View From The Trenches

U27111 at U27111 at
Sat Jul 15 21:42:51 EST 1995

>   Please!  You have some valid complaints, but when you suggest
>regulation as the cure-all for everything you are playing with

Well, first I don not claim accrediting labs is a cure all... just
a step in the right direction.  Second, why not?  It's appears to
work for clinical labs and forensic laboratories... why should
research be excluded in an attempt to maintain *some* type of

>   The increase in misconduct in science has nothing to do with

I 'think' what you are trying to say here is that regulations will
have nothing to do with decreasing misconduct?

Or are you saying that the increase in misconduct has nothing to do
with the lack of regulations?

>Whatever "good old days" you might want to return to (and
>actually, the researchers of the past were not uniformly honest
>either), they are not the result of having stormtroopers marching
>up and down the hallways.  They might better be credited to a
>scholastic system that is very cunningly designed to
>remove the temptations for misconduct, yet encourage innovative


The only 'good ole days' I *may* like to see is a repeal of the
Patent Reform Act for a starters... and then maybe a way to lessen
the amount of sloppiness which goes on in science (ie. lab
accreditation and certification).

I'll leave stormtroopers to the Star Wars sequels...

>Specifically, if you recall, the THEORY of getting grants is that
>after going through a wringer of educational steps to prove your
>prowess, you finally acquire a tenured position with which you
>have complete security --- THEN, you propose a good idea, and get
>money based on what is written in the grant (which actually
>represents a body of work in and of itself, but not work that can
>be falsified).  NEVER does it come to an interaction where a
>person receives money because they obtained a particular result
>--- because it is impossible to prevent cheating with certainty
>when the cheater knows what he is doing better than anyone else in
>the world!!!
>   Unfortunately, various attempts to transfer business ethics
>(which work just great for getting holes dug, since you can't lie
>about whether you dug a hole) to science have subverted this

Define 'business ethics'.... are you talking about the competition
for funding?  I'm sorry... but I don't get what you are talking
about here?

>The need to get "preliminary data" for a grant application, before
>a deadline!, is a recipe for disaster.

But preliminary data for a grant is the last vestige for
'attempting' to assure monies is being spent efficiently.  If you
think the honor system doesn't work with the presentation of
preliminary data... imagine it without that - we wouldn't even have
something to, at the very least, 'check' on validity.  I agree a
lot gets through - but can you imagine the amount of garbage
'theories' which may get funded otherwise.

That's why I think we do need lab accreditation and personal
certification - to help tighten the reign on these types of fraud
as well.

>Is it fraud if you don't run a gel twice because you don't have
>time, and get money based on a first result that might be

Well... is it considered fraud when you knowingly abuse the
scientific process (ie. sloppiness) in this manner?

>Is it fraud if you don't run a gel twice because you're AFRAID the
>second result might not agree with the first?

Same as above.

>Is it fraud if you run a gel twice, get two different answers, and
>come up with some half-bogus explanation for the second, then
>discard it and not mention it in the grant?


>Is it fraud to run the gel under conditions that you know might
>lead to a misleading result, and not to mention the way in which
>the result might be misleading?


>Once you tie money to data obtained, you have a slippery slope
>greased all the way to Hell.

But unfortunately, not many makes it too Hell - they go through
hell and high water to continue to prove their positions, NO MATTER
WHAT.  And it can take years to never before the rest of the
community discover the difference.

>   A similar set of arguments can be made concerning patents.  But
>they are the least of the reasons why a patent system is a poor
>replacement for basic science.

So how do we fix this?  What's your suggestions.

>   Had the scholastic system been maintained intact, then the only
>incentive left for fraud would be to gain fame.  But since fraud
>WILL inevitably be detected (unless it happens to be right --- not
>a sure bet!) the person who desires fame will stay honest.

But how long is inevitably and why should terminally ill people
waiting to die also have to wait until these frauds are detected
before research gets back on track (usually well after they are
dead and buried)?

And how honest *is* Bob Gallo?

>   As it is, however, we see cases described every month in The
>Scientist and so forth describing the reputation that is truly
>devastating to one's research --- not to be dishonest, but to be
>a whistleblower!  Because the dishonest scientist now brings money
>to his institution, and is a valuable asset, while the
>whistleblower is a dangerous liability who, _due to regulation_,
>represents the potential loss of a great deal of money!  And there
>is the sad state of affairs which you lament.


