Polyunsaturated Fatty Acids and Risk of Breast Cancer
Polyunsaturated Fatty Acids and Risk of Breast Cancer
In this meta-analysis dietary intake of marine n-3 polyunsaturated fatty acids (PUFA), but not alpha linolenic acid (ALA), was associated with a lower risk of breast cancer. Fish consumption was not associated with risk. Dose-response analyses indicated a 5% lower risk of breast cancer per 0.1g/day or 0.1% energy/day increment of dietary marine n-3 PUFA, but no significant trend for ALA or fish intake. To the best of our knowledge, this is the first time meta-analysis has systematically and quantitatively evaluated the association between intake of fish and n-3 PUFA and risk of breast cancer.
Ecological studies and prospective cohort studies have suggested an inverse association between fish consumption and risk of breast cancer and mortality. Marine n-3 PUFA (EPA, DHA, and DPA) are abundant in fish fat, and a few large prospective cohort studies and case-control studies have reported an inverse association with risk of breast cancer. These findings all agreed with one meta-analysis based on studies that assessed the biomarkers of intakes of dietary fatty acids, which suggested a potential protective effect of marine n-3 PUFA on breast cancer. Another systematic review suggested that there was no protective association between n-3 PUFA and breast cancer. Further evidence has been published since then, and quite a few of the newly published prospective studies indicated potential protective effects of n-3 PUFA on breast cancer. Our meta-analysis based on this evidence, together with previous publications, supports a protective role of marine n-3 PUFA on the incidence of breast cancer.
Subgroup analysis indicated that the protective effect of marine n-3 PUFA was more evident in Asian countries than in the US or European countries. Fish intake also tended to be associated with a lower risk of breast cancer in Asian populations, rather than in western populations. This could be because typical fish intake is much higher in Asian populations than in western populations. Therefore fish intake in these western populations might be too low to detect an expected protective effect. Furthermore, in North America and some European countries, a large proportion of intake of marine n-3 PUFA probably comes from fish oil supplementation in the form of capsules, thereby contributing to the different effects on risk. In addition, the protective effect of fish intake might be attenuated or even reversed by other constituents in fish, such as organometallics and pesticides. Taken together, these factors could explain our finding of an overall null association between fish intake and risk of breast cancer, which was in line with a previous large pooled analysis.
Tissue n-3 PUFA concentrations, compared with dietary assessment, might provide a more accurate estimation of intake. Subgroup analysis for marine n-3 PUFA and risk, however, indicated that the summary risk estimate for studies with dietary information was similar to that of tissue biomarkers, which further confirmed the robust results of the present meta-analysis. Further concerns regarding n-3 PUFA and risk are menopausal status and hormone receptor status (oestrogen receptor and progesterone receptor). Previous cohort studies indicated that the protective effect of marine n-3 PUFA against breast cancer was more evident in postmenopausal women than in premenopausal women. Our meta-analysis confirmed that n-3 PUFA intake was significantly inversely associated with breast cancer in postmenopausal but not in premenopausal women; this could mean that any benefit of marine n-3 PUFA is usually after long term exposure, which could be observed best at the postmenopausal period because breast cancer is a disease with a long latency between exposure and development. Another explanation could be related to the different effects of body fat on premenopausal and postmenopausal risk of breast cancer. One recent meta-analysis suggested that high BMI tends to be protective against premenopausal breast cancer but is a risk factor for postmenopausal breast cancer, and an interaction between BMI and menopausal status on breast cancer has been proposed. Therefore, marine n-3 PUFA could influence risk through BMI, which was supported by our subgroup analysis. The inverse association between marine n-3 PUFA and risk was greatly attenuated in studies that adjusted for BMI compared with studies without such adjustment. Most of the studies that investigated marine n-3 PUFA intake and postmenopausal breast cancer, however, did adjust for BMI, and an overall significant inverse association between marine n-3 PUFA and risk still existed in these postmenopausal studies. This suggests that the effect of marine n-3 PUFA on risk was partly independent of BMI, and a more precise mechanism for this discrepancy remains to be investigated. In addition, only a few studies examined the influence of oestrogen receptor and progesterone receptor status on the association between fish intake and risk of breast cancer. Stripp and colleagues reported an adverse effect of fish consumption only for oestrogen receptor positive breast cancer in a cohort of postmenopausal women. Two other studies based on large prospective cohorts found no evidence for the influence of hormone receptor status on n-3 PUFA and risk. More prospective studies are warranted to investigate the impact of hormone receptor status.
