Individually tailored medicines have been "over hyped" and are still many years away, leading scientists say.
Research may be able to reduce the number of adverse reactions to drugs
A report from the Royal Society said they were still at least 15 to 20 years away.
It is known that some drugs work more effectively in some people than in others. Research is being carried out to find out why.
Working group chairman Sir David Weatherall said there were many gaps in understanding of genetics and disease.
Personalised medicines, or pharmacogenetics, refers to treating the same disease differently in particular groups of people.
It also encompasses the identification of genes that make drugs effective or harmful and assessing how people can differ in susceptibility to infectious diseases, so that new medicines or vaccines can be developed.
At the moment, personalised medicines are being investigated for the treatment of some cancers and the use of a genetic test with a drug so it is only given to patients with the right genes for it to be effective.
'Groundwork necessary now'
But the report says that, for further development of such drugs to take place, there needs to be more knowledge of the relevant science among healthcare professionals, and more researchers who have specialised in the areas required to carry out pharmacogenetic research.
It calls for financial incentives to be introduced to encourage research in the UK and Europe and to encourage pharmaceutical companies to work on developing pharmacogenetic drugs which are likely to have relatively small potential markets.
Sir David said: "Personalised medicines show promise but they have undoubtedly been over-hyped.
"With the human genome sequenced, some people are expecting personalised medicines within a few years, but the reality is still many years away.
"There are some examples around today, but the complex multiple causes of diseases mean it will be at least 15 to 20 years before a patient's genetic make-up is a major factor in determining which drugs they are prescribed."
But he said the groundwork was needed now if that timescale was to be achieved.
He added: "Investment needs to come from both the public and private sectors.
"For new drugs we would like to see better use made of the genetic data collected by pharmaceutical companies during clinical trials. Also, monitoring should continue once products are on the market with the aim of linking genetic variability with clinical outcome.
"We believe this kind of work should be mandated at the national and European level as an extension to the current system of clinical trials."
And once personalised medicines do become routine, new challenges will emerge, including how the health service will manage their provision and how data on what drugs work for which people, the report adds.
Toxic effect risk
Sir David added: "Pharmacogenetics may prove valuable in the fight against the big killers worldwide, such as malaria, tuberculosis and HIV.
"Research is needed to establish the cost-effectiveness and clinical value of this approach for developing countries."
Professor John Toy, medical director at Cancer Research UK, said: "Unfortunately many cancer patients don't respond well to the treatments that they receive. Instead of benefiting from these medicines, patients sometimes suffer only the toxic side effects.
"If medicines were able to target an individual patient's cancer, they would work more effectively."
He added: żEfforts are going towards designing personalised medicines with some early success.
"Herceptin is an example of this approach it targets the HER2 protein which is found only on the cells of around one in four women with breast cancer.
"However, this science of pharmacogenetics is extremely complex and it is likely to be several years before many patients can expect to receive bespoke cancer treatments."