A man's right to know
A new study is raising valid questions about the screening and
treatment of prostate cancer. Less testing, however, is not the
answer.
A Times Editorial
Published May 29, 2004
--------------------------------------------------------------------------------
A new study on prostate cancer screening has renewed a useful debate
about what men should know and when and how they should be treated.
Oddly enough, though, some health experts are arguing that the study,
which found an alarming number of older men with serious cancer not
caught through the blood tests, suggests there should be less testing
and not more. Their logic is unhealthy.
The study, reported in the New England Journal of Medicine, raises
questions about the range currently considered normal. The researchers
performed biopsies on the prostates of 2,950 men, ages 62 to 91, whose
level of prostate specific antigen was measured at below 4.0 nanograms
per milliliter of blood - which is considered safe. But the biopsies
revealed cancer in 449 of the men and aggressive cancer in 67 of them.
The conclusion one would draw is more obvious than the course of
corrective action. The threshold is so high that it misses nearly one
in seven older men who take the test, but some of these men may end up
not needing treatment or even a biopsy. The reason is that some forms
of prostate cancer are not pernicious, and some forms of treatment can
cause impotence, incontinence and other complications. Biopsies
themselves can sometimes lead to bleeding or infection.
Ned Cologne, chairman of the U.S. Preventive Services Task Force, puts
it this way: "We are missing these prostate cancers, but we still
don't have an answer to the fundamental question, which is: If we
found them and treated them, would we make men live longer? That's
that missing piece of evidence."
Cologne's point is an important one, but not relevant to the test. We
know that early detection helps in fighting all forms of cancer, and
that the rate of prostate cancer has declined since the advent of PSA
blood tests (though other factors, such as improved treatment,
certainly contribute). So whether the medical community should use
these results to lower the threshold at which biopsies are routinely
ordered is a fair question. Whether prostate cancer in general is too
aggressively treated is also a source of continuing concern. But the
notion that blood test results, in themselves, can be harmful is
paternalistic at best.
Prostate cancer kills roughly 29,000 men every year, second only to
lung cancer, and men have a right to know. The PSA test is far from
perfect, to be sure, but it provides some basis from which men can
then, with the consultation of their doctors, carefully weigh other
risk factors and the full range of options available to them. Do they
have a family history of prostate cancer? Are they more advanced in
age? Has their PSA level increased sharply over time? That's the
nature of thorough, cautious, preventive medicine - and just what the
doctor should order.
[La
c palmer - 29 May 2004 17:59 GMT
well said.........
~ curtis
knowledge is power - growing old is mandatory - growing wise is optional
"Many more men die with prostate cancer than of it. Growing old is
invariably fatal. Prostate cancer is only sometimes so."
Leonard Evens - 29 May 2004 20:39 GMT
> A man's right to know
> A new study is raising valid questions about the screening and
[quoted text clipped - 53 lines]
> nature of thorough, cautious, preventive medicine - and just what the
> doctor should order.
I'm glad the Times on its editorial page has corrected the impression
which Gina Kolata's front page article on the subject may have left.
She covered much of the same ground but with much greater emphasis on
critics of PSA testing. The editorial rightly remarks that there are
issues about the effectiveness of testing and about how much prostate
cancer is treated unnecessarily, but that critics using this particular
study to push anti-testing and anti-treatment agendas are abusing it.
Since the Times editorial staff presumably doesn't have any prostate
cancer specialists, I wonder if someone got in touch with them and read
them the riot act about their coverage.
I've now read both the article in the New England Journal of Medicine
and the editorial in that magazine commenting on it. I am still not
sure what it all means. One could conclude, for example, that biopsies
for men with PSAs over 4.0 won't produce significantly more prostate
cancer diagnoses than biopsies for all men. Under current standards,
apparently only about one quarter to one third of all biopsies yield a
diagnosis of cancer, and in this study one in six men whose PSA never
exceeded 4.0 had positive biopsies after 7 years. In other words, it
almost seems as though those of us diagnosed through biopsy were just
lucky that something precipitated the biopsy and our cancers were
diagnosed (or as some critics would maintain, we were unlucky).
