Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 May 1.
Published in final edited form as: Eur J Cancer Prev. 2019 May;28(3):188–195. doi: 10.1097/CEJ.0000000000000455

Effects of Fish Oil Supplementation on Eicosanoid Production in Subjects at Higher Risk for Colorectal Cancer

Maya N White 1, Martha J Shrubsole 1,2,3,4, Qiuyin Cai 1,2,4, Timothy Su 1,2,4, Jennings Hardee 5, John-Anthony Coppola 6, Sunny S Cai 7, Stephanie M Martin 1,8, Sandra Motley 1,2, Larry L Swift 9, Ginger L Milne 10, Wei Zheng 1,2,3,4, Qi Dai 1,2,4, Harvey J Murff 1,3,8
PMCID: PMC6440821  NIHMSID: NIHMS973571  PMID: 30640206

Abstract

Objective

Fish oil supplementation may represent a potential chemopreventive agent for reducing colorectal cancer risk. Fish oil’s mechanism of action is unknown but presumed to be related to eicosanoid modification. The purpose of this study was to evaluate the effects of fish oil supplementation on levels of urinary and rectal eicosanoids.

Methods

We conducted a randomized, double-blind, controlled trial of 2.5 grams of fish oil per day compared to olive oil supplementation over a six month period. Study participants had a history of colorectal adenomas. Randomization was stratified based on the gene variant rs174535 in the fatty acid desaturase 1 enzyme (FADS1), which affects tissue levels of arachidonic acid.

Results

A total of 141 subjects were randomized. Urinary prostaglandin E2 metabolite (PGE-M) was measured at baseline, 3 months, and 6 months and rectal prostaglandin E2 (PGE2) at baseline and 6 months. Repeated measures linear regression was used to determine the effect of the intervention on each outcome measure. Overall, fish oil supplementation was found to reduce urinary PGE-M production compared to olive oil (P = 0.03). Fish oil did not reduce rectal PGE2 overall, however it did significantly reduce PGE2 in the subgroup of participants not using aspirin or NSAIDs (P=0.04). FADS1 genotype did not seem to modify fish oils effects on PGE2 production.

Conclusions

We conclude that fish oil supplementation has a modest but beneficial effect on eicosanoids associated with colorectal carcinogenesis, particularly in those not taking aspirin or NSAIDs.

Keywords: Colonic neoplasms, fish oil, omega-3 fatty acids, eicosanoids, prostaglandin E2, controlled clinical trials, randomized

Introduction

An important link between cancer and inflammation involves cyclooxygenase-2, (COX-2) the inducible form of cyclooxygenase, which is overexpressed in multiple malignancies (Wang, Honn et al. 2007). In the COX pathway, the n-6 polyunsaturated fatty acid (PUFA) arachidonic acid (AA) is released from cellular membranes and then enzymatically converted into various bioactive lipid molecules including prostaglandin E2 (PGE2). PGE2 promotes carcinogenesis through stimulating cellular proliferation, inhibiting apoptosis, promoting angiogenesis, and/or enhancing invasiveness (Wang and Dubois 2006). Up-regulation of COX-2 occurs in 50% of colon adenomas and 85% of colon cancers and is considered a key and early oncogenic event in colorectal carcinogenesis (Gupta and Dubois 2001).

Eicosapentaenoic acid (EPA), a n-3 long-chain (LC)PUFA found in fish oil, is converted to eicosanoids through the same enzymatic pathways as AA but produces series-3 prostanoids that have less inflammatory actions due to lower receptor affinities when compared with AA-derived series-2 prostanoids (Calder 2002, Wada, DeLong et al. 2007). In addition, EPA reduces tissue levels of AA through directly displacing AA from the sn2 position (second carbon) in the glycerol backbone in cell phospholipid membranes and inhibits COX enzymes. (Garg, Sebokova et al. 1988, Ferretti, Judd et al. 1991, Blair, Prakash et al. 1993, Smith 2005). Several animal and human studies have demonstrated that increased dietary intake of n-3 LCPUFAs are associated with reduced tissue levels of AA and reduced PGE2 production (Boudreau, Chanmugam et al. 1991, Ferretti, Judd et al. 1991, Blair, Prakash et al. 1993, Wander, Hall et al. 1997, Broughton and Wade 2002, Trebble, Wootton et al. 2003).

Tissues stores of AA come from both exogenous sources (i.e. diet) as well as endogenous sources (the conversion of linoleic acid to AA). Multiple genome-wide association studies (GWAS) have reported associations between single nucleotide polymorphisms (SNPs) in enzymes responsible for AA biosynthesis pathway and tissue levels of AA. Variants in a single gene, FADS1 (which encodes Δ5-desaturase enzyme) has been found to explain up to 28% of the additive variance in tissue levels of AA. Specifically, homozygotes for the T allele in the rs174537 FADS1 SNP have lower fatty acid desaturase activity and subsequently lower tissues levels of AA (Schaeffer, Gohlke et al. 2006, Rzehak, Heinrich et al. 2009). A recent GWAS study of loci associated with CRC risk found that SNPs in FADS1 are associated with colorectal cancer (CRC) risk (Zhang, Jia et al. 2014).

We conducted a randomized, double blind, controlled trial of marine-derived n-3 LCPUFA (fish oil) supplementation compared to olive oil supplementation on markers of colorectal proliferation and apoptosis in subjects with known prior history of colorectal adenomas. Because of the hypothesis that genetically-determined tissue levels of AA might be associated with an increased risk for CRC, participants were stratified based on rs174537 FADS1 genotype. In this report, we describe the results of the intervention on the secondary outcomes of levels of urinary and rectal eicosanoids.

