'Desk-top' machine for personalised prescriptions

Japanese companies have built a desk-top machine that allows doctors to check their patients' DNA before writing a prescription, the journal Nature reports. The device, which they say will be on sale for five million yen (£25,000) in a year's time, uses a single drop of blood and delivers results in less than an hour. The news comes shortly after the UK's Royal Society published a report claiming that although 'personalised medicines' have a promising future, it will be at least another 15-20 years before their use is widespread.

Pharmacogenetics - the tailoring of medicines to a person's genetic make-up - has often been touted as the future of healthcare, but some scientists have expressed reservations about whether such technology will ever live up to the hype, what it might cost, and how it would impact on healthcare systems.

The Japanese machine was developed by genomics researchers at the Institute of Chemical and Physical Research (RIKEN), printing company Toppan and scientific equipment firm Shimadzu. Takaaki Sato of Shimadzu told Nature that the key component of the device is a 'chip' that can analyse DNA directly from a blood sample, without the need for purification. It will initially be used to look for variations in genes that affect a person's response to an antibody called irinotecan - which can cause hearing loss in some patients - and to the commonly used drug warfarin.

Millions of people take warfarin to prevent harmful blood clots after a heart attack, stroke or major surgery. But the proper dose can vary greatly between individuals, in a way that is difficult to predict. Too high a dose can cause excessive bleeding in some people, while in others, too little a dose could allow dangerous blood clots to form. Currently, doctors use information such as a patient's sex, age, weight and medical history to set the dose, a process that can take months to get exactly right. A genetic test bcould help speed up this process. However, UK geneticist David Weatherall, who worked on the Royal Society report, is sceptical, saying that at least two different genes are involved in processing warfarin - a process that is not well understood.

Weatherall says that the only way to solve these problems is to test each drug independently in large populations monitored over several years. 'There is still a huge gap between the scientists who do this kind of work and its application for practical purposes', he said. Sato agrees that at the moment, the machine will prove most useful for research. But he says that given the response so far, there is 'no way' that 15 years will pass before doctors are using such devices for day-to-day diagnosis and treatment.