This article was downloaded by: [Bibliothèques de l'Université de Montréal] On: 02 December 2014, At: 08:52 Publisher: Routledge Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Nutrition and Cancer Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/hnuc20
Fish Consumption and Lung Cancer Risk: Systematic Review and Meta-Analysis a
a
b
a
Jian Song , Hong Su , Bao-long Wang , Yang-yang Zhou & Liang-Liang Guo
a
a
Department of Epidemiology and Biostatistics, School of Public HealthAnhui Medical University, Hefei, , Anhui Province, China b
Anhui Provincial HospitalAnhui Medical University, Hefei, Anhui Province, China Published online: 07 Apr 2014.
To cite this article: Jian Song, Hong Su, Bao-long Wang, Yang-yang Zhou & Liang-Liang Guo (2014) Fish Consumption and Lung Cancer Risk: Systematic Review and Meta-Analysis, Nutrition and Cancer, 66:4, 539-549, DOI: 10.1080/01635581.2014.894102 To link to this article: http://dx.doi.org/10.1080/01635581.2014.894102
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Nutrition and Cancer, 66(4), 539–549 C 2014, Taylor & Francis Group, LLC Copyright ISSN: 0163-5581 print / 1532-7914 online DOI: 10.1080/01635581.2014.894102
Fish Consumption and Lung Cancer Risk: Systematic Review and Meta-Analysis Jian Song and Hong Su Department of Epidemiology and Biostatistics, School of Public Health, Anhui Medical University, Hefei, Anhui Province, China
Downloaded by [Bibliothèques de l'Université de Montréal] at 08:52 02 December 2014
Bao-long Wang Anhui Provincial Hospital, Anhui Medical University, Hefei, Anhui Province, China
Yang-yang Zhou and Liang-Liang Guo Department of Epidemiology and Biostatistics, School of Public Health, Anhui Medical University, Hefei, Anhui Province, China
There is evidence pointing to a possible role of diet on cancer etiology. Prior studies evaluating the relationship between fish consumption and lung cancer risk reported inconclusive results. The aim of this study was to achieve a comprehensive assessment of the relationship between fish consumption and lung cancer risk through systematic review and meta-analysis. Case control and cohort studies up to September 1, 2012 about fish consumption and lung cancer risk were confirmed by an online search. Separate relative risk (RR) or odds ratio (OR) estimates with 95% confidence interval (CI) of the relationship between lung cancer risk and fish consumption level from the included articles were combined by Stata11.0 software. Publication bias was evaluated by Egger’s linear regression test and funnel plot. Twenty articles (17 case-control and 3 cohort studies) comprising 8799 cases of lung cancer and 17,072 noncases were included in the final analysis. The pooled results from all studies indicated that high fish consumption was significantly associated with a decreased risk of lung cancer (pooled RR: 0.79; 95% CI: 0.69–0.92). There was heterogeneity among the studies (I2 = 73%, P < 0.05). Pooled RR in case control and cohort studies were 0.76 (95% CI: 0.63–0.91) and 0.95 (95% CI: 0.73–1.24), respectively. Omission of any single study had little effect on the combined risk estimates. This article had no publication bias. This study identifies a significant association between fish consumption and lung cancer, confirming a protective role of fish in lung cancer. More well-designed prospective studies are required to further verify the effect of fish consumption on lung cancer.
