Wired.com, of all places, has an excellent article about the early diagnosis of cancer: Why Early Detection is the Best Way to Beat Cancer by Thomas Goetz. The problem is that early detection is a way to beat cancer, but it almost certainly is not the best way. First of all the article is very good and well worth reading. It is far more sophisticated than virtually anything one sees in the lay press.

The best way to beat cancer is not by treating it, no matter how early in its course, but by preventing it. Mr Goetz comes close to this approach in his discussion of Pap smears and cervical cancer – one of medicine’s greatest triumphs. Regular Pap smears result in the treatment of lesions of the cervix before they become malignant. Now with the understanding of the role of human papilloma virus in the pathogenesis of the disease new prevention strategies are possible.

The problem with early diagnosis of cancer is that it may not work in all forms of the disease. In some cancers the opportunity for cure may be gone as soon as the first malignant cell appears. Also early diagnosis of cancers that may cause little or no harm carries a risk benefit ratio that is unfavorable. Prostate cancer is a disease where the value of early diagnosis is fiercely debated. Goetz is fully aware of this problem.

Colon cancer is another disease where early diagnosis makes a positive difference, but it’s also a disease we know how to prevent. Colonoscopy, though inconvenient and expensive, often finds premalignant lesions the removal of which prevents the later development of cancer.

Screening for disease has a long and mostly unsuccessful history. Despite much effort there are only about three screening tests that are of unequivocal value. Two are mentioned above (Pap smears and colonoscopy) the remaining one is blood pressure measurement. Virtually everything else is debatable. For example do regular mammograms (though the procedure is standard of care) convey a benefit over self examination? Is the decline in breast cancer mortality due to earlier diagnosis or better treatment? The answer to these questions is still debatable.

Similarly, regular physical exams in asymptomatic patients convey no benefit. This doesn’t mean that in the future we won’t get new screening tests or procedures that are medically useful and cost effective. It just emphasizes how difficult it’s likely to be to get these tests and procedures.

Goetz discusses at length screening for ovarian cancer. He presents calculations that define the size of of an ovarian cancer small enough to cure if it could be diagnosed. The numbers given are entirely made up or if you prefer entirely an assumption. There’s no way to know at what size an ovarian cancer is too big to cure. It is possible that some ovarian cancers may never be small enough to cure. Thus its hard to know how valuable early diagnosis of this cancer will turn out to be.

But even if I’m completely wrong about all of this it should be obvious that of the two alternatives – treating a cancer that has already appeared versus preventing the cancer from developing in the first place – the latter is far better than the former. Given that for many cancers we can’t do either, I’d put my research money on prevention. No matter where you stand on the issue read Goetz’s article – it’s a gem.