Elsevier

Environmental Research

Volume 166, October 2018, Pages 175-204
Environmental Research

Review article
The additive to background assumption in cancer risk assessment: A reappraisal

https://doi.org/10.1016/j.envres.2018.05.015Get rights and content

Abstract

The assumption that chemical and radiation induced cancers act in a manner that is additive to background was proposed in the mid-1970s. It was adopted by the U.S. Environmental Protection Agency (EPA) in 1986 and then subsequently by other regulatory agencies worldwide for cancer risk assessment. It ensured that cancer risks at low doses act in a linear fashion. The additive to background process assumes that the mechanism(s) resulting in induced (i.e., treatment related) and spontaneous (i.e., control group) cancers are identical. This assumption could not be properly evaluated due to inadequate mechanistic data when it was proposed in the 1970s. Using the findings of modern molecular toxicology, including oncogene activation/mutation, gene regulation, and molecular pathway analyses, the additive to background assumption was evaluated in the present paper. Based on published studies with 45 carcinogens over 13 diverse mammalian models and for a broad range of tumor types compelling evidence indicates that carcinogen-induced tumors are mediated in general via mechanisms that are not identical to those affecting the occurrence of the same type of spontaneous tumors in appropriate control groups. These findings, which challenge a fundamental assumption of the additive to background concept, have significant implications for cancer risk assessment policy, regulatory agency practices, as well as fundamental concepts of cancer biology.

Introduction

This paper assesses a critical, but overlooked area of cancer risk assessment (i.e., cancer dose-response assessment), the additive to background assumption, that essentially ensures low dose linearity in the estimates of carcinogen exposure risks. This assumption was proposed for application to cancer dose-response assessment by Crump et al. (1976). A decade later it was incorporated into governmental risk assessment policy and practices during 1986 (Anderson, 1983, Crump, 1984, U.S. Environmental Protection Agency (EPA), 1986) and has continued to the present (U.S. Environmental Protection Agency (EPA), 2005, European Food Safety Authority (EFSA) Scientific Committee et al., 2017). This assumption was proposed during the mid 1970s when it was not possible to assess its scientific validity with the oncogene revolution starting in the mid-1980s and the continued clarification of molecular mechanisms for spontaneous and induced tumors to the present. It is now possible to evaluate the scientific validity of the additive to background assumption. The present paper demonstrates that the additive to background assumption that spontaneous and induced tumors occur via identical mechanisms is not compatible with the vast body of modern molecular findings. Prior to assessing the additive to background hypothesis, a brief historical reconstruction of how linearity at low dose was adopted for cancer dose-response assessment by U.S. regulatory agencies during the 1970s is presented, providing the necessary scientific and regulatory contexts and introduction needed to assess the additive to background assumption.

Section snippets

The Thanksgiving Cranberry Scare of 1959

Within five years following the National Academy of Sciences (NAS) Biological Effects of Atomic Radiation (BEAR) I Genetics Panel report (NAS/NRC, 1956) recommending the use of linear dose response modeling in risk assessment, Nathan Mantel and Raymond Bryan (1961) would publish their landmark paper on cancer risk assessment. The modestly entitled paper, Safety Testing of Carcinogenic Agents, was based on the use of the tolerance distribution probit dose response model. The probit model was

Ensuring linearity at low dose

The origin of the additive to background cancer risk concept was first suggested by Platt (1955) in a letter-to-the-editor of the Lancet. He proposed that the principal effect of a carcinogenic agent is likely to alter the functioning of a cell in such a manner that its clonal expansive descendants proliferate at a greater rate than other nearby cells. He then suggested that if aging were seen as an extended process of numerous cell divisions then cancerous processes would be a function of such

Discussion

The decision to adopt additive to background by the U.S. EPA for cancer risk assessment was influenced by the U.S. NAS SDWC (NAS, 1977) based on the Crump et al. (1976) paper, as well as the OSHA (1980) Carcinogen Hearings and Hoel (1980). Schneiderman, a member of that SDWC, summarized its recommendations in a follow-up paper, indicating that known carcinogens have never induced cancers in people that have never been seen before (i.e., novel cancers) (Schneiderman and Brown, 1978). With this

Conclusions

The question proposed in this paper is that posed by Hoel (1997), that is:

“Whether or not the original simple idea of background additivity is consistent with today's biology and whether the concept, if true, has any value for quantitative risk estimation.”

  • 1.

    Multiple, complementary and converging lines of evidence indicate that the “original simple idea” of additive to background for the induction of tumors that became incorporated into regulatory agency (i.e., EPA) cancer risk assessment is not

Declaration of Interest

None.

