SYSTEMATIC REVIEW

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils: analysis of 495,321 participants C. Lovegrove,1 K. Ahmed,1 B. Challacombe,2 M. S. Khan,2 R. Popert,2 P. Dasgupta1

1

SUMMARY

Review criteria

Introduction: Fish-oils have a potential role in inflammation, carcinogenesis inhibition and favourable cancer outcomes. There has been increasing interest in the relationship of diet with cancer incidence and mortality, especially for eicosapantaenoic acid (EPA) and docosahexaenoic acid (DHA). This systematic-analysis of the literature aims to review evidence for the roles of dietary-fish and fish-oil intake in prostate-cancer (PC) risk, aggressiveness and mortality. Methods: A systematicreview, following PRISMA guidelines was conducted. PubMed, MEDLINE and Embase were searched to explore PC-risk, aggressiveness and mortality associated with dietary-fish and fish-oil intake. 37 studies were selected. Results: A total of 495,321 (37-studies) participants were investigated. These revealed various relationships regarding PC-risk (n = 31), aggressiveness (n = 8) and mortality (n = 3). Overall, 10 studies considering PC-risk found significant inverse trends with fish and fish-oil intake. One found a dose–response relationship whereas greater intake of long-chain-polyunsaturated fatty acids increased risk of PC when considering crude odds-ratios [OR: 1.36 (95% CI: 0.99–1.86); p = 0.014]. Three studies addressing aggressiveness identified significant positive relationships with reduced risk of aggressive cancer when considering the greatest intake of total fish [OR 0.56 (95% CI 0.37–0.86)], dark fish and shellfish-meat (p < 0.0001), EPA (p = 0.03) and DHA (p = 0.04). Three studies investigating fish consumption and PC-mortality identified a significantly reduced risk. Multivariate-OR (95% CI) were 0.9 (0.6–1.7), 0.12 (0.05–0.32) and 0.52 (0.30–0.91) at highest fish intakes. Conclusions: Fish and fish-oil do not show consistent roles in reducing PC incidence, aggressiveness and mortality. Results suggest that the specific fish type and the fish-oil ratio must be considered. Findings suggest the need for large intervention randomised placebo-controlled trials.

Introduction The role of fish-oils in preventing and treating disease has been investigated for over 50 years (1). Fish-oil is known to be a common source of omega3 fatty acids and evidence suggests favourable outcomes relative to cancer symptoms such as fatigue and cachexia (2,3). Long-chain omega-3 and omega6 polyunsaturated fatty acids (PUFAs) form important components of phospholipid cell membranes, influencing permeability, structure, fluidity and cell– cell interactions (4). Fatty-acid families compete for degradation by lipoxygenase and cyclooxygenase enzymes, yet, because of differing structures, reaction end-products show heterogeneity. Typically, cell ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105. doi: 10.1111/ijcp.12514

Recent studies have indicated that fish oil may be associated with an increased risk of prostate cancer, though other studies have presented conflicting results. Studies have included fish and fish oils to examine this relationship.

MRC Centre for Transplantation, NIHR Biomedical Research Centre, King’s Health Partners, King’s College London, London, UK 2 Departments of Urology and Nephrology & Renal Transplantation, Guy’s & St Thomas’ Hospital, London, UK

Message for the clinic Great inconsistence in study design limits metaanalysis. When considering fish consumption there is no direct link with prostate cancer incidence but relationships with reduced mortality have been demonstrated. Different fish oils present differing results regarding reductions in prostate cancer risk, aggression and mortality, thus more extensive research is required before consideration in nutritional therapy and cancer prevention.

Correspondence to: Kamran Ahmed, MRC Centre for Transplantation, NIHR Biomedical Research Centre, King’s College London, London SE1 9RT, UK Tel.: +020 7188 5669 Fax: +020 7188 5660 Email: [email protected]. uk Disclosures No authors have a competing interest to declare, nor have they received any direct support from organisations or institutions for the submitted work. There have been no relationships or positions held which should influence the submission of this article.

membrane derived fatty acids are metabolised to eicosanoids; a term encompassing leukotrienes, prostaglandins, thromboxanes and hydroxyeicosatetraenoic acids (5,6). Omega-6 fatty acids release pro-inflammatory mediators whereas omega-3 fatty acids are converted to potent anti-inflammatory mediators and compete with omega-6 molecules for enzymatic degradation, further reducing inflammatory effects (7). Inflammation is proposed as one mechanism for carcinogenesis, promoting cell proliferation, oxidative damage and reduced DNA repair (6). Such changes do not subside without consequence and may be responsible for neoplastic growth. In addition to cancer propagation, inflammatory markers are involved

87

88

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils

in other physiological pathways, contributing to cachexia, anorexia, fever and drug metabolism (8,9). Omega-3 fatty acids derived from fish, eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), have been employed therapeutically in several diseases with variable success. Benefits have been noted in morbidities where inflammation is also implicated, for example cardiovascular disease, insulin resistance and colitis (10–13). Results contribute to public health campaigns and promotion of supplement use by the nutrition industry. Within the UK in 2009, fish-oil was found to have the greatest share in the market (£139.1 million) and be the largest markets internationally in Denmark and Finland (14,15). With increased emphasis on longevity and quality of life, the dietary supplement industry was valued at £738 million after 4% growth in 2012 and is projected to grow to £788 million by 2017, indicating scope for incorporating research into product design (16). Research has examined the relationship of dietary fish and the constituent fish-oils (EPA and DHA) in prostate cancer risk, aggressiveness and mortality. Studies report a spectrum of results. Szymanski et al. found limited evidence for an association between dietary fish intake and prostate cancer (17). Others reached the same conclusion regarding fish-oils and prostate cancer risk (18–21). However, these studies have not accounted for cancer staging nor investigated whether the type of fish consumed or the fishoil considered is important. This systematic review aims to examine current evidence for the role of dietary fish and different fish-oils in prostate cancer risk (incidence by histological diagnosis or PSA, prostate specific antigen), aggressiveness (staging and grading) and mortality (hospital records and cancer registries). In using these outcome measures, the association of fish and fish-oil intake with different phases of disease may be estimated.

Box 1 Search strategy example

MeSH • Neoplasms/or prostatic neoplasms • Eicosapantaenoic acid • Docosahexaenoic acid • Fish-oil Keywords • Incidence • Risk

Eligibility criteria and study selection These excluded studies which did not address human subjects but which examined cell lines or animal models. Studies not relevant to prostate cancer incidence, aggressiveness (stage and grade) or mortality were disregarded. Non-original research, editorials and reviews were also excluded. Duplicate findings were eliminated.

Data handling and analysis Using an Excel sheet, data were extracted from studies. Characteristics examined included study design, number of participants and identified cases of prostate cancer, outcome measures, additional variables, risk of bias and statistical analyses employed. Results of studies with similar outcome measures were compared on their odds, risk or hazard ratios and p-values. Correlations were summarised to present an overview of current evidence. Prostate cancer risk, aggressiveness and mortality were chosen as the outcome measures of interest. No statistical analysis was performed.