>   Your Stupid Lab Tricks are immensely entertaining, but prove

Thank you...

They were only meant to be entertaining... to lighten the mood?
And nothing more really.

>   As to the FDA:  I cannot believe that you do not see the
>advantage to having a drug available for some people to try a few
>years earlier than it would be now.  The disadvantage is to the
>few who suffer adverse reactions, who would have taken the risk
>voluntarily rather than face certain death.  The advantage
>is to all the rest of us, who would benefit if we fought the
>diseases that attack us as if it were a war, in which casualties
>were unfortunate but inevitable, rather than prohibiting acts of
>bravery as our first priority.

[Reference for this section provided by the article "The rise and
fall of AZT" by Simon Garfield, v.40 World Press Review, July '93,

In March of 1987 AZT was first available by prescription.  "Rumors"
of this drug as a possible treatment for AIDS went as far back as
1985... with 'great' claims of it's efficacy.  Thus, when the drug
first hit the market (taking only 20 short months to pass FDA
approval)... nearly everybody with AIDS demanded to use it (as well
as those who were HIV+).

By April of 1993, it was discovered these great claims were
becoming more and more questionable.... with the release of the
preliminary results from the Concorde Study (an Anglo-French,
three-year clinical study on the effectiveness of AZT).  For the
Concorde study showed that HIV+ persons, prior to symptoms of
AIDS... that AZT had no effect what-so-ever on their progression to
disease or mortality rates (with those on AZT actually having a
'slightly' higher one).  In addition, the study found that those on
AZT suffered from greater side effects.

AZT was first 'invented' in 1964 as possible agent against
leukemia.  It did not work and the drug was shelved.  Over the
years the drug had been tried on various types of cancer... but it
never worked.  In 1984 it was discovered to kill HIV in culture.

16 weeks into the first clinical trial on a group of 282 AIDS
patients (a trial which was proposed to last 24 months)...
Burroughs-Welcome gave a press conference were they claimed there
was such a pronounced difference between the AZT group and placebo
group that they had decided to stop the study and put all the dying
people on AZT.... for "ethical" reasons.

However, John Lauritsen, author of the book, "Poison By
Prescription: The AZT Story (1990), obtained test documents which
showed a very different story than that proposed by Burroughs-
Welcome.  He discovered that yes indeed at week 16 into the study
the ratio between deaths on placebo verses deaths on AZT was 19:1
(as reported by Burroughs-Wellcome).  However, the next week data
showed a 23:3 ratio... and if the trial continued, this difference
may have narrowed even further.  And as we now know, AZT is only
effective for a limited period of time... something we would've
discovered earlier if this trial was allowed to continue.

In a statement released by Burroughs-Wellcome, they have decided to
mostly ignore the results of the Concorde Study, claiming "other"
clinical trials have shown the effectiveness of AZT on asymptomatic
HIV+ patients.

In this same statement, they went on to explain how effective AZT
now was in reducing the risk of transmitting HIV to babies of HIV+
pregnant women.  ["Burroughs-Wellcome statement on publication of
Concorde Study"  AIDS Weekly, April 18, 1994, p.18(2).]

And again, the preliminary results from this study was so
pronounced that the trial was halted and all the women were started
on AZT.... citing a transmission rate of 8.3% when using AZT,
compared to a 25.5% transmission rate for those babies of women on
placebo. ["AIDS in newborns" Pediatrics for Parents, April 1994,

The FDA has just recently approved the use for AZT on pregnant HIV+
and their newborns.  ["Approval for new use of AZT"  AIDS Weekly,
Aug. 8, 1994, p.19(2).]

Here again, John Lauritsen had something to contribute....

From: laurit at (John Lauritsen)
Subject: LANCET editorial on AZT and pregnant women.
Date: 7 Aug 1994 20:11:15 -0500

The LANCET, widely regarded as the world's most prestigious medical
journal, published an editorial in its issue of 23 July 1994:
"Zidovudine for mother, fetus, and child: hope or poison".