Funding Sources

This work was supported by the US Air Force Office of Scientific Research (AFOSR FA9550-13-1-0047) and ExxonMobil Foundation (S18200000000256). The U.S. Government is authorized to reproduce and distribute for governmental purposes notwithstanding any copyright notation thereon. The views and conclusions contained herein are those of the author and should not be interpreted as necessarily representing policies or endorsement, either expressed or implied. Sponsors had no involvement in study

References (186)

  • A.G. Knudson

    Mutation and human cancer

  • D.P. Lovell

    Dose-response and threshold-mediated mechanisms in mutagenesis: statistical models and study design

    Mutat. Res.

    (2000)
  • R.W. Lutz et al.

    Statistical procedures to test for linearity and estimate threshold doses for tumor induction with nonlinear dose-response relationships in bioassays for carcinogenicity

    Regul. Toxicol. Pharmacol.

    (2002)
  • R.R. Maronpot et al.

    Mutations in the ras proto-oncogene: clues to etiology and molecular pathogenesis of mouse liver tumors

    Toxicology

    (1995)
  • H. Marxfeld et al.

    Differentiation of spontaneous and induced mammary adenocarcinomas of the rat by gene expression profiling

    Exp. Toxicol. Pathol.

    (2006)
  • W.S. Abbott

    A method of computing the effectiveness of an insecticide

    J. Econ. Entomol.

    (1925)
  • R.E. Albert

    Carcinogen risk assessment in the US Environmental Protection Agency

    Crit. Rev. Toxicol.

    (1994)
  • R.E. Albert et al.

    Considerations relating to the formulation of limits for unavoidable population exposures to environmental carcinogens

  • E. Albert et al.

    Rationale developed by the Environmental Protection Agency for the assessment of carcinogenic risks

    J. Natl. Cancer Inst.

    (1977)
  • E.L. Anderson

    Quantitative approaches in use to assess cancer risk

    Risk Anal.

    (1983)
  • M. Anderson et al.

    Oncogenes in mouse liver tumors

    Comp. Mol. Carcinog.

    (1992)
  • C.H. Anna et al.

    Ras proto-oncogene activation in dichloroacetic acid-, trichloroethylene- and tetrachloroethylene-induced liver tumors in B6C3F1 mice

    Carcinogenesis

    (1994)
  • Anonymous

    Scientific bases for identification of potential carcinogens and estimation of risks

    J. Nat. Cancer Inst.

    (1979)
  • P. Armitage et al.

    The age distribution of cancer and a multistage theory of carcinogenesis

    Br. J. Cancer

    (1954)
  • P. Armitage et al.

    A two stage theory of carcinogenesis in relation to the age distribution of cancer

    Br. J. Cancer

    (1957)
  • E.I. Azzam et al.

    Low-dose ionizing radiation decreases the frequency of neoplastic transformation to a level below the spontaneous rate in C3H 10T1/2 cells

    Rad. Res

    (1996)
  • Beninson D.J., 1988. Practical implications of the linear non-threshold dose-response relationship. In: Proceedings of...
  • P.E. Blackshear et al.

    Spontaneous mesotheliomas in F344/N rats are characterized by dysregulation of cellular growth and immune function pathways

    Toxicol. Pathol.

    (2014)
  • P.E. Blackshear et al.

    Gene expression of mesothelioma in vinylidene chloride-exposed F344/N rats reveal immune dysfunction, tissue damage, and inflammation pathways

    Toxicolgic Pathol.

    (2015)
  • H.F. Blum

    Cell Proliferation and the Growth of Tumor Masses

    (1959)
  • F. Blum et al.

    Relationships between dosage and rate of tumor induction by ultraviolet radiation

    J. Natl. Cancer Inst.

    (1942)
  • K.T. Bogen

    Linear-no-threshold default assumptions are unwarranted for cytotoxic endpoints independently triggered by ultrasensitive molecular switches

    Risk Anal.

    (2017)
  • W.R. Bryan et al.

    Quantitative analysis of dose-response data obtained with three carcinogenic hydrocarbons in strain C3H male mice

    J. Natl. Cancer Inst.

    (1943)
  • E.J. Calabrese

    Hormesis: why it is important to toxicology and toxicologists

    Environ. Toxicol. Chem.

    (2008)
  • E.J. Calabrese

    The road to linearity: why linearity at low doses became the basis for carcinogen risk assessment

    Arch. Toxicol.

    (2009)
  • E.J. Calabrese

    Origin of the linearity no threshold (LNT) dose-response concept

    Arch. Toxicol.