Results Study selection

Methods Literature search In November 2013, CL and KA conducted a literature search using PubMed, Ovid MEDLINE(R) (1946 to November Week 2 2013) and Embase (1980–2013 Week 46). Box 1 provides an example of a search strategy used. Publication type was limited to randomised clinical trials, case series and comparative studies available in English and related to human subjects. References were extracted from meta-analyses and systematic reviews. PRISMA guidelines were followed (22).

The literature search yielded 137 titles. Seventeen references not identified by literature searching, were extracted from reference lists. After removing non-relevant article types, 67 results were subjected to further investigation. Thirty were excluded for not addressing outcomes of interest. Finally, 37 articles were included in the study (Figure 1). Appendices 1 and 2 denote studies chosen for inclusion. Results are described, examining three outcome measures of interest (prostate cancer risk, aggressiveness and mortality) relative to: ‘fish intake’ and ‘fishoil intake’.

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils

Potentially relevant titles and abstracts found by electronic literature search and extraction from systematic reviews and meta-analyses (n = 154)

Exclusion of inappropriate formats and duplicated results (n = 87)

Chosen for full text review and quality screening (n = 67)

Exclusion of irrelevant texts (n = 30)

Studies included in systematic review (n = 37)

Figure 1 Study selection

Prostate cancer risk (Table 1) Fish intake Case–control studies. Thirteen case–control studies addressed risk of prostate cancer according to fish intake (23–35). Eight reported no significant relationship (24– 27,30–32,34). However, some studies analysed different types of fish including Mina et al. (31). While the Canadian, population-based case–control study found no associations when considering fresh and canned fish, the 3141 participants demonstrated a significantly decreased risk of prostate cancer with preserved fish intake of one to three times per month [OR 0.78 (95% CI 0.64–0.95)]. A significant inverse trend was also demonstrated in a study of 955 men, finding that in those with a high BMI, consumption of fish and shellfish reduced prostate cancer risk (24) and another study of 1799 prostate cancer cases in Canada, where participants consuming the most smoked fish showed a lower OR [0.7 (95% CI 0.4–1.2) p = 0.002] (27). This was not dose-dependent such as the findings of Amin et al. (23) and Jain et al. (28) (p = 0.017 and p = 0.05, respectively). ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

Three case–control studies demonstrated increased risk of prostate cancer with incremental fish consumption. Jian et al.’s analysis found a dose–response relationship with greatest salted fish consumption showing more than a twofold increased risk of prostate cancer compared with lowest intake [OR 2.12 (95% CI 1.17–3.86) p = 0.014] (29). Ukoli et al. reported a significantly increased risk when comparing prostate cancer risk between participants in first and fourth quartiles of all fish intake: OR 1.16 (95% CI 0.50–2.68), p < 0.04 (33). Fradet et al. considered different types of fish; prostate cancer cases had a significantly lower mean intake of dark fish (p < 0.0001) and shellfish (p = 0.0002) but not white fish (p = 0.31), tuna (p = 0.14) or fried fish (p = 0.15) (35).

Cohort studies. Eleven cohort studies considered the relationship between fish intake and prostate cancer risk (36–46). Nine found no significant relationship, including Allen et al.’s analysis of broiled fish (36,37,39–45). However, Allan et al. demonstrated that those with near daily fish intake (excluding broiled fish) had a

89

408 8117 1012

156 797

Case–control Case–control

Case–control

Case–control

Case–control

Case–control

Case–control

Case–control

Case–control

Case–control

Case–control

Case–control Case–control

Case–control

Case–control

Case–control

Amin et al. (23) Brasky et al. (52)

Chen et al. (24)

Dahm et al. (47)

Deneo-Pellegrini et al. (25) Fernandez et al. (26) Fradet et al. (35)

Hu et al. (27)

Jain et al. (28)

Key et al. (30)

Kristal et al. (51)

Mannisto et al. (49) Mina et al. (31)

Newcomer et al. (50) Norrish et al. (48)

Sung et al. (32)

270

198 3141

1197

656

1253

6838

2023

718

917 3038

404

Case–control

Jian et al. (29)

Participants

Design

Study

Table 1 Studies addressing prostate cancer risk

90

317

67

198 1534

605

328

617

1799

506

127

175

962

237

386 1674

130

Cases

180

480

156

198 1607

592

328

636

5039

478

3220

233

1061

481

531 1364

274

Non-cases/ controls

Incidence of PCa compared with RBC fatty-acid composition Incidence of PCa by specialist diagnosis compared with EPA and DHA of RBCs Incidence of PCa by histological confirmation

Incidence of PCa by clinical/radiological/biochemical/ histological confirmation Incidence of PCa according to history of PSA, digital rectal examination and medical records PCa diagnosis by cancer registry Incidence of PCa by histopathological confirmation

Incidence of PCa by histopathological confirmation

Incidence of PCa by histological confirmation

PCa incidence by histological stage compared with diet composition

Incidence of PCa by histological confirmation

PCa incidence and phospholipid consumption according to plasma fatty-acid content Incidence of PCa by histological confirmation

Incidence of PCa by histological confirmation

Incidence by PSA score PCa incidence and aggressiveness according to plasma EPA and DHA

Incidence of PCa by histopathological report

Outcome measure of interest

No significant association between EPA and DHA consumption and PCa risk The intakes of fatty acids were not associated with PCa risk No significant relationship between fresh and canned fish intake and PCa risk. Significantly reduced risk of PCa with increased proportion of dietary fact from fish sources and preserved fish consumption of 1–3 per month No significant relationship between RBC total omega-3 fatty acids, EPA, or DHA and PCa risk Significantly reduced risk of PCa with increased RBCs’ EPA or DHA content No significant relationship between fish intake and PCa risk

PCa cases showed a significantly increased mean consumption of dark meat fish and shellfish when compared with controls Significant difference between mean EPA or DHA consumption in cases and controls with controls demonstrating higher intakes No significant relationship between total fish consumption and PCa risk Decreased risk of PCa with increase consumption of smoked fish Significantly lower OR for PCa risk observed with highest quartile of intake of fish No significant relationship between fish intake and PCa risk

No significant relationship between fish intake and PCa risk

Significantly lower risk of PCa with increased consumption of salted fish Reduced risk of PCa with increased consumption of fish Significant association between increased plasma DHA and increased risk of total PCa. Non-significant association between increased plasma EPA and incidence of total PCa No significant relationship between fish intake and PCa risk Increased fish intake associated with significantly reduced risk of PCa in men with high BMI Increased long-chain PUFA content was associated with increased risk of PCa No significant relationship between fish intake and PCa risk

Summary of result

90 Systematic review of prostate cancer risk and association with consumption of fish and fish-oils

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

3246 18,115

Case–control

Cohort

Villeneuve et al. (34) Allen et al. (37)

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

20,167

2892 58,279 7999 6272

Cohort

Cohort

Cohort

Cohort

Cohort

Cohort Cohort

Cohort

Cohort

Leitzmann et al. (53)

Le Marchand et al. (40) Mills et al. (41)

Park et al. (42)

Rohrmann et al. (43) Schuurman et al. (44) Severson et al. (45) Terry et al. (46) 466

174

199 642

4404

180

198

2965

2161

2727 2482

196

1623

56

Cases

5826

7825

3693 57,637

78,079

20,118

44,901

18,006

139,793 45,400

17,919

1623

268

Non-cases/ controls

Incidence of PCa by linkage to national cancer register

Incidence of PCa according to histological diagnosis

Incidence of PCa according to cancer registry Incidence of PCa linked to cancer registry

Incidence of PCa according to diagnosis in medical notes Incidence of PCa (self-reported or linked to tumour registries)

Incidence of PCa according to national cancer registers Incidence of PCa according to review of medical records Incidence of PCa according to hospital records and pathology reports Incidence of PCa by medical records and pathology reports Incidence of PCa according to histological diagnosis

Incidence of PCa according to medical records/death certificate information/pathology specimens/active tissue registry report/autopsy diagnoses

Incidence of PCa by histological confirmation

Incidence of PCa according to PSA and digital rectal examination

Outcome measure of interest

PCa, prostate cancer; PSA, prostate specific antigen; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; ALA, alpha-linoleic acid.

