Real-time diagnosis in the ICU

Professor Keith Stanley is not afraid to admit being happy at getting his hands dirty by joining the dark side, also known as the commercial sector. Having spent a distinguished career as a 'pure' molecular biologist, three years ago he joined Australian manufacturer Corbett Life Science as the director of its molecular research division.

He still maintains his purity as head of the inflammation and cancer research group at Sydney's Centre for Immunology, a joint centre between St Vincent's Hospital and the University of New South Wales, where he is an adjunct professor, but in the meantime is directing research into improving the applications of PCR systems for Corbett.

One of those new applications is in microbiology, a topic he will discuss at the upcoming Australian Society for Microbiology (ASM) conference in Adelaide.

Corbett's main product is the well-known Rotor-Gene, which last year won a prestigious Frost & Sullivan award for technology innovation. This nifty little machine has over the years been refined to be adaptable for new technologies hitting the market, such as multiplexing and high resolution melting (HRM), which have greatly improved the efficiency of single nucleotide polymorphism (SNP) analysis, amongst other applications.

The main application for microbiology that he is working on is species identification through PCR in real-time, as opposed to the lengthy process of blood culturing.

It is an application that has developed from his work on refining PCR for its main roles in gene scanning, SNP genotyping and identification of predisposition genes.

In 2004, Stanley and Corbett Research received a federal START grant to develop an instrument for analysing cancer biopsies. That project is going along nicely, with a company called AusDiagnostics offering inexpensive products to the research community, as well as for clinical research and, shortly, diagnostics.

It was while conducting this project that applications for microbiology became apparent, Stanley says.

"Initially we were trying to develop a turnkey solution for human cancer analysis," he says.

"It all comes from the fact that for instance in breast cancer, only about 10 per cent of women benefit from chemotherapy, so the question is can you distinguish that 90 per cent who don't actually benefit at all, to save them the trauma of having the treatment.

"What we had to do was to be able to use paraffin-formaldehyde fixation of starting material, that was one of the challenges, and the second was we had to profile a large number of genes, as there are quite a few that are important. And we wanted to develop technology where basically you put a sample in and press a button and it gives you an answer.

"Essentially what we have done is developed a protocol for extracting RNA from formaldehyde fixed sections, and the main difference from other products is that it is much quicker and the quality is much the same."

Stanley and his team have developed a multiplexed system for PCR, the main feature of which is that it requires very little input material.

"We actually have molecular sensitivity and we can multiplex a lot of genes from a small sample," he says.

"In the human cancer situation, we can go down to single cells and measure at least 36 genes in a sample. And from a single paraffin section we can get 10 assays out, so we can measure about 300 genes.

"That's the basis of what we have done and we are about to bring out a robot that will semi-automate the whole process and make it easier to run these assays."

Species identification

One of the great leaps forward in contemporary real-time PCR analysis and high resolution melting is the ability to distinguish a single base pair change in an amplicon of 100 to 200 base pairs.

For microbiologists, instead of looking at expression profiles, species identification is the key.

"We can say here's a blood culture, what bacteria were present in it and what resistance genes were they carrying," Stanley says.

"And you can do treatment prediction in a similar way to what we are trying to do with cancer. The advantage that we have is that we can do 12 or 24 different genes simultaneously rather than having to do one at a time."

Stanley is collaborating with Dr Jon Iredell and Professor Tania Sorrell at Westmead Hospital's Institute of Clinical Pathology and Medical Research to develop the real-time PCR system in a clinical setting - in the intensive care unit, to be exact.

"The stance we have taken is let's work from blood culture, which is what people normally do, to ascertain whether it is gram-positive, negative or a fungal infection. We have designed our assays to work on those three areas and identify either fungus or common gram-positive or gram-negative bacteria. So it will fit in very smoothly with the current triage in a microbiology lab.

"The advantage is that PCR is much quicker - we can get an answer within 90 minutes, whereas if you have to spread the plates it can take 16 or 24 hours for the bug to grow. And then you won't necessarily know what drug resistance genes they might contain.

"We can say do you have MRSA, do you have vancomycin resistance, do you have aminoglycoside resistance, do have an IMP type carbapenemase - all of those resistances you can tell in one go.

"What we hoping for with our molecular sensitivity is that we can go from an EDTA blood sample and cut out the culture as well. We haven't tried that yet but we think we should be able to do it."

The ASM conference is being held in Adelaide from July 9 to 13. See www.asm2007.org for more.