Kathryn Barry

Ph.D., M.P.H.

University of Maryland School of Medicine

Assistant Professor

PLCO (Learn more about this study)

PLCO-260

Feb 23, 2017

BMI, weight change and the risk of colorectal cancer in the PLCO Cancer Screening Trial

Colorectal cancer remains among the most commonly diagnosed cancers worldwide, currently ranking third in men and second in women (Ferlay et al, 2015). It has long been known that a variety of dietary and other lifestyle factors (e.g., physical activity, alcohol consumption, dietary fiber and intake of red and processed meat) are associated with the risk of colorectal cancer, and a role of obesity has also been documented in many epidemiological studies (Jochem and Leitzmann, 2016). Previous studies, which have often used body mass index (BMI) as a measure of overweight (BMI>=25 kg/m^2) and obesity (BMI>=30 kg/m^2), have tended to observe stronger associations for men than women and for cancers of the colon than the rectum (World Cancer Research Fund/American Institute for Cancer Research, 2011). However, the majority of studies to date have evaluated obesity at only one point in time (Jochem and Leitzmann, 2016). In recent years, there has been growing interest in the role of weight change over the lifecourse in colorectal carcinogenesis, spanning from the development/recurrence of colorectal adenoma to colorectal cancer. Notably, accumulating evidence suggests that weight gain during early adult life may be more important than weight gain later in life in the risk of colorectal cancer (Karahalios et al, 2015; Schlesinger et al, 2015).

In the present study, we propose to conduct logistic regression models to estimate Odds Ratios (OR) and 95% confidence intervals (CI) for the risk of incident and recurrent colorectal adenoma associated with BMI at age 20, BMI at age 50 and change in weight (both increase and decrease) from age 20 to age 50 and from age 50 to baseline (ages 55-74) in the PLCO Cancer Screening Trial. In addition, we propose to use Cox regression models to estimate hazard ratios (HR) and 95% CI for the risk of colorectal cancer associated with BMI at age 20, BMI at age 50 and change in weight during the same time periods. We will also separately evaluate the association between weight loss and risk of colorectal adenoma (both incident and recurrent) and colorectal cancer among the subgroups of overweight and obese participants at age 20 or age 50, respectively. We will adjust for a variety of potential confounders in the models, including age, sex, smoking, alcohol consumption, family history of colorectal cancer, dietary factors and trial arm. In addition, we will separately evaluate associations for men and women, and we will also separately evaluate associations by anatomic subsite (proximal colon, distal colon and rectum) and trial arm. We will compute p-values for interaction (i.e. between change in BMI and sex or trial arm) using likelihood ratio tests comparing nested models with and without the respective interaction terms. We anticipate that findings may provide further insight into the role of obesity and weight change in colorectal carcinogenesis.

The aim of the proposed project is to evaluate the association between change in weight (both increase and decrease) from age 20 to age 50 and from age 50 to age at baseline (55-74) in PLCO and the risk of incident and recurrent colorectal adenoma and incident colorectal cancer and to separately assess associations by gender and anatomic subsite. We also aim to separately evaluate the association between weight loss and the risk of incident and recurrent colorectal adenoma and incident colorectal cancer among overweight and obese individuals.

Evaluating the role of parity in relation to obesity among women in the study.