82,483

20,316

47,866

142,520 47,822

Cohort Cohort

Allen et al. (36) Augustsson et al. (38) Chavarro et al. (39)

324

Case–control

Ukoli et al. (33)

Participants

Design

Study

Table 1 Continued

found between total fish intake and PCa risk found between total fish intake and PCa risk

found between total fish intake and PCa risk association between fatty-acid intake and PCa

Increased total fish consumption associated with decreased risk of PCa

No relationship found between total fish intake and PCa risk

No relationship No significant risk No relationship No relationship

No relationship found between total fish intake and PCa risk

EPA and combined EPA + DHA showed significantly reduced risk of total PCa No relationship found between total fish intake and PCa risk

No relationship found between intake of broiled fish and PCa risk Fish and total fish consumption associated with significantly increased risk of PCa No relationship found between total fish intake and PCa risk Increased total fish consumption associated with decreased risk of PCa No relationship found between total fish intake and PCa risk

Cases consumed more fish than controls but this was not significant On comparing 1st and 4th quartiles of fish intake, there was a significantly increased risk of PCa with increased fish consumption No significant relationship between fish intake and PCa risk

Summary of result

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils 91

92

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils

significantly increased risk of prostate cancer over those whose intake totalled less than twice per week [RR = 1.18 (95% CI 0.83–1.67) p = 0.03]. Two studies contradicted these results. Augustsson et al.’s follow-up of 47,822 men found total fish consumption over three times weekly corresponded to a lower risk of prostate cancer (38). Similarly, a prospective study in Swedish men found that those who ate no fish had a two to threefold increased incidence of prostate cancer over those with ‘moderatelarge’ intake; p = 0.05 (46).

Fish-oil intake Case–control studies. Seven case–control studies addressed prostate cancer risk relative to EPA and DHA consumption (35,47–52). One study, of 962 cases and 1061 controls, found that increasing plasma long-chain omega-3 fattyacids concentration was related to greater prostate cancer risk. This was significant when comparing highest and lowest quintiles [OR: 1.36 (95% CI: 0.99–1.86); p = 0.041] (47). Brasky et al.’s 2013 research supports this (52). Other studies demonstrated decreased prostate cancer risk associated with fish-oil intake. Fradet et al. reported a significantly lower mean intake of EPA and DHA in cases of prostate cancer compared with controls (p = 0.0007 and p = 0.0005, respectively) (35). Norrish et al. found an inverse relationship with erythrocyte EPA content (p = 0.03), although estimates of daily EPA and DHA consumption in their food frequency questionnaire did not show the same association (48). Four articles, found no association between individual fatty-acid intake (as reflected by erythrocyte membrane composition and serum cholesterol esters) and prostate cancer risk (49–52). This included Brasky et al. Although they identified that increasing plasma concentrations of EPA was not related to total risk (p = 0.08), it was noted that higher plasma DHA correlated with increased incidence (p = 0.009). In addition, they also found that prostate cancer cases had a significantly higher plasma DHA (p = 0.006) and EPA (p = 0.03) than controls.

Cohort studies. Two studies examined DHA and EPA intake and prostate cancer risk (42,53). One demonstrated a significantly reduced risk with increased EPA consumption (p = 0.02) and EPA + DHA (p = 0.04) across quintiles, with Q5 reporting risk ratios of 0.88 (95% CI 0.76–1.01) for EPA and 0.89 (95% CI 0.77–1.04) for EPA + DHA (53). Consumption of DHA was not associated with prostate cancer risk.

Park et al.’s multi-ethnic cohort study found no significant relationship between consumption of DHA or EPA with overall risk (42).

Prostate cancer aggressiveness (Table 2) Fish intake Case–control

studies. One case–control study addressed fish intake and disease severity. Fradet et al. revealed a high association between dark meat fish and of shellfish with reduced risk of aggressive prostate cancer (Gleason score ≥ 7, TNM stage ≥ T2c, or PSA at diagnosis > 10 ng/ml) (p < 0.001) (35). Cohort studies. Two cohort studies considered dietary fish and prostate cancer severity (38,42). Augustsson et al. found that fish consumption more than three times per week was related to reduced risk of metastatic disease compared with lower intake; multivariate OR 0.56 (95% CI 0.37– 0.8) (38). Conversely, Park et al.’s 8-year follow-up of almost twice as many men (82,483 participants) found no association (42).

Fish-oil intake Case–control studies. Six studies considered EPA and DHA with prostate cancer aggressiveness (35,47,48,52,54,55). Norrish et al. found a significant inverse relationship between erythrocyte EPA or DHA content and advanced prostate cancer risk (p = 0.03 and p = 0.04, respectively) (48). Though findings by Shannon et al. did not demonstrate a relationship with total risk, alpha-linoleic acid: EPA erythrocyte membrane ratio was significantly related to reduced risk of low-grade cancer [OR for highest vs. lowest tertile = 0.53, 95% CI (0.25–1.11)] (55). Another study identified a significant difference between higher quartiles of EPA and DHA consumption and lower risk of aggressive prostate cancer, p < 0.0001 in both instances (35). Conversely, Brasky’s study involving 1658 cases and 1803 controls found that those with serum DHA above the first quartile had increased risk of highgrade prostate cancer (p = 0.04) (54). Results for EPA and total EPA + DHA did not show an association with risk of low-grade or high-grade cancer. Similarly, Dahm et al. failed to find any relationship between increased plasma long-chain omega-3 fatty acids and advanced stage (p = 0.650) or high-grade cancer (p = 0.520), despite significant findings related to increased risk of cancer overall (p = 0.041) (47). A further study by Brasky et al. demonstrated a ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

Case–control

Case–control

Case–control

Case–control

Case–control

Case–control

Cohort

Cohort

Brasky et al. (54)

Brasky et al. (52)

Dahm et al. (47)

Fradet et al. (35)

Norrish et al. (48)

Shannon et al. (55)

Augustsson et al. (38)

Park et al. (42)

82,483

47,822

310

797

1012

2023

3038

3461

Participants

4404

2482

127

317

506

962

1674

1658

Cases

78,079

45,400

183

480

478

1061

1364

1803

Non-cases/ controls

Incidence of PCa, stage and grade according to review of medical records Incidence of PCa, grade and stage (self-reported or linked to tumour registries)

Incidence of PCa and grade of disease by specialist diagnosis against EPA and DHA intake assessed by RBC biomarkers Incidence of PCa by biopsy diagnosis

PCa incidence and severity compared with phospholipid consumption according to plasma fatty-acid composition PCa incidence and severity by histological stage and diet composition

PCa incidence and aggressiveness according to plasma EPA and DHA

Serum phospholipid fatty-acid levels in cases of different grades of PCa

Outcome measure of interest

PCa, Prostate Cancer; PSA, prostate specific antigen; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; ALA, alpha-linoleic acid.

