|Year : 2016 | Volume
| Issue : 4 | Page : 173-176
Review of the linear nonthreshold model
Badanidiyoor S Rao
Former Head, Radiological Physics and Advisory Division, BARC, Mumbai, Maharashtra, India
|Date of Web Publication||13-Feb-2017|
Badanidiyoor S Rao
Former Head, Radiological Physics and Advisory Division, BARC, Mumbai, Maharashtra
Source of Support: None, Conflict of Interest: None
|How to cite this article:|
Rao BS. Review of the linear nonthreshold model. Radiat Prot Environ 2016;39:173-6
Low-level radiation risk has been a controversial subject for decades. Uncertainties in the risk estimates resulting from a number of confounding factors and statistical limitations in the low-dose region have provided a fertile ground for debate. Scientists can be divided into a number of groups with diverse viewpoints. These include those who support supra-linear response, bystander signaling,, those who believe in the linear nonthreshold model (LNT), those who believe in the possibility of a practical threshold  of a few 100 mSv, and radiobiologists who believe in the theory of “hormesis” (protection by low levels of radiation).,,,, Events and ideas that have led to this controversy are outlined here.
The ability of radiations to cause tissue injury was recognized soon after the discovery of X-rays and radioactivity. Since the main application of X-ray was for diagnostic radiology, the early effects seen were erythema, skin irritation, and dermatitis, as a result of large skin doses from soft X-rays. Even though the International X-ray and Radiation Protection Committee (IXRPC) was set up in 1928 to provide adequate protection to radiation workers, due to lack of human data, the effects considered were mostly deterministic in nature. The dose limit was approximately 1 Sv (currently estimated based on the restriction in working hours). Subsequent to the data on radiation injuries in human beings and from model experimental systems, in 1934, the IXRPC reduced the limits to 500 mSv/year (threshold-based estimate). A further reduction of the dose limit to 3 mSv/week was recommended by the International Commission on Radiological Protection (ICRP) in 1951. Subsequent follow-up of the atomic bomb survivors by Atomic Bomb Causality Commission brought to light the ability of radiations to induce leukemias and some other solid cancers. Subsequent follow-up of the atomic bomb survivors progressively brought down the effective dose limits to 20 mSv/year averaged over a 5-year period. Muller's work demonstrating the ability of ionizing radiations to induce mutations in fruit flies (Drosophila melanogaster), brought to focus the ability of radiations to induce stochastic effects. Alteration of genetic information was believed to be the mechanistic bases for the induction of stochastic effects. The biological effects of ionizing radiation (BEIR) I Committee  of the US National Academy of Sciences reviewed these data and recommended to switch over to the LNT model from the threshold model for the estimation of risk of both radiation carcinogenesis and genetic effects. The United Nations Scientific Committee on the Effects of Atomic Radiation  report in 1958 also endorsed the LNT model which led to its acceptance by the ICRP.
By then, most of the data on radiation carcinogenesis were related to the induction of leukemias which have a short average latent period of 5–7 years. Only a few of the solid cancers had appeared during the first decade. Even though the leukemia incidence showed a linear quadratic type of response, data points at low doses were fraught with uncertainties. There was hardly any evidence of increase at doses <100–200 mSv. As for the solid cancers, in spite of inconsistency at low doses, response at high doses was fairly linear. It was believed by many that because of the statistical problems and a number of confounding factors, low-dose effects may be undetectable. Hence, the option chosen was to extrapolate the effects seen at high doses linearly to low doses. This model simplifies the calculations in radiation protection work related to optimization. The ICRP  has clearly stated the possibilities of uncertainties in the low-dose region and labeled the risk “inferential.” Even though many epidemiological studies were carried out to directly assess the risk of radiation carcinogenesis at low doses, these studies were either flawed or lacked the statistical power to detect any risk. Further, the high background incidence of carcinogenesis and genetic disorder in human beings renders it impossible to assess the risk at low doses with cohort size less than a few million. It must be borne in mind that radiation is a minor player in the induction of cancers or genetic disorders and contributes to only a small fraction of the natural incidence.
