Radiation's Big Lie
Did a Nobel laureate knowingly lie about the dangers of radiation in 1946?
Steven Cherry: Hi, this Steven Cherry for IEEE Spectrum’s “Techwise Conversations.” This is show number 68.
In 1946 Herman Muller was awarded the Nobel Prize in Medicine for work done on spontaneous gene mutation, including the effects of X-rays.
So Muller was the obvious guy to go to when right around the same time, the National Academy of Sciences formed a committee to offer expert advice to the government on the biological effects of atomic radiation.
In his Nobel acceptance speech and in the NAS committee meetings, Muller argued there are no safe levels of radiation exposure, a position the Academy came to adopt. That in effect influenced official policies toward radiation for decades.
The problem is, Herman Muller knowingly lied. So says Edward J. Calabrese, a professor of toxicology at the University of Massachusetts School of Public Health. He bases that opinion on a wealth of materials including formerly classified files and, recently, some overlooked correspondence between Muller and one of his coresearchers.
Professor Calabrese is the author of over 600 journal papers and more than a dozen books. He’s board certified in toxicology, is the editor in chief of the journal Human and Ecological Risk Assessment, is a member of three different toxicology societies, as well as the Society of Risk Analysis, and he’s my guest today. Ed, welcome to the podcast.
Edward Calabrese: Thank you very much.
Steven Cherry: Let’s start by noting that this question of whether there are benign doses of radiation is still hotly disputed, and no pun intended. And not just radiation, but I guess also the chemicals that we’re surrounded by. Can you take us through the two points of view?
Edward Calabrese: Yes, it’s—I would have to say, yes, it is hotly disputed today and has been for quite a while. And this is true for radiation as well as for chemicals, and it typically focuses in on the area of cancer-causing chemicals, as well as the ionizing radiation being a carcinogen. The issue is noncarcinogenic affect. Generally speaking, the federal government in its regulatory processes has assumed that that acts via a threshold mechanism, whereas the cancer endpoint has been assumed to be a nonthreshold, linear dose-response relationship with the implication that there is no safe level of exposure.
Steven Cherry: So if there is a threshold, then below that threshold, that would be considered a safe level of radiation exposure.
Edward Calabrese: That’s the interpretation; that’s correct.
Steven Cherry: So now you claim that Muller had himself done research that supported the idea that there are safe levels of radiation exposure. What was the research?
Edward Calabrese: I actually—in Muller’s own groundbreaking work that led to his Nobel Prize, his own dose responses really did not support a linear dose-response relationship. But he was particularly interested in trying to better define the nature of the dose response in the low-dose zone, and, in fact, he got several individuals to work on that project and inspired other people, obviously, in the wake of his groundbreaking discovery in 1927. But the data that emerged was not very clear. Some data actually seemed to support a threshold, and some clearly did not, but the problem with the work that Muller and the people of that era were doing was that the doses of radiation that they were using were extraordinarily high and have nothing to do at all with anything that we might view as remotely low dose today. In fact, they were probably exposing fruit fly germ cells at a level that would be equivalent to something like about a thousand X-rays in three and half minutes, at the lowest dose that they were testing.
Steven Cherry: So most of his research had to do with high doses of radiation and had nothing to do with this threshold question and what little research he did along those lines was ambiguous at best. Is that fair to say?
Edward Calabrese: Well, I think it was—that’s probably true, over the course of 20 years or so, from 1927 to let’s say, 1947, the levels of radiation that they investigated essentially got lower and lower and lower and I can’t say became extremely relevant to today’s issues but got out of the extremely irrelevant stages in terms of the high levels that they were studying back in the late 20s and mid 30s. But nonetheless, they were still very, very high, and basically, he was still in search of trying to understand the nature of the dose response in the low-dose zone. And so the data were not very clarifying at that point because people hadn’t made the effort to try to understand lower dose rates. And that effort was really improved by the Manhattan Project in this country, back in the 1940s when the U.S. government attempted to get a better handle on lower dose rates and what they might show in terms of a dose-response relationship for critical endpoints.