Material and Methods

Study Participants

Participants were recruited from individuals who were identified as colorectal adenoma cases within the Tennessee Colorectal Polyp Study (TCPS) or were identified through electronic medical record reviews to have undergone a colonoscopy at Vanderbilt University Medical Center and diagnosed with an adenoma. Eligible participants were between the ages of 40 and 80 years, had a past medical history of one or more adenomas, and had a known genotype for the rs174535 SNP of FADS1. Participants were excluded if they had a previous resected colorectal cancer; congestive heart failure or coronary artery disease; inflammatory bowel disease; any cancer (except non-melanoma skin cancer); advanced kidney disease; cirrhosis; were pregnant or breast feeding; using fish oil supplements or anti-coagulants; or allergic to fish products.

The Fatty Acid Desaturase Activity, Fish Oil, and Colorectal Cancer Prevention study was approved by the Vanderbilt University Medical Center Institutional Review Board and registered at ClinicalTrials.gov (NCT01661764).

Study Intervention

Participants were randomized to either 3 capsules of the fish oil-sourced supplement Lovaza® each containing 465 mg of EPA and 375 mg DHA; (total daily dose of 1395 mg EPA plus 1125 mg DHA) or 3 capsules of an olive oil supplement with each capsule contained 1,000 mg of olive oil. Olive oil was chosen as the control arm as prior studies have found no effect on colonic eicosanoid production(Hillier, Jewell et al. 1991). Both capsules were similar in size, shape and appearance. The rationale for this dose is that it was similar to us clinical trials that have successfully lowered indexes of rectal mucosal proliferation and is well tolerated without any reported adverse effects.(Bartoli, Palozza et al. 1993, Anti, Armelao et al. 1994) Medications were dispensed by the Vanderbilt University Investigational Drug Service (IDS).

Randomization, Allocation, Adherence and Blinding

Randomization was performed according to a permuted block randomization scheme stratified on FADS1 rs174537 genotypes (GG, GT, TT). Randomization proceeded within these three strata with a block size of balancing interval, varying randomly according to the outcome of a computer generated random number. The randomization code was kept by the IDS. Participants and study personnel were unaware of the randomization list or treatment assignment. To assess the study blinding, subjects were asked at the end of the study which treatment they thought they had received. Adherence was assessed by weekly telephone calls, pill counts, and red blood cell n-3 LCPUFA measurements.

Study Procedures

Study participation lasted 24 weeks with three in-person visits: at baseline, 12 weeks, and 24 weeks. At each visit, participants arrived after an overnight fast, and 20 ml of blood and 30 ml of spot urine was collected. Eight rectal epithelial biopsies were collected at baseline and at 24 weeks. Participants did not undergo a bowel preparation and samples were obtained using an anoscope and a 2.4 mm 160 cm Captura® biopsy forceps. Biopsies were taken circumferentially approximately 8 cm from the anal verge using biopsy forceps. Two samples were placed in formalin and six were snap frozen in liquid nitrogen.

Laboratory Measurements

Red Blood Cell Phospholipid Membrane Determination

Total lipids were extracted from 200 μl of double-washed packed red blood cells (RBC) using the method described by Folch et al (Folch, Lees et al. 1957). Phospholipids were isolated using thin layer chromatography on Silica Gel 60 A plates and fatty acids methylated using BF3 /methanol (Morrison and Smith 1964). The methylated fatty acids were analyzed by gas chromatography using an HP 7890A gas chromatograph equipped with flame ionization detectors and a capillary column (SP2380, 0.25 mm × 30 m, 0.20 μm film, Supelco, Bellefonte, PA). Helium was used as a carrier gas. Fatty acid methyl esters were identified by comparing the retention times to those of known standards. Inclusion of the internal standard, dipentadecanoyl phosphatidylcholine (C15:0), permitted quantitation of the amount of phospholipid in the sample. Fatty acid values are presented as percentage of total RBC membrane phospholipid fatty acid content. The lowest level of detection for individuals’ fatty acids is less than 0.5% of the total profile.

Rectal Epithelial Cell Eicosanoid Measurement

Tissue was added to 5mL of ice-cold methanol containing indomethacin to inhibit COX activation and 1.0 ng of the internal standard [2H4]-PGE2 (Cayman Chemicals, Ann Arbor, MI). The sample was then extracted and analyzed using GC/NICI-MS as described previously (Liu, Laidlaw et al. 2012, Boutaud, Sosa et al. 2016). The inter-day variability for each assay is < 10%. The precision for each assay is ±5% while the accuracy for each assay is 95%.

Urinary Prostaglandin E2 Metabolite Determination

To quantify endogenous PGE2 production, the urinary PGE2 metabolite PGE-M (11 alpha-hydroxy-9,15-dioxo-2,3,4,5-tetranorprostane-1,20-dioic acid) level was measured using liquid chromatography/tandem mass spectrometric method previously described (Mohebati, Milne et al. 2013). In our laboratory the coefficient of variation for samples analyzed in multiple batches was 7.2%.