Uncertainty in the risk estimate at low doses stimulated a great deal of research activity to understand the mechanism of induction of damage at cellular level. Some of these studies , pointed to bystander signaling and genomic instability, thus suggesting new mechanisms for radiation damage. Several other researchers observed the phenomenon of adaptive response  as well as the possible beneficial effects of low-level radiation which may result in the reduction of background incidence of carcinogenesis. This phenomenon was called “hormesis.”,, This hypothesis was also put forth to explain the reduction in the incidence of cancer in atomic bomb survivors  at low doses. In the recent past, many studies have demonstrated the phenomenon of adaptive response and hormesis at cellular level in vitro, but human epidemiological studies in high natural background radiation areas (HNBRAs), do not show any significant reduction in cancer incidence. Nearly 10,000 Taiwanese people, who received an estimated average dose of 400 mSv because of living in Co-60-contaminated buildings, did not show an increased incidence of cancers.
Few other scientists suggested a possible threshold for the induction of cancers. Even though the ICRP 103 (2007) made a reference to many epigenetic factors, due to lack of consistency, reproducibility, and conditional expression of the aforesaid phenomena, the commission did not favor to consider these in risk assessment and radiation protection.
Human epidemiological studies in HNBRA , of Kerala on the west coast of India and Guangdong province in China  showed no evidence for enhanced cancer risk or genetic risk from low doses that are 3 to 4 times higher than normal background radiation areas (NBRA). These populations had received doses of 100–200 mSv more than the populations in NBRA. In the light of failure to detect excess cancers in HNBRA, it can be argued that radiation is a minor contributor to the induction of carcinogenesis as well as genetic effects as compared to many known and unknown genotoxic agents. The absence of increase in the effects of radiation is not surprising given the many confounding factors and statistical problems in detecting the excess risk at low doses. The data can still be treated as not inconsistent with risk estimates based on extrapolation from higher doses. Many arguments such as “Absence of evidence is not the evidence of absence” and “What we do not detect at low doses may very well be present” can be cited. Further, the supra-linear response seen in some of the low-dose points may also be due to the confounders.
However, it is important to point out many limitations of the LNT model. First of all, the uncertainties in extrapolating from high doses to low doses, and second, the dose and dose rate effectiveness (DDREF) of 2 used to assess the risk at low doses and low dose rates., Uncertainty and underestimation of the dose may also result in higher risk estimates. Many surrogate factors might have also contributed to excess risk. It is not unlikely that the induction of radiation effects may be linear quadratic owing to the presence of cellular repair mechanisms. In that case, the risks at low doses and low dose rates may be very small as the alpha component tends to zero at low doses and low dose rates. This cannot be adequately addressed by a DDREF of 2. In addition to intracellular DNA repair, defense mechanisms such as apoptosis and terminal differentiation may also play an important role. At present, it is well known that mutated cells will have to go through a series of changes to attain frank malignancy, and even after this, elimination of malignant cells by immune system is possible. As for the genetic effects, selection process works against mutated cells to participate in reproduction. Low recoverability factor of radiation-induced mutations, preimplantation death, and early spontaneous abortions are among the many natural defense mechanisms. Even though the induction of mutations and stochastic effects is closely related, there are many uncertainties in extrapolating cellular level effects to the stochastic effects in multicellular organisms. In the absence of human data, extrapolating the genetic risk seen in rodents to human beings has its own limitations. Hence, it may be argued that the biological defenses at low doses have not been adequately addressed in the LNT model.
