Numerical model of Rn mortality

[David Scherer <scherer@uiuc.edu>]

Wade Patterson recently posted the abstract of Ken Bogen's numerical (CD-2)
model on Rn-induced lung cancer mortality.  A portion of that abstract is
quoted here:

>The CD2 model used adapts a widely applied, mechanistic, 2-stage
>stochastic model of carcinogenesis to realistically account for interrelated
>cell killing and mutation (both assumed to have a LN dose-response),
>and incomplete exposure of stem cells.

If I understand this correctly, this model uses a linear models for cell
mutation and for cell killing due to alpha radiation (Rn progeny).  In
addition, the study added a parameter to account for lower doses to some
stem cells.  The combined effect can be made to fit with Cohen's data and
miner mortality data using the parameters values that are consistent with
other biological data.

Quote:

>The CD2 fit was shown to: (i) to be consistent with the combined data;
>(ii) to have parameter values all consistent with biological data; and
>(iii) to predict inverse dose-rate-effects data for CP and other
radon-exposed >miners, despite the fact that optimization had not involved
any of these
>dose-rate data.

As I understand the last statement, Dr. Cohen's suprizing results are
predicted without optimizing a dose-rate-effect parameter (a DREF?).  I
take it that the dose-response result is due to the particluar combination
of mutation, cell killing, and dose distribution that goes on with
Rn-progeny exposure (i.e. alpha radiation).

This hardly seems to be the death-knell of the LNT model, especially for
low LET radiation.  The combination of these factors could be quite
different in that situation.  In fact, the model _explicitly_ uses a linear
model for radiation mutation leading to cancer induction.  I would like to
hear from someone who has read the entire preprint to see whether my
understanding is borne out by the full report.

Regards,

Dave Scherer
scherer@uiuc.edu