Friday, August 29, 2014

PSA: Beyond Logic

In a recent Medscape article on the PSA debate the author states:

Cost is another issue. It is estimated that to prevent a single cancer death through screening and treatment costs more than $5 million. Could those dollars be better used to have a greater impact on society? Because although we may prevent a man from dying from prostate cancer, in many cases that doesn't mean that he is going to live a whole lot longer than he would have lived anyway. We have a net dilemma in trying to resolve these controversies. They are not going to be resolved tomorrow. Ultimately, we can hope that 2 things occur. One, going forward, it would be great if the gene tests that are in development turn out to be able to tell us which men need to be diagnosed and treated and which men can be spared. Two, we need to ensure that a similar fiasco does not occur. By that, I mean another screening test for early cancer that is not specific for that cancer being developed, without the right studies being done before it is used. We must avoid approving tests when we don't know for certain that they will be associated with more benefit than harm.

Let me now address each of these assertions:

1. Cost: This calculations makes a set of assumptions which in my opinion as discussed before herein are incorrect. We know the mortality rate of PCa and the costs associated with it. We also know that PSA velocity and % Free are excellent when combined with PSA. What we do not know is when we see a Gleason 6+ cancer whether it is aggressive or indolent. That is a genetic problem. Thus lacking such knowledge it is impossible to make a financial assertion of this type.

2. Genetic Tests: BRCA is a reasonable test for Breast cancer. The problem is that many PCa are not BRCA like, they may be methylation like and thus one needs to sample the actual cancer cells, not just one but many, and many individually to seek a stem cell as well. Not a simple problem.

3. Many tests lack great sp0ecificity. Welcome to testing. We cannot only use tests that  are perfect, medicine is never really like that. We never can give an answer to a patient who says "Doc, how long do I have?". No one patient is the same to all others.

Thus when we see statements like this we should take them with a grain of salt. We use the tools at hand and learn, many of the tools are poor, but they do work, somewhat. Perfection we leave to zealots.