Effective thresholds for induction of skeletal malignancies by radionuclides.

Our analysis of data from the beagle project completed at the University of Utah has provided some comparisons that appear to be useful in testing the model proposed by Raabe of effective thresholds for induction of skeletal malignancy by bone-seeking radionuclides in beagles. Raabe's model predicted that cumulative skeletal doses of less than about 0.9 to 1.4 Gy from alpha emitters or 28 to 70 Gy from beta emitters deposited in the skeleton require a long enough time for bone cancer expression that the dog's natural lifespan would be exceeded before the tumor appeared. Results from the Utah beagle project seem to confirm these projections for 226Ra, 228Ra and, perhaps, for 90Sr. The lowest doses at which malignant bone tumors were observed in animals injected with these radium isotopes were about 0.9 Gy (226Ra) and 3 Gy (228Ra). For the beta emitter, 90Sr, the lowest doses at which bone tumors were seen were about 18, 50, and 70 Gy with an expectation for naturally occurring tumor of about one. Twenty-six of the two hundred and thirty-three Utah beagles given monomeric 239Pu that developed skeletal malignancies had doses between 0.02 and 0.51 Gy (80 of these dogs had skeletal doses of less than 0.9 Gy). Three dogs of 54 given 241Am with doses lower than 0.9 Gy had bone tumors at 0.23, 0.56, and 0.88 Gy with the expectation of about one naturally occurring case. For 25 animals injected with 228Th at skeletal doses below 0.9 Gy, one bone tumor dog had a dose of about 0.4 Gy, and the expectation of a dog with natural tumor among the group was only about 0.38. Five beagles of 74 given 224Ra with resulting doses of less than 0.9 Gy died with skeletal malignancy at 0.32 Gy or less with an expectation for non 224Ra induced tumor of about one. It appears that Raabe's proposal might be confirmed for some but not all of the radionuclides used in the Utah studies. Models presented in earlier papers by Raabe provide results that are somewhat different from his recent abstract and compare more favorably with those cited herein for Utah dogs. Re-examination of our data for these analyses has suggested a novel concept for calculation of carcinogenic dose to endosteal bone surfaces.