Overdiagnosis in common cancers

Overdiagnosis means that a cancer is diagnosed in someone who would never have had symptoms or come to any harm had the cancer not been found. This is clearly a very tricky conversation, and one that most of us find uncomfortable. We worry about cancers not being found, not about cancers that are identified, but never really needed to be.

A recent study from Australia examined 30 years of health care data and concluded that 18% of cancers found in women and 24% of cancers found in men fell into this category. The cancers included breast, thyroid, renal, prostate, and melanoma. Most were found from common screening tests (think prostate specific antigen, or PSA, for men or mammograms for women). Some were found during physical exams, and some were incidental findings on scans or studies ordered for entirely different reasons.

If you like numbers, here are some more: The overdiagnosed cases in women accounted for 22% of all breast cancers and 73% of all thyroid cancers found. In men, the stats are 42% of prostate cancers, 58% of melanomas, and 73% of thyroid cancers. This seems like a lot.

Does this matter? Why does it matter? Clearly it matters a great deal if a cancer is missed, and someone blithely goes on with her/his life until that same cancer is found at a later and more dangerous time. The authors contend that the issue is due to the fact that scans and other studies sometimes find cells that look like cancer, but don't end up acting that way.

Here, at least for me, is where things really begin to be complicated. We have done a good job in the United States and other well-resourced nations in public health education. Most people now are aware of the importance of early detection for most cancers and that the best chance of that is through screening. This report suggests that a fair number of the cancers identified could just as easily be left alone and that the person would have exactly the same health and outcomes — while avoiding the toxicity and unpleasantness of cancer treatment. An easy example to consider is the rising number of PSA tests, and the existing option, for some men, to choose watchful waiting rather than surgery or radiation therapy with the serious side effects those treatments often bring. Some well-informed men opt not to have their PSA measured, and it is hard to argue that this is often not a wise choice.

Let us think about mammograms. We know that most suspicious areas seen on these scans turn out not to be cancer. Some are found to be abnormal cells, such as atypical hyperplasia, that have the possibility of eventually evolving to a malignancy. The catch is that does not always happen. How do we know which areas should be excised to spare a woman a future breast cancer and which could be ignored? Further along the breast cancer spectrum is ductal carcinoma in situ (DCIS); some doctors feel that this condition should not be called cancer. We know that some DCIS will eventually break through the ducts and become invasive breast cancer, but we also know that some won't. Personally, I suspect that most women don't want to take that chance.

But should we be offered that chance if something were found that, in the experts' best guess, never would cause us any harm? There certainly are people who would opt to avoid the known side effects of cancer treatment, and there are plenty of others who would choose to proceed. Once more, I don't have any clear answers here, but this is all food for thought and issues to consider and more reasons to hope that, in the future, the best choice will always be clear.