Design

Study

Table 2 Studies addressing and prostate cancer aggressiveness

ALA: EPA associated with a significant reduction in risk of lowgrade disease Significantly reduced risk of metastatic PCa with increased fish intake No significant association between fish consumption and grade or stage of PCa

Significant association between increased consumption of dark fish meat and reduced incidence of aggressive prostate cancer High intake of EPA and DHA significantly associated with reduced risk of aggressive PCa Increased EPA or DHA associated with significantly decreased risk of high-grade disease

No associations of low-grade PCa with any fatty-acid measure DHA was significantly associated with increased risk of highgrade PCa Significant association between increased plasma DHA and increased risk of low-grade PCa but non-significant relationship regarding high-grade cases Significant association between increased plasma EPA and incidence of low-grade PCa No association between EPA or DHA and stage or grade of PCa

Summary of result

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils 93

PCa, prostate Cancer; PSA, prostate specific antigen; EPA, eicosapentaenoic acid; DHA, docosahexaenoic acid; ALA, alpha-linoleic acid.

5826 466 Cohort Terry et al. (46)

6272

Cohort

5589

21

5568

PCa mortality according to hospital records and pathological reports PCa mortality according to death certificate classification Incidence of PCa and mortality by linkage to national cancer register Pham et al. (56)

This systematic review highlights the effects of fish and fish-oil on prostate cancer risk, aggressiveness and mortality using up-to-date evidence. When considering overall risk of prostate cancer, dietary fish consumption showed no regular significant trend, likely a result of studies examining ‘total’ fish intake and not fish type. Several studies identified significant relationships between cancer risk according to class of fish, but using questionnaires imposes limits on the number of components assessed (27,29,31,33). These relationships may be attributed to variable lipid, triglyceride and phospholipid compositions of different fish species or prepa-

Participants

Discussion

Design

No studies addressed the role of fish-oil intake and prostate cancer mortality.

Study

Fish-oil intake

Cases

Non-cases/ controls

Outcome measure of interest

sumption and prostate cancer-associated mortality (39,46,56). All found significantly decreased risk of death from prostate cancer with increased fish intake. Pham et al. [crude HR 0.10 (95% CI 0.03–0.25)] and Terry et al. (multivariate RR p-value = 0.01) found consistent inverse associations between mortality in cases consuming high quantities of fish vs. those consuming lower quantities of fish. Chavarro et al. demonstrated the same relationship for total fish, dark meat fish and seafood omega-3 fatty acids (p = 0.04).

18,006

Fish intake Cohort studies. Three studies considered fish con-

2161

Prostate cancer mortality (Table 3)

20,167

significantly lower RR for organ-confined prostate cancer with increasing EPA intake (p = 0.03), though not for DHA or total EPA + DHA. Higher total EPA + DHA was related to lower RR in advanced prostate disease on considering age-adjusted RR (p = 0.04), but this was not significant after multivariate adjustment (p = 0.08).

Cohort

Summary of result

Cohort studies. Leitzmann et al. (53) reported

Chavarro et al. (39)

significant relationship between both plasma EPA and DHA and increased risk of low-grade cancer (though this did not apply to high-grade cases); (p = 0.048 and p = 0.008, respectively) (52). On comparing mean plasma phospholipid content between cases and controls, low-grade cancer cases demonstrated a significantly higher plasma EPA and DHA over controls (p = 0.02 for both) and highgrade cases showed significantly raised DHA levels (p = 0.009).

Increased fish intake associated with significantly lower risk of PCa mortality Increased fish intake associated with significantly lower risk of PCa mortality Increased fish intake associated with significantly lower risk of PCa mortality

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils

Table 3 Studies addressing prostate cancer mortality

94

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils

ration methods altering carcinogenic properties (57,58). In analysing omega-3 fatty acids, EPA and DHA demonstrated significant inverse relationships with prostate cancer risk in two studies (48,53). These support the mechanism for omega-3 fatty acids altering eicosanoid production, inhibiting in vitro prostate cancer cell growth (59). However, Dahm et al. only considered long-chain-PUFAs, implying that further studies are necessary to appreciate the implications of individual fatty-acid compounds in vivo (47). This hypothesis is supported by the findings of Brasky et al. who additionally demonstrated conflicting results whereby mean DHA was significantly higher in all prostate cancers though EPA was not (52). Brasky’s study from 2013 attracted media attention with the conclusion that omega-3 fatty acids increased risk of prostate cancer risk. However, on examination of results it is apparent that risk varies according to the phospholipid examined, suggesting that individual fatty-acids result in different physiological effects. Results indicate that total intake is not the only consideration in prostate cancer risk and more research is necessary to identify the mechanisms of fish and specific fish-oils and their independent implications in carcinogenesis. The systematic review demonstrates several strengths. Firstly, it encompasses a large cohort of participants (495,123) from 37 studies extracted systematically from a variety of databases, providing a reliable overview of evidence. Previous studies have permitted limited conclusions, many being conducted in western populations with relatively low fish intakes. However, this analysis includes evidence from worldwide populations including Nigeria, Sweden, Japan, Taiwan and differing ethnicities within countries. The proportion of fish in the diet ranges greatly between these groups. In the case of prostate cancer mortality and fish intake, similar results were observed in USA, Japan and Sweden. Heterogeneous populations enable a large overview, comparing results from close and distant geographical locations, allowing the conclusion that fish consumption is an important factor in mortality. Inconsistent findings regarding prostate cancer aggressiveness are likely explained by heterogeneous study designs. Few articles identified individual types of fish, preventing firm conclusions from being drawn. Variable results were found between studies of similar populations; for example Shannon et al. and Brasky et al. examined men over the age of 50 in the USA (54,55). However, they found a reduced risk of low-grade cancer and no correlation, respectively. Other articles considering individual fatty acids were also contradictory in their results. While Leitzª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

