Revision 1
March 19, 1998
by Radiation, Science, and Health, Inc.,
Edited by J. Muckerheide

1.2.4
Radium Body-Burden

1.2.4.1
Bone and Nasal Cancers

"This paper presents dose-response data from US female workers who were exposed to radium through the painting of luminous dials. Lognormal analyses were done for radium-induced bone sarcomas and head carcinomas after the respective dose populations were determined to be lognormally distributed. The geometric mean and standard deviation for each dose population were used to construct lognormal distributions for comparisons. Luminisers with average measured skeletal doses below 10 Gy (1391 subjects) showed no incidence of cancer. A primary purpose of this paper is to illustrate the strong case that ^226,228Ra is representative of radionuclides that have in humans a’threshold’ dose, or a dose below which there should be little concern for regulation. It is time to evaluate the data objectively instead of formatting the extrapolation scheme beforehand and forcing the data to fit a preconceived pattern such as linearity through the dose-effect origin. It is also time to re-evaluate (again) variations in background radiation levels throughout the world and to cease being concerned with, and regulating against, miniscule doses for which no biomedical effects on humans have ever been satisfactorily quantified."

"Lognormal analyses were performed for head carcinomas and for all bone sarcomas, separately and combined (sarcomas, n = 46; carcinomas, n = 19; total primary tumours, n = 65). When a primary cancer of each type was observed in a single subject (three cases), two separate primary lesions were tallied in the analyses (i.e. a carcinoma and a sarcoma were tallied for the same skeletal dose). For graphic comparison of the data with the calculated lognormal distribution, the number of tumours (either sarcomas, carcinomas, or a combination) per 10 Gy interval was totalled. The fraction this number represented of the subjects with skeletal doses above 10 Gy (154 total) was obtained for plotting in figures 1-3. No tumours were reported below 10 Gy, the lower dose limit used for the tumour analysis".

"Figure 3 depicts the dose-response data for individuals with bone sarcomas and/or head carcinomas, expressed as tumour incidence versus dose. The curves were constructed as log-normal distributions by using the means and standard deviations calculated as part of the ratiometric analyses for all of the variables.

"In another set of data (table 5) dose was determined in measured luminisers who expressed no radium-related cancerous condition. Because of the large number of subjects (1391) in this population with average calculated skeletal doses below 10 Gy, they were not listed individually."

Table 5. Summary of subjects with average skeletal doses below 10 Gy and no skeletal tumours.

a Of the 154 subjects with doses greater than 10 Gy. 62 had skeletal tumours. Three of these had multiple primary cancers, bringing the total cancer subjects to 65 for computational purposes.

"...Arguments can be made that an alpha emitter in mineral bone does not have the opportunity to cause effects on tissues that are sufficiently active to be carcinogenic. The practical threshold can easily reside within this explanation; if other radiation sources behave similarly within their target organs, they might represent other cases for a policy that a dose is below regulatory concern. Part of the reason that such threshold doses have not become visible is the devotion to linearity (classical or linear-quadratic) that is so prevalent in the radiation protection community. To say that all radiation is harmful and to use the collective dose concept is easy. One problem facing regulatory agencies and radiation protection bodies is the restrictions sometimes placed on them, by themselves or by others, through application of preconceived interpretation and application of scientific data. A classical example of this is excerpted below from the Federal Register containing the US Environmental Protection Agency’s (EPA’s) proposed drinking water standards for radionuclides including radium (EPA 1991). This report contains questions and advice from Scientific Advisory Board/Radiation Advisory Committee (SAB/RAC) of the EPA and EPA answers. The following is quoted directly:

‘The watch dial painter data indicate that the incidence of bone sarcomas may follow a dose-squared response, especially at higher exposures. EPA policy, supported by recom- mendations of SAB/RAC, is to assess cancer risks from ionising radiation as a linear response. Therefore, use of the dial painter data requires either deriving a linear risk coefficient from significantly non-linear exposure-response data, or abandoning EPA policy and SAB/RAC advice in this case.’

"This excerpt exemplifies how the need to satisfy certain ground rules (the use of linear modelling with what the EPA referred to as dose-squared radium dial painter data) forces the rulemakers to use only data that tend to give the desired result, or to use dicta for the interpretation of scientific data. One can only marvel that such results are generally acceptable by the standard-setting community. As a consequence of the above EPA policy, some of the present standards for radium, including that for ²²⁶Ra, are based on data from ²²⁴Ra injections. Radium-224 has a half-life of 3.6 days; therefore, it behaves as a surface-seeker rather than as a volume-seeker like ²²⁶Ra."
 

[Editor’s note: Emphasis added.]