Researchers say a decades-old Canadian study that influenced breast cancer screening policy found significant flaws in its approach, leading to a “substantial impact” on disease outcomes and possibly contributing to hundreds of “preventable deaths” each year.
Two studies — collectively known as the Canadian National Breast Screening Study (CNBSS) — found that mammograms for women in their 40s did not reduce breast cancer death rates.
The study was conducted in the 1980s and published in 1992.
But in comments published this week in the Journal of Medical Screening, researchers from four Canadian institutions and Harvard Medical School say how participants were selected for the study’s control or screening groups may have influenced the results.
They say more recent findings suggest that mammography screening has benefits for women under 50, including a 2014 observational study that found giving mammography to people in their 40s was associated with a 44 percent reduction in breast cancer deaths.
The Canadian Task Force on Preventive Health Care does not currently recommend mammograms for people ages 40 to 49 unless a pre-existing factor puts them at higher than average risk, such as if a family member has breast cancer or if they have the BRCA gene.
The Task Force said on Wednesday that their guidelines are “not scheduled for an immediate update.”
But Martin Yaffe, lead author of the commentary and senior scientist at the Sunnybrook Research Institute in Toronto, believes change is needed. He estimated that the CNBSS’s influence on policy may have contributed to the preventable deaths of more than 400 Canadian women each year.
“The idea of doing this trial was great, but the way it was designed and the way it was conducted really doesn’t make the results believable,” he said. “And basing policy on that is simply inappropriate.”
dr. Brenda Wilson, co-chair of the Task Force on Preventive Health Care, said the organization’s recommendations, last updated in 2018, “have been recognized as the best in the world.” She added that the group “conducts rigorous, detailed evidence reviews to formulate guidelines.”
“If there are substantial changes to that evidence, the Task Force will update a full review of the evidence, including new evidence,” she said in a statement.
The Public Health Agency of Canada released a statement Thursday saying it funds the Task Force and referred to the body as “arms-length from the government”.
“It would be inappropriate for PHAC to lead the Task Force on which studies should or should not be included in their guidelines,” the agency said.
“The Task Force assesses the strength of all the evidence used to develop their recommendations. The study described in the commentary is one of eight studies included in the Task Force’s guidelines.
The main problem with the 1992 study, Yaffe said, was that participants were given clinical breast exams at 14 of the 15 trial sites before being assigned to the screening or control groups. That approach, he said, may have unintentionally affected the results.
He said the nurses who performed the physical breast exams assigned participants to their groups by writing their names in an open book. If a nurse felt lumps in a woman’s breasts — which could indicate advanced cancer — she probably would have put her in the screening group earlier “for the sake of the patient, with all goodwill, but not understanding how clinical trials work.” .”
“There was a huge imbalance in the number of advanced cancers found on the mammography side of the study compared to the control side … so (the study) found no benefit from screening,” Yaffe said.
“They even found that more women died on the mammography side than on the control group, which was bizarre because every other study in older women had shown a benefit from mammography.”
Yaffe said he suspected the study’s methods had been flawed for years, but the “smoking gun” didn’t come until March when eyewitness statements from a staffer at one of the trial sites confirmed that the randomization may have been flawed.
The size of the study — nearly 90,000 participants — gave the findings weight, Yaffe said, one reason it may have influenced policy around the world.
“But if the trial was done poorly, if it was done a long time ago with approaches that are no longer used, it may not be relevant,” he said.
PHAC said mammogram screening policies are “under the purview of provinces and territories.” Jurisdictions can use the Task Force’s guidelines, but they also establish their own screening programs.
Ontario, for example, does not screen patients under the age of 50, but Nova Scotia residents between the ages of 40 and 49 can self-refer for annual mammograms. British Columbia says it “encourages” people aged 40-49 to talk to their doctor about the benefits and limitations of mammography. If screening is chosen, it will be available every two years.
Jurisdictions screen at-risk patients under the age of 50, and people who suspect something is wrong can get a mammogram referral from a doctor.
dr. Jean Seely, co-author of the most recently released commentary and chief of breast imaging at Ottawa Hospital, said the policy should be updated to allow all women aged 40 or older to undergo screening mammograms.
“Screening saves lives,” Seely said in a release. “There is a five-year survival rate of 98 percent for localized breast cancer when detected early.”
This report from The Canadian Press was first published on November 25, 2021.