10 March 2022

Less risk of breast cancer overdiagnosis

Cancer Clinical

The overdiagnosis of breast cancer through mammographic screening may not occur as frequently as feared, according to new research.


The overdiagnosis of breast cancer through mammographic screening may not occur as frequently as some research has suggested, a new study has found.

Previous studies have estimated that as many as 30% of all breast cancers in the US were incorrectly identified as potentially malignant and patients were subjected to unnecessary treatment – however the researchers in this study estimated the number to be about half that.

The study, published in the Annals of Internal Medicine has been welcomed by Australian researchers, who say the results are similar to Australian estimates and reinforce the importance of breast cancer screening, while recognising the possibility and potential impact of overdiagnosis.

Professor Paul Glasziou, director of Bond University’s Institute for Evidence-Based Healthcare, said the paper provided “interesting modelling,” which explicitly allowed for a non-progressive cancer.

“I think the importance this modelling is that it will allow us to understand the implications for overdiagnosis of different configurations of screening (age, frequency, etc),” he told The Medical Republic. “There have been a range of overall estimates (of overdiagnosis) – from 1% to 31% – so their estimate of 15% is in the middle of that range, and comparable to our estimate of 22%.”

Screen-detected cancer cases determined to be overdiagnosed included 6.1% that were indolent preclinical cancer and 9.3% that were found to be progressive preclinical cancer in women who would have died of an unrelated cause before clinical diagnosis.

Professor Glasziou said this split was supported in Australian studies.

“For us, the equivalent split is into 9% and 13% rather than 6% and 9% – so very similar,” he said.

As part of the recent US study, researchers studied data from Breast Cancer Surveillance Consortium facilities to estimate the rate of breast cancer overdiagnosis in contemporary mammography practice among almost 36,000 women aged 50 to 74 years which included over 82,600 mammograms and detected 718 breast cancers.

They determined that in a two-yearly screening program about one in seven cases of breast cancer would be overdiagnosed. This equated to about 15%, half the number previously reported in other studies.

“Overdiagnosis is considered one of the chief harms of mammography screening, because of the burden of the negative consequences that are associated with unnecessary tests and treatments,” the study author said. 

“We have known for a long time that the most prominent estimates of breast cancer overdiagnosis in the US were unrealistically high. We hope that our overdiagnosis estimate of 15% amongst screen-detected cancers will be adopted as a new standard and will lead to better-informed decision making around breast cancer screening, follow-up testing and treatment.”

Researchers estimated about a third of overdiagnosed cancers were non-progressive, small tumours that would never have grown or caused a problem, and the remainder were progressive, but particularly in older women, would not have led to symptoms within the women’s lifetime.

Epidemiologist Professor Alexandra Barrett, who researchers cancer overdiagnosis at Sydney University, said there were different methods of estimating the potential rate of overdiagnosis through breast cancer screening.

“I’ve done two different methods so far. One of them got around 30 to 40%. The other one got around 20%,” she said.

“So, I would have thought that 15% is not completely out of the ballpark.”

She said the study also highlighted the differences in overdiagnosis rates by age groups, including 11% at age 50 and 23% at age 74. It also demonstrated the inevitability of overdiagnosis in screening for several cancers.

“This is inevitable with screening. We see it in screening for prostate cancer, lung cancer, thyroid cancer and breast cancer. All these cancers can be detected and are preferentially detected by screening,” she said. 

Professor Barrett was involved in a large, randomised trial published in The Lancet in 2015, in which researchers told women who were turning 50, and would be soon invited to screening, about the risk of overdiagnosis. 

“What we found was that compared to the control group, the women who received the decision aid that explained overdiagnosis were better informed,” she said.

“They were slightly lower in their intention to get screened. But after two years, the participation rate and screening were actually the same in both groups. So, we know that telling people doesn’t cause them to stop. It might make a few stop, but the vast majority people will continue to be screened but the difference is they will know (of the risks of overdiagnosis).”

She said the study served to highlight the importance of informed consent in breast cancer screening.

“I do think that there’s a need in the professional community for more acceptance of what evidence shows about breast cancer screening, because I don’t think even now, despite mountains being published on it, that that knowledge that breast cancer can be overdiagnosed, just like prostate cancer can, but at a lower frequency,” she said. 

“We just need to recognise that this is this is inherent in a cancer screening program.”

Annals of Internal Medicine 2022, online 1 March

The Lancet 2015, online 18 February