -

a drug chosen on the basis of a patient's individual genetics.

The most famous example of Personalised Medicine is Genentech's

Herceptin. This novel anti-cancer drug failed its initial clinical

trials as it worked in only a fraction of patients. Normally such

drugs are abandoned on efficacy and cost-effectiveness grounds.

However Genentech performed further work and discovered that the

drug worked extremely well in a subset of patients whose cancers

over-expressed the protein HER-2. The drug is now considered a vital

treatment for these women (who are identified on the basis of a

pharmacogenetic test). The lead researcher in this study Professor

Yu, University of Texas commented :

"In the past, selecting a particular chemotherapy

drug has been a lottery. Patients had no way of knowing if they

would benefit. The new generation of cancer drugs is providing targeted

treatment with fewer side effects and giving patients a better chance

that a particular treatment will work for them".

Figure 4: Monoclonal antibodies targeting a HER2 protein over expressing cell

Pharmacogenetics may also open up new avenues of research by allowing

medicines to be designed based on an understanding of disease at

a genetic level. At Cambridge, many leading researchers are working

to decipher the molecular & genetic basis of complex diseases.

Professor Carlos Caldas, Department of Oncology, is looking to generate

molecular classifications that correlate with distinct clinical

outcomes using expression profiling and other techniques. The predictive

value of these indicators is then tested on tissue samples from

large patient cohorts. The eventual aim is to be able to genetically

assess an individual patient and use molecular classifications to

inform therapy choices. Meanwhile, Dr Sabine Bahn at the Institute

for Biotechnology is investigating the genetic basis of neuropsychiatric

diseases such as Schizophrenia and Bi-polar disorder. Drug treatment

for mental illness is currently hit and miss, and patients often

discontinue treatments due to severe side-effects. Dr Bahn's particular

focus is therefore on identifying biomarkers and genetic traits

which will lead to the development of pharmacogenetic tests to guide

drug use, as well as offering new targets for drug development.

Also at the Institute for Biotechnology, researchers in Professor

Chris Lowe's lab are developing the diagnostic technology which

is so important for this field. One problem with current tests is

their complexity which prevents them being used in a doctor's surgery

for example. New technology based on holograms allows a test result

to be obtained quickly for a particular analyte, with the result

being shown as a visible hologram.

This technology is now being commercialised by the spin-out Smart

Holograms Ltd.

Figure 5: Sensor holograms have many applications in the Life Sciences industry, in particular in Diagnostics and Medical Devices

In addition to improving drug safety, efficacy, and highlighting

new drug targets, the science of pharmacogenetics may benefit the

pharmaceutical and biotech industries in other ways. The current

pharma business model is based around producing blockbuster drugs

generating massive revenues of >$1billion per year. This model

exists because of the low success rate and high costs of taking

compounds from discovery through to market: only 9% of pre-clinical

compounds ever get to market and the cumulative cost (including

failures) of getting one drug to market is estimated to be in the

region of $1.8 billion dollars. This model is now under pressure

because of increased regulatory hurdles, and the fact that much

'low-hanging fruit' has been picked. Pharmacogenetics could address

some of these problems. For example, clinical trials could be shorter

and cheaper if patients likely to suffer adverse reactions were

identified and excluded from trials. Because of the blockbuster

model, big pharma has been reluctant to embrace pharmacogenetics

to date. In contrast, Biotech companies have been quick to exploit

the new technology, and are happy with smaller revenues of, for

example, $300 million for a drug which works on a small number of

people.

Figure 6: Gene-based approaches introduce a new language of "probability" and "susceptibility" to medical care.

It is worth remembering that, in the face of all this remarkable

science, personalised medicine has major ethical, legal and social

implications too. Particularly in areas such as patient consent

and genetic testing. For example: who owns genetic information?;

will small groups of patients with certain genetic profiles be left

without treatments because drug companies are not willing to develop

drugs for them?; should testing be performed when no treatment is

available?; should pharmacogenetic tests be available over-the-counter

or only in doctor's surgeries?

The Cambridge Genetics Knowledge Park (CGKP) and the Dept of History

and Philosophy of Science, Cambridge focuses on issues such as these

and you can read a previous article on ethics and pharmacogenomics

by clicking the appropriate link to 'other articles by the author'

below.

For general information, contact:

Cambridge Computational Biology Institute,

Centre for Mathematical Sciences,

Wilberforce Road,

Cambridge CB3 0WA

- August 2005

About the Author

Dr Karen Smith is a neuroanatomist. She was the Business Director of the Cambridge Computational Biology Institute and is now the Director of Bioprocess Leadership, Biochemical Engineering, University College London.