And it was really the work of Kurt Stern and two of his colleagues, Warren Spencer and Ernst Caspari, who were key colleagues with him. But also, Muller was a hired paid consultant to Stern during this time period, and during the world war period, Muller was a visiting professor at Amherst College in the same town that I live in and would make trips up to Rochester, and actually provided the fruit fly strain with which Stern and his coworkers would actually do their experiments with. The issue that emerged was that Stern and Muller operated under the belief that there was a linear dose-response relationship and that in the mutations that they were going to be evaluating, they expected that there would be, even at low doses, a linear dose response. And in fact, in the first study that was undertaken—a very large study, really done by this solo Dr. Warren Spencer—he reported what he believed was a strong support for a linear dose response. But this was to a very acute exposure to ionizing radiation and that is that the entire exposure was given over a period of several minutes, whereas the key question wasn’t so much acute exposure, it was actually more chronic exposure.
As it turns out in the chronic study, which was done far better in terms of laboratory improvements and research methodology than the earlier study by Spencer, what they found was quite a surprise. They found that the expected linear relationship was not supported, and what they observed was, in fact, something that would be more supportive of a threshold dose-response relationship. And this created a conflict—not for the actual lab researchers like Caspari, but actually it created a conflict for his boss, Kurt Stern, who truly was thinking in terms of support for the linear perspective. And so Stern tried to convince Caspari that the reason why his study showed no support for the linear model was because his control group values were reading artificially high or unexpectedly high or aberrantly high. And so basically he wanted to really discount his findings.
And so Caspari, good researcher that he was, he dug out the literature, he got lots of unpublished findings from Muller, and he put together a case that his own boss was wrong. And that is that his control group wasn’t aberrantly high; his control group was in the expected range of that which had been reported by many other excellent researchers for this type of fruit fly strain. And ultimately, what happened was that he got Stern to back down. And they accepted his findings and ultimately this supported the threshold dose response. Well, as it would go forward, the—and this is where Muller comes into the question—Stern contacted Muller and he said, Would you mind reviewing this big manuscript by Caspari that supports a—essentially, a threshold? And Muller said that he’d be available, and they sent it to him. And they got him—they sent it to him on November 6, 1946. And on November 12 he writes back to Stern indicating that he got it; he essentially went over the paper and he saw that this was in fact a major challenge to his perspective, that the results were contrary to what he would have thought would have happened, that in fact, he couldn’t challenge the paper because Caspari was an excellent researcher, they needed to replicate this again, and that this was really a very significant challenge to the belief that there was a linear dose response, because this study that was challenging it was the best study to date, and it was looking at the lowest dose rate that had every been looked at in such a study. It was actually the best test for what the nature of the dose response would be.
And so, what happened was that one month later, Muller goes to Stockholm to get his Nobel Prize—now, Muller and Stern and Caspari, and maybe a few others, but these are the people who actually know the data. But Muller goes, and he makes his Nobel Prize speech, and what he does is he tells the group of all the scientists, the dignitaries, the press, everybody who is there, and the world wide attention he’s gonna receive from this award—he tells them that one can no longer accept any consideration for the use of a threshold model, that all you can really accept now is the acceptance of a linear dose-response model. He said there’s no longer any room for the other alternative. And yet Muller had actually seen, one month before, the results of a study that he was a consultant on, a study that he provided the fruit fly strain on, and a study that was actually the best to that point in showing that in fact no support for the linear model but support for a threshold model. Yet he had the audacity to actually go in front of all the microphones, all these dignitaries, at the place where he was going to get to most significant award in science, and actually mislead, deceive the audience by saying there was no longer any possibility to believe this when in fact he had seen the future, and the future said this is a possibility. Now he could have easily gone in to the audience and said, This is a critical area, and in fact we don’t know what the answer is, and we need to do more research to try to figure this thing out; we need stronger, more rigorous efforts because the implications are important, what the nature of the dose response in the low-dose zone is. He could have said all of that; it would have been intellectually honest and actually the appropriate thing to say, but no, that’s not what he says. He tries to actually mislead the audience by saying there’s not even a remote possibility that this alternative exists, and yet he has seen it.