Submitted 29 November 2012; accepted in final form 24 November 2013. Address correspondence to Hong Su, Department of Epidemiology and Biostatistics, School of Public Health, Anhui Medical University, 69 Meishan Road, Hefei 230032, Anhui Province, China. Tel: +86 551 63869140; Fax: +86 551 63869140. E-mail: [email protected]
BACKGROUND Lung cancer is the most common malignant tumor, accounting for 1.1 million deaths annually all over the world (1). In China, the mortality of lung cancer increased by approximately 1.5 times in recent 20 yr (2). Despite continuous improvement in its treatment, the survival rate of lung cancer patients has not been substantially improved. Consequently, identification of modifiable risk factors for lung cancer is very important as it may lead to prevention opportunities. It is known that lifestyle plays a significant role in the cause of lung cancer. For example, smoking and alcohol are the most well known risk factors for lung cancer (3,4), whereas green tea is a protective factor (5). Emerging evidences have suggested that dietary factors play critical roles in tumor development and have received increasing research attention recently (6). Fish is part of the usual diet of most people worldwide, which is rich in n-3 polyunsaturated fatty acids (n-3PUFA). Epidemiological studies have shown that the incidence of malignant tumors is closely related to the type and quantity of fat intake. N3PUFA, an essential part of cell membranes, is believed to be involved in many biochemical reactions (7). High consumption of food rich in n-3PUFA regularly can significantly reduce the risk of tumor occurrence (8). Prior studies have indicated that fish may be a protective factor for lung cancer mainly by the following ways: damage to the mitochondrial membrane of tumor cells (9), influence to the metabolism of polyunsaturated fatty acids metabolism (10), and the inhibition of tumor angiogenesis (11). Briefly, it may lead to tumor cells apoptosis and inflammatory decrease (12). Fish oil with an increased n-3PUFA could be contributed to the prevention of lung carcinogenesis in mice (13). The relationship between fish consumption and human cancer risk has been extensively reported by meta-analysis, such as
539
540
J. SONG ET AL.
Downloaded by [Bibliothèques de l'Université de Montréal] at 08:52 02 December 2014
colorectal (14), gastric (15) and bladder cancer (16). Previous studies evaluating the relationship between fish consumption and lung cancer risk did not reach consensus. The inconsistent results in published studies may be because of the small sample size and lower statistical power in individual studies. Metaanalysis is a powerful method that generates a best estimate from a number of individual studies (17). To our best knowledge, a comprehensive meta-analysis quantitatively assessing the relationship between fish consumption and lung cancer risk has not been reported. Therefore, the aim of this study is to obtain a more accurate relationship between fish consumption and lung cancer risk through meta-analysis. METHODS Search Strategy and Study Selection Relevant studies written in English or Chinese up to September 1, 2012 were retrieved by an online search in PubMed, China National Knowledge Infrastructure and WanFang databases. We used the following keywords: “lung neoplasms” or “lung cancer” or “lung carcinoma” and “fish” or “seafood”. References in original studies were also manually reviewed. To guarantee the high quality of this study, all the included studies must meet the following criteria: (1) the study was either a case control or cohort study; (2) the exposure was fresh fish consumption in daily diet, salted fish and fish oil supplement consumption were excluded; (3) relative risk (RR) or odds ratio (OR) with 95% confidence interval (CI) for the highest vs. non/lowest fish consumption levels could be obtained from the studies; (4) animal experiments and reviews were excluded; (5) for the duplicate articles, the newest or most informative one was selected. All the included articles were confirmed by 2 authors (Jian Song and Yang-yang Zhou). Data Extraction Two authors (Jian Song and Yang-yang Zhou) extracted the following information from the included studies independently: first author’s name, publication year, geographical region, study design (case control or cohort study), sample size, years of follow up, exposure of fish consumption, risk estimates with 95% CI for the highest