The editorial, which ran for over two pages, was implicitly
critical of the zeal with which preliminary results of ACTG 076
were interpreted in the U.S., by the Pediatric AIDS Foundation,
other medical groups, and branches of the Public Health Service.
The editorial states:

     In the UK, reaction has, on the whole, been cautious about the
     interpretation of results from a relatively small trial before
     full peer review.

Throughout the editorial, emphasis is placed on the fact that the
results are only "preliminary", that they represent only an
"interim analysis".  The point is made: "The full report of this
trial is awaited."

In ACTG 076 oral AZT (zidovudine) or placebo was given to pregnant
women between 14 and 34 weeks' gestation.  80% of those in the AZT
group also received AZT intravenously during labor. The babies
received oral AZT or placebo for six weeks.

The LANCET gives the following account of the findings from ACTG

     At the first scheduled interim analysis, 13 (8.3%) babies in
     the zidovudine group and 40 (25.5%) in the placebo group were
     infected with HIV (at least one viral culture) and the trial
     was stopped.  The estimated difference in the percentage of
     infected children between the two groups was 17.2% (95% CI
     8.9-25.5%), which is equivalent to a 67.5% reduction in
     transmission risk.

This statement "(95% CI 8.9-25.5%)" may require interpretation.  In
plain English it means that, with a statistical confidence level of
about 95%, the "true" difference between the two groups is
somewhere between 8.9% and 25.5%.  In other words, from a
statistical standpoint, the figure is highly unstable.

However, even accepting the results at face value, it is clear that
three out of four of the babies who did *not* receive AZT (or
zidovudine) were nevertheless born without HIV infection.

The LANCET editorial makes this point when addressing the crucial
issue of long-term side effects:

     The most worrisome aspect is the possibility of long-term
     adverse effects on children exposed to zidovudine during fetal
     life, especially since the vast majority would not have been
     infected anyway.  A mechanism is needed in Europe, as in the
     USA, for long-term follow-up of all zidovudine-exposed
     children.  Meanwhile, the possibility of unknown long-term
     adverse effects must be weighed against the benefit of a
     considerable decrease in HIV transmission.

As AZT (or zidovudine) is a random terminator of DNA synthesis,
which is known to be both carcinogenic and mutagenic, it is
reasonable to assume that it will cause birth defects.  The
LANCET's emphasis on *long-term* adverse effects is prudent, as
some birth defects are not immediately apparent.  In the case of
children born to mothers who had taken the drug DES during
pregnancy, the birth defects did not manifest themselves until the
children reached adolescence.

While the LANCET editorial does not take a stand for or against
AZT, it fairly and dispassionately takes notice of AZT critics,
including SCAM (Steering Committee Against Malpractice), Peter
Duesberg, and myself.  (My book, _Poison By Prescription: The AZT
Story_, is cited in a footnote.)  The word "poison" in the
editorial headline signals an end to the honeymoon between AZT and
the medical press.

        "Sine ira et studio" -- Tacitus, Annales, I, 2
| John Lauritsen           author: The AIDS War (1993)           |
| 26 St. Mark's Place      Poison By Prescription: The AZT Story |
| New York City 10003      (1990)                                |
| Voice: (212) 674-3321    INTERNET: laurit at            |

DISCLAIMER:    I do not believe in the Duesbergian theory that HIV
               does not cause AIDS... but I do think Lauritsen
               does make some very good points concerning AZT.

Anywya, that's the problem with rushing drugs through the FDA in a

Plus... where is it written that unborn children and infants are
committed to preform 'acts of bravery' so that Burroughs-Wellcome
[now Galaxo] can attempt to stop the 24% drop in sells of AZT since
the publication of the Concorde Study?

>   As to regulation of radioactivity:  I have seen news reports of
>a case in which a major university was forced to spend $150,000
>recovering a package of P-32 from a landfill.  This is evidence of
>how perhaps well-meaning regulation leads to absurdity --- because
>such a quantity of P-32 is barely a danger at all, is
>well-contained in its shipping container, and WILL degrade within
>a year, which is more than you can say of anything else put into
>a landfill!
>Rules are no substitute for common sense.

I'll leave this one to be commented by Trivol... if he chooses.


More information about the Bioforum mailing list