Steven Cherry: Now, it’s hard for anybody, including a scientist, to sort of publicly change his mind or back down from what he thinks. But I guess you think that Muller also had some political reasons for ignoring this research.
Edward Calabrese: I’m not sure I’d call it political, but I guess it might be. But my sense of it was that Muller and many of his radiation geneticist colleagues of that era—they believed that they were the only ones who understood the significance of mutation and what it might mean for the human genome. And they were of the belief that there was, you know, a linear relationship, that’s what they, I think, actually believed or thought was true, and that they had great concerns with not just X-rays but with, let’s say, atmospheric testings of atomic bombs that was just now taking start. And they were concerned with the survival of the human genome. He wasn’t even focused on cancer so much; he was focused on the health of the genome. And I think that what he did and what people of that group did, that time period did, was they confused, you might say, in a philosophical side of life, the ends and the means. His end may have been a good end, and that was to try to protect the human genome, but to get there by actually deceiving the scientific community and the leaders, the political leadership, and the public, by withholding proper interpretations and correct information by misleading them—especially from that prestigious a perch within society—to me, was a real egregious thing to do.
Steven Cherry: So you think that the consequences are lasting even to this day.
Edward Calabrese: Oh, not only that; it was in 1977—the first ever National Academy of Sciences Safe Drinking Water Committee, that was to guide the EPA on cancer risk assessment for drinking water carcinogens, essentially—what they did was in their write-up, in their book called Drinking Water and Health, they specifically stated that they adopted the perspective of that 1956 committee on low doses of ionizing radiation and applied it to geno-toxic carcinogens. And essentially within two years of that—passing on that information, EPA came out with the first of many, many, many risk assessments and drinking water standards and so forth based upon low-dose linearity. And it was entirely going back to the historical roots of that first BEAR 1 [Biological Affects of Atomic Radiation] committee. And it really goes back to Muller and his leadership, and it was really through this careful manipulation, orchestration, leadership—he was able to have not only an immediate effect but really a very long historical reach.
Steven Cherry: Now we started this—we started this podcast by talking about that it is still disputed, this question of linearity versus threshold and isn’t it to, just a sort of a conservative safe approach to accept the linearity, even if we’re not sure it’s true?
Edward Calabrese: Well, that’s a policy issue, but I would say that when one looks at it, what you really want to do is—there are consequences for acting on linearity. For example, when you have to, let’s say, clean up a contaminated site to a profoundly clean level, that costs not just millions of dollars, sometimes many multiple billions of dollars, and you can multiply that by manyfold that goes around this country and the world, until—well, a lot of times you might be buying cleanup, but you basically can’t invest that money in other types of R&D projects, so that’s one thing. Another thing is, is by accepting the belief in low-dose linearity, certainly for radiation, what that did was it markedly changes the view in terms of how—I’m just gonna say low doses of X-rays could be used therapeutically. Back in the 1940s, low doses of X-rays were used; it’s a competing technology, but it was used to treat many anti-inflammatory conditions, including arthritis. And it actually had a pretty good record for doing so. And as it turns out, as a result of the belief in a linear dose-response relationship, that entire technology was actually removed from the competitive market and replaced by, I’m gonna just say NSAIDS and other types of drugs that themselves have a whole series of other potential biomedical problems with, but it has a broad range of issues.
Steven Cherry: Thanks for all your research, and thanks for being on the show.
Edward Calabrese: Thanks so much.
Steven Cherry: We’ve been speaking with University of Massachusetts professor of toxicology Edward J. Calabrese about whether there are benign doses of radiation and whether a Nobel Laureate lied about that in 1946. For IEEE Spectrum’s “Techwise Conversations,” I’m Steven Cherry.
Announcer: “Techwise Conversations” is sponsored by National Instruments.
This interview was recorded 29 September 2011.
Audio engineer: Francesco Ferorelli
Follow us on Twitter @Spectrumpodcast