Statistical Analysis

The primary outcome of the study was changes in rectal cell markers of proliferation and apoptosis with the secondary outcomes being changes in eicosanoid production. The primary outcomes are completing analyses and pathologic review and this paper reports our secondary outcomes. Sample size estimates were derived using previously described clinical trial information using similar primary outcomes and doses of fish oil supplements. Courtney et al found a change in the mean level of Ki67 labeling of −5.86 (SD ± 8.8) for individuals allocated to fish oil supplements (n=14) and 3.62 (SD ± 8.0) for the placebo group (n=14) (Cheng, Ogawa et al. 2003, Courtney, Matthews et al. 2007). With a sample size of 22 per group for each genotype would provide at least 90% of power to detect 1 SD (8.4) difference in ΔKi67 between the two groups within each genotype with type I error rate = 5%. Assuming approximately 10% of drop-out rate, the total of 150 patients (25 patients per group per genotype × 2 groups × 3 genotypes = 150) would provide power to detect a clinically significant difference between the intervention and placebo groups.

Baseline characteristics stratified by treatment allocation were compared using Wilcoxon rank order sum test for continuous variables and chi-square for categorical variables. We compared RBC membrane percentages of PUFAs at baseline stratified by genotype using Wilcoxon rank order sum test. Urinary PGE-M and rectal eicosanoids were right skewed and log transformed for analyses. We present medians and interquartile ranges for ease of interpretation. We used repeated measures linear regression to determine the effect of the intervention on urinary PGE-M levels. Models were adjusted for genotype and sample batch. For rectal eicosanoids, we used general linear regression and analysis of covariance to adjust for baseline rectal eicosanoid levels, genotype and batch number. We conducted secondary analysis stratifying the analysis by aspirin/NSAID use (regular use, no regular use), baseline RBC membrane n-3 LCPUFA content (below median, above median), baseline RBC membrane AA content (below median, above median), and FADS genotype (G/G, G/T, T/T). We calculated Spearman partial correlation coefficients between urinary PGE-M and rectal PGE2 adjusting for age, gender, aspirin/NSAID use, genotype and study allocation. All analyses were conducted using SAS v9.4.

Results

Three hundred and seventy-nine potential participants were assessed for eligibility and 141 were randomized to either fish oil (n = 70) or olive oil (n=71) (Figure 1). Sixty-two participants (89%) allocated to fish oil completed the study and 64 participants (90%) allocated to olive oil completed the study. Reasons for discontinuation are presented in Figure 1.

Figure 1.

Figure 1

Open in a new tab

Flow diagram of trial recruitment and retention

We were unable to obtain pill/bottle returns for 17 participants (13%). Of participants who did return unused pills or empty bottle, 92% (100/109) reported greater than 80% compliance. In the fish oil arm 33% of participants (20/61) believed they were taking fish oil, 28% (17/61) believed they were taking olive oil, and 39% (24/61) were unsure of their treatment allocation. For the olive oil arm 11% (7/64) reported they were taking fish oil, 33% (21/64) reported they were taking olive oil and 56% (36/64) were unsure of their treatment allocation.

There was no difference in age, gender, or baseline RBC membrane fatty acids between the two groups (Table 1). As expected, RBC membrane AA content was lower in individual who were homozygous for the T allele of the FADS gene (Table 2). In addition, RBC membrane content of linoleic acid, a precursor of endogenously synthesized AA, was greater in individuals who had the T allele.

Table 1.

Baseline participant characteristics

Baseline Characteristics Fish Oil (n = 70) Placebo (n = 71) P-value
Age, years, median (IQR) 57 (54, 64) 59 (55, 66) 0.37
Sex, female, n (%) 34 (48.6) 34 (47.9) 0.94
Regular aspirin or NSAID use, n (%) 47 (67.1) 37 (52.1) 0.07
Genotype, n (%)
 G/G 26 (37.1) 25 (35.2) 0.97
 G/T 25 (35.7) 26 (36.6)
 T/T 19 (27.1) 20 (28.2)
Red blood cell phospholipid fatty acids, as % of total
 Linolenic Acid, median (IQR) 11.2 (9.99, 12.6) 11.2 (9.95, 12.6) 0.89
 Arachidonic acid, median (IQR) 16.7 (14.7, 18.1) 17.2 (15.2, 18.5) 0.19
 Eicosapentaenoic acid, median (IQR) 0 (0, 0.57) 0.35 (0, 0.64) 0.29
 Docosahexaenoic acid, median (IQR) 3.48 (2.89, 4.51) 3.55 (2.82, 4.47) 0.93
Open in a new tab

P values for baseline differences determined using Wilcoxon Rank Order Sum for continuous variables and λ2 for proportions

Table 2.

Baseline Red Blood Cell Phospholipid Membrane PUFAs by Genotype

RBC Membrane % G/G (n = 51) G/T (n = 51) T/T (n = 38) P-value1
 Linolenic Acid 10.6 (9.39, 12.09) 11.4 (10.4, 12.6) 11.7 (10.7, 13.5) 0.004
 Arachidonic acid 17.7 (16.0, 18.9) 17.2 (14.8, 18.2) 15.4 (12.6, 16.9) 0.0004
 Eicosapentaenoic acid 0 (0, 0.58) 0.0 (0, 0.57) 0.47 (0, 0.68) 0.25
 Docosahexaenoic acid 3.69 (2.93, 4.54) 3.47 (2.81, 4.39) 3.21 (2.72, 4.60) 0.53
Open in a new tab
1

P values for baseline differences determined using Wilcoxon Rank Order Sum. Data presented as medians and interquartile ranges

Fish oil supplementation reduced urinary PGE-M in individuals allocated to the fish oil arm compared with the olive oil arm (P = 0.03) (Figure 2). The reduction in urinary PGE-M was greater at the 3-month measurement with a waning of the effect at the 6-month measurement period which was no longer statistically significant. We found no evidence of an interaction between fish oil supplementation and urinary PGE-M based on FADS genotype (Pinteraction = 0.39). We also found no evidence of effect modification between n-3 LCPUFA supplementation and urinary PGE-M based on NSAID use (Pinteraction = 0.64)

Figure 2.

