Weight and cancer risk

Friday February 15 2008

“Extra two stones boosts risk of cancer,” reads the headline in The Daily Telegraph . The newspaper adds that “the chance of developing five different types of cancer increases by 50 per cent if your weight goes up by more than two stone”.

The newspaper story is based on a review of previous studies  that presents good evidence for the link between weight and cancer risk. The results confirm some relationships between particular cancers and weight and provide an indication of the risk for less common cancers. The interpretation of the results is limited by weaknesses in the studies chosen for analysis. However, overall, this study had provided good evidence of the size of the link between increasing weight and cancer.

Where did the story come from?

Dr Andrew Renehan and colleagues from the University of Manchester, the University of Bristol and the University of Bern in Switzerland carried out this research. The study was partly funded by an award from the British Medical Association. It was published in the peer-reviewed medical journal: The Lancet .

What kind of scientific study was this?

The study was a systematic review of studies that looked at the effect that a 5kg/m² increase in body mass index (BMI) has on the risk of different types of cancer.

The researchers searched the global literature for studies that had followed a group of people over time and measured the rate of new cases of cancer (incidence). They only included studies that had also measured BMI at the start of the study and which had been published as a full report, i.e. they did not include letters, conference abstracts or studies. Other studies which grouped cancers together to report on rates, e.g. “all breast cancers” etc were also excluded from the analysis.

The researchers assessed the quality of each study they found and used these quality scores as part of the analysis. They pooled the studies using a technique called meta-analysis – a mathematical/statistical technique which combines the results of separate studies into one overall measure. From this, the researchers looked at the effect changes in BMI had on the risk of different cancers.

What were the results of the study?

The researchers found 141 articles (reporting 76 separate studies) that met the criteria for the meta-analysis. As some of these studies researched more than one group of people, they had a total of 221 datasets to combine. The studies were primarily from North America, Europe and Australia.

When they pooled the results, they found that in men, increasing BMI by 5kg/m² was strongly associated with increased risk in oesophageal, thyroid, colon, and renal cancers by 52%, 33%, 24% and 24% respectively. For other cancers in men, e.g. malignant melanoma, multiple myeloma, rectal cancer and leukaemia, increases in risk were significant but smaller. There was no increase in men’s risk of liver, gallbladder, pancreas, prostate and stomach cancers.

In women a 5kg/m² increase in BMI was associated with increases in risk of endometrial, gallbladder, renal and oesophageal cancers by 59%, 59%, 34% and 51% respectively. There were significant but weaker associations with leukaemia (17% increased risk), thyroid cancers (14%), postmenopausal breast cancer (12%), pancreas cancer (12%), colon cancer (9%) and non-Hodgkin’s lymphoma (7%). There was no increase in risk of cancers of the liver, stomach, ovaries, and rectum, or in risk of malignant melanoma.

The researchers also found that increasing BMI was more strongly associated with an increased risk of colon and rectal cancer in men than women; however, for renal cancer it was more associated with women. For many cancers, the association between increased BMI and risk of cancer were consistent across ethnicities; however, in some types of cancer there were conflicting or differing degrees of association. For example, increasing BMI increased postmenopausal breast cancer risk in women from the Asia-Pacific region, but not in other ethnic groups.

What interpretations did the researchers draw from these results?

The researchers conclude that increased BMI is associated with “increased risk of common and less common malignancies”. They say that for some cancer types, associations differ between sexes and populations of different origins.

What does the NHS Knowledge Service make of this study?

  • The technique used to pool the results (meta-analysis) can sometimes be problematic when the combined studies are different from each other due to factors such as different baseline populations, the way BMI is measured etc. In some cases, this pooling of data may not have been appropriate. Some results, e.g. the association with thyroid cancer and for women with endometrial, leukaemia, and lung cancer, should be interpreted with caution for this reason. 

  • The review gives a measure of the association between a factor (i.e. BMI) and the risk of cancer, it cannot establish which came first, the high BMI or the cancer. As such, it cannot prove that high BMI causes cancer. However, the researchers say that as increased BMI was associated with some cancers and not with others, this argues for a “possible causal link between increased BMI and the risk of developing some cancers”. 

  • Cancers are not caused by one factor. An individual may be at increased risk of cancer due to a combination of genetic, lifestyle and environmental factors. The associations between weight and cancer is not unexpected: oesophageal cancer and colon cancer are both known to be strongly linked with smoking and poor diet, which  may signify a poorer general lifestyle that goes hand-in-hand with obesity. 
  • The limitations with such a review generally lie with weaknesses in the studies included for analysis. Not all studies will have measured factors that may be responsible for the rates of cancers in their included populations, e.g. the use of hormone replacement therapy can effect rates of breast, endometrial and ovarian cancers. When studies are pooled, any potential bias that this introduces can add up. 
  • Importantly, the results don’t capture the effects that weight change over time may have on the risk of cancer (as it only considered the contribution of participants’ BMI at the start of the studies).

This systematic review and meta-analysis provides strong evidence of the strength of the relationship between increasing body mass index and the risk for different cancers. The results support previous reports that excess body increases the risk of some common cancers. The study also highlights an association with risk of some less common cancers , and raises questions that require further research including whether BMI is the best way to measure adiposity and  why there are differences in cancer rates between the sexes and between some ethnic groups.

Analysis by Bazian
Edited by NHS Choices