Research Profile - Adding up the unknowns
Dr. Gregory Zaric
Computer modelling can put a price on the use of new genetic tests to solve the mysteries of cancers of unknown origin.
The Canadian Cancer Society estimates that 177,800 new cases of cancer surfaced in Canada in 2011. Of those, between 3,500 and 9,000 likely were metastatic carcinomas with no clear source of origin and no apparent optimal treatment.
Classified by the cryptic term “cancers of unknown primary,” they are a nightmare to a patient, an enigma to an oncologist and a major expense to the health care system.
Currently oncologists who come up against cancers of unknown primary are “left to use very general treatment regimens known to work with different cancer types,” says the University of Western Ontario’s Dr. Gregory Zaric.
At a Glance
Who – Gregory S. Zaric, Canada Research Chair in Health Care Management Science, Associate Professor, Ivey School of Business and Schulich School of Medicine and Dentistry, University of Western Ontario.
Issue – Treating cancers of unknown origin, which constitute 2–5% of all new cases, is difficult. Without knowledge of the original source, oncologists are spending time and resources on potentially inappropriate general treatments. However, genetic tests can sometimes help identify the best treatment for a patient.
Approach – Dr. Zaric’s team is using computer modelling to calculate the cost-effectiveness of implementing two new genetic testing products. One, called the Tissue of Origin Test, is FDA approved in the United States while the other, CupPrint, is available in Europe.
Impact - Results of the study will be published and made available to policy makers to help them decide whether to fund genetic testing to diagnose cancers of unknown origin.
But this kind of generic therapy may be a distressing waste of a patient’s time and the health care system’s limited money. Likewise, there are also the high costs of the “head-to-toe” imaging that is typically deployed to try and identify the cancers.
Genetic testing holds the potential to solve the dilemma by changing how cancer is diagnosed and treated, but integrating the new technologies into established clinical practice is still a work in progress, says Dr. Zaric, a Canada Research Chair in Health Management Science.
Funded by CIHR, Dr. Zaric is leading a three-year investigation into the cost-effectiveness of using two recently developed genetic tests to identify primary tumours in metastatic carcinomas. In essence, the tests track the activity of a number of genes in a sample tumour in an effort to identify the primary cancer type. But the tests are not perfect – the margins for error (a false or inconclusive result) vary from about 12% to 17%.
“The tests can cost between $3,750 and $4,600,” says Dr. Zaric. “Do you spend money for everybody to identify about 85% of all patients and direct them to better treatment regimens that might lead to more desirable outcomes? We’re trying to capture that trade-off – the costs and the health effects involved.”
Dr. Zaric is using a computer-assisted process called decision analytic modelling to weigh the benefits of these tests against their costs.
“You can imagine it as a thought experiment in which you have one world where people would be treated according to status quo. We infer what that looks like by examining health administrative databases from across Canada. Then we can make a comparison with a hypothetical world in which this technology is available.”
To create that hypothetical world, Dr. Zaric merges data compiled during the genetic tests’ clinical trials with data from provincial health care records. Using this data, he estimates what the impact would be if patients with unknown cancers underwent the tests and received potentially more targeted treatments.
“We can take that information and share it with people in ministries of health who need to make a decision about whether to start paying for these tests. They can take our number and see how it compares against what they are currently paying for.”
For radiation oncologist Dr. George Rodrigues at the Lawson Health Research Institute and London Health Sciences Centre, this kind of research is vital.
“We know that not all cancers are the same,” says Dr. Rodrigues who is also a researcher with the project. “There are underlying genetic issues for not just the tumour but the patient in how they respond to treatments. We need this research because more of these clinical tests are coming into practice as we head more and more towards individualization of treatments.”
"There is a limited amount of money that Canadians can spend on health care. Ultimately what we're trying to figure out is: do these tests fall into the set that constitutes the best possible use of resources?"
– Dr. Gregory Zaric, University of Western Ontario