In contrast with marine n-3 PUFA, the effect of ALA, a plant based n-3 PUFA, on breast tumour growth is less clear, and we found no significant association. The explanation for the inconsistency among studies regarding dietary ALA and risk of breast cancer could be the different dietary sources. Thiebaut and colleagues showed that dietary ALA from fruit and vegetables and vegetable oils was inversely associated with risk but observed a positive association for ALA from nut mixes and processed meat. In addition, the biological effect of ALA on breast cancer per se might be not as strong as marine n-3 PUFA, as suggested by the summarised relative risk of biomarker data with no significant association in our study. Taken together, we found no significant protective association with ALA, and the inconsistent associations observed among previous studies might reflect different dietary/food patterns involving other nutrients related to risk of breast cancer.
The anticarcinogenic effects of marine n-3 PUFA are biologically plausible. Possible mechanisms include inhibition of eicosanoid derived from arachidonic acid, regulation of transcription factor activity, gene expression and activities of molecules involved in the signal transduction of cell growth, differentiation apoptosis, angiogenesis, and metastasis. In addition, marine n-3 PUFA could decrease the production of oestrogen, thus reducing oestrogen stimulated cell growth. Specifically, studies using cell lines and on rodent models have shown the protective effects of marine n-3 PUFA against breast tumour growth.
The present meta-analysis has several strengths. Firstly, the large sample size allowed us to quantitatively assess the association of fish and n-3 PUFA intake and risk of breast cancer, thus making it more powerful than any individual study. Secondly, the prospective nature of the included studies avoided the influence of recall and selection bias. Thirdly, we systematically reviewed and assessed the summarised association between breast cancer with different types of individual n-3 PUFA, including EPA, DHA, DPA, and ALA. These data gave a most comprehensive view of the association between n-3 fatty acids and risk based on the current evidence.
The meta-analysis does, however, also have several limitations. Firstly, different methods of assessment (diet and tissue biomarker) were used in the included studies, and the units were heterogeneous across different studies. Nevertheless, we used relative risks for the highest versus lowest category of n-3 or fish intake, which could, to some extent, reduce the bias caused by different units or exposure assessment methods. Furthermore, dose-response analysis supports our results. Secondly, available data on the individual n-3 PUFA, especially DPA, is rather limited. Therefore, future prospective studies are needed for the detailed analysis of association between individual n-3 PUFA and risk of breast cancer. Thirdly, the observational nature of the included studies makes it subject to the influence of residual confounders. In addition, possible language bias could occur because we excluded articles not in English. Our eligible articles, however, covered a wide range of non-English countries, such as countries across Europe and Asia, and the number of large cohorts in other non-English countries is limited.
Discussion
In this meta-analysis dietary intake of marine n-3 polyunsaturated fatty acids (PUFA), but not alpha linolenic acid (ALA), was associated with a lower risk of breast cancer. Fish consumption was not associated with risk. Dose-response analyses indicated a 5% lower risk of breast cancer per 0.1g/day or 0.1% energy/day increment of dietary marine n-3 PUFA, but no significant trend for ALA or fish intake. To the best of our knowledge, this is the first time meta-analysis has systematically and quantitatively evaluated the association between intake of fish and n-3 PUFA and risk of breast cancer.
Results in Relation to Other Studies
Ecological studies and prospective cohort studies have suggested an inverse association between fish consumption and risk of breast cancer and mortality. Marine n-3 PUFA (EPA, DHA, and DPA) are abundant in fish fat, and a few large prospective cohort studies and case-control studies have reported an inverse association with risk of breast cancer. These findings all agreed with one meta-analysis based on studies that assessed the biomarkers of intakes of dietary fatty acids, which suggested a potential protective effect of marine n-3 PUFA on breast cancer. Another systematic review suggested that there was no protective association between n-3 PUFA and breast cancer. Further evidence has been published since then, and quite a few of the newly published prospective studies indicated potential protective effects of n-3 PUFA on breast cancer. Our meta-analysis based on this evidence, together with previous publications, supports a protective role of marine n-3 PUFA on the incidence of breast cancer.
Subgroup analysis indicated that the protective effect of marine n-3 PUFA was more evident in Asian countries than in the US or European countries. Fish intake also tended to be associated with a lower risk of breast cancer in Asian populations, rather than in western populations. This could be because typical fish intake is much higher in Asian populations than in western populations. Therefore fish intake in these western populations might be too low to detect an expected protective effect. Furthermore, in North America and some European countries, a large proportion of intake of marine n-3 PUFA probably comes from fish oil supplementation in the form of capsules, thereby contributing to the different effects on risk. In addition, the protective effect of fish intake might be attenuated or even reversed by other constituents in fish, such as organometallics and pesticides. Taken together, these factors could explain our finding of an overall null association between fish intake and risk of breast cancer, which was in line with a previous large pooled analysis.
