Study Showing Long-Term Results after Prostactectomy

Thread view:

Enable EMail Alerts

Start New Thread

Thread rating:

Dave Perry - 31 Aug 2004 23:07 GMT

Dr. Bianco of Memorial Sloan-Kettering Cancer Center reported at the
annual meeting of the American Urological Association the results of a
study of 1716 prostatectomy patients who were followed for 1-20 years
(mean of 6 years). PSA levels rose in 16% of patients followed for 5
years, 22% followed 10 years and 27% followed 15 years. Prostate
cancer was the cause of death in 1% of patients at 5 years, 4% at 10
years, and 7% after 15 years. If those patients with PSA progression
of disease after prostatectomy, 23% died of PCa by year 10 and 39%
died by year 15. Also, patients with clinical stage T3, 56% remained
free of PSA progression after surgery at year 15 and 14% died of PCa.
Patients with Gleason score of 8-10 before surgery, 32% had no PSA
progression at year 10. Surgery seems to cure nearly 75% overall out
to 15 years. Finally, when PSA increased after surgery, it did so
within the first few years in 70% of cases and within 5 years in 90%
of cases. Reported in Internal Medicine News Vol. 37, No. 16.
Dave Perry

Reply to this Message

Leonard Evens - 01 Sep 2004 04:28 GMT

> Dr. Bianco of Memorial Sloan-Kettering Cancer Center reported at the
> annual meeting of the American Urological Association the results of a
[quoted text clipped - 12 lines]
> of cases. Reported in Internal Medicine News Vol. 37, No. 16.
> Dave Perry

Studies of this kind are always interesting, but you have to read them
carefully. Remember that they are written for a scientific audience who
will tend to know about the subtleties, but often they are taken up by
the lay press and the results can be misinterpreted.

There is one thing that sticks out. First, 15 years ago was 1989 and 20
years ago was 1984. Even 10 years ago was 1994. In addition, the
results are probably not completely current, which places the beginning
of the study even further back for some of the long term results. PSA
testing was first generally introduced in the US in the early 90s and it
took several years to become widespread. So many of these cases must
come from the pre PSA era. It is generally recognized that today with
PSA testing, prostate cancer is detected 5 or more years earlier than it
was back then. That means that more cases are discovered at an earlier
curable stage, and looking forward in the future, the figures will
probably look a lot better. Also, looking forward, even if you ignore
the previous comment, you have to start the clock 5 years earlier, and
that by iteslf shold improve the statistics.

Reply to this Message

Alan Meyer - 01 Sep 2004 04:57 GMT

> ... PSA levels rose in 16% of patients followed for 5
> years, 22% followed 10 years and 27% followed 15 years.

...

> Finally, when PSA increased after surgery, it did so
> within the first few years in 70% of cases and within 5 years in 90%
> of cases.

...

These numbers don't appear to add up.

If I'm reading correctly, at 5 years, 16% of patients
had rising PSA. At 15 years 27% of patients had
rising PSA.

It looks like 16/27 = 59% of patients with rising PSA
rose in the first five years.

But it also says that 90% of patients with rising PSA
had it rise in the first 5 years. Doesn't that contradict
the other numbers?

Leonard - you're a mathematician. Help us with this.

Alan

Reply to this Message

Leonard Evens - 01 Sep 2004 12:02 GMT

>>... PSA levels rose in 16% of patients followed for 5
>>years, 22% followed 10 years and 27% followed 15 years.
[quoted text clipped - 21 lines]
>
> Leonard - you're a mathematician. Help us with this.

Without reading the paper and doing the arithmetic, it is impossible to
know just what is going on. However, note that on the average patients
were followed for 6 years. So we are talking about different groups of
patients. It appears not to be the case that 16 percent of all patients
had rising PSA in the first five years and 27 percent of all patients
has rising PSA within 15 years. It may be the 70 percent and 90
percent numbers are estimates that they think would have applied if
every patient had been followed for the same period of time. But as I
said, I would have to read to paper. I am not likely to do that since
it is not available over the web, and I am not about to go down to the
NU Medical School to look it up any time soon.

> Alan

Reply to this Message