Feature: Automating the personalised medicine revolution

This feature appeared in the May/June 2010 issue of Australian Life Scientist. To subscribe to the magazine, go here.

You are diagnosed positive for HIV. No problem: your doctor orders your genome and that of the virus to be sequenced, and your response to the infection and treatment is predicted. A vaccine or other treatment strategy is then prescribed. All just for you.

This is the vision of personalised medicine. On paper, it looks enticing. But there are many technical hurdles to overcome before the idea can become reality. Researchers at a new research facility, the Institute for Immunology and Infectious Diseases (IIID) at Murdoch University in Perth, are working on bringing personalised medicine into the real world.

Growing out of the Centre for Clinical Immunology and Biomedical Statistics, established in 2000, this new purpose-designed facility sits within the University’s biotech and health sciences precinct. It will eventually house about 70 staff including clinicians, biostatisticians, computer scientists, laboratory scientists and other researchers. In its new digs, the IIID will continue and expand the landmark research work on HIV and hepatitis started by infectious diseases clinician and Institute Director, Professor Simon Mallal, almost 20 years ago at the Royal Perth Hospital.

The institute now runs five programs: HIV-host interactions, hepatitis C virus (HCV) infection, drug hypersensitivity reactions, antiretroviral drug toxicity and biostatistical methods. A unique resource in the field, the institute consists of a multidisciplinary team of leading scientists and clinicians with access to state-of-the-art facilities to integrate scientific knowledge and methods with clinical and diagnostic medicine to address questions of clinical importance and ultimately apply the answers to improving human health.

Head of R&D at the IIID is Dr Abha Chopra, who describes Mallal’s research as visionary. She joined Mallal six years ago, when the lab staff consisted of only four people, and took on the task of setting up and running an extensive array of high-throughput automation and robotics systems to enable the team’s ambitious research efforts.

“In broad terms, IIID’s research is focused on the practical applications of identifying a patient’s genetic profile and then determining the most appropriate strategy for treating the disease,” says Chopra.

This focus was advanced enormously five years ago when Mallal’s team was awarded $12 million in funding through the inaugural round of the Bill & Melinda Gates Foundation Grand Challenges in Global Health initiative to investigate the genetic influences on clinical outcomes in HIV and other viral diseases.

“A major focus of the viral work is investigating basic patient biology in terms of histocompatibility or human leukocyte antigen [HLA] typing and human genetics, and then adapting these two things to each other,” explains Chopra. “We basically want to know why the viruses are what they are, and why are they causing the disease. We know there is a strong genetic link.”

Mallal’s seminal work at Royal Perth Hospital on host and viral factors that generate genetic diversity was published in Science in 2002. It basically demonstrated for the first time that HIV diversity is largely driven and shaped by HLA diversity at the population level, and effectively opened up a whole new field of host-pathogen interactions.

“The work also generated a huge and valuable resource of patient data including genetic profiles with a particular focus on HLA typing and full-length HIV genomic sequencing from a large number of infected individuals, together with background histories,” says Chopra.

---PB---

Automating it

“Our original project with the Gates funding was to analyse clinical samples from around 1750 patients worldwide for HLA typing, genome sequencing and viral sequencing. This approach would enable us to get much more power in our analyses by handling large numbers of patients together in a reasonable period of time,” says Chopra.

“One of the biggest issues in these types of genetic analyses is keeping track of the specific patient samples and collating records and data, so high-throughput with robotics was really the only way to go. Once the record is entered for a given patient in our LIMS, everything to do with that individual can be automatically tracked and all results linked at any stage.”

Chopra looked at several automation systems available in Australia at the time, deciding eventually on the Biomek FX liquid handling platform systems from Beckman Coulter as the most reliable and versatile for the group’s current and future applications and goals. According to Chopra, it has turned out to be a very good partnership and they now have nine instruments from Beckman Coulter.

Over the past four years the team retrieved samples from labs and clinics and set up automated testing and validation systems in a world-class NATA accredited laboratory. “The Biomek systems can handle large number of samples at a time and what would have taken months 10 years ago is now being done within a day,” Chopra says.

“We started sequencing the whole genome of virus extracted from infected patients, as well as the patient’s DNA. Viruses such as HIV or HCV have very high mutation rates and hence are very diverse. They can vary across patients and over time or following drug therapy in the same patient, which is why the immune system is not able to recognise them and generate a quick response. We want to know the mechanism behind the virus’s ability to adapt genetically.”

Being able to detect or predict what the changes are likely to be with a given patient HLA type or other genetic make-up might enable the clinician to predict how that virus is going to change with a particular patient genotype, and target the prevention or treatment specifically for the patient.

If this approach is successful, a patient might be able to mount an immune defence against the attacking virus much earlier in the infection process by making the right antibodies. It basically is a personalised vaccine. According to Chopra, we already know that there are clear racial differences among HLA type and therefore among patient responses to a given virus, thus a candidate vaccine has to take all of this into consideration as well.

---PB---

Personalising it

Another major focus of the Institute’s research is looking at drug hypersensitivity in patients with infectious diseases, again starting with HIV patients. “We have been working on this for a long time,” says Chopra. “And we are now applying our high-throughput automation and robotics systems to look at large patient cohorts. We want results that can be eventually implemented routinely in the clinical setting.”

Early work in this area by Mallal established a link between drug toxicity and the genetics of both the virus and the patient. “So a patient with a certain genotype is more likely to develop a reaction at some point of taking the drug,” Chopra explains.

They went on to identify specific genetic regions implicated in the hypersensitivity reaction to abacavir, an early anti-retroviral agent that proved very effective against HIV, but also caused quite severe reactions in some patients, limiting its overall use. “We identified particular genotypes in patients that excluded use of this drug, while it was safe for all others. The findings meant that this drug could be used in some patients by finding out the genotype of those individuals who could not tolerate it.”

The group is gearing up for bigger research projects along this line with many new drugs emerging for viral diseases. Most of the genes implicated in hypersensitivity reactions are also turning out to be in the major histocompatibility region of the patient’s genome – a highly variable genetic region – and certain gene loci seem to be more susceptible then others to adaptation by the virus.

As with the immune response to the virus, they expect to find ethnic differences in drug hypersensitivity that are associated with a given HLA allele or haplotype association. An aim of this research program is therefore to develop predictive models of drug hypersensitivity based on their growing body of sequencing data that incorporates genetic, demographic and disease-related variables.

Once the move to the new building is complete, Chopra is looking forward to the next big venture – next-generation sequencing – using their brand new 454 Genome Sequencer FLX system from Roche. This will significantly advance their characterisation of the viral diversity, which Chopra describes as amazing.

“The viral genome can even change over the course of an infection within a patient. Drugs may then affect it also, which in turn might affect the virus resistance to the drug. Even in the host there is a lot of genetic pressure applied from the virus and we are interested in tracking all these changes, which means a whole lot of sequencing.

“The next-generation technologies allow us to take on such a big task. It not only makes these types of projects more feasible to do in the first instance, they also become much quicker (months instead of years), cheaper and more reliable, and this that the idea of personalised medicine in this field is getting closer very quickly.”

This feature appeared in the May/June 2010 issue of Australian Life Scientist. To subscribe to the magazine, go here.