A few weeks ago there was some minor media storm over a study claiming that High Fructose Corn Syrup is some kind of super food for cancer cells. A group of researchers in the US had taken cancerous pancreatic cells and fed them both fructose and glucose to see which one of these two types of sugar was a more effective nutrient for the cancer cells. The results were somewhat interesting in that pancreatic cancer cells seemed to grow and – more importantly perhaps – multiply faster on a fructose than on a glucose diet. In fact, some metabolic processes took place 120 – 150 percent faster when the cells were provided pure fructose rather than pure glucose. Those who have a sufficiently strong technical bent can read the study themselves.
So far, so what? This is where it’s getting interesting. Studies on cancer cell metabolism are a dime a dozen and you rarely read about them in the media – unless the authors go out of their way to make the story interesting for the media. In this case, the hook for the media was in the last paragraph:
Therefore, fructose is a particularly significant dietary sugar component with important implications for patients with cancer, particularly given the significant dietary change that has occurred in human fructose consumption since the mid- 20th century. Our findings provide important insights into recent epidemiologic studies that have identified refined fructose as an independent risk factor for pancreatic cancer, and identify fructose-mediated actions as a novel therapeutic cancer target.
What they are getting at is this: since the middle of the 20th century, so-called high-fructose corn syrup (HFCS) has become one of the most important industrial sugars in the United States (less so elsewhere, mostly because the use of HFCS in lieu of other sugars is a particular artefact of American agricultural policy). And for quite some time, food activists and nutritionists of varying credibility have claimed that HFCS is a particular horrid type of sugar that’s responsible for pretty much every human health problem under the sun (except those caused by meat, of course).
But so far, there has been very little proof that HFCS is indeed particularly bad for humans health – for all we know so far, HFCS is nutritionally pretty much indistinguishable from honey or maple syrup, and humans have been eating these for quite a long time. While HFCS may indeed be bad for our health, there is no reason to believe that is worse than honey, maple syrup, or the sugar contained in fruits (which is mostly fructose, by the way).
That’s why this study made such an impact. For the first time, there was a serious academic paper that showed that fructose in particular was likely to have a more negative impact on human health than glucose. And to make sure that the media got the message on this one, the corresponding author – a certain Anthony P. Heaney – told reporters
"I think this paper has a lot of public health implications. Hopefully, at the federal level there will be some effort to step back on the amount of high fructose corn syrup in our diets"
And because Ms. Fox – the Reuters in-house health and science reporter who wrote the story – has been following the HFCS debate as any decent health and science reporter should, she included plenty of background information on the issue, including the superficially frightening statistic that
U.S. consumption of high fructose corn syrup went up 1,000 percent between 1970 and 1990, researchers reported in 2004 in the American Journal of Clinical Nutrition.
Never mind that this is already 20-year-old data – lets assume that consumption of HFCS has at the very least stayed at that level since then (and for all I know, it may have increased) – if fructose is a particular risk factor for cancer, shouldn’t a 1,000 percent (ten-fold) increase in its consumption have resulted in a pretty good up-tick of overall cancer rates in the US? Well, at least in the case of pancreatic cancer, this seems not to be the case. According to the National Cancer Institute, pancreatic cancer incidents per 100,000 people haven’t changed much since at least 1986
Yes, it’s only one particular type of cancer – but it’s precisely the type of cancer the group has been studying, and considering how radically different various types of cancer are in their metabolism and pathology, it’s not really good science to assume that just because something happens in the case of pancreatic cancer cells that it would also happen in the case of, say, endometrial cancer. So – if Heaney’s team is right that fructose is a particularly potent food for pancreatic cancer cells, and if the consumption of fructose has really gone up by an entire order of magnitude over the last forty years, then it stands to reason that pancreatic cancer rates should have gone up significantly.
But maybe fructose doesn’t cause pancreatic cancer, only makes it worse? If by worse we mean more deadly, then the data does not bear that out, either:
And if the graphics are not clear enough, let’s quote the National Cancer Institute verbatim:
Unfortunately, overall pancreatic cancer incidence and mortality rates have changed very little throughout the past three decades.
Since I’m not really a scientist (though I can at times give the impression), I am hesitant to speculate too much about what this data might mean for the hypothesis of Heaney and Co. So far, however, I’m not convinced. It would be one thing if we had an unexplained epidemiological riddle, such as: why have pancreatic cancer rates gone up over the last few decades? Or – why have mortality rates for pancreatic cancer rates increased so much since about the 1970s? In that case these findings would be really interesting: ah – maybe the increased cancer/mortality rates can be linked to the way glucose and fructose are metabolized differently by cancerous cells…
Yet, there is no such epidemiological riddle. There is in fact nothing really interesting going on in the epidemiology of pancreatic cancer (other than the fact that African American males have more of it than White males).
However, Heaney et. al. claimed in their introduction that
Increased obesity due to increased total energy consumption and reduced activity now contributes to 15% of U.S. cancer deaths. […] In addition to higher fat intake, a large increase in refined carbohydrate intake has occurred, which itself has been hypothesized to be a risk factor for several cancers.
To support their claim, the authors cite two studies in particular:
Calle EE, Thun MJ. Obesity and cancer. Oncogene 2004;23: 6365–78.
