JESSICA LEEDER AND CARLY WEEKS
From Saturday's Globe and Mail
VANCOUVER and TORONTO For Canadian medical labs under fire for bungling tests and misdiagnosing cancer, there is a simple $600 solution.
It has been in the works for the past year, since a pair of quality-obsessed pathologists in Vancouver and Saskatoon combined with a determined technologist to pioneer a new, Web-based testing system they say will offer laboratories across the country a surefire way to improve accuracy at a bargain rate.
“It's innovative, it's unique in the world … it's led by people with the appropriate expertise,” said Blake Gilks, one of the founders and the head of anatomic pathology at Vancouver General Hospital.
The program, which has passed two pilot phases, is ripe to be implemented nationwide at a time when Canada's top pathology experts are calling for a national, government-backed strategy forcing labs in every province to routinely prove the accuracy of their most sensitive cancer tests. They say it is the only way to slash the dramatic inconsistencies in test quality from coast to coast.
And there's no time to waste: The way pathologists see it, they will play a growing role in cancer treatment as scientific developments widen their slate of sensitive tests.
“There is nothing nationwide and that's probably the most fundamental problem that we have,” said Sylvia Asa, pathologist-in-chief at Toronto's University Health Network, one of North America's foremost cancer research institutes. “There has to be some investments in programs like that.”
Until now, there has never been such a program to invest in. A firestorm has been raging for months, touched off by a Newfoundland and Labrador inquiry into breast-cancer testing errors that went undetected for eight years; during that time, hundreds of women were given incorrect results. Labs in Ontario, Manitoba and New Brunswick have launched subsequent probes into lab workers' errors.
However, because laboratory quality-control systems in Canada vary widely by province, it is impossible for officials to evaluate the true extent of bad testing.
Dr. Gilks said the new program could shed light on the depth of the problem. But the program has a problem of its own: It is now running off borrowed research dollars and the founders' personal accounts, and could flounder without official funding.
“I think there's a lack of commitment to quality assurance,” Dr. Gilks said. “We're so used to accepting lab tests as being accurate here that we assume it's just going to happen. It doesn't.”
Province by province
A Globe and Mail analysis of the various systems used to validate breast biomarkers – used by pathologists to break down the molecular characteristics of a cancer so treatments can be tailored to attack it – found broad discrepancies in safety measures taken by labs across Canada.
The easily bungled tests, which are part of a broader family called immunohistochemistry (IHC), have yet to be regulated. In Quebec and Eastern Canada, labs are not required to monitor their proficiency. In Ontario and Western Canada, provincial accreditation agencies require all IHC labs to participate in some kind of external quality-control program.
But concerns about whether errors are going undetected have prompted oversight agencies in several provinces, including Alberta and Saskatchewan, to begin informal reviews of their policies.
“It's somewhat mind-boggling to realize that in 2008 you could have provinces where there is no mandatory laboratory proficiency testing going on. I don't think that's acceptable,” said Dennis Kendel, registrar of the College of Physicians and Surgeons of Saskatchewan.
“The incredible patient harm that occurs when you don't get it right is something we need to do everything possible to prevent,” he said, adding: “Any process that involves human endeavour is going to have a certain error rate … we need to strive to bring it as close to zero as possible.”
That's where Dr. Gilks and his B.C. colleagues come in. Under their watch, rates of accuracy in B.C. are hovering around 95 per cent. That makes B.C. one of the best places in North America for women diagnosed with breast cancer, said Robert Wolber, who works with Dr. Gilks and is medical discipline leader for anatomic pathology at Vancouver Coastal Health.
“My goal is that through the next couple of years … we'll be able to say we do the best work on this anywhere in the world,” Dr. Wolber said. Along with his wife, technologist Beverley Wolber, and Dr. Gilks, he spearheaded an effort last year to force the province's accreditation agency to institute mandatory, provincewide IHC proficiency tests.
“Based on our own discipline, we realized we need to regulate,” Dr. Wolber said. “It's not because we're any smarter than anybody else. It's simply because we have a real sense of discipline around doing things right, as real science.”
