Summary and comments

The authors of these six papers provided interesting and thoughtful answers to ®ve questions concerning adaptive response (AR) and hormesis elicited by ionizing radiation (IR). The questions were: (1) How does the dose that induces AR relate to human and environmental (ecological) exposure? (2) What are the potential upand down-sides of having one's adaptive response induced? (3) Can the adaptive response be manipulated for medical and other bene®ts? (4) How does the adaptive response relate to the concept of hormesis?, and (5) Should knowledge of the adaptive response affect current risk assessment methods for carcinogenesis? The consensus response to Question #1 was that the doses known to induce the adaptive response (a few cGy delivered acutely or during a few hours) are several orders of magnitude higher than those likely to be encountered by most humans via acute exposures, and are, in fact, about equal to the average human life-time exposure. These doses are, therefore, irrelevant to induction of AR. Regarding Question #2 up and down side of AR stimulation and Question #3, can AR simulation be manipulated for medical and other bene®ts? The upside most mentioned was possible protection against radiation-induced tumors: the down side was that long-term AR, if such existed, might protect tumors against eradication by radiation therapy. However, the underlying theme in the responses to these two questions, and the others as well, was that much more knowledge of genetic mechanisms is needed before the effects of AR stimulation can be accurately predicted. Question #4 (relationship of AR to the concept of hormesis). [Stebbings de®ned hormesis, as the stimulatory effects caused by low levels of potentially toxic agents]. The general feeling was that AR and hormesis could have common features and mechanisms but what these might be can be determined only by continued detailed studies of the mechanisms that underlie AR and hormesis. These authors largely agreed that while there is substantial evidence for the existence of radiation hormesis, including features of certain dose response curves and epidemological studies, it suffers from an almost complete lack of mechanistic understanding. So while the consensus was that radiation AR response has been demonstrated beyond dispute (but not at all understood mechanistically), the majority view was that hormesis is a fascinating concept with tantalizing and highly important implications for radiation risk assessment, which will remain a concept until its underlying mechanisms are elucidated. Question #5: Should AR affect current risk assessment methods for carcinogenesis? Opinion was split on this important issue. Cai and Ikushima argued that the case for AR effects is so compelling that the linear, non-threshold dose response model should be reconsidered immediately. Olivieri, Rigaud, Skov and Mossman, on the other hand, felt strongly that changes in policy should not be made until reasonable progress had been made in unraveling mechanisms underlying AR. The majority point of view was summarized by Mossman and Ledesma: