Thursday, October 22, 2009

On Screening For Prostate and Breast Cancers

The American Cancer Society is modifying their previous whole-hearted support for screening in the case of certain types of cancers, including breast cancer:

The cancer society's decision to reconsider its message about the risks as well as potential benefits of screening was spurred in part by an analysis published Wednesday in The Journal of the American Medical Association, Dr. Brawley said.

In it, researchers report a 40 percent increase in breast cancer diagnoses and a near doubling of early stage cancers, but just a 10 percent decline in cancers that have spread beyond the breast to the lymph nodes or elsewhere in the body. With prostate cancer, the situation is similar, the researchers report.

If breast and prostate cancer screening really fulfilled their promise, the researchers note, cancers that once were found late, when they were often incurable, would now be found early, when they could be cured. A large increase in early cancers would be balanced by a commensurate decline in late-stage cancers. That is what happened with screening for colon and cervical cancers. But not with breast and prostate cancer.

Still, the researchers and others say, they do not think all screening will — or should — go away. Instead, they say that when people make a decision about being screened, they should understand what is known about the risks and benefits.

What is happening here? The researchers speculate that screening might pick up a lot of "innocent" tumors: tumors which might never grow or which might even go away on their own, but that dangerous tumors might grow too fast for screening to be of much use. This might mean that much of the treatment of cancers found in early detection could be excessive:

Overdiagnosis and overtreatment as a result of cancer screening are a major concern. It is estimated that for every 100 women who are told they have breast cancer, as many as 30 have cancers that are so slow-growing they are unlikely to be life-threatening.

In the case of prostate cancer, for every 100 men with diagnoses, as many as 70 have cancers that if left untreated would never have harmed them. Even for men with aggressive prostate cancer, whether screening improves the odds of survival remains a matter of debate.

That excessive treatment is not desirable should be obvious. Who wants to be cut or poisoned or radiated if it is not necessary for survival? Such excessive treatment also costs money. The obvious problem here is how to distinguish the "innocent" tumors from the ones which do demand immediate attention.

As an aside, the professional views on screening for cancers have always been much more complicated than the recommendations consumers get. To give you an example of that, suppose that some type of cancer can be detected very early with screening but that knowing about its existence doesn't ultimately help with treatment or survival rates. Suppose, also, that two individuals both develop the cancer at the same time and live equally long with it, but one gets screened early and the other one only gets diagnosed at a late stage in the disease. Any crude comparisons of these outcomes would then suggest that screening helps people with cancer live longer, when that is not in fact the case.

That example does not mean that screening wouldn't be beneficial if early detection allows for better intervention methods. But finding something early is not a treatment in itself.