Pharmacogenomics is a big word but its meaning is straightforward - in future, medicines will be cleverer.

Since the human genome was mapped two years ago, life sciences companies have been bustling away to put the identification of the human body's 30,000 genes to useful medical effect.

One of the most exciting development concepts is proving to be the tailoring of pharmaceuticals to a genetic profile.

"It turns the whole theory of traditional drug delivery on its head," says David Speechly, vice-president of Applera, the US group that owns Celera, which mapped the genome. "In the past, companies have proceeded serendipitously, developing a compound, then seeing what it could be used for."

Now, companies can take their knowledge of genes - and the disease-causing proteins that genes express - to tweak drug design. Hence the terms pharmacogenomics and, next in line, pharmacoproteomics.

Christine Soden, finance director of Oxagen, a private pharmacogenomics company, says: "Genetics allow us to understand disease far more fully." So-called diseases, such as cancer and asthma, she explains, are in fact woolly terms for many different diseases, which could have a range of genetic triggers. "Genetics can help to pinpoint the cause and support the design of a precisely targeted drug."

The endgame, most scientists agree, is a spaceage-sounding scenario of patients armed with genetic smart-cards visiting GPs for a personalised treatment.

Through pharmacogenomic profiling, scientists will be able to understand an individual's susceptibility to disease, their likely response to existing drugs and whether they would suffer side-effects. "The question is not if but when that will happen," says John Padfield, chief executive of Amersham Healthcare, the diagnostics company. "Personally, I think it's 20 years rather than five years away."

But in the meantime, diagnostics companies such as Amersham are set to benefit. Using knowledge of genes and proteins, they can help to diagnose a patient's likelihood of contracting a certain disease.

Mr Padfield cites NeoSpect, a diagnosis product that can detect the expression of the somatostatin protein in people at risk of developing malignant lung tumours. In theory, this could help doctors to establish before the event that a heavy smoker was liable to contract lung cancer.

Such technology is clever. But until it can be teamed with clever treatments, patients risk suffering. "Imagine you are 25 and we can tell you that you are going to get Alzheimer's but not when," says Mr Padfield. "That's just going to ruin your life."

That is why diagnostic companies see the greatest potential in teaming with big pharmaceutical companies to develop packages of complementary diagnostic and therapeutic tools to target certain illnesses in certain types of people.

Much of the technology to design a new generation of gene and protein-specific drugs is still only in the early stages of development.

Many big pharmaceuticals companies profess reticence to accelerate the process. "I know a lot of big pharma senior executives who are saying in public that personalised medicines will never happen because they cannot afford to give up the margins on bulk drug production and marketing," says one industry insider. "But, in private, they are working away to develop them because they know they have to. Competitive pressure and regulators will demand it."

Destabilising as the move to personalised medicines will be to the traditional drug market, there could also be clear commercial benefit for big pharma to draw from pharmacogenomics.

Several big-name drugs, though highly successful in some patients, have proved ineffective in others. Herceptin, the Roche treatment for breast cancer, is often cited. It is thought only to work in about 20 per cent of patients, because it acts on the Her2 protein, which is only a trigger for 20 per cent of cases.

Geno-typing technology would allow diagnostic and drug companies to pinpoint which segment of the patient population would respond to a drug like this and which would not. Similar technology could be applied to specify who will suffer side-effects from taking a drug.

That, say some, could prompt pharmaceutical companies to resurrect old drugs that had been shelved for their horrific side-effects.

Mr Speechly cites Baycol, the Bayer cholesterol-lowering drug, which was withdrawn last year after unexplained patient fatalities.

"If you know who reacts and who does not, who is exposed and who is not, who is going to have a side-effect and who is not, because of their genetic profile, then you can revitalise shelved drugs like these," he says.

EMAIL THIS PRINT THIS MOST POPULAR