Kaaks R, Lukanova A. Effects of weight control and physical activity Kaaks R, Lukanova A. Effects of weight control and physical activity. Ann N Y Acad Sci 2002;963:268–81.
My curiosity tweaked, I read the studies myself – and found something rather very odd: neither study as much as mentions carbohydrates, never mind refined carbohydrates. I kid you not: not a single mention of carbohydrates, refined or otherwise, in either the Calle and Thun, or the Kaaks and Lukanova paper. Not even the word sugar or starch is mentioned in either. Both papers discuss in great detail how overweight has been found to contribute significantly to various forms of cancer. Neither discusses in any significant detail the role of different nutrients.
In this case it’s really not necessary to have an advanced degree in science to recognize that it is not good science to claim that some studies have drawn a link between carbohydrates and cancer when these studies do not even discuss carbohydrates in passing. Not good science at all.
At this point, I couldn’t help myself but dig even more deeply into the references. I was now particularly interested in the claim that
Increased refined fructose consumption, in particular, has been highlighted as conferring greater pancreatic cancer risk than other sugars in several recent large epidemiologic studies
This was based on the following two studies:
Michaud DS, Liu S, Giovannucci E, et al. Dietary sugar, glycemic load, and pancreatic cancer risk in a prospective study. J Natl Cancer Inst 2002;94:1293–300
Larsson SC, Bergkvist L, Wolk A. Consumption of sugar and sugarsweetened foods and the risk of pancreatic cancer in a prospective study. Am J Clin Nutr 2006;84:1171–6.
I don’t know anything about Michaud and his colleagues, but I happen to know that Larsson and her team are pretty good when it comes to epidemiological cancer research. Based on their bibliography, they seem to enjoy nothing more than running countless correlation analyses of various Swedish Cohort Studies, looking for the most amazing connections between various substances and different cancers. In this particular case they were looking at findings which indicate that
hyperglycemia and hyperinsulinemia may be implicated in the development of pancreatic cancer
Frequent consumption of sugar and high-sugar foods may increase the risk of pancreatic cancer by inducing frequent postprandial hyperglycemia, increasing insulin demand, and decreasing
Very interesting indeed, and they promptly find exactly such a link. But – what does this have to do with the different metabolization of glucose and fructose by pancreatic cancer cells?
Notice also the fact that Larsson et. al. talk about sugar and high-sugar foods in general. They do not discuss fructose, or high-fructose corn syrup. In fact, in reference to a US study on pancreatic cancer risk and sugar intake they state explicitly that
Results of this US prospective study may not be directly comparable with the results in the current study, however, because sucrose is the sugar added as a caloric sweetener to soft drinks in Sweden, whereas high-fructose corn syrup is the major source of caloric sweeteners in soft drinks in the United States.
Sucrose. Not fructose. Never mind that sucrose is eventually broken down into fructose and glucose in the body before it is delivered as food to the various cells, carcinogenic or otherwise. The point here is that Healey et. al. are trying to show that high-fructose corn syrup is a contributing factor to pancreatic cancer based on a study that explicitly points out that it does not deal with high-fructose corn syrup, but sucrose. Very different.
At this point I did not expect too much from looking at the Michaud study. But it turned out to be a little better than the others: Michaud et. al. did indeed find a link between fructose intake and elevated risk of pancreatic cancer – among overweight women, particularly those who engaged in little physical activity. The link was not found among lean women who were physically active. The study did not include men (not because Michaud et. al. are sexist, but because the 1976 Nurses’ Health Study did not include men).
So, while the Michaud study does in fact find a link between fructose intake and cancer risk among some women, it is a far cry from what Heaney et. al. are claiming. Even more interesting, Michaud et. al. explicitly stress that
Although fructose intake may itself play an important role in the risk of pancreatic cancer, it may also be a marker of a high-sugar diet. More studies are needed to elucidate the precise role of fructose in pancreatic carcinogenesis
Have there been such studies since 2002? Possibly, but nothing decisive – at least not based on the citations included by Heaney et. al. There is plenty of discussion on the way cancer cells metabolise fructose, but the critical question of how fructose contributes to carcinogenesis is not addressed.
But if we do not know whether or not fructose is is a more important factor in carcinogenesis than other types of sugar, there is little point in discussing whether or not the consumption of high-fructose corn syrup should be reduced. Besides, there’s plenty of fructose in foods not made with HFCS. Even removing fructose from all foods would not make a difference since sucrose is broken down into fructose and glucose in the body, since our cells do not use sucrose.
As far as I’m concerned, the findings by Heaney et. al. have interesting implications for the treatment of pancreatic cancer – but in terms of epidemiology, they are likely meaningless. That’s not really a shocker, though, since in-vitro studies on pretty much anything are rarely of great epidemiological significance. They can be – but they rarely are.
The real shocker, however, is the absolutely dismal level of scientific integrity in this paper. It is hard to swallow that a paper that misrepresents its sources to such a degree made it through the peer review process at a major international cancer research journal. And it is completely beyond the pale that the authors of this paper had the audacity to go to the media with grandiose claims about the dangers of high-fructose corn syrup to the general population when there’s almost nothing in their paper to actually support this claim. What are these people thinking, and what exactly is it they are trying to achieve? Because if their goal is good science, then they have clearly failed.