Unfortunately, statistics can be quite subtle, and it is very easy to
mix apples and oranges when making such comparisons, so I have to think
about it some more and consult with my favorite mathematical
statistician. But, whatever the case for PSA testing, it is clear that
we also need other more specific tests for early prostate cancer. Some
already exist, and are used by good clinicians who don't realy on just
one number. For example, my cancer was initially diagnosed because of
PSA velocity, which I believe is more specific for prostate cancer. I
wonder how many men in the study would have been sent for biopsy if PSA
velocity had been used as a criterion. That could conceivably account
for some significant portion of the 15 percent. Also, we don't know
how many of the men in study would have been sent for biopsies in
subsequent years as their PSAs rose over 4.0. But current research
looking for other markers for prostate cancer needing treatment holds
promise that all this will be moot in a few years.
> [La
Danny McCarty - 29 May 2004 20:56 GMT
>Subject: Re: A Man's Right to Know (St.Petersburg Times Ediitorial on PCal)
>From: Leonard Evens len@math.northwestern.edu
[quoted text clipped - 97 lines]
>looking for other markers for prostate cancer needing treatment holds
>promise that all this will be moot in a few years.
So 25 % to 33 % of men with PSA > 4 have prostate cancer, and 17 % of men with
PSA < 4 have prostate cancer. Thus at least 42 % to 50 % have prostate
cancer. Doc can make a lot of money doing biopsies. I didn't read anything
about Gleason scores in there...
Don Coon - 29 May 2004 23:10 GMT
> >Subject: Re: A Man's Right to Know (St.Petersburg Times Ediitorial on PCal)
<SNIP>
> So 25 % to 33 % of men with PSA > 4 have prostate cancer, and 17 % of men with
> PSA < 4 have prostate cancer. Thus at least 42 % to 50 % have prostate
> cancer. Doc can make a lot of money doing biopsies. I didn't read anything
> about Gleason scores in there...
Huh? You can't add percentages. In this case, 25% + 17% does not equal
42%. The correct calculation is 25% times the % with PSA > 4 plus 17%
times the % with PSA < 4. For example, if the groups were by chance equal,
25% x50% + 17% x 50% = 21%.
Danny McCarty - 30 May 2004 21:46 GMT
>Subject: Re: A Man's Right to Know (St.Petersburg Times Ediitorial on PCal)
>From: "Don Coon" coondw_nospam@hotmail_dot_.com
[quoted text clipped - 17 lines]
>times the % with PSA < 4. For example, if the groups were by chance equal,
>25% x50% + 17% x 50% = 21%.
No. The only relevant objection is that not all men go in to get their PSA
checked, when/if they get regular physicals. If there is no difference between
the incidence of prostate cancer of men who do and do not get regular
physicals, then the PSA separates all men into two groups- those whose PSA is >
4 and those whose PSA is < 4. 4 is 4.0000000000000000000, and the set of males
with that PSA is empty. So, 42 to 50% of males, in the age range where PSA
tests are often given, have prostate cancer, according to the study. This is
stunning.
Don Coon - 30 May 2004 23:23 GMT
> >Subject: Re: A Man's Right to Know (St.Petersburg Times Ediitorial on PCal)
> >From: "Don Coon" coondw_nospam@hotmail_dot_.com
[quoted text clipped - 26 lines]
> tests are often given, have prostate cancer, according to the study. This is
> stunning.
Again, huh? ; )
Leonard Evens - 30 May 2004 23:41 GMT
>>Subject: Re: A Man's Right to Know (St.Petersburg Times Ediitorial on PCal)
>>From: "Don Coon" coondw_nospam@hotmail_dot_.com
[quoted text clipped - 31 lines]
> tests are often given, have prostate cancer, according to the study. This is
> stunning.
I'm sorry. I have to agree with Don that you haven't done the
arithmetic right. Percentages don't work that way. Suppose you had
2000 men divided into two equal groups of size 1000, call them A and
B. Suppose 17 percent of the men in group A had prostate cancer. That
would be 170 men. Suppose 25 percent of the men in group B also had
prostate cancer. That would be 250 men. So the total number of men in
both groups with prostate cancer would be 170 + 250 = 420. But the
total number of men in both groups is 2000, so the percentage in all is
420/2000 = .21 or 21 percent.