mann et al. concluded that increased EPA reduced prostate cancer risk and risk of localised disease, total EPA + DHA reduced overall risk and DHA was not implicated, Brasky et al. found several results indicating that DHA increased risk of high-grade prostate cancer (52–54). This suggests variation in carcinogenic mechanisms depending on the fish-oil examined and disease aggressiveness; hence contrasting results. However, case-identification methods can also account for identified differences. Few studies addressed mortality. Fish intake seems to be associated with reduced prostate cancer mortality (39,46,56). Though possible that confounding variables were at work (e.g. treatment received or cancer stage at diagnosis, health behaviours), use of large, heterogeneous populations from USA, Japan and Sweden add weight to this conclusion. This applies to total fish and individual fish types (dark fish meat, seafood). With further analysis, findings could be important in public health promotion campaigns. There was great diversity when measuring fish intake; some considered weight of fish consumed per week, others based analysis on servings per week or month and some used ‘high’ and ‘low’ classifications. Furthermore, the use of several detection methods by different research papers could alter validity and reduce the worth of conclusions. Histological diagnosis was used most frequently, though many used PSA testing. PSA’s high detection rate often identifies high PSA of benign prostate hyperplasia as ‘abnormally high’, diagnosing prostate cancers which often do not become clinically significant without testing (60). In addition, men with health-seeking behaviours are more likely to present for screening, thus studies are in danger of over-estimating the link between prostate cancer risk, aggressiveness or mortality and healthy lifestyle choices. This is an inherent bias of case–control studies. However, Chavarro et al. accounted for this; after eliminating cases identified by PSA testing, the inverse correlation of fish and fatty-acid intake with prostate cancer mortality remained (39). A further limitation is the potential confounding by other components of fish and fish-oils. Marine meat is a source of vitamins A, D, retinol and selenium, all of which have been linked to reduced prostate cancer incidence (61). The possibility that these affect carcinogenesis via biochemical mechanisms or incorporation of EPA and DHA into cell membranes cannot be eliminated. Furthermore, fish intake may be linked to consumption of another food (via healthy behaviour), not identified by studies but which could be involved in prostate risk, aggressiveness and mortality.

95

96

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils

Finally, many articles only assessed baseline consumption; consequently, no account was taken for changes in diet during the follow-up period confounding the results obtained. Studies were conducted over many years and public health campaigns and lifestyle changes can provoke alterations in dietary patterns. This was the rationale for several studies using erythrocyte membrane fatty-acid composition as a marker of dietary intake (55). This intermediate marker provides a slightly longer term overview of nutritional pattern than a single questionnaire at baseline. Nevertheless, it is possible that changes in fish consumption were not accounted for, detracting from the validity of any conclusions drawn. This systematic review faced limitations, mainly from differing designs and qualities of included studies. Meta-analysis and direct comparison was prevented by lack of homogenous parameters, thus statistical significance cannot be drawn. In addition, the heterogeneous studies did not all take account of the same confounding variables at work, for example age, family history and lifestyle factors such as physical exercise and diet which have been demonstrated to affect the risk of several cancers. While other systematic reviews and meta-analyses have considered only the effects of fish intake or fatty acids on prostate cancer risk, this systematic review links these related factors. By combining studies, conclusions regarding the link between dietary fish, fatty-acid composition and prostate cancer may be drawn.

Conclusions Evidence on the role of fish and fish-oils in prostate cancer risk is inconsistent. This systematic review demonstrates that total fish intake cannot be directly associated with prostate cancer risk but is implicated in mortality. No relationship with prostate cancer risk can be confirmed without further investigation into the role of fish types, their fatty-acid content,

References 1 Kromhout D, Bosschieter EB, de Lezenne Coulander C. The inverse relation between fish consumption and 20-year mortality from coronary heart disease. N Engl J Med 1985; 312(19): 1205–9. 2 Vaughan VC, Hassing MR, Lewandowski PA. Marine polyunsaturated fatty acids and cancer therapy. Br J Cancer 2013; 108(3): 486–92. 3 Hulbert AJ, Kelly MA, Abbott SK. Polyunsaturated fats, membrane lipids and animal longevity. J Comp Physiol 2013; 184(2): 149–66. 4 Das UN. Biological significance of essential fatty acids. J Assoc Physicians India 2006; 54: 309–19. 5 Wallace JM. Nutritional and botanical modulation of the inflammatory cascade–eicosanoids, cyclooxygenases, and lipoxygenases – as an adjunct in can-

6

7

8

9

preservation and cooking methods. Findings suggest the need for large intervention randomised placebocontrolled trials to assess the benefits in a systematic manner. EPA and DHA demonstrate relationships with reduced prostate cancer incidence and the conflicting findings regarding disease aggressiveness or the role of total long-chain PUFAs show those compounds’ ratios are implicated in mechanisms protective against prostate carcinogenesis; this is important and can be applied to the growing nutrition industry.

Author contributions Catherine Lovegrove was involved in the design of the study and acquisition of data, including its analysis and interpretation. She also drafted the article and has revised it in preparation for submission. Kamran Ahmed conceived the idea of the study and was involved in data analysis and article revision. Ben Challacombe was also involved in data interpretation and article revision. M Shamim Khan’s role was in data interpretation and manuscript drafting. Rick Popert was involved in data analysis and interpretation, assisting with the revisions to the manuscript. Prokar Dasgupta also conceived the study and had a role in analysing and interpreting the data collected.

Acknowledgements Financial support was received from the Department of Health via the National Institute for Health Research (NIHR) Biomedical Research Centre award to Guy’s & St Thomas’ NHS Foundation Trust in partnership with King’s College London and King’s College Hospital NHS Foundation Trust. Support was also received from the MRC Centre for Transplantation. Contributions were received from Mr S. Froghi and Dr H. Abboudi.

cer therapy. Integr Cancer Ther 2002; 1(1): 7–37; discussion Wall R, Ross RP, Fitzgerald GF, Stanton C. Fatty acids from fish: the anti-inflammatory potential of long-chain omega-3 fatty acids. Nutr Rev 2010; 68 (5): 280–9. Isobe Y, Arita M, Iwamoto R et al. Stereochemical assignment and anti-inflammatory properties of the omega-3 lipid mediator resolvin E3. J Biochem 2013; 153(4): 355–60. Slaviero KA, Clarke SJ, Rivory LP. Inflammatory response: an unrecognised source of variability in the pharmacokinetics and pharmacodynamics of cancer chemotherapy. Lancet Oncol 2003; 4(4): 224–32. Calder PC. n-3 polyunsaturated fatty acids, inflammation, and inflammatory diseases. Am J Clin Nutr 2006; 83(6 Suppl.): 1505S–19S.

10 Zulyniak MA, Perreault M, Gerling C, Spriet LL, Mutch DM. Fish oil supplementation alters circulating eicosanoid concentrations in young healthy men. Metabolism 2013; 62(8): 1107–13. 11 Fedor D, Kelley DS. Prevention of insulin resistance by n-3 polyunsaturated fatty acids. Curr Opin Clin Nutr Metab Care 2009; 12(2): 138–46. 12 Hu FB, Cho E, Rexrode KM, Albert CM, Manson JE. Fish and long-chain omega-3 fatty acid intake and risk of coronary heart disease and total mortality in diabetic women. Circulation 2003; 107(14): 1852–7. 13 Arita M, Yoshida M, Hong S et al. Resolvin E1, an endogenous lipid mediator derived from omega-3 eicosapentaenoic acid, protects against 2,4,6-trinitrobenzene sulfonic acid-induced colitis. Proc Natl Acad Sci U S A 2005; 102(21): 7671–6.

