Well, I would have been considered low risk with a T1c, 3+3 bx, and 7
psa. As I stated earlier, I'm not a gambling man, so I went ahead with
the surgery. The only "surprise" was the higher pathology GS which
confirmed my poor gambling luck if I had none WW or AS.

B.A.

Let me remind everyone first that, while I am pretty good at counting, I
am not a prostate cancer researcher, so I could easily miss important
points in any analysis I make.

I looked at the abstract.  I see some questions about this study.
First, it appears to be based not on raw data but on a model.  This is
not uncommon, but the modelling has to be done carefully, and it is
certainly possible to derive false conclusions that way.  Secondly,  I
don't know what the term 'conservative management' means.   (That was
the alternative to early treatment in an attempt to cure the disease.
Unfortunately, the news report used the term 'untreated', which means
something very different to me.)  If it means no treatment at all, then
1-2 percent disease specific mortality for men with Gleason 6 or below
is way out of line with the result I quoted from Scardino, and
personally, I don't believe it.  If they mean regular monitoring
followed by hormone therapy when necessary, their result is more
plausible.  (I think this is the more likely meaning since no one is
going to be left untreated these days once PSA starts rising rapidly.)
 It can easily take up to 10 years before a Gleason 6 case would
require hormone therapy, and it could take another 10 years before that
would fail.  Needless to say, a desire to avoid metastatic cancer and
hormone therapy is a rational basis for decision separate from mortality.

It should also be noted that they estimate that the disease specific 15
year mortailty for Gleason 7 cases is 9-31 percent (rather a large
range).  This compares to the over 60 percent 15 year disease specific
mortality for untreated Gleason 7 cases mentioned in Scardino's book and
based on the 1998 paper of Albertsen.  Their upper limit of 72 percent
for Gleason 8-10 cancers is more consistent with Albertsen's slightly
over 80 percent.

As various people have noted, cases diagnosed clincially as Gleason 6
frequently turn out to be Gleason 7 when the entire prostate is examined
after surgery.  That factor alone cast doubt on a model with suggests a
9 to 15 times increase in mortality when going from Gleason 6 to 7.

There is one other factor which could affect these estimates.   With PSA
screening, prostate cancer is diagnosed at least 5 years earlier than it
was earlier.  Also, presumably, more innocuous cancers are diagnosed
which with earlier methods would never have been noticed.  (We don't
really have good estimates of how many such there are, but I've seen
estimates of 15-20 percent, and for Gleason 6 cancers it could be
higher.)   Both these factors would lower the 15 year disease specific
mortality rate.   The second would because it would increase the size of
the pool of 'diagnosed cancers' by adding a significant number of
innocuous ones.  The first would because it would set the clock back by
something like five years.  So you  might have to compare Albertsen's
result at 10 years (something like 15 percent) or even earlier to the
British estimate at 15 years.   The figures still don't look plausible
unless you assume an extremely large number of innocuous cases diluting
the pool.  Also, this whole idea is based on the idea of early detection
through PSA screening, which is not common in England anyway.   I think
the more important factor is the difference between 'untreated' and
'conservatively managed'.

I will have to get hold of the paper and examine it carefully.

What you say is correct.  Even latitude has an effect, presumably having
to do with exposure to sunlight.   But it should be kept in mind that
the US figures are affected by the higher prevalence and greater
aggressiveness for African American men, a factor which is preumably
less significant in England.

You refer in several places to the Scandanavian study.  Note that I was
comparing the figures in the abstract for the Britich study with some
graphs in Scardino's The Prostate Book.  These are modified from the
following paper

Albertsen, P.C.,et.al, "Competing Risk Analysis of Men Aged 55 to 74
Years at Diagnosis Managed Conservatively for Clincially Localized
Prostate Cancer", MAMA 280, no. 11. (1998): 975-80.

They studied 767 men in the Connecticut Tumor Registry who were
diagnosed with prostate cancer between 1971 and 1984 and either not
treated or treated with HT and followed for 10 to 20 years.  They found
the number who died withn 15 years varied as follows:  Gl 2-4: 4-7 %, Gl
5 : 6 - 11 %,  GL 6: 18 to 30 %, Gl 7: 42 to 70 %, Gl 8-10: 60-80 %.

Although some of these men were treated with HT, I think this paper and
earlier papers of Albertsen are commonly used to estimate the death rate
from untreated prostate cancer.

There were also Scandanavian studies done in the late 80s, but I made no
reference to them.  I thought that those had found that there was little
difference between aggressive early treatment and WW, but those studies
have been largely rejected because they didn't follow up long enough.

There is also the recent work of Holmberg, et.al. comparing RP with WW,
but I think that is another matter.

That was one of my conjectures about how they got their result.  I would
like to see why they conclude that PSA screening results on the average
in detecting cancer 10 years earlier.  That would surprise me because I
thought it was more like 5 years.   But 10 years would certainly explain
the figures.   If you push back the results from Albertsen, et.al. some
10 years, you would be looking at his population at something like 5
years after diagnosis.  Even in the ppre PSA testing era, few men with
Gleason 6 cancers died within five years of diagnosis.

But in that case results like this, misconstrued in the media, can have
a miselading effect.   Suppose you tell a 55 year old man that he stands
very little chance of dying of his cancer before age 70, but you neglect
to tell him that the chances of his dying of it by age age are  20
percent or so.  And you also don't tell him that by age 70 there is a
reasonable chance he will be diagnosed with metastatic prostate cancer,
and be put on HT which may keep him alive another 10 years.  Finally you
also neglect to tell him that this applies to some average man and that
he may in fact die of it much earlier.  I think you are misleading him.

There is one additional difficulty with this sort of reasoning as it
applies to American men.  There is actually a distribution of times.
PSA testing may on average diagnosis 10 years earlier than previously,
but for some men it may be one or two years and for others it may be 15
years.  Perhaps their model takes this into account.  I will have to
look at the paper.

However, I do think models like this, as long as they aren't taken too
seriously do suggest that more men with Gleason 6 tumors can wait before
deciding on treatment.  I doubt, however, whether they can provide
clinically useful criteria which would help a doctor advice a specific
man considering his options.

I didn't say it came from either the news report or the scientific paper
it supposedly reports on.  It is based just on things I've read.  But I
just did a google search and found a discussion in the JAMA, 2002, pp.
969-70, which confirms this idea.  The discussion is about a
controversial paper by Stamey at Stanford and his claim that PSA testing
is worthless.  Great Britain and PSA testing is mentioned in passing.
I'm sure if I worked at it hard enough I could find other evidence.

Of course, this applies to the past.  It is possible that they are doing
more testing in Great Britain now.  But testing in the US started in the
late 80s and became quite common in the early 90s.  That is when my
physician started ordering PSA tests for me.

This is the same thing.  If you don't do PSA testing, then you are not
going to be in as good a position to treat prostate cancer early.

I was just quoting the news report.  It says nothing about the best
practice in both countries.  But assuming PSa testing in the UK is less
common than in the US, it does make a point.

True,

I agree.  That was my point.  PSA testing in the US started in earnest
15 years ago.

I wasn't trying to attack the NHS.  I'm sure they do what they well
pretty well.  I believe that life expectancy is greater in the UK by a
few years than it is in the US.   But the US is unique in that PSA
testing has been quite common for many years.  There is still a lot of
controversy, including in the US, about whether this does any good.  If
it does, then the US is ahead of most of the rest of the world in this
one regard.