Click here to view next page of this article


Prostate Cancer

Prostate cancer is second in mortality in men. It has really come on down past gastrointestinal malignancy as the second most significant malignancy in males. There are over 30,000 deaths a year from prostate cancer. About 47,000 with breast cancer. The incidence of death from prostate cancer, the mortality rate, actually has gone down a little bit last year for the first time. So again, this is a parallel to the breast cancer data. There has been a slight drop in mortality despite the rising incidence of cancer of the prostate.

We think that not everybody is at equal risk for prostate cancer. As in breast cancer, we have been able to elucidate some genetic and other factors that seem to be associated with an increased risk of disease. The most pertinent of these if family history. We know that if you have one first degree relative with prostate cancer it does increase your risk you certainly have screening in those patients. If you have more than one relative then you have an even bigger effect on risk. We also know that there are racial differences in prostate cancer. This might be due to diet or habits, or to other genetic factors. We know that the bottom line is that African-American men have a two-fold increased risk of prostate cancer compared to their Caucasian counterparts. We think that increased levels of androgens, particularly 5-alpha reductase.

Now prostate cancer does not have, at this point, any prevention data, like the data I showed you from the P-1 trial for breast cancer. We will conclude by looking at some prevention opportunities but all we have right now is secondary prevention or screening for this disease. Again, as in breast cancer screening, I think this is an area that has become somewhat controversial. I think that itís fair to say that the PSA determination is the most sensitive test that can be used. But the specificity of PSA, particularly in mild elevations, is not optimum because you will have mild elevations - at least in non-prostatic hypertrophy. Digital rectal examination should not be forgotten. Remember that the posterior lobes of the prostate.

Remember also that cancer is not usually associated with symptoms of BPH. Symptoms of BPH result from having hypertrophy of the glandular tissue in the periurethral area and cancer usually involves the posterior lobes. So most people with cancer and having symptoms of prostatism simply have both. They donít have prostatism symptoms.

This would be one algorithm for screening that I think is reasonable. Now there has been a lot of argument about PSA level for screening and particularly whether it actually translates into any mortality advantage. There is no question that it is a sensitive test but what does it all mean? A large study from Canada that appeared to show a mortality reduction through the use of PSA screening in Canadian men. If we assume for the moment that there is a value to PSA screening in terms of how you would use this test, I would recommend something like this algorithm. That is, once you are going to start screening patients, and that would vary perhaps with the age and risk that a patient has. If you are going to start screening and the patient is normal risk without a family history, you could probably start screening - if you were going to do this and you believe in it - at about age 50. For patients who have a family history and are at high risk probably start a screening type of algorithm at perhaps age 40. The combination again of digital rectal exam and PSA I think is the way to go. Clearly if both of these are normal you would just simply re-screen the patient a year later. What more often happens is you have a normal rectal exam but the PSA is abnormal.

I already alluded to this and that is the problem of early diagnosis. There are several problems posed by early diagnosis which really basically equates with someone who has a PSA level thatís elevated, gets worked up and is found to have the disease. The issues of early diagnosis are: first of all, does it really improve survival? This has been argued about. As I said, the data presented at _ this year suggests that there is a mortality reduction with PSA screening, but not surprisingly that study was jumped on as soon as it was out of the box and attacked by many statisticians as being seriously flawed. I think that usually until you have a second study that reproduces positive data itís hard to be certain that a single trial is definitive. So I think there are still issues about whether these lesions that are found in PSA elevation only will translate into improved survival with treatment. Therefore we have still this year one of the most accurate measurements for early detection, given the specificity issues with PSA. Can we distinguish clinically significant disease from that which is not significant? And finally, what is the most effective form of therapy for these early stage patients? Particularly those found by PSA? One of the things that turns out to be very useful in assessing what you are dealing with is the use of tumor grading. And there are multiple ways you can grade these tumors. You can look at flow psychometry, you can get flow levels, you can look at nuclear morphology but the architectural formation, the so-called Gleason system I think has been the most useful to date in establishing the grade.

Again, before we had the Gleason score and the PSA level, we just looked at volume of disease. You can see that these are the patients detected by TURP chips, small amounts of tumor involved. Much better prognosis than more extensive disease. These are all people with no palpable disease. This is multiple chips involved.

So if you have prostate cancer, these are the choices that you have available to you. You have observations, surgical radical prostatectomy, external radiation, radiation implants, androgen depletion, some combinations of the above. As Iíve already said, patients who have favorable characteristics might be good candidates for observation only. With regard to the other therapies, we have some data, although it tends not to be randomized and prospected. This is pooled data from series that have either done prostatectomy or radiation therapy.