Consumption of Fish and ω-3 Fatty Acids and Cancer Risk: An Umbrella Review of Meta-Analyses of Observational Studies

By: Keum Hwa Lee, Hyo Jin Seong, Gaeun Kim, Gwang Hun Jeong, Jong Yeob Kim, Hyunbong Park, Eunyoung Jung, Andreas Kronbichler, Michael Eisenhut, Brendon Stubbs, Marco Solmi, Ai Koyanagi, Sung Hwi Hong, Elena Dragioti, Leandro Fórnias Machado de Rezende, Louis Jacob, NaNa Keum, Hans J van der Vliet, Eunyoung Cho, Nicola Veronese, Giuseppe Grosso, Shuji Ogino, Mingyang Song, Joaquim Radua, Sun Jae Jung, Trevor Thompson, Sarah E Jackson, Lee Smith, Lin Yang, Hans Oh, Eun Kyoung Choi, Jae Il Shin, Edward L Giovannucci, Gabriele Gamerith
Published: 03 June 2020

ABSTRACT

Multiple studies have suggested that ω-3 fatty acid intake may have a protective effect on cancer risk; however, its true association with cancer risk remains controversial. We performed an umbrella review of meta-analyses to summarize and evaluate the evidence for the association between ω-3 fatty acid intake and cancer outcomes. We searched PubMed, Embase, and the Cochrane Database of Systematic Reviews from inception to December 1, 2018. We included meta-analyses of observational studies that examined associations between intake of fish or ω-3 fatty acid and cancer risk (gastrointestinal, liver, breast, gynecologic, prostate, brain, lung, and skin) and determined the level of evidence of associations. In addition, we appraised the quality of the evidence of significant meta-analyses by using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system. We initially screened 598 articles, and 15 articles, including 57 meta-analyses, were eligible. Among 57 meta-analyses, 15 reported statistically significant results. We found that 12 meta-analyses showed weak evidence of an association between ω-3 fatty acid intake and risk of the following types of cancer: liver cancer (n = 4 of 6), breast cancer (n = 3 of 14), prostate cancer (n = 3 of 11), and brain tumor (n = 2 of 2). In the other 3 meta-analyses, studies of endometrial cancer and skin cancer, there were no assessable data for determining the evidence levels. No meta-analysis showed convincing, highly suggestive, or suggestive evidence of an association. In the sensitivity analysis of meta-analyses by study design, we found weak associations between ω-3 fatty acid intake and breast cancer risk in cohort studies, but no statistically significant association in case-control studies. However, the opposite results were found in case of brain tumor risk. Although ω-3 fatty acids have been studied in several meta-analyses with regard to a wide range of cancer outcomes, only weak associations were identified in some cancer types, with several limitations. Considering the nonsignificant or weak evidence level, clinicians and researchers should cautiously interpret reported associations between ω-3 fatty acid consumption and cancer risks.