More PSA

There are some fascinating aspects to the PSA discussion. Many believe that testing for PSA finds disease that has no implication to health. That is, the small is not indicative of the large. A biopsy of the prostate showing cancer does not mean that the individual with that biopsy is at risk from cancer--but not always.
The graph below from the New England Journal of Medicine shows that the time after the introduction of PSA testing resulted in dramatic drop in prostate cancer deaths. (The other side of the argument is that many were treated who did not need it.)
Another thing from the graph: Look at the virtual non-impact of breast cancer survival with the introduction of mammography.

P.S.A., Prostate Cancer, and the NYT

The NYT and WashPo recently published the results of a review done by a task force on the value of P.S.A. testing. (P.S.A. is a protein --Prostatic Specific Antigen--made by the prostate that gets into the bloodstream depending on number of blood vessels, their permeability and age. It rises in many instances of prostate cancer, as well as other conditions that increase vasculature and vessel permeability.)The report is astonishing in a number of ways and offers remarkable insights into problems facing the United States and its fragile relationship with science.

The task force, The United States Preventive Services Task Force, is an independent panel of experts in prevention and primary care appointed by the Department of Health and Human Services. They offer opinions on medical practice in the United States. While these positions are unsolicited, they often determine the approach taken by Medicare with regard to standards in health care management and, significantly, payment for testing. (A few years ago the task force determined that mammograms should been done only every two years rather than the traditional one year. The limits of funding followed.) In this particular task force study, the complications of therapy for the prostate cancers discovered was seen as unjustified when compared to the improvement in survival rate, which was seen as slight. In essence, P.S.A. elevation was seen to result in complications from biopsies and eventually complication from therapy when the biopsy was positive and no advantage in life expectancy was seen over a ten year period.

Several of the studies were particularly interesting as they compared tested and untested groups rather than diseased and healthy groups--they used testing as a proxy for disease found. So the Scandinavian study compared groups tested for P.S.A. against those not tested for P.S.A. and compared their life expectancy over seven to ten years. Men declining biopsy as well as those with negative biopsies were included in the tested group. As elevated P.S.A. is associated with negative biopsy 70% of the time, the tested group give a muted idea of the disease. But most complications of biopsy--like blood in the semen at 50%--while harmless are reported as serious complications.

Another factor not considered is the target population of biopsy. Most agree that clinical prostate cancer, that is cancer that is symptomatic and/or can be appreciated on examination, does not limit the life expectancy in men over the age of 70 in most cases but does in those men under 70. However, P.S.A. elevation--when associated with prostate cancer--precedes clinical findings by 5 to 7 years. Thus the target group for physicians treating prostate cancer is men under the age of 65 or so. Another group is those with rapidly growing disease in the generally low risk populations as P.S.A. changes often reflect that. Targeting these groups is more than reasonable; in a civilized society that understands the implication of illness and feels a responsibility towards its victims, it is obligatory. Another important factor is the development of symptoms which can be decreased or eliminated by early detection; thus survival rate does not tell the whole story.

What is particularly galling about these haphazard investigations and reviews is that they are done by seemingly reasonable people, people who should know the limits of their studies and the difficulties inherent in making good scientific generalities. Perhaps the makeup of the task force is illuminating: There is no one on the task force collecting information on P.S.A. and prostate cancer who clinically evaluates or treats either.

In essence, for the sake of unbiased neutrality, expertise and knowledge on the task force were screened out.

Prostate Cancer and The New York Times

An article in the New York Times of August 30, 2010 discusses the diagnosis of prostate cancer through the lens of a new book entitled "Invasion of the Prostate Snatchers" by Ralph Blum (a cultural anthropologist) and Dr. Mark Sholtz (an oncologist). http://well.blogs.nytimes.com/2010/08/30/a-rush-to-operating-rooms-that-alters-mens-lives/?emc=eta1

The scene depicted is not pretty. The gist is that the vast majority of patients with prostate cancer are over treated, that 80% of the surgery done for the disease is unnecessary and that a large motive for treatment is financial remuneration. Radical prostatectomy, the commonest surgical treatment is said to lengthen the life expectancy of one in forty-eight patients.

As a disclaimer, I have not read this book but the picture presented here deserves analysis. Some background studies--
Random autopsies show a rate of unexpected cancer of the prostate in about 30% of men in the groups around 45, 55 and 65 years of age. It trends higher in the age of 75 and higher in the 80's.
PSA, a protein made in the prostate at a predictable rate, rises in situations where blood vessels are increased in number, fragility and porousness and, for one or more of these reasons, shows up elevated in some prostate cancers.
The diagnosis of prostate cancer has historically been made on physical examination where the disease is found as a hard lump on the prostate. The prostate, when the exam is normal but the PSA is elevated between 4 and 10, has a positive biopsy in about 30% of cases.
PSA elevation generally precedes clinically detectable prostate cancer by 5 to 7 years. In clinically detectable disease, life expectancy without treatment in men under about 70 is less than those with treatment over ten years; in men over 70 the life expectancy is about the same.

Clearly, 30% of men under the age of 70 are not dying of prostate cancer but if the disease that progresses to clinical findings can be identified in these men, that should help. A big problem is this group cannot be identified. PSA is generally seen as evidence of tumor activity but that may not be so. Biopsy of men with elevated PSA may find a disease that was sleeping and will sleep on. Some evidence of aggressiveness is often assumed from the Gleason Number, a rather subjective grading of the tumor tissue's histological deviation from the norm, and further hints of presumed behavior can be gleaned from volume of disease in the biopsy and multiplicity of sites of involvement, again generalities. Youth with the diagnosis is a negative.

There is nothing written in stone here: This is a disease of context, indeed many diseases are. Chicken pox is annoying in children, fatal in the elderly. A heart attack might pass unnoticed in an older man where the younger man is struck dead. Who survived the Black Death and why? Current medical thought sees prostate cancer as a mosaic, a collection of tendencies, clinical leanings and statistics from which decisions have to be made.

And it is hard. But articles like this make it harder. And presumed scientists often do not help. Last year a study showed up from Europe that looked at two groups of men, those in whom PSA studies were done and those who did not have the studies. Using this information as a proxy for prostate cancer and the value of finding it early with PSA testing, the scientists followed the two groups for up to seven years and found no significant difference in survival rate. Their conclusion: PSA testing--early detection of prostate cancer--was of no value. But PSA elevation precedes clinical disease by 5 to 7 years and clinical disease takes a while to kill. How was this study meaningful over such a short time frame? It wasn't. And the scientists knew it. Why, then, did they publish it? Why, indeed.

How the authors of this book determined that 80% of surgery was unnecessary is not clear from the article but it will be wonderful to learn. This difficult mosaic, which so many struggle with, will finally be made clear. At least I hope it will be--although I am not sure where these insights have been hiding. I hope this is not just another financial enterprise created to take advantage of the classical American under education in science, preoccupation with conspiracies, slavish devotion to sensationalism and confusion of disrespect with independence.