Pooling of Prospective Studies of Diet and Cancer

点击翻译按钮获取中文摘要

We propose to examine the association between dietary factors and risk of advanced prostate cancer, estrogen receptor negative breast cancer, and gliomas in the Pooling Project of Prospective Studies of Diet and Cancer (Pooling Project), a consortium of 27 prospective cohort studies. Each study met the following inclusion criteria: used a prospective study design, assessed usual diet using a validated dietary assessment method, and met the minimum number of incident cases. The studies were conducted in North America (n=18), Europe (n=6), Asia (n=2), and Australia (n=1) with 834,339 male (n=14 studies) and 1,364,231 female (n=23 studies) participants. During follow-up of these studies, 5,503 advanced prostate cancer (which will be analyzed separately as either stage C and D or as Gleason score >7), 5,484 estrogen receptor negative breast cancer, and 2,516 glioma cases were identified. For advanced prostate cancer, associations with dairy foods, calcium, vitamin D, meat, fat, fruits, vegetables, vitamin E, and body mass index will be examined. For estrogen receptor negative breast cancer, analyses will evaluate fruits, vegetables, and alcohol. For gliomas, associations with vitamins D, C, and E, processed meat, fish, fat, fruits, vegetables, and alcohol will be analyzed. Expansion of the Pooling Project to these cancers with intermediate or low incidence rates will take advantage of the statistical power of this project. In addition, because individual studies will not be required to have published on the specific associations being examined, publication bias will be avoided in these analyses. We will recode the primary data from each study using a standardized approach across studies so that the exposures, covariates, and outcomes will be defined and modeled consistently across studies. This standardized approach will reduce potential sources of between-studies heterogeneity. Analyses will also evaluate whether associationsare modified by other risk factors including sex, ethnicity, age, smoking habits, and other dietary factors to identify potential high-risk groups that may not otherwise be identified in smaller studies. Because the Pooling Project includes populations with different diets, the range of intakes that can be examined will be wider than in individual studies, further increasing the statistical power of the study. The annual meetings, and frequent communication among investigators, will facilitate the analyses and interpretation of the findings. These analyses will take full advantage of extensive data already collected to provide powerful insights into the relation between diet and the risk of advanced prostate cancer, estrogen receptor negative breast cancer, and gliomas, cancer sites that have been examined in few prospective studies. This project provides a highly cost-effective approach to evaluate these pressing issues in a detailed, standardized manner, to yield robust findings that have immediate clinical and public relevance.