vs. non/lowest fish consumption levels, adjusted variables, dietary assessment method. Because lung cancer is a rare disease, OR can be interpreted as RR. For simplicity, we reported all results as RR. Any disagreements were resolved by discussion. Statistical Analysis Study-specific RR and 95% CI for the highest vs. non/lowest fish consumption levels were extracted. If the study provided both overall and subgroup estimates, we included only the overall results. For example, the study that provided RR and 95% CI of squamous cell carcinoma (SCC) and adenocarcinoma (AD) and histology combined, only the combined data was included in our meta-analysis. Heterogeneity of effect size among the studies was estimated by using Q and I 2 statistics. Q statistics
at the P < 0.05 were considered to be significant. Random effects model and fixed-effects model were selected to calculate the summary estimated risks based on the heterogeneity among the studies. Publication bias was assessed by funnel plot and Egger’s linear regression test (18). P > 0.05 by Egger’s linear regression test was considered to be no publication bias. A sensitivity analysis was conducted to evaluate the stability of the result by removing each study separately. Subgroup analysis was performed according to the study design, geographical region, dietary information assessment, and gender. We also performed meta-regression analysis to explore the sources of heterogeneity among the studies. The following variables were included in the meta-regression analysis: study design (case control compared with cohort study), country (Asian compared with non-Asian), published year, and exposure assessment. All statistical analyses were conducted by Stata11.0. All the P values were 2-sided, P < 0.05 was considered as statistically significant. RESULTS Literature Search Six hundred thirty-four potentially eligible studies were identified initially (563 from PubMed, 71 from Chinese database). Most of the articles were excluded by reading the titles and/or abstracts, which were irrelevant to our analysis. The remaining 27 studies were assessed by reading the full text. Among them, 5 papers (19–23) evaluated the relationship between salted fish intake and lung cancer risk. Three articles (24–26) were reviews. Two articles (27,28) were excluded because of insufficient data. The endpoint in 3 articles was the death in lung cancer patients (29–31). In addition, 1 study (32) was added by reading pertinent references in the original articles. Eventually, 20 studies including 17 case control studies (32–48) and 3 cohort studies (49–51) were included. The detailed steps of study selection were shown in Fig. 1. 634 articles identified through searching online
Papers excluded by screening of titles and/or abstracts
31 papers selected for further assessment 13 papers excluded for not meeting the criteria (19–31) 1 paper included by hand searching (32)
20 papers included in this meta-analysis (32–51)
FIG. 1.
Flow chart of selection of included studies in meta-analysis.
541
FISH CONSUMPTION AND LUNG CANCER RISK
Downloaded by [Bibliothèques de l'Université de Montréal] at 08:52 02 December 2014
TABLE 1 The characters of the case control studies included in the meta-analysis
Country
Design
Case/control
Fish consumption levels
Odds ratio (95% confidence interval)
Author
Year
Lim (33)
2011 Singapore HBC
258/712
0.47 >10.00 (0.31–0.73) servings/wk vs. once wk vs. 2 servings/wk vs.
Nutrition and Cancer Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/hnuc20
Fish Consumption and Lung Cancer Risk: Systematic Review and Meta-Analysis a
a
b
a
Jian Song , Hong Su , Bao-long Wang , Yang-yang Zhou & Liang-Liang Guo
a
a
Department of Epidemiology and Biostatistics, School of Public HealthAnhui Medical University, Hefei, , Anhui Province, China b
Anhui Provincial HospitalAnhui Medical University, Hefei, Anhui Province, China Published online: 07 Apr 2014.
To cite this article: Jian Song, Hong Su, Bao-long Wang, Yang-yang Zhou & Liang-Liang Guo (2014) Fish Consumption and Lung Cancer Risk: Systematic Review and Meta-Analysis, Nutrition and Cancer, 66:4, 539-549, DOI: 10.1080/01635581.2014.894102 To link to this article: http://dx.doi.org/10.1080/01635581.2014.894102
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Nutrition and Cancer, 66(4), 539–549 C 2014, Taylor & Francis Group, LLC Copyright ISSN: 0163-5581 print / 1532-7914 online DOI: 10.1080/01635581.2014.894102
