Friday, April 27, 2007

4 Comments:

"The question still remains, however, whether or not finding slow growing prostate cancers early makes any difference."

Watchful waiting on prostate cancer is a fools game.

Most American men don't fall for the waiting game and only 13% of our prostate cancer victims die of the disease. Most of the watchful waiting propaganda comes from Europe where governments don't want to spend the money. So you end up with these death rates from prostate cancer:

France 49%
UK 57%

Sloan-Kettering has a nanogram on their web site that shows the expected cure rate if you enter PSA, Gleason score and the size of the cancer found in the biopsy needle. The cure rate starts at 96% which is not hard to achieve if the doctor is properly watching the PSA velocity and recommends immediate treatment. Increase any of those factors and the cure rate starts dropping, and it drops fast.

Tell a prostate cancer victim to wait and you are likely to condemn him to a painful death.

By  Jake, at 9:32 AM  

Jake,

That's true if the prostate cancer is aggressive. Most aren't. It's trying to figure out which ones are worth the radical prostatectomy and which aren't that's hard.

Do a radical prostatectomy on a man who has an indolent prostate cancer which isn't likely to grow and you may needlessly condemn him to a life of impotence and urinary incontinence.

It's impossible to tell who will have the aggressive cancer and who won't. The majority aren't aggressive, but you hate to make the wrong call on the minority that are. That's what makes it such a tough call. It remains to be seen whether or not this test will make those decisions any easier.

By  Sydney, at 11:36 AM  

I agree that a radical prostatectomy should be avoided if possible, but waiting is not the way to accomplish that goal. Early treatment is.

If you catch prostate cancer early enough, you can use computer focused radiation (especially proton radiation) to kill the cancer cells. The cure rate is better than the best surgeons achieve, and the side-effects are few if any.

By  Jake, at 12:52 PM  

Jake, you cannot just compare percentage of people die from those diagnosed to prove your point unless you know the extent of overdiagnosis. Ever learned division in elementary school? Like when your nominator is the same, but your denominator is larger, the result of the division is smaller? So even if screening had absolutely no effect (and we don't know at this point that it does), because the screening discovers a lot of cancers that would've never been discovered in person's lifetime if remained undetected, the number-people-who-die/number-of-people-diagnosed would still be smaller. Because you are dividing by larger number.

Let's say, for example, you have n cases of cancer that is going to spread and m cases of overdiagnosis, and let's take extreme case where screening has no effect i.e. the same number of people -x die in both cases. Then x/n is always larger than x/(m+n), so the rate looks better among screened population, but the number of people who die is still the same.

The only true measure of test's effectiveness is comparing how many people die from prostate cancer in randomized trial where one group is screened and another is not. There is no such data in case of prostate cancer and PSA.

Orac of Respectful Insolence had a couple of very informative posts on early detection - read them. And learn some elementary school math.

By  Anonymous, at 10:30 PM  

Post a Comment