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils

14 Brookes G. Economic Impact Assessment of the European Union (EU’s) Nutrition & Health Claims Regulation on the EU food supplement sector and market. In: EHCA, 2010. 15 NHS. Supplements: who needs them? In: NHS, 2nd edn., 2011. http://www.nhs.uk/news/2011/05May/ Documents/BtH_supplements.pdf. 16 Vitamins and dietary supplements in the United Kingdom: Euromonitor International, 2013. http:// www.euromonitor.com/vitamins-and-dietary-supplements-in-the-united-kingdom/report (accessed 11.02.2014). 17 Szymanski KM, Wheeler DC, Mucci LA. Fish consumption and prostate cancer risk: a review and meta-analysis. Am J Clin Nutr 2010; 92(5): 1223– 33. 18 Astorg P. Dietary N-6 and N-3 polyunsaturated fatty acids and prostate cancer risk: a review of epidemiological and experimental evidence. Cancer Causes Control 2004; 15(4): 367–86. 19 Brouwer IA. Omega-3 PUFA: good or bad for prostate cancer? Prostaglandins Leukot Essent Fatty Acids 2008; 79(3–5): 97–9. 20 Gerber M. Omega-3 fatty acids and cancers: a systematic update review of epidemiological studies. Br J Nutr 2012; 107(Suppl. 2): S228–39. 21 Sala-Vila A, Calder PC. Update on the relationship of fish intake with prostate, breast, and colorectal cancers. Crit Rev Food Sci Nutr 2011; 51(9): 855– 71. 22 Liberati A, Altman DG, Tetzlaff J et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ 2009; 339: b2700. 23 Amin M, Jeyaganth S, Fahmy N et al. Dietary habits and prostate cancer detection: a case-control study. Can Urol Assoc J 2008; 2(5): 510–5. 24 Chen YC, Chiang CI, Lin RS, Pu YS, Lai MK, Sung FC. Diet, vegetarian food and prostate carcinoma among men in Taiwan. Br J Cancer 2005; 93(9): 1057–61. 25 Deneo-Pellegrini H, De Stefani E, Ronco A, Mendilaharsu M. Foods, nutrients and prostate cancer: a case-control study in Uruguay. Br J Cancer 1999; 80(3–4): 591–7. 26 Fernandez E, Chatenoud L, La Vecchia C, Negri E, Franceschi S. Fish consumption and cancer risk. Am J Clin Nutr 1999; 70(1): 85–90. 27 Hu J, La Vecchia C, DesMeules M, Negri E, Mery L; Canadian Cancer Registries Epidemiology Research G. Meat and fish consumption and cancer in Canada. Nutr Cancer 2008; 60(3): 313–24. 28 Jain MG, Hislop GT, Howe GR, Ghadirian P. Plant foods, antioxidants, and prostate cancer risk: findings from case-control studies in Canada. Nutr Cancer 1999; 34(2): 173–84. 29 Jian L, Zhang DH, Lee AH, Binns CW. Do preserved foods increase prostate cancer risk? Br J Cancer 2004; 90(9): 1792–5. 30 Key TJ, Silcocks PB, Davey GK, Appleby PN, Bishop DT. A case-control study of diet and prostate cancer. Br J Cancer 1997; 76(5): 678–87. 31 Mina K, Fritschi L, Johnson KC; Canadian Cancer Registries Epidemiology Research G. An inverse association between preserved fish and prostate

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

32

33

34

35

36

37

38

39

40

41

42

43

44

45

46

cancer: results from a population-based case-control study in Canada. Nutr Cancer 2008; 60(2): 222–6. Sung JF, Lin RS, Pu YS, Chen YC, Chang HC, Lai MK. Risk factors for prostate carcinoma in Taiwan: a case-control study in a Chinese population. Cancer 1999; 86(3): 484–91. Ukoli FA, Taher K, Egbagbe E et al. Association of self-reported consumption of cooked meat, fish, seafood and eggs with prostate cancer risk among Nigerians. Infect Agents Cancer 2009; 4(Suppl. 1): S6. Villeneuve PJ, Johnson KC, Kreiger N, Mao Y. Risk factors for prostate cancer: results from the Canadian National Enhanced Cancer Surveillance System. The Canadian Cancer Registries Epidemiology Research Group. Cancer Causes Control 1999; 10 (5): 355–67. Fradet V, Cheng I, Casey G, Witte JS. Dietary omega-3 fatty acids, cyclooxygenase-2 genetic variation, and aggressive prostate cancer risk. Clin Cancer Res 2009; 15(7): 2559–66. Allen NE, Key TJ, Appleby PN et al. Animal foods, protein, calcium and prostate cancer risk: the European Prospective Investigation into Cancer and Nutrition. Br J Cancer 2008; 98(9): 1574–81. Allen NE, Sauvaget C, Roddam AW et al. A prospective study of diet and prostate cancer in Japanese men. Cancer Causes Control 2004; 15(9): 911– 20. Augustsson K, Michaud DS, Rimm EB et al. A prospective study of intake of fish and marine fatty acids and prostate cancer. Cancer Epidemiol Biomarkers Prev 2003; 12(1): 64–7. Chavarro JE, Stampfer MJ, Hall MN, Sesso HD, Ma J. A 22-y prospective study of fish intake in relation to prostate cancer incidence and mortality. Am J Clin Nutr 2008; 88(5): 1297–303. Le Marchand L, Kolonel LN, Wilkens LR, Myers BC, Hirohata T. Animal fat consumption and prostate cancer: a prospective study in Hawaii. Epidemiology 1994; 5(3): 276–82. Mills PK, Beeson WL, Phillips RL, Fraser GE. Cohort study of diet, lifestyle, and prostate cancer in Adventist men. Cancer 1989; 64(3): 598–604. Park SY, Murphy SP, Wilkens LR, Henderson BE, Kolonel LN. Fat and meat intake and prostate cancer risk: the multiethnic cohort study. Int J Cancer 2007; 121(6): 1339–45. Rohrmann S, Platz EA, Kavanaugh CJ, Thuita L, Hoffman SC, Helzlsouer KJ. Meat and dairy consumption and subsequent risk of prostate cancer in a US cohort study. Cancer Causes Control 2007; 18 (1): 41–50. Schuurman AG, van den Brandt PA, Dorant E, Goldbohm RA. Animal products, calcium and protein and prostate cancer risk in The Netherlands Cohort Study. Br J Cancer 1999; 80(7): 1107–13. Severson RK, Nomura AM, Grove JS, Stemmermann GN. A prospective study of demographics, diet, and prostate cancer among men of Japanese ancestry in Hawaii. Cancer Res 1989; 49(7): 1857– 60. Terry P, Lichtenstein P, Feychting M, Ahlbom A, Wolk A. Fatty fish consumption and risk of prostate cancer. Lancet 2001; 357(9270): 1764–6.

