Cancer Risk Assessment Concern Regarding the Publication “Random Threshold Model: A Low-Dose Radiation-Induced Risk Assessment Approach Considering Individual Susceptibility to Cancer”

In this letter I point out why there is a need for a radiation-dose-type change when employing the Random Threshold (RT) Model ¹ in assessing cancer risks for radiosensitive subpopulations. With the RT model there is a distribution of individual-specific threshold doses for radiation-induced cancer. For radiosensitive individuals, all thresholds are low, thus special consideration ¹ is given to this subpopulation. However, even with the invalid^2,3 LNT hypothesis as it relates to the total population, it is implied that there is a hyper-radiosensitive subpopulation that is assigned (but not proven) to have cancers induced by very low radiation doses. With an assigned LNT risk function, the dose that is assigned for a radiation-caused cancer risk of 0.001 supposedly causes cancer in 100% of the most radiosensitive 0.1% of the population. The assigned risk vs dose relationship for this supposedly hyper-radiosensitive subpopulation would also be LNT, but with a slope 1000-fold larger than for the total population. With all LNT-based cancer risk vs dose relationships, it is implied that there is a uniform distribution of individual-specific dose thresholds, with the lowest threshold dose (ie, population threshold) to an organ or tissue being from a single ionization.

The RT model was unfortunately employed ¹ using the hypothetical effective dose in millisievert, which is not organ/tissue specific. Note that the assigned effective dose (not a real dose) is derived from applying the tissue-sensitivity-related weighting factor (w_T) to the radiation-type-weighted (with w_R) organ/tissue-specific real absorbed dose (ie, equivalent dose is generated), and is intended for use with LNT risk functions. ⁴ For application of the RT model, it is more scientifically credible to use the organ/tissue-specific radiation absorbed dose (eg, in grays), as is done for threshold-based acute lethality risk assessment. ⁵ For characterizing acute lethality risks due to radiation-caused massive tissue damage, a Weibull distribution of individual-specific threshold absorbed dose has been assumed; with the distribution depending on radiation type as well as dose-rate pattern. ⁵ The absorbed dose is converted to the unitless normalized dose X (with X = 1 being the median lethal normalized dose) when addressing acute lethality risk for combined exposure to low and high linear-energy-transfer radiations and changing dose rate over time. ⁵