How self-interest and deception led to the adoption of the linear non-threshold dose response (LNT) model for cancer risk assessment

• Published in: The Science of the total environment Vol. 898; p. 165402
• Main Authors: Selby, Paul B., Calabrese, Edward J.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 10.11.2023
• Subjects: Cancer risk assessment; Mouse; Mutation; Radiation; Specific-locus test; Threshold dose-rate response; Mutation; Mouse; Specific-locus test; Threshold dose-rate response; Radiation; Cancer risk assessment
• ISSN: 0048-9697; 1879-1026
• DOI: 10.1016/j.scitotenv.2023.165402

This paper clarifies scientific contributions and deceptive/self-serving decisions of William L. Russell and Liane Russell that led to the adoption of the linear non-threshold (LNT) model for cancer risk assessment by the US EPA. By deliberately failing to report an extremely large cluster of mutations in the control group of their first experiment, and thereby greatly suppressing its mutation rate, the Russells incorrectly claimed that the male mouse was 15-fold more susceptible to ionizing-radiation-induced gene mutations as compared with fruit flies. This self-serving error not only propelled their research program into one of great prominence, but it also promoted the LNT-based doubling dose (DD) concept in radiation genetics/cancer risk assessment, by the US National Academy of Sciences (NAS) Biological Effects of Atomic Radiation (BEAR) I Genetics Panel (1956). The DD concept became a central element in their recommendation that regulatory agencies switch from a threshold to an LNT model. This error occurred because of a decision by W. Russell not to report that a large cluster of control group mutations found in an experiment for which preliminary results were reported in 1951. This failure to report that cluster and similar clusters continued throughout the careers of the Russells, resulting in massive overestimation of low dose radiation risks supporting the LNT. The Russell database and the repeated claim that those data show that there is no threshold dose rate for mutation in irradiated mouse stem-cell spermatogonia, have provided mechanistic validation supporting the epidemiological LNT hypothesis for radiation-induced leukemias and cancers. This reanalysis supports the threshold model for both males and females, thereby rebutting epidemiological extrapolations from the NAS and EPA claiming support for the LNT hypothesis for cancer risk assessment. The implications of the Russell errors/deceptions, how/why they occurred, and their impact upon society are enormous and need to be addressed by scientific/regulatory agencies, affecting regulatory and litigation activities. [Display omitted] •The scientific foundation of LNT for cancer risk assessment is presented.•The role of the massive mouse mutation data by the Russells is highlighted.•Discrepancies and errors with the Russell data are documented.•Correction of these errors supports a threshold dose response.•The threshold, not LNT, should have been used for cancer risk assessment