Fish Consumption and Lung Cancer Risk: Systematic Review and Meta-Analysis Jian Song and Hong Su Department of Epidemiology and Biostatistics, School of Public Health, Anhui Medical University, Hefei, Anhui Province, China
Downloaded by [Bibliothèques de l'Université de Montréal] at 08:52 02 December 2014
Bao-long Wang Anhui Provincial Hospital, Anhui Medical University, Hefei, Anhui Province, China
Yang-yang Zhou and Liang-Liang Guo Department of Epidemiology and Biostatistics, School of Public Health, Anhui Medical University, Hefei, Anhui Province, China
There is evidence pointing to a possible role of diet on cancer etiology. Prior studies evaluating the relationship between fish consumption and lung cancer risk reported inconclusive results. The aim of this study was to achieve a comprehensive assessment of the relationship between fish consumption and lung cancer risk through systematic review and meta-analysis. Case control and cohort studies up to September 1, 2012 about fish consumption and lung cancer risk were confirmed by an online search. Separate relative risk (RR) or odds ratio (OR) estimates with 95% confidence interval (CI) of the relationship between lung cancer risk and fish consumption level from the included articles were combined by Stata11.0 software. Publication bias was evaluated by Egger’s linear regression test and funnel plot. Twenty articles (17 case-control and 3 cohort studies) comprising 8799 cases of lung cancer and 17,072 noncases were included in the final analysis. The pooled results from all studies indicated that high fish consumption was significantly associated with a decreased risk of lung cancer (pooled RR: 0.79; 95% CI: 0.69–0.92). There was heterogeneity among the studies (I2 = 73%, P < 0.05). Pooled RR in case control and cohort studies were 0.76 (95% CI: 0.63–0.91) and 0.95 (95% CI: 0.73–1.24), respectively. Omission of any single study had little effect on the combined risk estimates. This article had no publication bias. This study identifies a significant association between fish consumption and lung cancer, confirming a protective role of fish in lung cancer. More well-designed prospective studies are required to further verify the effect of fish consumption on lung cancer.
Submitted 29 November 2012; accepted in final form 24 November 2013. Address correspondence to Hong Su, Department of Epidemiology and Biostatistics, School of Public Health, Anhui Medical University, 69 Meishan Road, Hefei 230032, Anhui Province, China. Tel: +86 551 63869140; Fax: +86 551 63869140. E-mail: [email protected]
BACKGROUND Lung cancer is the most common malignant tumor, accounting for 1.1 million deaths annually all over the world (1). In China, the mortality of lung cancer increased by approximately 1.5 times in recent 20 yr (2). Despite continuous improvement in its treatment, the survival rate of lung cancer patients has not been substantially improved. Consequently, identification of modifiable risk factors for lung cancer is very important as it may lead to prevention opportunities. It is known that lifestyle plays a significant role in the cause of lung cancer. For example, smoking and alcohol are the most well known risk factors for lung cancer (3,4), whereas green tea is a protective factor (5). Emerging evidences have suggested that dietary factors play critical roles in tumor development and have received increasing research attention recently (6). Fish is part of the usual diet of most people worldwide, which is rich in n-3 polyunsaturated fatty acids (n-3PUFA). Epidemiological studies have shown that the incidence of malignant tumors is closely related to the type and quantity of fat intake. N3PUFA, an essential part of cell membranes, is believed to be involved in many biochemical reactions (7). High consumption of food rich in n-3PUFA regularly can significantly reduce the risk of tumor occurrence (8). Prior studies have indicated that fish may be a protective factor for lung cancer mainly by the following ways: damage to the mitochondrial membrane of tumor cells (9), influence to the metabolism of polyunsaturated fatty acids metabolism (10), and the inhibition of tumor angiogenesis (11). Briefly, it may lead to tumor cells apoptosis and inflammatory decrease (12). Fish oil with an increased n-3PUFA could be contributed to the prevention of lung carcinogenesis in mice (13). The relationship between fish consumption and human cancer risk has been extensively reported by meta-analysis, such as