Tissue n-3 PUFA concentrations, compared with dietary assessment, might provide a more accurate estimation of intake. Subgroup analysis for marine n-3 PUFA and risk, however, indicated that the summary risk estimate for studies with dietary information was similar to that of tissue biomarkers, which further confirmed the robust results of the present meta-analysis. Further concerns regarding n-3 PUFA and risk are menopausal status and hormone receptor status (oestrogen receptor and progesterone receptor). Previous cohort studies indicated that the protective effect of marine n-3 PUFA against breast cancer was more evident in postmenopausal women than in premenopausal women. Our meta-analysis confirmed that n-3 PUFA intake was significantly inversely associated with breast cancer in postmenopausal but not in premenopausal women; this could mean that any benefit of marine n-3 PUFA is usually after long term exposure, which could be observed best at the postmenopausal period because breast cancer is a disease with a long latency between exposure and development. Another explanation could be related to the different effects of body fat on premenopausal and postmenopausal risk of breast cancer. One recent meta-analysis suggested that high BMI tends to be protective against premenopausal breast cancer but is a risk factor for postmenopausal breast cancer, and an interaction between BMI and menopausal status on breast cancer has been proposed. Therefore, marine n-3 PUFA could influence risk through BMI, which was supported by our subgroup analysis. The inverse association between marine n-3 PUFA and risk was greatly attenuated in studies that adjusted for BMI compared with studies without such adjustment. Most of the studies that investigated marine n-3 PUFA intake and postmenopausal breast cancer, however, did adjust for BMI, and an overall significant inverse association between marine n-3 PUFA and risk still existed in these postmenopausal studies. This suggests that the effect of marine n-3 PUFA on risk was partly independent of BMI, and a more precise mechanism for this discrepancy remains to be investigated. In addition, only a few studies examined the influence of oestrogen receptor and progesterone receptor status on the association between fish intake and risk of breast cancer. Stripp and colleagues reported an adverse effect of fish consumption only for oestrogen receptor positive breast cancer in a cohort of postmenopausal women. Two other studies based on large prospective cohorts found no evidence for the influence of hormone receptor status on n-3 PUFA and risk. More prospective studies are warranted to investigate the impact of hormone receptor status.
In contrast with marine n-3 PUFA, the effect of ALA, a plant based n-3 PUFA, on breast tumour growth is less clear, and we found no significant association. The explanation for the inconsistency among studies regarding dietary ALA and risk of breast cancer could be the different dietary sources. Thiebaut and colleagues showed that dietary ALA from fruit and vegetables and vegetable oils was inversely associated with risk but observed a positive association for ALA from nut mixes and processed meat. In addition, the biological effect of ALA on breast cancer per se might be not as strong as marine n-3 PUFA, as suggested by the summarised relative risk of biomarker data with no significant association in our study. Taken together, we found no significant protective association with ALA, and the inconsistent associations observed among previous studies might reflect different dietary/food patterns involving other nutrients related to risk of breast cancer.
The anticarcinogenic effects of marine n-3 PUFA are biologically plausible. Possible mechanisms include inhibition of eicosanoid derived from arachidonic acid, regulation of transcription factor activity, gene expression and activities of molecules involved in the signal transduction of cell growth, differentiation apoptosis, angiogenesis, and metastasis. In addition, marine n-3 PUFA could decrease the production of oestrogen, thus reducing oestrogen stimulated cell growth. Specifically, studies using cell lines and on rodent models have shown the protective effects of marine n-3 PUFA against breast tumour growth.
Strengths and Limitations
The present meta-analysis has several strengths. Firstly, the large sample size allowed us to quantitatively assess the association of fish and n-3 PUFA intake and risk of breast cancer, thus making it more powerful than any individual study. Secondly, the prospective nature of the included studies avoided the influence of recall and selection bias. Thirdly, we systematically reviewed and assessed the summarised association between breast cancer with different types of individual n-3 PUFA, including EPA, DHA, DPA, and ALA. These data gave a most comprehensive view of the association between n-3 fatty acids and risk based on the current evidence.
The meta-analysis does, however, also have several limitations. Firstly, different methods of assessment (diet and tissue biomarker) were used in the included studies, and the units were heterogeneous across different studies. Nevertheless, we used relative risks for the highest versus lowest category of n-3 or fish intake, which could, to some extent, reduce the bias caused by different units or exposure assessment methods. Furthermore, dose-response analysis supports our results. Secondly, available data on the individual n-3 PUFA, especially DPA, is rather limited. Therefore, future prospective studies are needed for the detailed analysis of association between individual n-3 PUFA and risk of breast cancer. Thirdly, the observational nature of the included studies makes it subject to the influence of residual confounders. In addition, possible language bias could occur because we excluded articles not in English. Our eligible articles, however, covered a wide range of non-English countries, such as countries across Europe and Asia, and the number of large cohorts in other non-English countries is limited.
