Thresholds or no threshoplds for inhaled alpha emitters?

Radsafers,

I would like to offer a few references that show apparent thresholds for lung cancer (or others) at doses quite respectable compared to regulatory limits and indoor exposures. There are of course many other publications on animal data but, generally, the cancer incidence in controls is low, very low, or zero, which makes risk calculations quite uncertain or impossible. The only study in which alpha dose rate could be adjusted independently of dose is the Morlier - Monchaux study (ref. below).

Animal data

Threshold for lung cancer after inhalation of Pu oxide at about 1 Gy :

Sanders et al. Lung tumor response to inhaled Pu and its implications for radiation protection. Health Phys. 55:455-462, 1988.

Threshold at about 3 Gy for bone cancer in dogs after injection of Ra 226 :

White et al. Bone sarcoma characteristics and distribution in beagles injected with radium 226. Radiat. Res., 137, 361-370 (1994).

No detectable effect ( no cancer in controls and exposed animals) :

Taylor et al. Comparative toxicity of 226Ra, 239Pu, 241Am, 249Cf, and 252Cf in C57BL/Do black and albino mice. Radiat. Res. 95, 584-601, (1983).

Influence of dose rate on lung cancer in rats exposed to radon progeny :

Morlier et al. Radiat Prot Dosim 56:93—97 (1994)

Monchaux et al. Effects of Radon and Radon Daughters in Rats. Envir. Health Perspectives, Vol. 102. No. 1, 1994.

The very low cancer incidence in controls (about 1%) makes the risk estimate at 25 WLM, 2 WL not quite significant at the 95% level when compared to controls, but the difference between the cancer rate at 25 WLM, 2 WL and 25 WLM, 100 and 150 WL is quite significant (RR = 4.2, 95% 1.3 - 14). Dose rate alone makes the difference.

Alpha exposures in animals – general observation :

We have collected virtually all published data on cancer induction at low doses. Eleven (65%) of the 17 studies on alpha radiation effects in which the cancer incidence in controls is between zero and 1 % in controls have a U-shaped dose response. With such a low background, no study can show a statistically significant difference but this is still a large proportion of U-shapes. Could a statistician tell us what the probability of finding such a high proportion of U-shapes?

Human data

Radium dial painters :

Rowland et al. Health Physics 44, Suppl. 1, pp. 15-31 (1983) Threshold for bone cancer at about 8 – 10 Gy

Thomas R.G. (1994) The U.S. radium luminisers: A case for policy of below regulator concern. Radiol. Prot., 14, 141-153.

Raabe et al. Health Physics 44, Suppl. 1, (1983) pp. 33 – 48. Latency time is dose-rate dependent, it exceeds the lifespan of animals when dose rate is low enough.

Thorotrast patients :

Andersson, M., Storm, H.H. (1992) Cancer incidence among Danish Thorotrast-exposed patients. J. Natl. Cancer Inst. 84,1318-1325. Threshold at about 2 Gy for liver cancer

Mayak Russian Pu and reprocessing plant :

Kreishmeier et al. Radiat Res 154:3-11 (2000). Mayak workers have about 28% less lung cancers than the average Russian population (1990-1994) and 30% less than the average US population (SEER 1983-87 data) at peak values of the cancer incidence at age attained curves. (They have 45% more lung cancers than the Swedes, 1983-1987 data), peak values of the attained age curves.

It seems that low dose rate alpha radiation is poorly effective at inducing cancer. Unquestionable thresholds are observed in man and in animals. Why should low dose rate radon progeny be more efficient for lung cancer induction?

Philippe Duport
International Centre for Low Dose Radiation Research
University of Ottawa
555 King Edward Ave.
Ottawa, ON, Canada, K1N 6N5
Tel: (613) 562 5800, ext. 1270
pduport@uottawa.ca