539
540
J. SONG ET AL.
Downloaded by [Bibliothèques de l'Université de Montréal] at 08:52 02 December 2014
colorectal (14), gastric (15) and bladder cancer (16). Previous studies evaluating the relationship between fish consumption and lung cancer risk did not reach consensus. The inconsistent results in published studies may be because of the small sample size and lower statistical power in individual studies. Metaanalysis is a powerful method that generates a best estimate from a number of individual studies (17). To our best knowledge, a comprehensive meta-analysis quantitatively assessing the relationship between fish consumption and lung cancer risk has not been reported. Therefore, the aim of this study is to obtain a more accurate relationship between fish consumption and lung cancer risk through meta-analysis. METHODS Search Strategy and Study Selection Relevant studies written in English or Chinese up to September 1, 2012 were retrieved by an online search in PubMed, China National Knowledge Infrastructure and WanFang databases. We used the following keywords: “lung neoplasms” or “lung cancer” or “lung carcinoma” and “fish” or “seafood”. References in original studies were also manually reviewed. To guarantee the high quality of this study, all the included studies must meet the following criteria: (1) the study was either a case control or cohort study; (2) the exposure was fresh fish consumption in daily diet, salted fish and fish oil supplement consumption were excluded; (3) relative risk (RR) or odds ratio (OR) with 95% confidence interval (CI) for the highest vs. non/lowest fish consumption levels could be obtained from the studies; (4) animal experiments and reviews were excluded; (5) for the duplicate articles, the newest or most informative one was selected. All the included articles were confirmed by 2 authors (Jian Song and Yang-yang Zhou). Data Extraction Two authors (Jian Song and Yang-yang Zhou) extracted the following information from the included studies independently: first author’s name, publication year, geographical region, study design (case control or cohort study), sample size, years of follow up, exposure of fish consumption, risk estimates with 95% CI for the highest vs. non/lowest fish consumption levels, adjusted variables, dietary assessment method. Because lung cancer is a rare disease, OR can be interpreted as RR. For simplicity, we reported all results as RR. Any disagreements were resolved by discussion. Statistical Analysis Study-specific RR and 95% CI for the highest vs. non/lowest fish consumption levels were extracted. If the study provided both overall and subgroup estimates, we included only the overall results. For example, the study that provided RR and 95% CI of squamous cell carcinoma (SCC) and adenocarcinoma (AD) and histology combined, only the combined data was included in our meta-analysis. Heterogeneity of effect size among the studies was estimated by using Q and I 2 statistics. Q statistics
at the P < 0.05 were considered to be significant. Random effects model and fixed-effects model were selected to calculate the summary estimated risks based on the heterogeneity among the studies. Publication bias was assessed by funnel plot and Egger’s linear regression test (18). P > 0.05 by Egger’s linear regression test was considered to be no publication bias. A sensitivity analysis was conducted to evaluate the stability of the result by removing each study separately. Subgroup analysis was performed according to the study design, geographical region, dietary information assessment, and gender. We also performed meta-regression analysis to explore the sources of heterogeneity among the studies. The following variables were included in the meta-regression analysis: study design (case control compared with cohort study), country (Asian compared with non-Asian), published year, and exposure assessment. All statistical analyses were conducted by Stata11.0. All the P values were 2-sided, P < 0.05 was considered as statistically significant. RESULTS Literature Search Six hundred thirty-four potentially eligible studies were identified initially (563 from PubMed, 71 from Chinese database). Most of the articles were excluded by reading the titles and/or abstracts, which were irrelevant to our analysis. The remaining 27 studies were assessed by reading the full text. Among them, 5 papers (19–23) evaluated the relationship between salted fish intake and lung cancer risk. Three articles (24–26) were reviews. Two articles (27,28) were excluded because of insufficient data. The endpoint in 3 articles was the death in lung cancer patients (29–31). In addition, 1 study (32) was added by reading pertinent references in the original articles. Eventually, 20 studies including 17 case control studies (32–48) and 3 cohort studies (49–51) were included. The detailed steps of study selection were shown in Fig. 1. 634 articles identified through searching online
Papers excluded by screening of titles and/or abstracts
31 papers selected for further assessment 13 papers excluded for not meeting the criteria (19–31) 1 paper included by hand searching (32)
20 papers included in this meta-analysis (32–51)
FIG. 1.
Flow chart of selection of included studies in meta-analysis.
541
FISH CONSUMPTION AND LUNG CANCER RISK
Downloaded by [Bibliothèques de l'Université de Montréal] at 08:52 02 December 2014
TABLE 1 The characters of the case control studies included in the meta-analysis
Country
Design
Case/control
Fish consumption levels
Odds ratio (95% confidence interval)
Author
Year
Lim (33)
2011 Singapore HBC
258/712
0.47 >10.00 (0.31–0.73) servings/wk vs. once wk vs. 2 servings/wk vs.