Pushing for a standard
The seeds of that discipline were planted in the mid-1980s in Winnipeg, where a young Beverley Wolber (then Dupuis) was studying an emerging science called immunohistochemistry. Lured to Vancouver to help explore the use of IHC for cancer diagnosis, she met and fell in love with Robert Wolber, who had arrived at Vancouver General Hospital from the University of Michigan to study IHC.
She also discovered there was potential for widespread discrepancies with the tests: After sending a batch of tumour slides to several labs, she was shocked to learn not all labs were getting the same outcome.
“That's when we first started to find out that there were different results between sites,” she said. To try remedying the inconsistencies, Ms. Wolber began holding regular meetings for technologists involved in processing the tests.
John Garratt, a technologist at Vancouver's Lions Gate Hospital, was a participant in those early trouble-shooting sessions, which dropped off over the years. However, Mr. Garratt took it upon himself to jump-start the meetings again a couple of years ago after his colleague Dr. Gilks developed an easy, cost-effective means of quality testing.
Using tissue “microarrays” – thin slices taken from a paraffin block studded with 200 tiny cores of different tumour tissues – Dr. Gilks figured out a simple but cutting-edge way of sending a slew of tumour tissues to labs; the sheer volume of test tumours on the slide meant researchers would have a greater sense of the lab's true proficiency than in the past, when surveys were based on only one or two tumours.
After test runs proved the microarrays useful for surveying quality – and in catching misinterpretations at labs in a matter of weeks instead of months or years – Dr. Gilks and Mr. Garratt, helped by the Wolbers, began lobbying the provincial accreditation agency to institute the program.
Now, all labs doing biomarker testing have their results checked every three months. “You don't want to find out a year after you've been testing a whole bunch of women that you've got it wrong,” Dr. Wolber said.
Mr. Garratt said he was motivated to join the effort because patients should “know that professionals are looking after them well. I know it sounds a bit hokey, but it's just, what would you want your own wife or daughter to have? It's people's lives and health.”
Not long after helping to get the provincial program up and running, Dr. Gilks received a serendipitous phone call from Emina Torlakovic, a Saskatoon-based pathologist appointed to chair a Canadian Association of Pathologists committee pushing for national standards for IHC testing.
Dr. Torlakovic's drive to create a separate national quality-control agency to monitor labs inspired Dr. Gilks and Mr. Garratt – who is in charge of collating all the B.C. labs' test results into charts for statistical analysis – to join her effort to create a program that would link the provinces.
Their brainchild has aced two test surveys (results are slated for publication in the Canadian Medical Association Journal next month) in which volunteer labs from across the country participated.
Called Canadian Immunohistochemistry Quality Control, the program is built on a skeleton of B.C.'s provincial system – and on the founders' own time. To expand nationally, Mr. Garratt says labs would need to pay about $200 a survey, or $600 a year for three of them.
For now, when participants complete the free tests, the results are mailed back to Vancouver, where Mr. Garratt plugs them into a statistical grid the group has jokingly dubbed Garratt-o-grams. The stained slides are then passed on to Dr. Torlakovic in Saskatoon, who, using borrowed Web space, scans images of them onto the Internet for participants to view and compare.
Dr. Torlakovic, who moved to Saskatchewan from Norway after she helped set up an external quality-control agency for Scandinavian countries, said theirs is the only independent quality-control program that posts results of the test surveys online. That has promoted some friendly competition among laboratories, but also vital exchanges of protocols between labs in need of better techniques.
“I would hope that it becomes the Canadian standard,” Saskatchewan's Dr. Kendel said.
Although only 17 labs participated in the most recent survey, dozens more have since contacted Dr. Torlakovic about signing on, providing proof of the need for a properly funded national quality-control agency. “There should be one program available to everyone in Canada that has a Canadian edge,” she said.
Still, ramping up for the program's third pilot run, its creators are worried they've approached the limit of how far they can stretch borrowed equipment, no funding and limited volunteer time.
“It's not sustainable unless there's a way of resourcing it,” Mr. Garratt said. “It's frustrating knowing it could just disappear… Why should somebody have different outcomes for their cancer just because they live somewhere else in this country?”