97

47 Dahm CC, Gorst-Rasmussen A, Crowe FL et al. Fatty acid patterns and risk of prostate cancer in a case-control study nested within the European Prospective Investigation into Cancer and Nutrition. Am J Clin Nutr 2012; 96(6): 1354–61. 48 Norrish AE, Skeaff CM, Arribas GL, Sharpe SJ, Jackson RT. Prostate cancer risk and consumption of fish oils: a dietary biomarker-based case-control study. Br J Cancer 1999; 81(7): 1238–42. 49 Mannisto S, Pietinen P, Virtanen MJ et al. Fatty acids and risk of prostate cancer in a nested case-control study in male smokers. Cancer Epidemiol Biomarkers Prev 2003; 12(12): 1422–8. 50 Newcomer LM, King IB, Wicklund KG, Stanford JL. The association of fatty acids with prostate cancer risk. Prostate 2001; 47(4): 262–8. 51 Kristal AR, Cohen JH, Qu P, Stanford JL. Associations of energy, fat, calcium, and vitamin D with prostate cancer risk. Cancer Epidemiol Biomarkers Prev 2002; 11(8): 719–25. 52 Brasky TM, Darke AK, Song X et al. Plasma phospholipid fatty acids and prostate cancer risk in the SELECT trial. J Natl Cancer Inst 2013; 105(15): 1132–41. 53 Leitzmann MF, Stampfer MJ, Michaud DS et al. Dietary intake of n-3 and n-6 fatty acids and the risk of prostate cancer. Am J Clin Nutr 2004; 80(1): 204–16. 54 Brasky TM, Till C, White E et al. Serum phospholipid fatty acids and prostate cancer risk: results from the prostate cancer prevention trial. Am J Epidemiol 2011; 173(12): 1429–39. 55 Shannon J, O’Malley J, Mori M, Garzotto M, Palma AJ, King IB. Erythrocyte fatty acids and prostate cancer risk: a comparison of methods. Prostaglandins Leukot Essent Fatty Acids 2010; 83(3): 161–9. 56 Pham TM, Fujino Y, Kubo T et al. Fish intake and the risk of fatal prostate cancer: findings from a cohort study in Japan. Public Health Nutr 2009; 12 (5): 609–13. 57 Passi S, Cataudella S, Di Marco P, De Simone F, Rastrelli L. Fatty acid composition and antioxidant levels in muscle tissue of different Mediterranean marine species of fish and shellfish. J Agric Food Chem 2002; 50(25): 7314–22. 58 Sugimura T, Wakabayashi K, Nakagama H, Nagao M. Heterocyclic amines: mutagens/carcinogens produced during cooking of meat and fish. Cancer Sci 2004; 95(4): 290–9. 59 Rose DP. Effects of dietary fatty acids on breast and prostate cancers: evidence from in vitro experiments and animal studies. Am J Clin Nutr 1997; 66 (6 Suppl): 1513S–22S. 60 Smith M, Hogan S, Gray H, Edwards C, Kelley A. Swamp Blues [sound recording]. London: Sonet, 1978. 61 Chan JM, Gann PH, Giovannucci EL. Role of diet in prostate cancer development and progression. J Clin Oncol 2005; 23(32): 8152–60.

Paper received March 2014, accepted July 2014

Case–control

Case–control

Deneo-Pellegrini et al. (25)

Hu et al. (27)

Case–control

Chen et al. (24)

Case–control

Case–control

Amin et al. (23)

Fernandez et al. (26)

Study Design

Study

6838

3347

408

718

917

Participants

Appendix 1. Case-Control Studies Included

Appendices

1799

127

175

237

386

Cases

5039

3220

233

481

531

Non-Cases

Servings of total fish, fresh fish, smoked fish

Servings of fish per week

Intake of fish

Intake of fish and shellfish

Servings of fish per week

Fish/fish-oil measure

Risk of prostate cancer

Risk of prostate cancer

Risk of prostate cancer

Risk of prostate cancer

Risk of prostate cancer

Outcome of interest

Age Ethnicity Education Family history Smoking Alcohol Vasectomy Cystitis/prostatitis Age Education Marital status Parenthood status Religion Occupation Income BMI Physical activity Smoking Alcohol Coffee Tea Chemical exposure Hospital Age Residence Urban/rural status Education Income Family history Age Sex Area of residence Education Alcohol consumption Tobacco smoking BMI Age BMI Nationality Prostate cancer stage Smoking status

Variables

Unconditional logistic regression Interaction analyses

Multiple logistic regression analysis Chi-square test

Chi-squared test

Multiple logistic regression analysis

Multiple logistic regression analysis

v2 test

Statistical analysis

98 Systematic review of prostate cancer risk and association with consumption of fish and fish-oils

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

Study Design

Case–control

Case–control

Case–control

Case–control

Study

Jain et al. (28)

Jian et al. (29)

Key et al. (30)

Mina et al. (31)

Appendix 1. Continued

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

3141

656

404

1253

Participants

1534

328

130

617

Cases

1607

328

274

636

Non-Cases

Intake of fresh and canned fish, preserved fish

Frequency of fatty fish intake

Intake of salted fish per day

Intake of fish per day

Fish/fish-oil measure

Risk of prostate cancer

Risk of prostate cancer

Risk of prostate cancer

Risk of prostate cancer

Outcome of interest

Energy intake Dietary intake Fish supplements Age Weight Energy intake Family history Smoking status Rectal examination in last 5 years Vasectomy Age BMI Physical activity Caloric intake Dietary components Residence Education Income Family history Alcohol consumption Smoking status Tea consumption Height BMI Age of school leaving Social class Smoking Family history of prostate cancer Fish-oil Multivitamin use Mean daily nutrient intake Age Smoking BMI Race Fat intake Energy intake Servings of fresh/

Variables

Multiple logistic regression analysis

Multiple logistic regression analysis

t-Test Chi-square test Multiple logistic regression analysis

Multiple logistic regression analysis

Statistical analysis

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils 99

Study Design

Case–control

Case–control

Case–control

Study

Sung et al. (32)

Ukoli et al. (33)

Villeneuve et al. (34)

Appendix 1. Continued

3246

324

270

Participants

1623

56

90

Cases

1623

268

180

Non-Cases

Servings of fish per week

Intake of fish, seafood

Servings of fish per week

Fish/fish-oil measure

Risk of prostate cancer

Risk of prostate cancer

Risk of prostate cancer

Outcome of interest

canned fish Servings of preserved fish Age Height Weight Ethnicity Education Spouse’s education Marital status Religion Occupation Family income BMI Diet Smoking Domestic work Dietary components Exercise Residency Age Recruitment site Education Income History of BPH Obesity status Anthropometry including BMI Dietary components Family history Race Fat consumption Tomato intake Energy intake Smoking history Coffee intake Tea intake Alcohol consumption Anthropometry including BMI Physical activity

Variables

Multivariate logistical regression analysis

Chi-square test Unconditional logistic regression analysis

Multivariate conditional logistic regression analysis Hierarchic modelling

Statistical analysis

100 Systematic review of prostate cancer risk and association with consumption of fish and fish-oils

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

Study Design

Case–control

Case–control

Case–control

Case–control

Study

Brasky et al. (54)

Brasky et al. (52)

Dahm et al. (55)

Fradet et al. (35)

Appendix 1. Continued

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

944

2023

3038

3461

Participants

466

962

1674

1658

Cases

478

1061

1364

1803

Non-Cases

Intake of fatty acids

Plasma fatty-acid composition

Plasma fatty-acid composition

Serum phospholipid fatty-acid composition

Fish/fish-oil measure

Risk of prostate cancer Risk of localised prostate cancer Risk of advanced prostate cancer Risk of low-grade prostate cancer Risk of high-grade prostate cancer Risk of prostate cancer

Risk of overall prostate cancer Risk of low-grade prostate cancer Risk of high-grade prostate cancer