    (2013)
  • E.J. Calabrese

    Preconditioning is hormesis part II. How the conditioning dose mediates protection: Dose optimization within temporal and mechanistic frameworks

    Pharmacol. Res.

    (2016)
  • E.J. Calabrese

    Post-conditioning hormesis creates a “subtraction to background” disease process: biological, aging, and environmental risk assessment implications

    J. Cell Commun. Signal.

    (2018)
  • E.J. Calabrese et al.

    Hormesis: the dose-response revolution

    Annu. Rev. Pharmacol. Toxicol.

    (2003)
  • R. Carson

    Silent Spring

    (1962)
  • M. Cazorla et al.

    Ki-ras gene mutations and absence of p53 gene mutations in spontaneous and urethane-induced early lung lesions in CBA/J mice

    Mol. Carcinog.

    (1998)
  • B. Chen et al.

    Dose-dependent ras mutation spectra in N-nitrosodiethylamine induced mouse liver tumors and 4-(methylnitrosamino)−1-(3-pyridyl)−1-butanone induced mouse lung tumors

    Carcinogenesis

    (1993)
  • J. Chinsky et al.

    Comparison of chemically induced and spontaneous murine thymic lymphomas in RF and AKR mice: differential expression of c-myc and c-myb

    Proc. Natl. Acad. Sci. USA

    (1985)
  • J. Cornfield

    Carcinogenic risk assessment

    Science

    (1977)
  • D.M. Costle

    Environmental protection Agency. water quality criteria. Request for Comments

    Part V. Fed. Regist.

    (1979)
  • M. Crawford et al.

    Low-dose linearity: the rule or the exception?

    Human. Ecol. Risk Assess.

    (1996)
  • K.S. Crump

    Commentary to article by Heitzmann and Wilson

    BELLE Newsl.

    (1997)
  • K.S. Crump

    Bogen's critique of linear-no-threshold default assumptions

    Risk Anal.

    (2017)
  • K.S. Crump et al.

    Fundamental carcinogenic processes and their implications for low dose risk assessment

    Cancer Res.

    (1976)
  • T.R. Devereux et al.

    Role of ras protooncogene activation in the formation of spontaneous and nitrosamine-induced lung tumors in the resistant C3H mouse

    Carcinogenesis

    (1991)
  • Cited by (19)

    • Linear non-threshold (LNT) fails numerous toxicological stress tests: Implications for continued policy use

      2022, Chemico-Biological Interactions
      Citation Excerpt :

      Such a conclusion has profound implications for the risk assessment process [112] - see Ian Purchase discussion). Furthermore, exposure to carcinogens can markedly alter the occurrence of spontaneous tumors, reducing/preventing some background tumors and further challenging the concept and meaning of background tumor incidence [81]. Although this analysis raised many new questions for cancer risk assessment, it established that the additive-to-background assumption is not a general one, is often incorrect, should not have been adopted by EPA for cancer risk assessment, and lacks scientific validity.

    • Cover up and cancer risk assessment: Prominent US scientists suppressed evidence to promote adoption of LNT

      2022, Environmental Research
      Citation Excerpt :

      Over the past decade, one of us (EJC) has published a series of papers that document the occurrence of numerous scientific errors, striking ideological biases at the highest scientific levels, and deliberate misrepresentations of the genetic toxicology research record, with a goal to establish and sustain the linear non-threshold (LNT) dose response model for cancer risk assessment. These numerous irregularities and falsifications of the research record now serve as the historical foundation of cancer risk assessment in the US and worldwide (Calabrese, 2011, 2012, 2015a,b, 2016a,b, 2017a,b, 2018a,b, 2019a,b,c, 2020, 2021a,b,c). Despite being founded in 1970 and now more than five decades in operation, the United States Environmental Protection Agency (US EPA) has served as an unwitting vehicle to implement such scientific deceptions due to its failure both to explore the historical foundations of cancer risk assessment, much of which occurred prior to its creation, and to take action to correct the errors once the ramifications were understood (Table 1).

    • Resolving an Open Science-policy question: Should the LNT still be an omnibus regulatory assumption?

      2022, Science of the Total Environment
      Citation Excerpt :

      Background and exposure cancer mechanisms should be identical. Calabrese (2018) published a detailed evaluation of additivity, employing the methods and findings of molecular toxicology, including oncogene activation/mutation, gene regulation and molecular pathway analysis. Using published findings for 45 carcinogens, over 13 mammalian bioassays, and different tumor types, he found that carcinogen-induced tumors (in dichotomous studies of exposed versus controls) are mediated by different mechanisms from those affecting spontaneous tumors.

    • Stem cells and hormesis

      2022, Current Opinion in Toxicology
    View all citing articles on Scopus
    View full text