UBC brings genomics to primary care
L-R: Martin Dawes and Chris Phillips, a participant in the TreatGx pilot study. Photo credit: Martin Dee
The act of prescribing a medication seems so simple, at least to a patient.
The physician, after arriving at a diagnosis, reaches for a pad, scribbles a few lines, and utters a few words of instruction and caution (something like “not on a full stomach”). Then it’s off to the pharmacy.
But consider these statistics:
- In 2008, UBC researchers found that 7 per cent of visits to Vancouver General Hospital’s emergency department were due to harmful side effects of a prescribed drug or an incorrect prescription, either in the type of drug or the dosage.
- 14 percent of particularly vulnerable patients in Scotland were given high-risk prescriptions that should have been avoided, according to a 2011 study.
- A 2014 study in Ireland found that 35 per cent of the typical general practitioner’s prescriptions for older people were “potentially inappropriate,” meaning the risks outweighed the benefits.
An unnerving reality emerges from these numbers: Prescribing medication is not as straightforward as patients might assume. Sometimes, the drug prescribed is simply not effective. Sometimes, it’s even harmful.
The “X factor”
So many factors come into play: not only the obvious ones, such as the patient’s age, weight and known allergies, but also other drugs the patient might be taking, and whether those drugs might react badly with each other.
Then there is the latest evidence about each drug’s pros and cons, which are constantly being revised by drug companies, regulatory authorities and academic journals, but which are not easily synthesized for front-line physicians.
And lastly, there is “X factor” of a patient’s particular biological make-up – particularly their genes.
A single mutation can spell the difference between a drug being useful, or being deadly. One mutation on a gene called CYP2C19 causes rapid metabolizing of two commonly used anti-depressants, diminishing the drugs’ effectiveness; another mutation on that same gene could have the opposite effect – slowing down metabolization, thus heightening the risk of side effects. Another mutation causes codeine to be rapidly converted to morphine, posing a danger to babies drinking breast milk from women taking this common over-the-counter drug.
More than 150 medications approved by the U.S. Food and Drug Administration (FDA) now carry such pharmacogenomic warnings. But without knowing a patient’s genetic make-up, a doctor cannot act upon them.
A screenshot of the TreatGx software.
Decisions guided by software
Starting this summer, the Faculty of Medicine will try to make that information more available – and more actionable.
“TreatGx” uses a sampling of genetic data – culled from patients’ saliva samples – to try to reduce harmful drug reactions in people with one of 10 common conditions, including depression, hypertension, high cholesterol, gout and asthma.
TreatGx uses specially-designed software that considers the usual details about a patient: age, weight, blood pressure, allergies, other medical conditions, other drugs they might be taking. It also includes the latest reliable evidence about the safety and effectiveness of the more than 250 drugs involved in the treatment of those 10 conditions.
But layered on top of those inputs is genetic data about the patient in question – specifically, whether he or she has any of the 33 variations, spread across five genes, that might render a drug ineffective or even harmful.
Rarely is pharmacogenetic data obtained from patients who don’t have a life-threatening disease, predominantly cancer. TreatGx is one of the first attempts at providing that information to primary care physicians, in a clinically useful manner, for more common illnesses.
Three years in the making, TreatGx was conceived and designed by Martin Dawes, Head of UBC’s Department of Family Practice, working with colleagues from UBC’s Personalized Medicine Initiative, a not-for-profit UBC spinoff.
“We’re automating the tasks that can be automated,” Dr. Dawes says. “All of the information that underpins medication choices is out there, but it’s impossible for physicians to hold it all in their heads. We’re trying to simplify the process – and enhancing it with powerful genetic data that physicians, until recently, haven’t had at their disposal.”
This is where the future of medicine is going.”Bruce Hobson
The software, including the data about each patient’s genetic profile, is initially being provided to a dozen family physicians and pharmacists around British Columbia to help in their prescription decisions for 250 patients. If that three-month feasibility study goes smoothly, TreatGx will be expanded to cover 26 diseases, and will be marketed directly to consumers.
An avalanche of information
TreatGx’s use of genetic information would not have been feasible 10 or even five years ago, but rapidly improving technology has made genetic sequencing dramatically faster and cheaper. The first sequencing of the human genome, completed in 2000, took 10 years and $3 billion. Now it costs around $1,000 and takes a day or two. The analysis to be performed in TreatGx, limited to the 33 specific sites on a person’s genome, costs about $200.
Hagit Katzov-Eckert, a member of the TreatGx team, prepares samples for analysis. Photo courtesy of GenomeBC
That technological progress – not just in sequencing people’s genomes, but in the explosion of information about the function of individual genes, and the harnessing of computer hardware and software to produce useful information – is a major reason why some health researchers, including Pieter Cullis, Director of UBC’s Life Sciences Institute, are predicting the dawn of “personalized medicine.”
“An avalanche of molecular-level analyses of the bits and pieces that make up you is coming online,” Dr. Cullis, a Professor in the Department of Biochemistry and Molecular Biology, wrote in his new book, “The Personalized Medicine Revolution” (Greystone). “Within the next five years, increasingly comprehensive molecular tests will be available that will tell you, with ever-improving accuracy, what is wrong with you.”
The TreatGx trial is receiving funding from Genome BC, the Rx&D Health Research Foundation (a consortium of Canadian pharmaceutical companies) and additional contributions from pharmaceutical companies. It will be operated and marketed by GenXys Health Care Systems, a UBC spin-off company, which will conduct the genetic testing of participants’ saliva samples at its lab in the Faculty of Pharmaceutical Sciences building.
Some doctors might see TreatGx as undermining their autonomy, says Bruce Hobson, who has worked for 33 years as a family physician in Powell River. But if TreatGx’s recommendations are trustworthy, current and easily accessible during patient visits, the tool could remove much of the uncertainty surrounding prescription decisions, he says.
“This is where the future of medicine is going,” says Dr. Hobson, a UBC Clinical Instructor who leads a B.C. group working to encourage physicians’ use of electronic medical records and the sharing of that data to improve patient care. “Sometimes we feel very much alone when prescribing drugs. Having the most information possible when making those decisions, so that each decision is what is best for the patient in front of me, is the way things need to go.”