The impact of lifetime overweight and obesity on cancer incidence–A population-based study in Icelandic women

Abstract

Introduction The impact of risk factors on cancer development largely depends on time. In addition to the level of exposure, the time spent exposed or the cumulative exposure might provide a better measure of the cancer risk given a risk factor. In a large cohort of women, we assess the relationship between different patterns of overweight and obesity during adulthood and the incidence of three types of cancer that previously have been found to be associated with BMI; postmenopausal breast, colorectal and endometrial cancer. Methods The study population consisted of 90,299 women who attended organized nationwide screening for cancer of the cervix and breast in Iceland from between 1979 and 1989, corresponding to about 70% of the national adult female population. Data on reproductive factors and anthropometry were available at more than 300,000 occasions and information on cancer incidence was obtained through record linkage with the population-based Icelandic Cancer Registry. Repeated measurements of height and weight during follow-up were used in a growth curve model to predict individual-specific trajectories of body mass index (BMI) and to compute cumulative measures of overweight and obesity. The cumulative time spent above a given BMI value (e.g., 25 for the time spent with overweight and 30 for obesity) and cumulative overweight/obesity-years, taking into account the time spent as well as the degree of overweight/obesity, between ages 20–50 (between early adulthood and menopause), were calculated. To investigate the impact of these time-mediated and cumulative measures of overweight and obesity on the occurrence of postmenopausal breast, endometrial and colorectal cancer, Cox regression models were fitted, adjusting for important confounders. Results The mean BMI at 20 years was 21.9 and BMI was found to increase on average by 0.15 kg/m2 by year. Out of all women included, 64,015 reached age 50 during follow-up and among those, 2886 developed postmenopausal breast cancer. In addition, 891 colorectal and 404 endometrial cancers were observed in this subset of women. Taking into account the effects of relevant confounding factors, overweight and obesity before age 50 were associated with an increased risk of developing postmenopausal endometrial cancer (HR = 1.42, 95% CI: 1.04–1.94 and HR = 1.87, 95% CI: 1.13–3.10, respectively), with a 3% increase in risk for every additional year spent with obesity. The risk of colorectal cancer associated with obesity was HR = 1.34 and increased by 1% for every additional year spent with obesity. In contrast, there was no evidence for any statistically significant effect of mean BMI, overweight duration and cumulative exposure to overweight and obesity during adulthood on the risk of developing postmenopausal breast cancer. Conclusions The role of overweight duration in the aetiology of cancer differed between the three types of cancers. The results indicate that weight control is relevant at all ages. Furthermore, exploring time-dependent and cumulative effects of overweight and obesity increases our understanding of time-varying exposures and their association with disease occurrence. This knowledge is vital for the development of effective and targeted cancer prevention strategies in different stages of the life course.