Figure 2

Open in a new tab

Concentrations of Urinary PGEM at baseline, 3-months, and 6-months

aLinear regression comparing baseline to 3-month values adjusted for baseline value, genotype, and sample batch, bLinear regression comparing baseline to 6-month values adjusted for baseline value of urinary PGEM, genotype, and sample batch; cRepeated measures analysis of variance adjusted for genotype and sample batch

Overall, we found no changes in rectal mucosal levels of PGE2 after 6 months of supplemental fish oil (Figure 3). Rectal PGE2 did decrease over the course of the study in the fish oil arm compared to the olive oil arm however the difference was not statistically significant. We also compared baseline and 24-month changes in rectal PGE3, PGD2 and PGD3. For each measured rectal eicosanoid there was a decrease from baseline in the fish oil arm but the difference was not statistically significant (Figure 3). Similarly to urinary PGE-M we found no evidence of an interaction between fish oil supplementation and rectal eicosanoids by genotype (data not shown).

Figure 3.

Figure 3

Open in a new tab

Figure 3a–d: Changes in Rectal Eicosanoids after 6-Months of n-3 LCPUFA Supplementation Linear regression comparing baseline to 6-month values adjusted for baseline value of rectal eicosanoid, genotype, and sample batch. Panel a: Changes in mean log transformed rectal biopsy PGE2 concentrations; Panel b: Changes in mean log transformed rectal biopsy PGE3 concentrations; Panel c: Changes in mean log transformed rectal biopsy PGD2 concentrations; and Panel d: Changes in mean log transformed rectal biopsy PGD3 concentrations

We did find a statistically significant interaction between NSAID use and fish oil supplementation (Pinteraction = 0.008). In participants who did not use regular NSAIDs over the course of the study (n = 57, fish oil group = 23, olive oil group = 34), supplemental fish oil reduced rectal PGE2 levels from 3.10 ± 0.32 (log means ± S.E.) at baseline to 2.54 ± 0.28 at 6 months (18% reduction) compared with a change of 2.80 ± 0.23 to 3.09 ± 0.19 in participants allocated to placebo (10% increase) (P = 0.04). In contrast, participants who did use regular NSAIDs (n = 84, fish oil group = 47, olive oil group = 37), supplemental fish oil did not significantly reduce rectal PGE2 levels compared with placebo; levels went from 2.34 ± 0.22 at baseline to 1.84 ± 0.19 at 6 months, compared with a change of 1.61 ± 0.38 to 1.41 ± 0.36 in participants allocated to placebo (P = 0.19). We found no evidence of interaction with other rectal eicosanoids (data not shown).

We found a small but statistically significant positive correlation with an increase in urinary PGE-M being correlated with an increase in rectal PGE2 levels (adjusted correlation coefficient = 0.25 95% CI 0.06, 0.42, p-value = 0.01).

Over the course of the study RBC membrane percentages of n-3 LCPUFA increased by 76% compared to 6% for individuals allocated to olive oil (Figure 4). Similarly individuals allocated to the fish oil arm had a statistically significant decrease in RBC membrane AA content.

Figure 4.

Figure 4

Figure 4

Open in a new tab

Figure 4a–b: Changes in red blood cell phospholipid fatty acid content based on study arm

Discussion

We found in a randomized controlled trial that fish oil reduced urinary PGE-M levels compared with olive oil. This effect was strongest during the first three months of the study but we found no effect at 6 months. Changes in rectal mucosa PGE2 were smaller and significant only in individuals not using NSAIDs. We also found that the FADS1 polymorphism rs174537 did not appear to interact with fish oil intake even though this variant did influence AA levels. Finally, rectal mucosa PGE2 and urinary PGE-M had only a negligible correlation.

PGE-M, the major urinary metabolite of PGE2, is an established marker of systemic PGE2 (Sterz, Scherer et al. 2012, Wang and DuBois 2013). Several case-control studies have found higher levels of PGE-M in subjects with high-risk adenomas or known CRC than in controls (Cai, Gao et al. 2006, Johnson, Schmidt et al. 2006, Shrubsole, Cai et al. 2012, Bezawada, Song et al. 2014, Davenport, Cai et al. 2016). A recent meta-analysis concluded that PGE-M levels are strongly associated with an increased risk for multiple small tubular adenoma, advanced adenoma, and CRC, and that PGE-M is the most promising urinary biomarker for CRC risk assessment and screening (Altobelli, Angeletti et al. 2016).

PGE-M may also be an important marker to identify subsets of patients who would benefit the most from interventions targeted toward reducing inflammation. A nested case-control study found that regular NSAID use was associated with a reduction in adenoma risk in subjects with high baseline PGE-M but not in those with low PGE-M (Bezawada, Song et al. 2014). This suggests that PGE-M could help with both evaluating risk for CRC and assessing response to interventions – it has the potential to become a “risk-response” biomarker, similar to the role the lipid panel plays in the field of cardiovascular medicine in helping assess risk as well as evaluating response to statin therapy (Colbert Maresso, Vilar et al. 2014). Our finding that fish oil supplementation significantly decreased urinary PGE-M levels in our study population – which was by design a higher risk group in that all subjects had known history of adenomas – further supports the potential use of urinary PGE-M as a response marker. However one concern was the relatively weak correlation between rectal PGE2 and urinary PGE-M seen in our study. Currently underway is the seAFOod polyp prevention trial, a large randomized controlled trial to examine whether EPA, alone or in combination with aspirin, prevents colorectal adenomas (Hull, Sandell et al. 2013).

