[Excerpt] "INTRODUCTION Despite the nearly universal adoption of the linear nonthreshold dose response model (LNT) as the primary basis for radiation protection standards for the past half century, the LNT remains highly controversial and a contentious topic of discussion among health physicists, radiation biologists, and other radiological scientists. Indeed, it has been pointed out that the LNT has assumed the status of a paradigm, synonymous with an ideal, standard, or paragon or perhaps to some, a sacred cow.1 Reduced to its very basics, the LNT postulates that every increment of ionizing radiation dose, however small, carries with it a commensurate increase in the chance or risk that the exposed individual will suffer some undesirable radiation effect, and that the risk thus incurred is directly proportional or linearly related to the dose. The specific effects are termed “stochastic,” which has been defined as “of a random or statistical nature.”2 Stochastic or probabilistic effects of radiation may occur as a result of low doses and are generally taken to be cancers (including leukemias) and genetic defects in the progeny. The severity of these radiation-induced stochastic effects, should they occur, are independent of the dose that produced them; thus, even though the likelihood or probability of an occurrence may be small to negligible, any and all manifestations of a radiation induced stochastic effect will have equal severity. By contrast, higher doses of radiation are known to produce characteristic somatic or deterministic effects including erythema, epilation, sterility, diminution of blood cell counts, cataracts and, in very high exposures, acute and chronic radiation syndromes. Such frank biological effects are nonstochastic in nature (in fact, they were at one time termed “nonstochastic effects”) and will always be manifested once a particular minimum dose – i.e., a “threshold” – has been received. The severity of the effect is related to the dose. Below the threshold dose there will be no demonstrable effect; as the dose increases beyond the threshold, so does the severity of the effect, or the degree of harm. It bears repeating that the LNT is specifically applicable to the so-called stochastic effects of cancer and genetic defects in the progeny, and refers only to low doses and presumably low dose rates of ionizing radiation. What constitutes a low dose is open to interpretation. Many authors and publications simply use the term “low dose” without definition or further explanation. Indeed, there is disagreement among radiological health scientists as to just what constitutes a “low dose.” This is evident from the numerous radioepidemiologic studies that have been carried out over the years, and the application of the LNT down to doses that are fractions of the natural background radiation levels. The authoritative International Commission in Radiological Protection (ICRP) Publication 60, 1990 Recommendations of the International Commission in Radiological Protection, indicates that stochastic effects occur at “. . .doses well below the thresholds for deterministic effects,”3 and that for most tissues (Paragraph 58), severe effects are unlikely at dose rates less than about 0.5 Gy (Gray) y-1.4 The ICRP report further lists thresholds for various deterministic effects. The lowest threshold so listed is for temporary sterility in the male, given as a single acute dose to the testes of 0.15 Gy. Generally, however, the term low dose, as applied to radiation induced stochastic effects, is taken to be in the neighborhood of 0.1 Gy (10 rad) above the natural background dose acquired by an individual. In a recent report, the National Council on Radiation Protection and Measurements (NCRP) published human studies of cancer risks from “low radiation doses,” including in their range doses of several tens of Gy and noting that almost all risk coefficients for stochastic effects have been obtained from individuals whose doses have exceeded 0.1 Gy,5 and, most recently concluded, perhaps somewhat equivocally, that the exact shape of the dose response relationship for radiation induced carcinogenesis in humans at doses below about 0.05 to 0.1 Sv (Seivert) is not known but that there is sufficient experimental evidence to suggest that a threshold is unlikely to exist.6 This, coupled with several recent Position Statements by the Health Physics Society and the international conference “Bridging Radiation Policy and Science,” which pointedly noted that for lifetime doses below 10 rem (equivalent to 0.1 Gy for low Linear Energy Transfer (LET) radiation) stochastic effects are negligible or nonexistent, might by implication suggest that this level might serve as the boundary for what defines low dose. Although it should be noted that a case could be made for defining the 0.1 Gy dose level as either the upper or lower boundary for low dose.7 At least one investigator has proposed in a recent paper examining whether low-level ionizing radiation causes cancer that the upper limit for low dose be specified as 0.1 Gy.8 The purpose of the above discussion is to illustrate the underlying controversy and confusion that surrounds the LNT today, as well to underscore the lack of precision that sometimes accompanies the arguments of both the proponents and opponents of the LNT. Given that the LNT is a low dose phenomenon, there needs to at least be consensus on what is low dose, and such a consensus needs also to include consideration of other relevant and important factors such as the dose rate and specific stochastic end point (i.e., type of cancer or mutation). With this as a backdrop, the historical development and gyrations that led to the LNT as it is currently applied (or, some would say, misapplied) in radiological protection can be examined in the context of current scientific thinking with respect to radiation effects. It is not the purpose of this paper to endorse any particular position or to take sides but rather to present the story in a factual and fair minded manner. Hopefully, what follows will successfully achieve this goal. Thus, this paper will briefly review the scientific bases and supporting studies that led to the development and acceptance of the LNT in health physics. It will briefly touch on such topics as hormesis and other studies, such as the classic work of the late Robley Evans, that clearly demonstrate a threshold and nonlinear response for certain stochastic effects such as osteogenic sarcoma, along with the plethora of studies that suggest or have been interpreted to indicate that for at least some end points (i.e., cancers), response to ionizing radiation is consistent with the LNT model."
Ronald L.Kathren, Historical Development of the Linear Nonthreshold Dose- Response Model as Applied to Radiation, 1 PIERCE L. REV. 5 (2002). Available at http://scholars.unh.edu/unh_lr/vol1/iss1/5