The fact that overweight and obesity represent a serious health problem in many Western industrialized nations is probably widely known. But is an elevated BMI even associated with the risk of cancer? Various epidemiological studies have been devoted to this question in the past.
(ag) A new study in the Lancet now tested the extent to which weight, or obesity, is actually associated with cancer incidence. The authors examine the association between body mass index (BMI) and the most common carcinomas in a broad systematic evaluation. They also included potential confounding variables in the analysis. According to the researchers, such a study has great relevance in light of rising obesity rates worldwide.
Studies to date unsatisfactory
Many previous studies on this topic have lacked statistical power or have insufficiently or inconsistently considered confounding factors. Little is known to date about the importance of important factors, such as smoking status and age, that could significantly alter the association between BMI and cancer. Also, they mainly used BMI values based on self-reporting, which probably does not do complete justice to the actual value. Nevertheless, a meta-analysis from 2008 already demonstrated a strong association between BMI and carcinomas of the esophagus, thyroid, colon, kidneys, endometrium, and gallbladder. The authors now wanted to verify these findings themselves using epidemiological data.
Method
The association (linear and non-linear) between BMI and the 22 most common cancers was examined. In addition, they tested whether the results changed when factors such as gender, menopausal and smoking status, and age were modified.
A full 5.24 million individuals were included in the cohort study (based on digitized primary care records from the United Kingdom). They had a sufficiently long follow-up period and had no prior cancer diagnosis. BMI data were available from all. On average, follow-up ended after 7.5 years and mean BMI was 25.5 kg/m2.
The weight plays an important role
166 955 individuals from the sample developed one of the 22 relevant carcinomas. Thereby, BMI was associated with 17 cancer types according to the non-linear analysis, but there were large differences according to cancer type (in the direction and strength of association).
With each 5 kg/m2 increase in BMI, the risk of uterine, gallbladder, kidney, cervical, thyroid cancer, and leukemia also increased linearly, controlling for all potential confounding factors. The associated values were:
- Uterine cancer: hazard ratio [HR] 1.62; 99% CI 1.56-1.69; p<0.0001
- Gallbladder cancer: 1.31; 1.12-1.52; p<0.0001
- Renal cancer: 1.25; 1.17-1.33; p<0.0001
- Cervical cancer: 1.10; 1.03-1.17; p=0.00035
- Thyroid cancer: 1.09; 1.00-1.19; p=0.0088
- Leukemia: 1.09; 1.05-1.13; p≤0.0001
BMI was overall highly significantly positively associated with liver, colon, ovarian, and postmenopausal breast cancer, but these nonlinear effects varied by BMI level and individual characteristics.
An inverse association with risk of prostate and premenopausal breast cancer was seen in the overall population and in nonsmokers. The values in the overall population were HR 0.98; 0.95-1.00 (prostate) and HR 0.89; 0.86-0.92 (breast cancer). This showed that the risk of premenopausal breast cancer increased up to 22 kg/m2 but decreased at higher levels. The same was true for prostate cancer. The authors suggest that this circumstance is due to delayed cancer diagnosis in overweight people (which other studies also show).
Such an inverse association was initially also found in lung and oral cavity cancer – but disappeared when only the non-smoking subgroup was considered. Thus, the smoking status is more likely to play the supporting role here.
Assuming a causal association, this means that 41% of uterine and 10% or more of gallbladder, kidney, liver, and colon cancers can be attributed to overweight and obesity.
Causal relationship possible
The authors calculated that a mean BMI increase of 1 kg/m2 in the overall population would result in 3790 additional patients per year developing one of the 10 positively associated cancers. The researchers conclude that BMI is associated with cancer risk. Assuming causality, many cancers are attributable to overweight and obesity. However, the heterogeneity of effects also shows that different mechanisms and subgroups need to be considered for the specific cancer types, he said.
Source: Bhaskaran K, et al: Body-mass index and risk of 22 specific cancers: a population-based cohort study of 5.24 million UK adults. The Lancet 2014; 384(9945): 755-765.
InFo Oncology & Hematology 2014; 2(7): 2