Risk of low-grade prostate cancer Risk of high-grade prostate cancer

Outcome of interest

Age Ethnicity

Marital status Education Income Age Education Race Physical activity Smoking status Alcohol consumption BMI Diabetes history Family history Clinical stage Biopsy history Age Race Education BMI Smoking status Alcohol consumption PSA Finasteride use Aspirin use Diabetes history Family history Clinical stage SELECT Trial intervention assignment Age BMI Smoking status Alcohol intake Physical activity Education Stage of cancer Grade of cancer

Variables

Unconditional logistic regression models

TT reduction Multiple logistic regression analysis

Geometric means Cox proportional hazards model Test for linear trend Multiple logistic regression analysis

Chi-square test F-test Wilcoxon rank-sum test Multiple logistic regression analysis

Statistical analysis

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils 101

Case–control

Case–control

Case–ontrol

Mannisto et al. (49)

Newcomer et al. (50)

Norrish et al. (48)

310

797

223

396

1197

Participants

127

317

67

198

605

Cases

183

480

156

198

592

Non-Cases

Erythrocyte cell membrane fattyacid composition

Erythrocyte cell membrane fattyacid composition

Erythrocyte cell membrane fattyacid composition

Serum fatty-acid composition

Intake of fatty acids

Fish/fish-oil measure

Risk of prostate cancer Risk of advanced prostate cancer Risk of prostate cancer Risk of Gleason 6 prostate cancer Risk of Gleason 7 prostate cancer

Risk of prostate cancer

Risk of prostate cancer

Risk of localised prostate cancer Risk of regional/ distant prostate cancer

Outcome of interest

BMI, body mass index; PSA, prostate specific antigen; BPH, benign prostate hyperplasia; NSAID, non-steroidal anti-inflammatory.

Case–control

Case–control

Kristal et al. (51)

Shannon et al. (55)

Study Design

Study

Appendix 1. Continued

Family history Smoking status BMI PSA test history Serum PSA Clinical stage Histological grade Diet composition Age Race Family history Education BMI PSA testing history Stage of disease Diet composition Age BMI Alcohol consumption Smoking history Location Education Diet composition Age Ethnicity BPH history Family history BMI Age Socioeconomic status NSAID use Diet composition Age BMI Ethnicity Smoking status Education Marital status NSAID use Family history

Variables

Chi-square test Fisher’s test Spearman’s rank Multiple logistic regression analysis

Multivariate logistic regression

Paired t-test Spearman’s rank Unconditional logistic regression models

Paired t-test Chi-square test Conditional multivariate logistic regression analysis

Multiple logistic regression analysis

Statistical analysis

102 Systematic review of prostate cancer risk and association with consumption of fish and fish-oils

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

Study design

Cohort study

Cohort study

Cohort study

Cohort study

Cohort study

Study

Allen et al. (37)

Allen et al. (36)

Augustsson et al. (38)

Chavarro et al. (39)

Leitzmann et al. (53)

Appendix 2. Cohort Studies Included

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

47,866

20,167

47,822

14,2520

18,115

Participants

2965

2161

2482

2727

196

Cases

44,901

18,006

45,400

13,9793

17,919

Non-cases

Fatty-acid intake

Servings of total fish, tuna, dark fish meat, other fish, shrimps/ lobster/scallop, seafood n-3 fatty acids

Servings of fish per month

Intake of fish and fish products, fatty fish, white fish

Servings of fish/ broiled fish/total fish per week

Fish/fish-oil measure

Risk of prostate cancer Risk of local prostate cancer

Risk of prostate cancer

Risk of prostate cancer Risk of advanced prostate cancer Risk of metastatic prostate cancer

Risk of prostate cancer

Risk of prostate cancer

Outcome of interest

Age Calendar period Residence Radiation exposure Education Marital status Smoking BMI Dietary components Age Height Weight BMI Smoking status Physical activity Education Marital status Country of residence Dietary components Age Dietary components Multivitamin use Supplement use PSA testing Vasectomy history Rectal examinations Smoking status BMI Physical activity Age Height BMI Dietary components Alcohol consumption Smoking status Multivitamin use Physical activity Age BMI Height Family history

Variables

Multivariate logistic regression analysis Cox regression analysis

Cox regression analysis Multivariate logistic regression analysis

Multivariate logistic regression analysis

Multivariate logistic regression analysis Cox regression analysis Chi-square test

Multivariate logistic regression analysis

Statistical analysis

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils 103

Cohort study

Cohort study

Cohort study

Mills et al. (41)

Park et al. (42)

Pham et al. (56)

Cohort study

Cohort study

Le Marchand et al. (40)

Rohrmann et al. (43)

Study design

Study

Appendix 2. Continued

3982

5589

82,483

20,316

Participants

199

21 deaths

4404

180

198

Cases

3693

5568

78,079

20,118

Non-cases

Weekly fish intake

Total fish intake

Intake of EPA, DHA, fish and shellfish

Frequency of fish intake

Total fish intake

Fish/fish-oil measure

Risk of total prostate cancer Risk of high stage prostate cancer Risk of low stage prostate cancer

Risk of prostate cancer

Risk of all prostate cancer Risk of non-highgrade or localised prostate cancer

Risk of prostate cancer

Risk of prostate cancer

Risk of advanced prostate cancer

Outcome of interest

Diabetes history PSA screening history Smoking history Physical activity Dietary components Ethnicity Weight Height Smoking status Alcohol consumption Dietary components Education Marital status Age at marriage Smoking history Alcohol consumption Enlarged prostate Dietary components Family religion Age Ethnicity BMI Education Multivitamin use Smoking status Family history Dietary components Age Smoking status Alcohol consumption Employment Dietary components History of diabetes Cohabitation Age BMI Energy intake Fat intake Dietary components Alcohol consumption Supplement use

Variables

Wald test Cox regression analysis Multivariate logistic regression analysis

Chi-square test ANOVA Cox regression analysis

Cox regression analysis

Cox regression analysis

Multivariate logistic regression analysis

Statistical analysis

104 Systematic review of prostate cancer risk and association with consumption of fish and fish-oils

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

ª 2014 John Wiley & Sons Ltd Int J Clin Pract, January 2015, 69, 1, 87–105

Cohort study

Cohort study

Cohort study

Schuurman et al. (44)

Severson et al. (45)

Terry et al. (46)

6272

79,99

58,279

Participants

BMI, body mass index; PSA, prostate specific antigen.

Study design

Study

Appendix 2. Continued

466

174

642

Cases

5826

7825

57,637

Non-cases

Frequency of fish intake

Weekly fish intake

Total fish intake

Fish/fish-oil measure

Risk of prostate cancer Risk of prostate cancer mortality

Risk of prostate cancer

Risk of prostate cancer

Outcome of interest

Race Smoking status PSA/digital rectal examination Vasectomy Family history Dietary components Age Family history Education Smoking status Residence Education Occupation Marital status Generation Physical activity Number of children Dietary components Age BMI Sedentary lifestyle Low socioeconomic status Smoking history Alcohol consumption Dietary components

Variables

Cox regression analysis

Multivariate logistic regression analysis

Multivariate logistic regression analysis

Statistical analysis

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils 105

Systematic review of prostate cancer risk and association with consumption of fish and fish-oils: analysis of 495,321 participants.

Fish-oils have a potential role in inflammation, carcinogenesis inhibition and favourable cancer outcomes. There has been increasing interest in the r...
197KB Sizes 0 Downloads 6 Views