A big advantage of this technology is that one could look at multiple markers in a single reaction

Professor Richard Cote

They are found at elevated levels in the blood of men who are developing prostate cancer.

The device can detect these proteins at levels 20 times smaller than those produced by prostate cancer.

This makes it much more sensitive that the current test used to detect the disease.

The developers, from the University of Berkeley, believe the device could also be useful in diagnosing other diseases such as Aids and breast cancer which produce a characteristic pattern of proteins in either the blood or urine.

Diving board

The device works by inducing the proteins to stick to and bend a microscopic cantilever - essentially a diving board the size of a hair.

The higher the concentration of the protein being measured, the greater the deflection of the cantilever. This deflection can be measured using a laser.

Optimism must be tempered by the realisation that screening, for prostate cancer at least, is a complex issue

Cancer Research Campaign

The cantilevers are made from silicon nitride using techniques identical to those employed by the semiconductor industry to make microprocessors.

The Berkeley team have found a way to put several hundred cantilevers onto a single silicon chip.

This could make it possible to test for many different chemicals at the same time.

Professor Richard Cote, of the Keck University of Southern California, said: "A big advantage of this technology is that one could look at multiple markers in a single reaction, whereas currently available assays require a separate reaction for each analyte."

"In a very short period of time, we've really made enormous advances. I'd be surprised if this doesn't become a viable assay system within the next three to five years."

Welcome development

A spokesperson for the Cancer Research Campaign said any technique that helped to accurately detect cancers at an earlier stage was to be welcomed.

"This team's development of a method for analysing multiple proteins simultaneously on a single chip could in time lead to a fast, sensitive and relatively cheap cancer detection system.

"But this has yet to be proven. Optimism must be tempered by the realisation that screening, for prostate cancer at least, is a complex issue."

The spokesperson said: "The authors' claim that this technique is able to detect 'levels 20 times lower than the clinically relevant threshold' says more for its superior sensitivity than for its ability to detect more prostate cancers.

"The majority of marginally raised PSA levels detected in non-symptomatic men are not due to prostate cancer but to other diseases.

"The success of this test as a screen for cancer depends a great deal on its specificity not being sacrificed for increased sensitivity."

The research is published in the journal Nature Biotechnology.