Sorry, LNT-deniers.
The LNT model of radiation carcinogenesis isn't likely to change anytime soon. It is the best model of radiation risk of cancer versus dose. LNT doesn't "assume" the risk from low doses is absolutely linearly extrapolated from higher doses, as many LNT deniers like to frame the model. It is based on historical evidence, and it could also be called Linear-Quadratic (curved line instead of straight). But just like one ignores the curvature of the Earth when measuring a room for carpet, we can ignore the L-Q curvature in the low dose domain.
The LNT model started with early observations of increased cancer with radiologists, radiology patients, and others working with radiation. Recall that between 1895 and 1900 Roentgen, Becquerel, and Curie made all their discoveries regarding radiation. Radiology became a popular tool of medicine, with radiologists often intentionally exposing themselves to reduce fears of the new technology in their patients. Skin erythema was the initial measure used to determine radiation's safety (by then many radiologists had lost fingers and hair and started experiencing leukemia).
Researchers initiated animal testing in order to better understand the health effects of radiation. It was with this research that LNT was first formulated. Dose rate effects were recognized as well as differences in LET radiations. These results were consistent with the human results of the time. But the human data was lacking in statistical certainty and there's always uncertainty in drawing conclusions about humans from animal studies.
It is because of this history that we don't perform experimental research on humans. We know radiation is carcinogenic (risk to humans not well characterized at that time). Instead, we have to wait for accidental or other means of exposure in order to better characterize the risk to humans.
Enter WWII and the bombings of Hiroshima and Nagasaki. This event provided an opportunity to study a large number of humans exposed to various levels of radiation. That study is still ongoing but clearly shows LNT. As with all epidemiological studies (and particularly non-experimental ones) one can only statistically discern an increased incidence of cancer due to excessive dose down to a certain dose level. At some point, one can't say whether there really is a statistical increase in cancer or not. This can also be true of any two dose levels, say 1 Sv and 1.1 Sv. This is a statistical detection threshold and depends on the number of individuals exposed and the degree of dose separation one is trying to ascertain an effect difference in.
Another approach that we can employ today in order to better understand the risk relationship is using molecular biology. The human genome was only decoded in 2000 so this approach is only a bit over a decade old. It has given us some insights (like which genes are mutated in certain cancers), but has provided little information on overall risk per dose. And it is unlikely that this approach will provide an informed risk anytime soon. Why? Because cancer is a complex phenomenon involving genes, chromosomes, signaling, the immune system, etc. Cancer is like climate and molecular biology is like studying climate by looking at individual raindrops or sunlight photons. There are plenty of biological nonlinearities at play, negative and positive feedback loops, epigenetics, etc. It is ordered chaos, just like life itself (and climate) is ordered chaos.
Instead we need a chaos systems approach to risk. But we don't really care about the risk of cancer on a molecular level, we really need think only about the risk relative to background radiation. This is about 1 mrem/day. And we get cancer in a background of 1 mrem/day (yes, there are other carcinogens too, but ionizing & UV radiation are the most ubiquitous carcinogens our species has been exposed to over our long history. As far as we can tell we've always had cancer).
Is there any significant difference in the underlying ordered chaos at doses of 1 mrem/day compared with higher doses? I don't see evidence of it. And it will take that sort of evidence to really undermine LNT.
Let's recall what 1 mrem/day really is. 1 mrem is 0.1 ergs of energy per gram of tissue. Let's say you have 70,000 grams of tissue and 1 eV= 1.6E-12 erg, 1 day = 86,400 seconds, and generalize a photon energy of 300,000 eV.
Do the math and I get about 170,000 photons per second (if all the photon energy is absorbed, which it isn't, which means there are more photons per second). There are something like 50 trillion cells in the body. Even at this low dose the statistics (related to risk) is astounding. Note that it only takes about 34 ev per ionization, so one photon could result in over 8,000 ionizations. A few of these will occur within the cell nucleus, most outside of it. Those ionizations outside of the nucleus can still be detrimental to DNA (via the formation of reactive oxidizing species) or can result in other modification like epigenetic or cell signaling modifications.
An LNT denier would have to provide evidence of some biological phenomenon that doesn't work at background radiation levels (we get cancer) and also doesn't work at radiation levels where epidemiology can statistically discern a cancer risk increase. Yet this phenomenon just so happens to work within the window between these two mileposts.
Underlying all of this is evolutionary biology. In order to survive, DNA-based life has to be mutable to adapt to changing conditions. Cancer is the evolutionary survival of certain cells relative to their neighbors. Speciation (evolution by natural selection) deals with reproductive success in changing environments. As long as humans can maintain somatic genetic fidelity long enough to reproduce, it becomes inefficient for the integrated cellular system to maintain that fidelity into old age. This is why cancers usually develops in older folks. So relative to background levels, why would cells evolve mechanisms to maintain fidelity above the levels they are typically exposed to? That would be very inefficient. It is more reasonable that in increments of background level dose rates and total dose, the cancer risks increases.
No comments:
Post a Comment