FW: Some reasons not to believe the linear no-threshold hypothesi s

["Dukelow, James S Jr" <jim.dukelow@pnl.gov>]

This is a reposting.  My previous attempt was truncated by the pernicious
RADSAFE "From" bug.

I sent the following to RISKANAL, a mailing list for risk professionals, in the
context of a general discussion of the LNTH.

It seems relevant to the current discussion on RADSAFE, also.
-----Original Message-----
From:	Dukelow, James S Jr 
Sent:	Thursday, March 16, 2000 3:10 PM
To:	'riskanal@lyris.pnl.gov'
Subject:	A partial response to Finkel and Dankovic

Some reasons not to believe the linear no-threshold hypothesis.

First, and most flip, it has a lot of the flavor of homeopathy.

Second, it is so implausible as to strain credulity.  Living cells and 
living organisms have a variety of essentially independent schemes for 
preventing cancer.  Human cells have several different DNA-repair 
mechanisms.  If these fail to repair damaged DNA, there are mechanisms 
for identifying the failure and commanding cell suicide.  If this fails, 
immune system surveillance and response may destroy the proto-tumor.  A 
small solid tumor cannot grow past a certain point without inducing the 
body to build it a network of blood vessels to provide nourishment and 
remove wastes.  A solid tumor may need to metastasize before it can 
finally kill its host.  All of these lines of defense work with a 
greater or lesser degree of effectiveness, but with sufficient 
effectiveness that a majority of humans live long enough to die of 
something else. 

Considered from the mathematician's or engineer's point of view, you can think
of 
the organism as a black box, with some carcinogenic exposure as input 
and a probability of cancer induction and eventual fatality down the 
road as output and with a half-dozen to a dozen internal "knobs" 
representing the performance of the body's anti-cancer defenses.  The 
idea that these knobs happen to be set so that the output (probability 
of cancer fatality) is linear in the input (exposure to the carcinogen) 
strains credulity. 

This is particularly the case if you take natural selection into 
account.  Theodosius Dobzhansky wrote, "Nothing makes sense in biology, 
except in the light of evolution." 
 
In that spirit, the values of the internal knobs (i.e., the fidelity and 
reliability of the anti-cancer systems) have been "set" by evolution.  I 
haven't been able to imagine any evolutionary advantage resulting from a 
linear no-threshold dose-response relationship.  On the other hand, a 
dose-response relationship that provides a positive threshold below 
which there is no cancer induction for "normal" levels of carcinogenic 
insult has an obvious competitive advantage.  The corollary is that the 
"threshold" organisms will out-compete the LNT organisms.

The case becomes stronger if you take time into account.  Cancer is 
primarily a disease of old age, which is more than coincidently the time 
when our immune systems begin running down.  Evolutionary pressure and 
natural selection will be indifferent to what happens to us once we get 
past the age of reproduction, since the way it works is by controlling 
the structure of the next generation.  The best example of this is 
probably the relatively large number of species that die immediately 
after reproducing. 

Thus, our defense mechanisms against cancer (and other diseases, for 
that matter) work to protect us through the age of reproduction and 
don't "care" what happens to us after that.  Many of the relatively rare 
number of cancers in children and adults of reproductive age can be tied 
to defects in one or another of the body's defense mechanisms.  These 
deaths, those before reproduction, at least, represent a selection 
against those defects and a maintenance of the full array of defense 
mechanisms in the next generation. 

Third, the carcinogenic insults from radiation and other carcinogens 
tend to be irrelevant to the real battle going on -- the body's war 
against oxygen.  If the cells and the immune system are unable to 
suppress the oxidative damage to DNA caused by free radicals and other 
oxidants inextricably tied to our metabolism, the effect of the similar, 
but much smaller amounts of damage caused by low levels of carcinogens 
won't have any impact on survival. 

Some defenders of LNTH argue that the damage to DNA from radiation is 
qualitatively different and more dangerous than that caused by oxidants 
arising from normal metabolism.  I find that argument unpersuasive.  One 
of the major reasons is the existence of Deinococcus radiodurans.  D. 
radiodurans was discovered in the 50's, growing in a can of meat that 
had been heavily irradiated.  Subsequent experimentation has established 
that D. radiodurans can survive acute exposures to 1.5 to 3 megaRads.  
The LD50 for human acute whole-body irradiation is on the order of 400 
Rads, 3 to 4 orders of magnitude smaller.  D. radiodurans appears to 
survive these enormous exposures by carrying a number of copies of its 
genome and having particularly effective DNA repair mechanisms.  Now the 
interesting question is, How did D. radiodurans come by this 
extraodinary ability to survive irradiation?  It is never exposed to 
anything close to a megaRad of ionizing radiation in it's natural 
environment.  The answer seems to be that it is exposed to other 
environmental insults that damage DNA, and the armamentarium it has 
evolved to protect it against those works quite nicely against 
radiation. 

Fourth, even some of the strong proponents of the LNTH don't really 
believe it and/or they admit that there is no sound scientific evidence 
supporting its application in the low dose region. 

Finally, there is a whole lot of scientific evidence that contradicts 
the LNTH.  I have always taught my students that a single counterexample 
is sufficient to disprove a theorem that makes some claim about how all 
things of a certain class behave.  That rule doesn't seem to hold in the 
world of "regulatory science" that sets radiation standards.

As good a single source for the evidence against LNTH as any is the web 
site of the Biological Effects of Low Level Exposures organization at 
<www.belleonline.com>.  All thirty or so of their newletters are online. 

Finally, in response to Dave Dankovic's questions:  A good place to 
start would be to quit using LNTH in the low exposure domain where there 
is no scientific support for it.  That means we quit believing (and 
acting on) the idea that 250 million people exposed to 1 millirem of 
radiation from Chernobyl equals 250,000 person-rems, which "everyone 
knows" means 1,250 extra cancers.  Jackie Kittrell was deeply offended by 
the aspirin analogy, but it is perfectly apt.  Aspirin is safe (even 
therapeutic) at low doses and poisonous at higher doses.  It's a concept 
that goes back at least 450 years to Paracelsus.  Similarly, the 
scientific evidence that radiation is safe at low doses and dangerous at 
higher doses is pretty convincing, to me, at least. 

Best regards.

Jim Dukelow
Pacific Northwest National Laboratory
Richland, WA
jim.dukelow@pnl.gov

These comments are mine and have not been reviewed and/or approved by my 
management or by the U.S. Department of Energy.

************************************************************************
The RADSAFE Frequently Asked Questions list, archives and subscription
information can be accessed at http://www.ehs.uiuc.edu/~rad/radsafe.html