We’ve talked extensively over the last few years about obesity, metabolic syndrome, fasting, and type 2 diabetes. These are extremely important in the development of cardiovascular disease (heart attacks and strokes). This addresses the number 1 killer of Americans, but we have not yet touched upon the very close #2 killer of Americans – the big C – Cancer.
You can see that heart disease, stroke and cancer really dwarf the other causes of death. It is important to realize that many of the other major causes of death also have a metabolic base. This includes diabetes, Alzheimer’s disease, liver disease, and kidney disease. This means that fully 6 of the top 10 causes of death are metabolic in nature. The other major causes are infectious, smoking related (lung disease) and suicide/ accidents.
The links between obesity and cancer are quite preliminary because the obesity epidemic really only get started in 1977 or so. Prior to that, obesity was stable in prevalence so there was nothing to compare to. The other problem is that there is a prevailing view of cancer as a genetic disease that permeates current scientific thinking. For various reasons which we will get into later, Cancer, as a whole, is almost certainly NOT a genetic disease.
First of all, what do I mean by ‘Cancer’? Cancer is not a single disease. There are multiple types of cancer, all of which are different. For example, there are the common cancers, such as breast, colorectal, prostate, skin, pancreatic, liver etc. There are blood cancers such as the various leukemias and lymphomas. They are all different, but they also share some common characteristics. It is these common characteristics that I want to discuss.
It is like discussing professional sports. You can say that football, soccer, hockey, fencing and baseball are all different but they can all be considered together in general. All involve some sort of competition and physical skill despite all the differences. Similarly, cancer shares many commonalities. One of the most cited article in oncology (the study of cancer) is the classic Weinberg paper detailing 8 common characteristics. We will cover that in detail later, too.
The links between cancer and obesity was really solidified in 2003 with a large scale epidemiological study published in the NEJM. This was a huge prospective cohort study called the Cancer Prevention Study II. This means that participants were identified and enrolled as healthy people and then followed to see what happened to them. It started in 1982 and require 77,000 volunteers simply to enroll all the participants who numbered over 1 million. In 1984, 1986 and 1988, volunteers would personally call these million participants to see who died and why. That is truly mindboggling. After 1988, national databases made collecting this data much easier. The variable of interest was death from cancer.
Overall, the increased risk of cancer does not appear with modest degrees of increased body fat. There is no increase in risk in the ‘overweight’ group (BMI 25-30). From there it gradually rises and the relative risk for all cancers is 1.52 for those in the ‘morbidly obese’ (BMI >40) group. In plain English, that means that morbid obesity is associated with a disastrous 52% increased risk of dying from cancer compared to if you are merely overweight or normal weight. Certain cancers are worse than others, including liver cancer where your risk is increased by 452%!
The news is almost certainly worse than that, though. Lung cancer, shows an inverse relationship. The relative risk is 0.67 meaning that obese people have 33% LESS lung cancer. But that is highly likely due to the well known effect of weight loss and cigarette smoking. Since lung cancer is among the biggest cancer killers, this means that the 52% increased risk is almost certainly an underestimate. If you remove all the smokers from the cohort, then you start to see a positive association with weight and cancer even in the ‘overweight’ category. In non smoking women the relative risk went up to 1.88 in the BMI>40 or an 88% increased risk of cancer.
Similarly, due to the well known phenomenon of cancer cachexia (the tendency for advanced cancer patients to lose their appetite and weight), there is a tendency towards weight loss, which will similarly obscure the true link between obesity and cancer. This effect will, once again, lead to an underestimation of the risk.
From this data, researchers calculate something called the population attributable fraction (PAF), based upon how strong the association and the prevalence of obesity. This gives an estimate of how much obesity contributes to cancer in general. In men, the estimate is 4.2-14.2% and women it is 14.3-19.8%. In other words, about 15% of cancer overall can be directly attributed to obesity. Keep in mind that this is an estimate that only includes up to 1998 data. Since obesity has continued to surge upwards in the last 19 years, this fraction is certainly too low at present obesity rates. For some cancers, the risk is even higher. For endometrial cancer, for example, the PAF is an estimated 56.8%.
So, which cancers are the most linked to obesity? Breast cancer was one of the first cancers to be linked. Epidemiologic studies since the 1970s have consistently found this link, both with cancer occurrence and prognosis. In postmenopausal women, rates of breast cancer increase by 30-50% with increase body weight. Some studies have found that central adiposity is an additional risk factor but others have not. Morbidly obese women have breast cancer death rates 3 times higher than those that are very lean. The reasons for this link is not entirely known, but one strong hypothesis is that adipose tissue may increase estrogen effects. Endometrial cancer may show a similar link for the same reason.
However, other cancers are also strongly linked despite estrogen playing little or no role in cancer development. For example, adenocarcinoma (a type of cancer type) of the esophagus also shows a massive 52.4% PAF in the US, for reasons that are not yet clear. Kidney cancer is also massively increased in the obese. Pancreatic cancer, colorectal cancer, liver and gallbladder cancer all show lower degrees of association, but still significant nonetheless.
Certain cancers are not related to obesity at all. Lung cancer, shows little association with obesity, which makes sense since smoking plays a dominant role. Cervical cancer similarly shows no association. Again this makes sense since Human Papilloma Virus is thought to play the starring role. But ovarian cancer, and prostate cancer also show no evidence that obesity has any influence.
The bottom line is that obesity is an important factor in cancer in general, although not all cancers. Clearly, cancer is a multi-factorial disease, meaning that many different factors determine its course. This is similar to cardiovascular disease, where it is well known that there is no single cause of disease. Smoking, genetics, sex, menopausal state, inflammation, diet, exercise, stress, obesity, diabetes, high blood pressure, metabolic syndrome all play a role in its development. This does not minimize the importance of any of these factors, but simply means we must accept that multiple pathways are important. In heart disease, this is well established dogma.
However, in cancer, there has been a pervasive consensus that it is caused by a single problem – mutations, and that everything that causes cancer does so through genetic mutations. This is certainly true for some things such as ionizing radiation causing cancer. However, this so-called somatic mutation theory (SMT) is almost certainly wrong, for reasons that we will go into later in detail. The strong association with the obesity is also a great example. This is rarely discussed given that obesity accounts for 20-30% of common cancers (PAF). Becoming obese does not cause genetic mutations. Fat cells are not mutagenic. But it certainly does open the door to considering the hormonal/ metabolic side of cancer. Because if metabolic disease plays a key role in certain cancers, then prevention of those disease will depend upon reversing those metabolic defects. Again, a new hope arises.
Continue to Cancer Part 2 – Hyperinsulinemia and cancer