Screening distorts diagnosis of cancer

What's in a name? Prof Alexandra Barratt says we need to reconsider the word cancer.

Earlier this year, leading American cancer scientists called for a set of changes to deal with the problem of over-diagnosis and over-treatment caused by cancer screening.

Efforts to raise awareness and detect cancer early through screening have led to significant increases in early-stage cancer diagnoses, without delivering a proportionate drop in advanced cancers.

After three decades of breast cancer screening, the United States has seen a large increase (100%) in the rate of early-stage breast cancer, but only a small decrease (8%) in the rate of advanced or late-stage cancer.

This strongly suggests overdiagnosis is the culprit, and could account for nearly a third of all newly-diagnosed breast cancers.

In response to this emerging problem of screening programmes detecting harmless cancers, the scientists made a set of recommendations to address and mitigate overdiagnosis.

One of their recommendations is a major rethink about the use of the word cancer when talking about screen-detected abnormalities.

The word cancer, they wrote in the Journal of the American Medical Association (Jama), should be reserved for life-threatening cancers, that is, ''lesions with a reasonable likelihood of lethal progression if left untreated''.

Early cancers and pre-cancers (abnormal cells that could turn cancerous) found by screening tests, such as mammograms and PSA tests, should be renamed without (scary) words such as carcinoma or neoplasia in their title.

They suggested they could be renamed IDLEs - indolent lesions of epithelial origin.

Chief medical officer of the American Cancer Society, Otis W. Brawley, noted: ''We need a 21st-century definition of cancer instead of a 19th-century definition of cancer, which is what we've been using.''

One example of a cancer that's a candidate for a name change is ductal carcinoma in situ (DCIS) of the breast.

Diagnosis of DCIS, which itself does not metastasise or cause death but can be a precursor of invasive breast cancer, has soared since screening began. It now accounts for about 20% of the breast cancers found by screening, compared with about 2% of all breast cancers in pre-screening days.

DCIS is treated very much like breast cancer so it's understandable that women diagnosed with DCIS may not understand that they don't have invasive breast cancer. Renaming it might reduce this confusion.

In other suggestions for understanding and mitigating overdiagnosis, the scientists recommended the establishment of registries of IDLE lesions to record detailed information about the diagnosis (including pathology and molecular biology) and treatment of screen-detected early cancers.

The information collected and analysed would be invaluable to help us understand how fast and how often IDLEs progress to advanced cancer, or regress.

It would help us assess new screening tests, and evaluate the new molecular and genetic tests that aim to distinguish indolent screen-detected cancers from their more aggressive relatives.

These registries of screen-detected abnormalities could extend the work of existing cancer registries, which already collect limited information about all cancer cases in order to monitor trends in diagnoses and mortality rates of different cancers.

The panel also said it was essential the community (both medical professionals and citizens) recognise that overdiagnosis exists, and start to talk about it with more understanding.

The National Cancer Institute scientists note: ''Physicians, patients, and the general public must recognise that overdiagnosis is common and occurs more frequently with cancer screening. Overdiagnosis, or identification of indolent cancer, is common in breast, lung, prostate, and thyroid cancer. Whenever screening is used, the fraction of tumors in this category increases.''

Overdiagnosis is one reason why five-year-survival rates are misleading when it comes to cancer screening. By adding harmless cancers to the number of cancers diagnosed, overdiagnosis ensures five-year-survival rates improve, even if just as many people still die from cancer.

Until doctors and citizens alike have a better understanding of overdiagnosis, we are at risk of being misinformed and misinforming others about cancer screening.

Cancer screening may deliver benefits, but we must also recognise its harms, risks, and opportunity costs.

The National Cancer Institute has provided suggestions about how to go forward. We should pick up the ball and run with it.

Alexandra Barratt is a professor of public health at the University of Sydney.

Add a Comment