Several interventional studies have examined the effect of fish oil on production of eicosanoids and other inflammatory markers, with mixed but overall promising results. In 40 healthy male volunteers, supplementation of 15 grams/day of fish oil for 10 weeks led to a 14% reduction in urinary PGE-M compared with 15 grams/day of placebo oil (Ferretti, Judd et al. 1991). In a crossover feeding trial, 10 healthy male volunteers had a 24% reduction in PGE-M when fed a salmon-rich diet versus a placebo diet with identical macronutrient breakdown (Ferretti 1991). Another crossover trial found that rectal PGE2 levels were significantly lower after 4 weeks of fish oil supplementation compared with corn oil supplementation (435.5 vs 671.5 pg/mg wet tissue, p <0.05). A third crossover trial found that urinary excretion of thromboxane B2, a metabolite of platelet activator thromboxane A2 which is also derived from AA, was significantly reduced with a salmon diet compared with a reference diet (Prakash, Nelson et al. 1994). A study examining varying doses of fish oil supplementation found that PGE2 in peripheral mononuclear cells was decreased significantly for all doses of fish oil (0.3 g, 1g, 2g of EPA plus DHA), with a trend toward a dose response (Trebble, Wootton et al. 2003).

Several meta-analyses have been published on this topic (Jiang, Li et al. 2016). Five of the included studies reported PGE2 changes in response to fish oil supplementation; only two of these showed reduction in PGE2 (Andrade, Ribeiro et al. 2007, Tartibian, Maleki et al. 2011). However, of the three studies that did not show a significant reduction in PGE2, one involved healthy participants (Rees, Miles et al. 2006), and the other two included very limited patient populations, namely patients on hemodialysis (Peck, Monsen et al. 1996) and patients with rheumatoid arthritis (Park, Lee et al. 2013).

In our study, fish oil only statistically significantly lowered rectal mucosa PGE2 levels in non-users but not in users of aspirin/NSAIDs. Of note the baseline PGE2 levels were lower in the aspirin/NSAID users than in non-users. After fish oil supplementation for 24 weeks, rectal mucosa PGE2 levels in the non-users of aspirin/NSAID approached but did not reach the baseline levels in aspirin/NSAID users. This suggests that rectal PGE2 is reduced with fish oil supplementation, but this effect is not seen with NSAID use. Intriguingly our baseline rectal PGE2 levels were lower in aspirin/NSAID users who were allocated to olive oil supplementation. This difference was not statistically significant (P = 0.14) but may have impacted our stratified results. As this was a secondary analysis and participants were not randomized based on NSAID use, this difference in baseline rectal PGE2 status may be related to confounding factors. We found no differences in the groups based on age, body mass index, sex, smoking status, or educational level (data not shown) however an unmeasured confounders may have also influenced the results.

Strengths of the study included the randomized, controlled design. We had good compliance with the study medication as determined by pill counts and changes in RBC membrane PUFAs. Our participant demographics are similar to patient populations that would likely benefit most from improved colon cancer prevention strategies – namely middle-aged and older patients who have risk factors for CRC. The study has some weaknesses. Most notably, the results we report are of a secondary outcome of our study, and our study might have been underpowered to detect an overall effect on rectal PGE2. We did not collect rectal biopsies at 12 weeks and cannot report whether rectal PGE2 levels changes similarly to urinary PGE-M, with an initial strong decrease followed by a waning effect. Another weakness was our choice of FADS polymorphism to stratify the participant. While this polymorphism is associated with endogenous AA production, it has not been associated with CRC in GWAS studies. Our choice of dose for n-3 LCPUFA supplementation could have also impacted the results as limited data exists to determine if a dose-response effect exists between n-3 LCPUFA supplementation and eicosanoid production. Finally, the control arm included olive oil supplementation, which may have had an impact on eicosanoid production.

In conclusion, this study showed that supplemental fish oil reduced urinary PGE-M levels; however this reduction was seen at 3 months and not at 6 months. It would be important to replicate our findings as this could indicate that studies of a shorter duration might suggest a stronger eicosanoid modifying effect of fish oil than might exist over longer periods of time. Additionally, we found a very weak correlation between urinary PGE-M levels and rectal PGE2 levels suggesting that generalized PGE2 production only weakly represents colonic production. Finally, our results suggest that fish oil is more effective in reducing rectal PGE2 in individuals not using NSAIDs. These results might be better delineated after the results of the seAFOod Polyp Prevention Trial are reported.

Supplementary Material

Supplemental Data File _.doc_ .tif_ pdf_ etc._

Acknowledgments

Source of Funding:

Financial Support: This study was supported through the National Institute of Health grants R01CA160938, R01CA143288, P50CA95103, and R01CA97386. Surveys and sample collection, processing, and preparation for this study were conducted by the Survey and Biospecimen Shared Resource, which is supported in part by P30CA068485. Fatty acid analyses were performed by the Vanderbilt Diabetes Research and Training Core Lipid Core lab supported by DK20593. The study was supported in part by the Vanderbilt CTSA grant UL1TR002243. The content of this paper is solely the responsibility of the authors and does not necessarily represent the official views of the National Cancer Institute or the National Institutes of Health.

Footnotes

Conflicts of Interests

Disclosure of potential conflict of interest: The authors report no conflict of interests to disclose.

