TO THE EDITOR: In their Invited Perspective, Duncan et al. (1) continue a defense of the linear no-threshold (LNT) model for low-dose radiation (LDR) but do not respond to Siegel et al. (2) regarding important issues within the Biologic Effects of Ionizing Radiation (BEIR) VII report. This usually means that the authors concur with those contents, or do not find them objectionable. Here are 2 concerns:
Both Siegel et al. (2) and the National Research Council (3) agree that at low doses in the range of 0−100mSv, there are no data supporting the LNT model. BEIR VII uses data to support the LNT model (4,5) down to about 20 mSv, but Siegel et al. demonstrate the BEIR VII effort shows the failure of the LNT model in the 0- to 100-mSv range. Duncan et al.'s (1) nonresponse to Siegel et al. (2) seems a tacit admission of BEIR VII’s failure to make a valid claim for linearity in the “low-dose range” of 0–100 mSv.
Siegel et al. (2) emphasize “at relatively low doses, there is still uncertainty as to whether there is an association between radiation and disease, and if there is an association, there is uncertainty about whether it is causal or not” (3). Duncan et al. (1) ignore this observation, which is key to their claims about the risks of low-dose CT scans.
More recent understandings of mutations disclose a substantial number of spontaneous, endogenous double-strand breaks (DSBs) (EDSBs), and further studies of the close fidelity of DSB repairs between EDSBs and radiation-induced DSBs (RIDSBs) for low doses/dose rates (as with CT scans) demonstrate that there can be no identifiable, increased CT-induced cancer risk compared with the background risk from spontaneous EDSBs in the whole body. This results from the body’s adaptive responses to LDR.
Many CT scans produce doses less than 10 mSv, most are less than 20 mSv, and all are low in the LDR range. For a typical, low-dose CT scan covering 10% of the body, current literature shows that such low doses affect only DNA in a small fraction of cells in the target mass/volume. The RIDSBs from those are only about 3 in 1 million of the spontaneous EDSBs occurring in the body over the same time. Un- or misrepaired RIDSBs from higher doses are about 0.001% of the un- or misrepaired EDSBs in the body over the same time. For an essentially equal repair fidelity of RIDSBs and EDSBs, as discussed previously (6), un- or misrepaired RIDSBs are only about 0.0003% of un- or misrepaired EDSBs in the body over the same time. Further, all un- or misrepaired DSBs still require other low-probability events (which are also addressed by adaptive response) to arrive at some cancerous prelude.
Finally, the U.S. government has recently reported that cancer incidence declined by about 1%/y, and cancer mortality declined by about 1.6%/y over recent years, whereas CT usage has expanded, in support of increasing early detection and decreasing cancer mortality. Duncan et al. (1) repeat the words that “a threshold requires processes that leave no cells harboring DNA mutations” (3). Contradictorily, Duncan et al. (1) then cite how DNA errors of EDSB repair can lead to inactivating tumor suppression genes through premalignant lesions. These are obviously background, spontaneous DNA events, and with large contributions of EDSBs harboring DNA mutations, the fallacy of the quotation (3) is apparent: large, spontaneous, EDSB backgrounds exist in the body due to its metabolism, environments, and other factors; thresholds exist because LDR stimulates adaptive responses to remove IRDSBs and EDSB backgrounds, an enhanced dose response that reduces the body’s inventory of potential cancer precursors.
Footnotes
Published online Jul. 20, 2018.
- © 2018 by the Society of Nuclear Medicine and Molecular Imaging.