As described earlier, many of the human epidemiological studies fail to provide any association between radiation exposure and cancer risks as well as genetic risks., Even the progenies of atomic bomb survivors  exposed to 500 mSv do not show any increase in the incidence of genetic disorders as compared to the children of unexposed parents. Following the Chernobyl accident, many European populations failed to show an increased incidence of cancers. Recently, a few epidemiological studies on radiation workers and emergency workers showed an association between the doses received and excess incidence of cancers. Three meta-analyses based on pooling of 22 case–control studies , suggest an association between lung cancers and radon exposures above 100 Bq/m 3. Retrospective studies , on radiation workers in 15 countries also suggest the possibility of a small excess risk among radiation workers. Statistical uncertainties in the data points are large and do not exclude the possibility of the absence of an increase. In the light of many confounders, it is difficult to demonstrate a causal relationship between the radiation and excess cancers. Furthermore, presence of some surrogate factors leading to such excess risk cannot be ruled out.
The idea of a possible threshold for the induction of stochastic effects is not without basis. There is a possibility of an adaptive response by sustained stimulation of repair pathways. Low-dose chronic exposures may modulate gene expression resulting in the elimination of modified cells by apoptosis and other mechanisms. The possibility of chronic exposures to stimulate the immune response cannot be ruled out. It can be hypothesized that modulation of repair pathways may result in more of the damage to go through the error-free recombination repair route. If this can repair the damage induced by other environmental carcinogens and mutagens, low-level radiation can be protective. This has not been clearly seen in human populations exposed to low levels of radiation. Such hypothesis needs further investigations by molecular biological research involving proteomics and genomics.
Even though the LNT model is convenient for calculations related to radiation protection problems, it comes with many disadvantages too. Overestimation of risk creates fear and psychological pressure among exposed populations. This may also compel the reduction of dose limits to unrealistic levels. Interventional strategies to prevent insignificant doses can result in enormous wastage of resources in the event of nuclear disaster. For example, the evacuation of people to avert small doses may result in less good and more harm by the way of psychological trauma. Destruction of agricultural products, livestock, milk, and dairy products can result in secondary problems such as malnourishment and financial crisis. Psychological pressure may result in abortion of desirable (planned) pregnancies. Low-dose limits can also create difficulties in complying with dose limits in complex operational conditions.
From the preceding discussion, it is clear that any of the suggested models for the estimation of radiation risk at low levels is fraught with many uncertainties. The most important reason for this being the many confounding factors which constitute the major source of risk. The radiogenic cancers are practically indistinguishable from those caused by other genotoxic agents. Low-dose epidemiological studies will not provide reliable information until we can specifically identify and distinguish a radiogenic cancer. Until then, the LNT model might be used exclusively to fix the dose limits for occupationally exposed and the general public. However, as cautioned by the ICRP 103 (2007) general recommendations, risk assessments to large populations exposed to very small doses are not justified. Such estimates are full of uncertainties and not consistent with many observations. From the past experiences based on the interventions in the management of disasters, the risk and harm should be carefully assessed and balanced. Exposure to 5–10 times the background radiation levels may not result in any increase in cancer incidence and should not elicit panic and overreactions. Such undetectably small risks must be put in the proper perspective and do not warrant the kind of publicity and attention they invariably attract, eventually triggering panic and confusion in the minds of the public. The problems and means of communicating with public regarding the health effects of radiation have been discussed by Wrixon  earlier. Hasty countermeasures may result in definite harm for imaginary benefits. Here, the LNT model should be used with caution. It is important not to stretch the LNT model beyond its scope. Overemphasis on the risks of low-level radiation may dampen the expansion of radiation industry and technology, thus depriving the human beings from the benefits they derive from it.
| References|| |
Goldberg Z, Lehnert BE. Radiation-induced effects in unirradiated cells: A review and implications in cancer. Int J Oncol 2002;21:337-49.
Mothersill C, Seymour CB. Bystander and delayed effects after fractionated radiation exposure. Radiat Res 2002;158:626-33.
1990 recommendations of the international commission on radiological protection. Ann ICRP 1991;21:1-201.
Hoel DG, Li P. Threshold models in radiation carcinogenesis. Health Phys 1998;75:241-50.