References

  1. Altobelli E, Angeletti PM, Latella G. Role of Urinary Biomarkers in the Diagnosis of Adenoma and Colorectal Cancer: A Systematic Review and Meta-Analysis. J Cancer. 2016;7(14):1984–2004. doi: 10.7150/jca.16244. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Andrade PM, Ribeiro BG, Bozza MT, Costa Rosa LF, Tavares do Carmo MG. Effects of the fish-oil supplementation on the immune and inflammatory responses in elite swimmers. Prostaglandins Leukot Essent Fatty Acids. 2007;77(3–4):139–145. doi: 10.1016/j.plefa.2007.08.010. [DOI] [PubMed] [Google Scholar]
  3. Anti M, Armelao F, Marra G, Percesepe A, Bartoli GM, Palozza P, Parrella P, Canetta C, Gentiloni N, De Vitis I, et al. Effects of different doses of fish oil on rectal cell proliferation in patients with sporadic colonic adenomas. Gastroenterology. 1994;107(6):1709–1718. doi: 10.1016/0016-5085(94)90811-7. [DOI] [PubMed] [Google Scholar]
  4. Bartoli GM, Palozza P, Marra G, Armelao F, Franceschelli P, Luberto C, Sgarlata E, Piccioni E, Anti M. n-3 PUFA and alpha-tocopherol control of tumor cell proliferation. Mol Aspects Med. 1993;14(3):247–252. doi: 10.1016/0098-2997(93)90011-2. [DOI] [PubMed] [Google Scholar]
  5. Bezawada N, Song M, Wu K, Mehta RS, Milne GL, Ogino S, Fuchs CS, Giovannucci EL, Chan AT. Urinary PGE-M levels are associated with risk of colorectal adenomas and chemopreventive response to anti-inflammatory drugs. Cancer Prev Res (Phila) 2014;7(7):758–765. doi: 10.1158/1940-6207.CAPR-14-0120. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Blair IA, Prakash C, Phillips MA, Dougherty RM, Iacono JM. Dietary modification of omega 6 fatty acid intake and its effect on urinary eicosanoid excretion. Am J Clin Nutr. 1993;57(2):154–160. doi: 10.1093/ajcn/57.2.154. [DOI] [PubMed] [Google Scholar]
  7. Boudreau MD, Chanmugam PS, Hart SB, Lee SH, Hwang DH. Lack of dose response by dietary n-3 fatty acids at a constant ratio of n-3 to n-6 fatty acids in suppressing eicosanoid biosynthesis from arachidonic acid. Am J Clin Nutr. 1991;54(1):111–117. doi: 10.1093/ajcn/54.1.111. [DOI] [PubMed] [Google Scholar]
  8. Boutaud O, I, Sosa R, Amin T, Oram D, Adler D, Hwang HS, Crews BC, Milne G, Harris BK, Hoeksema M, Knollmann BC, Lammers PE, Marnett LJ, Massion PP, Oates JA. Inhibition of the Biosynthesis of Prostaglandin E2 By Low-Dose Aspirin: Implications for Adenocarcinoma Metastasis. Cancer Prev Res (Phila) 2016;9(11):855–865. doi: 10.1158/1940-6207.CAPR-16-0094. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Broughton KS, Wade JW. Total fat and (n-3):(n-6) fat ratios influence eicosanoid production in mice. J Nutr. 2002;132(1):88–94. doi: 10.1093/jn/132.1.88. [DOI] [PubMed] [Google Scholar]
  10. Cai Q, Gao YT, Chow WH, Shu XO, Yang G, Ji BT, Wen W, Rothman N, Li HL, Morrow JD, Zheng W. Prospective study of urinary prostaglandin E2 metabolite and colorectal cancer risk. J Clin Oncol. 2006;24(31):5010–5016. doi: 10.1200/JCO.2006.06.4931. [DOI] [PubMed] [Google Scholar]
  11. Calder PC. Dietary modification of inflammation with lipids. Proc Nutr Soc. 2002;61(3):345–358. doi: 10.1079/pns2002166. [DOI] [PubMed] [Google Scholar]
  12. Cheng J, Ogawa K, Kuriki K, Yokoyama Y, Kamiya T, Seno K, Okuyama H, Wang J, Luo C, Fujii T, Ichikawa H, Shirai T, Tokudome S. Increased intake of n-3 polyunsaturated fatty acids elevates the level of apoptosis in the normal sigmoid colon of patients polypectomized for adenomas/tumors. Cancer Lett. 2003;193(1):17–24. doi: 10.1016/s0304383502007176. [DOI] [PubMed] [Google Scholar]
  13. Colbert Maresso K, Vilar E, Hawk ET. Urinary PGE-M in colorectal cancer: predicting more than risk? Cancer Prev Res (Phila) 2014;7(10):969–972. doi: 10.1158/1940-6207.CAPR-14-0215. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Courtney ED, Matthews S, Finlayson C, Di Pierro D, Belluzzi A, Roda E, Kang JY, Leicester RJ. Eicosapentaenoic acid (EPA) reduces crypt cell proliferation and increases apoptosis in normal colonic mucosa in subjects with a history of colorectal adenomas. Int J Colorectal Dis. 2007;22(7):765–776. doi: 10.1007/s00384-006-0240-4. [DOI] [PubMed] [Google Scholar]
