You ignornace is showing up once again.  The fact that life expectancy is
going up in industrialized nations is not due to the survival of 3% of
prostate cancer non-survivors after 10 years for a short period of time.
The gradual decline of mortality rates for nations as a whole is not
affected by what you post...it could not even be detected.  But as life
expectancy goes up, disease death rates have to decline.  You also don't
know the difference between age-specific rates and overall rates either.
Nor do you bother to notice that many nations do NOT test non-symptomatic
men for prostate cancer, but also have longer life expetancies.  I suggest
you get a good book on life tables for a start and maybe even read the
article you cite.

The issue is whether or not declining mortality due to prostate cancer
provides contrary evidence to the assertion that biopsy in itself
increases the risk of metastatic prostate cancer.

George's argument, I agree, seems strange.   It is almost as if he is
arguing that a decline in overall mortality causes a decline in
mortality due to specific diseases, e.g., prostate cancer, rather than
vice versa.   It is very hard to see how one could determine the truth
of such an assertion.  There could be indirect ways that might happen,
but I don't see how you could separate out the factors specific to
prostate cancer, such as early detection and better methods of treatment
from such unspecified other effects.

But let's get back to the original question.  Death rates due to
prostate cancer definitely declined in the US after the introduction of
PSA screening.  I haven't been able to find comparable data for other
countries,  but Walsh does maintain that this didn't happen to the same
degree in Sweden, where PSA screening has not been common.  Suppose that
is the case.  Can we now say that the risk of a biopsy causing
metastatic prostate cancer is negligible on the basis of this evidence.
 I think not because the PSA screening, by leading to earlier treatment
of prostate cancer may reduce the risk of metastatic prostate cancer
more than a biopsy increases it.  Indeed many people believe that early
screening and treatment do signficantly reduce the risk of metastatic
prostate cancer, and a lot of preliminery evidence supports it.   But it
is still possible to doubt it without being intellectualy bankrupt.  In
any case, the best we can say is that the whole process, including PSA
testing, biopsy where indicated, and early treatment, reduces the risk
of metastatic prostate cancer.  It could be that there is still some
significant risk of biopsy resulting in metastatic prostate cancer but
that this is more than counterbalanced by the other factors which reduce
that risk.

On the other hand, if it is true that early detection and treatment have
no signmificant effect on outcome, as George appears to believe, then
your argument is inevitable.   In that case, metastatic cancer rates
should go up in screened populations compared to unscreened populations
if it is true that biopsies signifcantly increase the risk of metastatic
cancer.   As best I can tell, no study has indicated such an increased
risk of metastatic prostate cancer in a screened population.

It would be nice if someone came up with a clear definitive study which
could isolate the different factors and tell use clearly what the risk
of metastatic prostate cancer from biopsy really is, and how it depends
on a variety of factors.   But it is not clear to me how you could do
such a study in the real world with real patients.  Clearly few medical
experts think this is a serious risk on the basis of what they know, and
I at least am not going to try to second guess them.

The Homlberg, et. al. studies do appear to show in the Swedish context
that radical prostatectomy reduces the risk of prostate cancer mortality
and the risk of metastatic cancer when  compared to WW followed by
hormone therapy when needed.   But since prostate cancer is not usually
discovered through PSA screening in Sweden, this wouldn't seem to bear
on the issue of whether biopsy by itself increases the risk.  All the
men in the Holmberg study persumably had biopsies.

P.S.  I just did a Medline search the risks of prostate biopsy and found
nothing within recent years suggesting an increased risk of metastatic
prostate cancer from biopsy.  I had previously found one or two reports
of cancer cells being found along the biopsy track, but that is not the
same as an increased risk of metstatic cancer because of reasons given
previously.  There was also a report of metastatic prostate cancer
having been transmitted through a heart transplant, but transplant
patients are subjected to strong immune suppressant drugs.   So that
raises the possiblity that people with severe immune definciency might
develop metastatic prostate cancer because some cells escape in a biopsy
and later become metastatic.  Prostate cancer for such people is
probably not at the top of their list of concerns.

 As general mortality goes down,  disease-specific mortality must decline
too.  Prostate cancer is just one type of mortality.