Yumie Takata

PhD

Vanderbilt University

Research Fellow

PLCO (Learn more about this study)

PLCO-11

Jan 24, 2013

Calcium intake and lung cancer: effect modification by sex and smoking status

Lung cancer is the leading cause of cancer deaths in the world and in the United States. Although cigarette smoking is the major cause, up to 25% of lung cancer cases are non-smokers, and non-smoking-related risk factors, including dietary factors, are not well understood.

In the Shanghai Women’s Health Study, we recently observed a strong inverse association between dietary calcium intake and lung cancer risk among never smokers. Our finding is generally consistent with two previous cohort studies conducted in the United States and Germany that reported an inverse association between calcium intake and lung cancer risk. However, a US case-control study found a positive association among men and current smokers, but not among women or former smokers.

Calcium intake level varies by population and its absorption is influenced by food and supplement intakes, vitamin D status (circulating 25-hydroxyvitamin D concentration), and the major determinants [e.g., race/ethnicity (skin color) and physical activity], as well as menopausal status. However, very few cohort studies alone have sufficient statistical power to investigate the role of calcium intake in lung cancer among never smokers while taking covariates into consideration. A pooling project with increased statistical power and a wide range of calcium intake levels will allow an in-depth investigation of the association.

The major objectives of the proposed pooling analysis, which was approved by the secretariats of the NCI Cohort Consortium in 2012, are to comprehensively investigate whether calcium intake is associated with lung cancer risk and whether any such associations are modified by smoking status or gender. We plan to include multiple prospective cohorts, including PLCO in the analysis. Data on dietary and supplement calcium intakes, major food sources of calcium (e.g., dairy products), smoking status and history, lung cancer incidence and mortality, along with other major covariates, will be compiled from participating cohorts. Two approaches will be considered in data analysis: 1) meta-analysis using study-specific risk estimates (hazard ratios and 95% confidence intervals) obtained from multivariate Cox proportional hazards regression and 2) aggregated individual data analysis with adjustment for cohort. Analyses will be stratified by smoking status, gender, race/ethnicity, and factors that may influence calcium absorption (e.g., major determinants of vitamin D status and menopausal status).

1. To investigate whether calcium intake is associated with lung cancer risk or mortality and whether any such associations are modified by smoking status or gender.

2. To evaluate whether the associations between calcium intake and lung cancer among female never smokers are modified by race/ethnicity, major determinants of vitamin D status, menopausal status, and hormone therapy use.

3. To explore whether calcium intake is associated with tumor characteristics of lung cancer (e.g., stage and histology) among female never smokers.

4. To prospectively evaluate the association of tea and coffee consumption with lung cancer risk and survival by gender and smoking status in the international lung cancer consortium. (Xiao Ou Shu)