Obesity and cancerBMJ 2007; 335 doi: https://doi.org/10.1136/bmj.39384.472072.80 (Published 29 November 2007) Cite this as: BMJ 2007;335:1107
- Eugenia E Calle, managing director of analytic epidemiology
Obesity is an important cause of type 2 diabetes mellitus, hypertension, and dyslipidaemia. The adverse metabolic effects of excess body fat accelerate the development of atheroma and increase the risk of coronary heart disease, stroke, and early death. The association between adiposity and cancer, however, is less well known. In this week's BMJ, Reeves and colleagues report a large prospective cohort study from the United Kingdom—the million women study—which assesses the association between body mass index (BMI) and cancer incidence and mortality.1
In 2002, the International Agency for Research on Cancer (IARC) convened an expert panel—which would draw on epidemiological, clinical, and experimental data—to evaluate the link between weight and cancer.2 It concluded that some colon cancers, postmenopausal breast cancers, endometrial cancers, kidney cancers, and adenocarcinomas of the oesophagus could be prevented by avoiding weight gain. Since the IARC report, many observational studies have investigated the association between adiposity and cancer. The results indicate that more cancers are probably linked to obesity than was thought originally, including adenocarcinoma of the gastric cardia, gallbladder cancer, liver cancer, pancreatic cancer, haematopoietic cancers, and advanced prostate cancer.3 4
Reeves and colleagues' study evaluates the effect of BMI on the incidence of cancer and mortality from cancer in more than a million women aged 50-64. Increasing BMI was associated with significantly increased incidence of postmenopausal breast cancer, endometrial cancer, kidney cancer, and adenocarcinoma of the oesophagus, in agreement with the IARC review. Higher BMI was also significantly related to the risk of leukaemia, multiple myeloma, non-Hodgkin's lymphoma, pancreatic cancer, and ovarian cancer.
These findings are generally in agreement with accumulated evidence to date. Most available studies of the relation between haematopoietic cancers and BMI—although smaller than the current study—have reported increases in the risk of non-Hodgkin's lymphoma, multiple myeloma, and leukaemia.3 4 5 6 7 Relative risks from these studies have generally been between 1.2 and 2.0.
Recent studies also suggest that high BMI is associated with increased risk for pancreatic cancer, with relative risk estimates for obesity generally between 1.5 and 2.0.3 4 8 9 However, some studies have found smaller positive associations. Evidence indicates that the association between adiposity and pancreatic cancer is non-linear, and increased risk is not seen until BMI reaches 30. Chronic hyperinsulinaemia and glucose intolerance may contribute to an increased risk of pancreatic cancer. A recent study suggests that people with insulin resistance who are in the highest quarter of fasting concentrations of serum glucose and insulin have more than double the risk of pancreatic cancer than those in the lowest quarter.10 Another study found that a tendency towards central (versus peripheral) weight gain was associated with a 45% increase in risk of pancreatic cancer after adjustment for the independent effects of general adiposity.11 The variability in estimates of risk associated with BMI for pancreatic cancer may partly result from using BMI, rather than a measure of central adiposity, as the measure of exposure.
Reeves and colleagues' study found no association between BMI and colorectal cancer in postmenopausal women—who comprised most of the women studied. Studies in different populations have consistently found that obesity is a stronger predictor of colorectal cancer in men than in women. The reasons for this sex difference are unclear. One hypothesis is that central adiposity, which occurs more often in men, is a stronger predictor of colon cancer risk than peripheral adiposity or general overweight. Recent prospective cohort studies examining the predictive value of various anthropometric measurements for the risk of colon cancer4 12 found that waist circumference was an independent risk factor for colon cancer that was stronger than BMI. This association was seen in both women and men. Thus, abdominal obesity is probably a more important predictor of colon cancer than general overweight; this might explain the differences in the findings of the UK study.
Substantial observational evidence suggests that increasing adiposity—both overall and central—is associated with increasing risk of many cancers. The strongest empirical support for mechanisms to link obesity and cancer risk involves the metabolic and endocrine effects of obesity, and the alterations they induce in the production of peptide and steroid hormones.3 The worldwide obesity epidemic shows no signs of abating, so insight into the mechanisms by which obesity contributes to the formation and progression of tumours is urgently needed, as are new approaches to intervene in this process.
This article was posted on bmj.com on 6 November 2007.
Competing interests: None declared.
Provenance and peer review: Commissioned; not externally peer reviewed.