Yumie Takata

Oregon State University

Associate Professor

PLCO (Learn more about this study)

PLCO-973

Apr 25, 2022

BMI, its dietary determinants and lung cancer risk

Lung cancer is the leading cause of cancer deaths in the US and the world and its survival rates are poor. Smoking is the major risk factor for lung cancer. Given that smokers tend to have a lower body mass index (BMI), previous studies reported inverse associations between BMI and lung cancer risk among smokers, which were also reported among non-smokers. These results contradict positive BMI associations reported with other cancer sites and chronic diseases. Aside from genetic factors, BMI is reflective of energy balance and diet plays a major role. Current evidence for lung cancer prevention by the World Cancer Research Fund and American Institute for Cancer Research includes foods and nutrients that are typically associated with lower BMI as protective factors and red and processed meats and alcoholic beverages that are typically associated with higher BMI as risk factors. Within the PLCO study, previous publications reported whole grains as a protective factor and soft and alcoholic drinks as risk factors for lung cancer as well as the inverse association between BMI and lung cancer risk.
Besides dietary factors, the notion of the low BMI among smokers is not always supported by previous studies; not all smokers lose weight or avoid weight gain due to smoking and some smokers do gain weight. Further, it is possible that previous studies have overestimated the BMI and lung cancer association by not explicitly accounting for competing risk of first new diagnosis of obesity-related cancers (i.e., esophageal adenocarcinoma, multiple myeloma, and cancers of the thyroid, gallbladder, stomach, liver, pancreas, kidney, postmenopausal breast, ovaries, uterus, colon and rectum) during the follow-up period, although participants diagnosed with cancer prior to study entry were excluded.
Hence, there is a gap in knowledge regarding what and how much dietary factors have contributed to the reported BMI associations for lung cancer. In this project, we will test the specific aims below. Given the availability of data on incidence of cancer of all sites, comprehensive and detailed dietary intake data, and a wide range of cancer risk factors including smoking and body weight histories, the PLCO study is well-suited for our project. In our statistical analysis, we will use multivariate linear or logistic regression models (Aim 1) and multivariate Cox regression models (Aim 2). BMI will be modeled as continuous or categories such as WHO BMI category. Dietary variables will be modeled as food groups or nutrient intakes from foods as well as overall dietary indices such as the Healthy Eating Index. Finding from our project will help to elucidate factors contributed to the paradoxical BMI association for lung cancer and to clarify and formulate future cancer prevention recommendations on diet and BMI.