Redpath JL, Antoniono RJ. Induction of an adaptive response against spontaneous neoplastic transformation in vitro
by low-dose gamma radiation. Radiat Res 1998;149:517-20.
Calabrese EJ. Origin of the linearity no threshold (LNT) dose-response concept. Arch Toxicol 2013;87:1621-33.
Scott BR. Residential radon appears to prevent lung cancer. Dose Response 2011;9:444-64.
Doss M. Evidence supporting radiation hormesis in atomic bomb survivor cancer mortality data. Dose Response 2012;10:584-92.
Cohen BL. Lung cancer rate vs. mean radon level in U.S. counties of various characteristics. Health Phys 1997;72:114-9.
X-ray and Radiation Protection. Recommendations of the 2nd
International Congress of Radiology, 1928. Circular No. 374 of the bureau of standards, US Government Printing Office. Br J Radiol 1928;1:359-63.
Muller HJ. Artificial transmutation of the gene. Science 1927;66:84-7.
Muller HJ. Advances in radiation mutagenesis through studies on Drosophila
. Prog Nucl Energy 6 Biol Sci 1959;2:146-60.
US National Academy of Sciences (NAS) Committee on Biological Effects of Ionizing Radiation (BEIR I)/Genetic Panel 1956.
UNSCEAR. Sources, Effects and Risks of Ionising Radiation. United Nations Scientific Committee on the Effects of Atomic Radiation. 1959 Report to the General Assembly. United Nations, New York; 1959.
Pierce DA, Shimizu Y, Preston DL, Vaeth M, Mabuchi K. Studies of the mortality of atomic bomb survivors. Report 12, Part I. Cancer: 1950-1990. Radiat Res 1996;146:1-27.
ICRP. The 2007 Publication of the International Commission on Radiological Protection. ICRP Publication 103. Oxford, New York: Pergamon Press; 2007.
Nair RR, Rajan B, Akiba S, Jayalekshmi P, Nair MK, Gangadharan P, et al.
Background radiation and cancer incidence in Kerala, India-Karunagappally cohort study. Health Phys 2009;96:55-66.
Nair RK, Amma NS, Gangadharan P, Padmanabhan V, Jayalekshmi P, Jayadevan S, et al
. Epidemiological study of cancer in high background radiation area in Kerala. In: Wei LX, Sugahara T, Tao ZF, editors. High Levels of Natural Radiation: Radiation Dose and Health Effects. Amsterdam: Elsevier; 1997. p. 271-6.
Chen WL, Luan YC, Shieh MC, Chen ST, Kung HT, Soong KL, et al.
Effects of cobalt-60 exposure on health of Taiwan residents suggest new approach needed in radiation protection. Dose Response 2006;5:63-75.
Tao ZF, Kato H, Zha YR, Akiba S, Sun QF, He WH, et al
. Study on cancer mortality among the residents in high background radiation area of Yangjiang, China. In: Wei LX, Sugahara T, Tao ZF, editors. High Levels of Natural Radiation: Radiation Dose and Health Effects. Amsterdam: Elsevier; 1997. p. 249-54.
Darby S, Hill D, Auvinen A, Barros-Dios JM, Baysson H, Bochicchio F, et al.
Radon in homes and risk of lung cancer: Collaborative analysis of individual data from 13 European case-control studies. BMJ 2005;330:223.
Cardis E, Vrijheid M, Blettner M, Gilbert E, Hakama M, Hill C, et al.
Risk of cancer after low doses of ionising radiation: Retrospective cohort study in 15 countries. BMJ 2005;331:77.
Lubin JH. Studies of radon and lung cancer in North America and China. Radiat Prot Dosimetry 2003;104:315-9.
Direct estimates of cancer mortality due to low doses of ionising radiation: An international study. IARC Study Group on Cancer Risk among Nuclear Industry Workers. Lancet 1994;344:1039-43.
Wrixon AD. What can we tell people about the health effects of radiation exposure? Radiat Prot Environ 2016;39:117-21.