  15. Davenport JR, Cai Q, Ness RM, Milne G, Zhao Z, Smalley WE, Zheng W, Shrubsole MJ. Evaluation of pro-inflammatory markers plasma C-reactive protein and urinary prostaglandin-E2 metabolite in colorectal adenoma risk. Mol Carcinog. 2016;55(8):1251–1261. doi: 10.1002/mc.22367. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Ferretti A, Judd JT, Ballard-Barbash R, Nair PP, Taylor PR, Clevidence BA. Effect of fish oil supplementation on the excretion of the major metabolite of prostaglandin E in healthy male subjects. Lipids. 1991;26(7):500–503. doi: 10.1007/BF02536593. [DOI] [PubMed] [Google Scholar]
  17. Ferretti A, Nelson GJ, Schmidt PC. Salmon-rich diet inhibits arachidonate cyclooxygenation in healthy men. J Nutr Biochem. 1991;2:547–552. [Google Scholar]
  18. Folch J, Lees M, Sloane Stanley GH. A simple method for the isolation and purification of total lipides from animal tissues. J Biol Chem. 1957;226(1):497–509. [PubMed] [Google Scholar]
  19. Garg ML, Sebokova E, Thomson AB, Clandinin MT. Delta 6-desaturase activity in liver microsomes of rats fed diets enriched with cholesterol and/or omega 3 fatty acids. Biochem J. 1988;249(2):351–356. doi: 10.1042/bj2490351. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gupta RA, Dubois RN. Colorectal cancer prevention and treatment by inhibition of cyclooxygenase-2. Nat Rev Cancer. 2001;1(1):11–21. doi: 10.1038/35094017. [DOI] [PubMed] [Google Scholar]
  21. Hillier K, Jewell R, Dorrell L, Smith CL. Incorporation of fatty acids from fish oil and olive oil into colonic mucosal lipids and effects upon eicosanoid synthesis in inflammatory bowel disease. Gut. 1991;32(10):1151–1155. doi: 10.1136/gut.32.10.1151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hull MA, Sandell AC, Montgomery AA, Logan RF, Clifford GM, Rees CJ, Loadman PM, Whitham D. A randomized controlled trial of eicosapentaenoic acid and/or aspirin for colorectal adenoma prevention during colonoscopic surveillance in the NHS Bowel Cancer Screening Programme (The seAFOod Polyp Prevention Trial): study protocol for a randomized controlled trial. Trials. 2013;14:237. doi: 10.1186/1745-6215-14-237. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Jiang J, Li K, Wang F, Yang B, Fu Y, Zheng J, Li D. Effect of Marine-Derived n-3 Polyunsaturated Fatty Acids on Major Eicosanoids: A Systematic Review and Meta-Analysis from 18 Randomized Controlled Trials. PLoS One. 2016;11(1):e0147351. doi: 10.1371/journal.pone.0147351. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Johnson JC, Schmidt CR, Shrubsole MJ, Billheimer DD, Joshi PR, Morrow JD, Heslin MJ, Washington MK, Ness RM, Zheng W, Schwartz DA, Coffey RJ, Beauchamp RD, Merchant NB. Urine PGE-M: A metabolite of prostaglandin E2 as a potential biomarker of advanced colorectal neoplasia. Clin Gastroenterol Hepatol. 2006;4(11):1358–1365. doi: 10.1016/j.cgh.2006.07.015. [DOI] [PubMed] [Google Scholar]
  25. Liu T, Laidlaw TM, Feng C, Xing W, Shen S, Milne GL, Boyce JA. Prostaglandin E2 deficiency uncovers a dominant role for thromboxane A2 in house dust mite-induced allergic pulmonary inflammation. Proc Natl Acad Sci U S A. 2012;109(31):12692–12697. doi: 10.1073/pnas.1207816109. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Mohebati A, Milne GL, Zhou XK, Duffield-Lillico AJ, Boyle JO, Knutson A, Bosworth BP, Kingsley PJ, Marnett LJ, Brown PH, Akpa EG, Szabo E, Dannenberg AJ. Effect of zileuton and celecoxib on urinary LTE4 and PGE-M levels in smokers. Cancer Prev Res (Phila) 2013;6(7):646–655. doi: 10.1158/1940-6207.CAPR-13-0083. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Morrison WR, Smith LM. Preparation of Fatty Acid Methyl Esters and Dimethylacetals from Lipids with Boron Fluoride--Methanol. J Lipid Res. 1964;5:600–608. [PubMed] [Google Scholar]
  28. Park Y, Lee A, Shim SC, Lee JH, Choe JY, Ahn H, Choi CB, Sung YK, Bae SC. Effect of n-3 polyunsaturated fatty acid supplementation in patients with rheumatoid arthritis: a 16-week randomized, double-blind, placebo-controlled, parallel-design multicenter study in Korea. J Nutr Biochem. 2013;24(7):1367–1372. doi: 10.1016/j.jnutbio.2012.11.004. [DOI] [PubMed] [Google Scholar]
  29. Peck LW, Monsen ER, Ahmad S. Effect of three sources of long-chain fatty acids on the plasma fatty acid profile, plasma prostaglandin E2 concentrations, and pruritus symptoms in hemodialysis patients. Am J Clin Nutr. 1996;64(2):210–214. doi: 10.1093/ajcn/64.2.210. [DOI] [PubMed] [Google Scholar]
  30. Prakash C, Nelson GJ, Wu MM, Schmidt PC, Phillips MA, Blair IA. Decreased systemic thromboxane A2 biosynthesis in normal human subjects fed a salmon-rich diet. Am J Clin Nutr. 1994;60(3):369–373. doi: 10.1093/ajcn/60.3.369. [DOI] [PubMed] [Google Scholar]
