AACR 2014

The relationship between obesity and cancer, and other dietary factors

Dr Walter C. Willett - Harvard School of Public Health, Boston, USA

Can you tell us a little about the impact of diet on cancer?

Cancer is definitely a challenging topic to study and studying the causes of cancer is very difficult because what we have seen is that some factors operate very early in life, in childhood, early adult life; other factors may be operating at age 70 or 75 even in causing cancer so we really have to take into account, if we’re looking at diet, the whole life course. We have learned quite a bit though, one of the big factors that has emerged is that being overweight or obese is an important cause of many different cancers.

What about obesity and breast cancer?

During the premenopausal years overweight and obesity is not a risk factor but after menopause it is. So it turns out that gain of weight during adult life, and unfortunately in the US and I think in many other countries, many people it’s almost the norm to put on a pound or two a year, which doesn’t seem like a lot but after twenty or thirty years, by the time a woman reaches menopause, she can be thirty or forty pounds overweight. That will bump up breast cancer in quite an appreciable way.

Is this because of oestrogen?

It’s quite clear that a lot of the reason for the excess risk for breast cancer among overweight women is the extra oestrogen, that the fat cells are actually generating oestrogen. So an obese woman has about three times higher circulating oestrogens in her blood compared to a woman who is lean.

What do women need to do about this, apart from not get overweight?

The best thing, indeed, would be to not get overweight, not wait until we’re fifty and thirty or forty pounds overweight. But when we first put on three or four pounds, that’s the time to make some change in our lifestyle. That can be putting in a little more physical activity into our daily life and also being careful about what we eat. That, of course, is a big topic in itself and the conventional wisdom had been just avoid dietary fat, just avoid calories from fat and load up on all the carbohydrate you want to. It turns out that was really bad advice, that those extra carbohydrates can lead to being overweight as well.

So doctors need to find out ways psychologically of persuading their patients to avoid high carbohydrate meals.

That’s right. There have been now quite a few studies in the last ten years about what types of diets are most effective in helping people control weight. There may be different solutions for different people depending on the rest of their lifestyle and their personal preferences but overall it does seem something that’s like a Mediterranean diet is most effective because it’s healthy calories, it’s a relatively modest balance between fat and carbohydrate and it’s healthy forms of fat.

Let me ask you about colorectal cancer because you’ve been talking here at the AACR, one of the topics you talked about was folate in colorectal cancer. There’s a link, isn’t there? Any chance of preventing colorectal cancer or even treating it when you’ve got it?

Yes, we’re not 100% sure of the link between folate and colorectal cancer but I’d say I’m 95% sure about that now, specifically that low folate is a risk factor. Probably the best way still is to eat lots of fruits and vegetables, those are the main sources of folate, but I also think taking a multiple vitamin as a safety net is a good thing to do.

A prudent thing to do.

And that will make sure that we get enough folate and help us out with some other gaps that a lot of people do have in their diet.

What impact does alcohol have on different types of cancer?

Alcohol has long been established as a cause of oesophageal cancer, oral cancer, laryngeal cancer, but that’s usually with fairly high doses. What’s emerged, though, is that even very modest doses of alcohol will increase risk of breast cancer, we can see even with one drink every other day now there is a small but significant increase in breast cancer risk and the risk goes right up with higher doses. It doesn’t matter whether it’s wine, beer, liquor, it’s really alcohol in any form that counts. I should mention that there is benefit of moderate alcohol consumption for heart disease so there are some risks and trade-offs here. My advice if a woman does want to drink – keep it modest, buy a really good glass of wine and enjoy it but have it in a small amount of a wonderful wine.

What about things which were conventionally understood to be good for us, such as dairy?

Dairy is really interesting, we don’t have all the final answers on that yet. We, of course, do need some calcium and we have been advised to consume in the US three servings a day of dairy. It turns out we really don’t need that much for healthy bones to begin with. We are seeing with that high dairy intake, three or more servings a day, some increases in risk of prostate cancer and also some likely increased risks of endometrial cancer as well. So for dairy you don’t have to eliminate it but I think it’s a wise thing to be on the lower side, one or two servings a day. Interestingly there seems to be some special benefits of having some in the form of yoghurt as it relates to weight gain and diabetes risk.

How do you deal with the temporal aspects of these problems?

The temporal relationship, the timing, between what we eat and the diagnosis of cancer, it’s likely to be very long in some cases. We know that some cancers involve events that occur over a lifetime. Where there was most clear from the atomic bombing of Japan in World War II that if women were irradiated from that bombing while they were children or young adults a few decades later there was a higher risk of breast cancer. But if they were aged over 40 their breasts were almost immune to that radiation, there was very minimal, if any, increase in breast cancer risk later in life which really does tell us that our tissues are particularly susceptible while we’re at a younger age. So if there’s going to be a protective factor of diet it’s got to be present during that early period of life. Now that we are starting to look at diet during high school and breast cancer risk we are seeing things that we didn’t see during mid-life – that red meat consumption, for example, is more clearly related to breast cancer if it’s red meat consumed during high school years.

What would be your take-home message?

Despite some of the challenges, I think there are a number of things that we can use to help us guide our decisions if we want to be as healthy as possible. First of all it would be to maintain a healthy weight, or as close as we can to a healthy weight, throughout our life. That will help, we’ve known for a long time, diabetes and cancer risk but also it’s quite clear it will help reduce risk of many cancers now. Let me say that again. We’ve known for a long time that keeping our weight close to the ideal weight will reduce the risk of diabetes and cardiovascular disease, what’s new is that this will also reduce the risk of many important cancers as well. It’s a good idea to eat an abundant amount of fruits and vegetables, it will probably have a pretty small impact on cancer but it will benefit risk of cardiovascular disease. There are also some indications that the type of fat, which we know to be very important for preventing heart disease, may have some benefits for cancer as well, specifically less animal fats, more vegetable fats, just so that they’re not hydrogenated.