Re: Alpha radiation as tests of LNT

[Bernard L Cohen <blc+@pitt.edu>]

Bernard L. Cohen
Physics Dept.
University of Pittsburgh
Pittsburgh, PA 15260
Tel: (412)624-9245
Fax: (412)624-9163
e-mail: blc+@pitt.edu


On Thu, 1 Aug 1996, Dick King wrote:

> >> 
> >> Date: Wed, 31 Jul 1996 13:55:12 -0400 (EDT)
> >> From: Bernard L Cohen <blc+@pitt.edu>
> >> Subject: Re: Alpha radiation as tests of LNT
> >> 
> >> 	The risk from radon is determined from data on miners who 
> >> contracted lung cancer from high exposures in mines. LNT simply says that 
> >> the risk of lower exposures is linearly related to the dose.Nothing else 
> >> is involved.Presumably, if a single alpha particle can kill a cell, the 
> >> number of cells killed is also linearly related to dose.
> 
> This doesn't follow at all.
> 
> It may be that a single alpha particle can kill a cell with a very low
> probability because there is one tiny target that can be killed with a single
> hit, but that there is a much bigger nonlinear target, such as the DNA, that
> needs to take more than one hit before the repair can take place.
> With low doses the main kill mechanism is the single hit mechanism, but damage
> is very light.  High doses are nonlinear.

--- I was assuming that the principal mechanism for cell killing by alpha 
particles is by a single alpha particle (which can easily provide 2 or 
more "hits"in a single cell nucleus). This is surely true up to 
very high doses; that's why it is often assumed that low-LET may be 
linear-quadratic, but high-LET must be simply linear for cancer 
initiation which is presumed to require two hits.
> 
> Note carefully that even if the number of cells killed is linear in the dose
> that doesn't mean that health effects are proportional to dose.  If you kill
> 50kg of cells in one person the person dies.  If you kill 50mg of cells taken
> at random in each of a million people probably almost nobody would notice.
> Each person would lose one or two pixels of vision, perhaps, and a pixel of
> skin sensation somewhere, etc. but cells die all the time.

----- I meant that if the underlying cancer causation risk is assumed to 
be linear, cell killing will leave the total cancer causation risk linear.

> 
> -dk
>