  31. Rees D, Miles EA, Banerjee T, Wells SJ, Roynette CE, Wahle KW, Calder PC. Dose-related effects of eicosapentaenoic acid on innate immune function in healthy humans: a comparison of young and older men. Am J Clin Nutr. 2006;83(2):331–342. doi: 10.1093/ajcn/83.2.331. [DOI] [PubMed] [Google Scholar]
  32. Rzehak P, Heinrich J, Klopp N, Schaeffer L, Hoff S, Wolfram G, Illig T, Linseisen J. Evidence for an association between genetic variants of the fatty acid desaturase 1 fatty acid desaturase 2 ( FADS1 FADS2) gene cluster and the fatty acid composition of erythrocyte membranes. Br J Nutr. 2009;101(1):20–26. doi: 10.1017/S0007114508992564. [DOI] [PubMed] [Google Scholar]
  33. Schaeffer L, Gohlke H, Muller M, Heid IM, Palmer LJ, Kompauer I, Demmelmair H, Illig T, Koletzko B, Heinrich J. Common genetic variants of the FADS1 FADS2 gene cluster and their reconstructed haplotypes are associated with the fatty acid composition in phospholipids. Hum Mol Genet. 2006;15(11):1745–1756. doi: 10.1093/hmg/ddl117. [DOI] [PubMed] [Google Scholar]
  34. Shrubsole MJ, Cai Q, Wen W, Milne G, Smalley WE, Chen Z, Ness RM, Zheng W. Urinary prostaglandin E2 metabolite and risk for colorectal adenoma. Cancer Prev Res (Phila) 2012;5(2):336–342. doi: 10.1158/1940-6207.CAPR-11-0426. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Smith WL. Cyclooxygenases, peroxide tone and the allure of fish oil. Curr Opin Cell Biol. 2005;17(2):174–182. doi: 10.1016/j.ceb.2005.02.005. [DOI] [PubMed] [Google Scholar]
  36. Sterz K, Scherer G, Ecker J. A simple and robust UPLC-SRM/MS method to quantify urinary eicosanoids. J Lipid Res. 2012;53(5):1026–1036. doi: 10.1194/jlr.D023739. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Tartibian B, Maleki BH, Abbasi A. Omega-3 fatty acids supplementation attenuates inflammatory markers after eccentric exercise in untrained men. Clin J Sport Med. 2011;21(2):131–137. doi: 10.1097/JSM.0b013e31820f8c2f. [DOI] [PubMed] [Google Scholar]
  38. Trebble TM, Wootton SA, Miles EA, Mullee M, Arden NK, Ballinger AB, Stroud MA, Burdge GC, Calder PC. Prostaglandin E2 production and T cell function after fish-oil supplementation: response to antioxidant cosupplementation. Am J Clin Nutr. 2003;78(3):376–382. doi: 10.1093/ajcn/78.3.376. [DOI] [PubMed] [Google Scholar]
  39. Wada M, DeLong CJ, Hong YH, Rieke CJ, Song I, Sidhu RS, Yuan C, Warnock M, Schmaier AH, Yokoyama C, Smyth EM, Wilson SJ, FitzGerald GA, Garavito RM, Sui de X, Regan JW, Smith WL. Enzymes and receptors of prostaglandin pathways with arachidonic acid-derived versus eicosapentaenoic acid-derived substrates and products. J Biol Chem. 2007;282(31):22254–22266. doi: 10.1074/jbc.M703169200. [DOI] [PubMed] [Google Scholar]
  40. Wander RC, Hall JA, Gradin JL, Du SH, Jewell DE. The ratio of dietary (n-6) to (n-3) fatty acids influences immune system function, eicosanoid metabolism, lipid peroxidation and vitamin E status in aged dogs. J Nutr. 1997;127(6):1198–1205. doi: 10.1093/jn/127.6.1198. [DOI] [PubMed] [Google Scholar]
  41. Wang D, Dubois RN. Prostaglandins and cancer. Gut. 2006;55(1):115–122. doi: 10.1136/gut.2004.047100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Wang D, DuBois RN. Urinary PGE-M: a promising cancer biomarker. Cancer Prev Res (Phila) 2013;6(6):507–510. doi: 10.1158/1940-6207.CAPR-13-0153. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Wang MT, Honn KV, Nie D. Cyclooxygenases, prostanoids, and tumor progression. Cancer Metastasis Rev. 2007;26(3–4):525–534. doi: 10.1007/s10555-007-9096-5. [DOI] [PubMed] [Google Scholar]
  44. Zhang B, Jia WH, Matsuda K, Kweon SS, Matsuo K, Xiang YB, Shin A, Jee SH, Kim DH, Cai Q, Long J, Shi J, Wen W, Yang G, Zhang Y, Li C, Li B, Guo Y, Ren Z, Ji BT, Pan ZZ, Takahashi A, Shin MH, Matsuda F, Gao YT, Oh JH, Kim S, Ahn YO, Chan AT, Chang-Claude J, Slattery ML, Gruber SB, Schumacher FR, Stenzel SL, Casey G, Kim HR, Jeong JY, Park JW, Li HL, Hosono S, Cho SH, Kubo M, Shu XO, Zeng YX, Zheng W Genetics, C. Epidemiology of Colorectal Cancer, S. Colorectal Transdisciplinary, R. Colon Cancer Family. Large-scale genetic study in East Asians identifies six new loci associated with colorectal cancer risk. Nat Genet. 2014;46(6):533–542. doi: 10.1038/ng.2985. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Data File _.doc_ .tif_ pdf_ etc._
NIHMS973571-supplement-Supplemental_Data_File___doc___tif__pdf__etc__.docx (128.7KB, docx)

RESOURCES