Background:

Whether circulating polyunsaturated fatty acid (PUFA) levels are associated with pancreatic cancer risk is uncertain. Mendelian randomization (MR) represents a study design using genetic instruments to better characterize the relationship between exposure and outcome.

Methods:

We utilized data from genome-wide association studies within the Pancreatic Cancer Cohort Consortium and Pancreatic Cancer Case–Control Consortium, involving approximately 9,269 cases and 12,530 controls of European descent, to evaluate associations between pancreatic cancer risk and genetically predicted plasma n-6 PUFA levels. Conventional MR analyses were performed using individual-level and summary-level data.

Results:

Using genetic instruments, we did not find evidence of associations between genetically predicted plasma n-6 PUFA levels and pancreatic cancer risk [estimates per one SD increase in each PUFA-specific weighted genetic score using summary statistics: linoleic acid odds ratio (OR) = 1.00, 95% confidence interval (CI) = 0.98–1.02; arachidonic acid OR = 1.00, 95% CI = 0.99–1.01; and dihomo-gamma-linolenic acid OR = 0.95, 95% CI = 0.87–1.02]. The OR estimates remained virtually unchanged after adjustment for covariates, using individual-level data or summary statistics, or stratification by age and sex.

Conclusions:

Our results suggest that variations of genetically determined plasma n-6 PUFA levels are not associated with pancreatic cancer risk.

Impact:

These results suggest that modifying n-6 PUFA levels through food sources or supplementation may not influence risk of pancreatic cancer.

Pancreatic cancer remains one of the deadliest cancers (1). Polyunsaturated fatty acids (PUFA), linked to the inflammatory process, may influence pancreatic cancer development (2). However, evidence from epidemiologic studies is inconsistent (3). For example, associations with n-3 PUFA intake were inverse, positive, or null, and associations with n-6 PUFA intake were positive or null across different studies. Conventional epidemiologic study designs may suffer from methodologic limitations, such as reverse causation, selection bias, and uncontrolled confounding (4). We, therefore, conducted a Mendelian randomization (MR) analysis using genetic variants as instrumental variables. Higher proportions of variance of n-6 PUFA levels were explained by variants compared with n-3 PUFA levels (5, 6). We thus focused on n-6 PUFA in our analysis.

Instrumental variables

We identified SNPs associated with plasma or red blood cell (RBC) levels of n-6 PUFAs [linoleic acid, arachidonic acid, adrenic acid, gamma linolenic acid (GLA), and dihomo-gamma-linolenic acid (DGLA)] from the genome-wide association studies (GWAS) catalog and from published literature (up to November 2018; ref. 6). We selected SNPs associated at P < 5 × 10−8 that were independent from each other (r2 < 0.1). For correlated SNPs, the SNP with a lower P value was selected unless an independent association was reported, in which case both were selected. We used estimates of association with plasma PUFA levels for our analyses. For SNPs initially reported to be associated with RBC PUFA levels, we checked their associations with plasma levels in the GWAS conducted by the Cohorts for Heart and Aging Research in Genomic Epidemiology Consortium (CHARGE; ref. 6); if estimates were significant (P < 0.05) we included them in our analyses.

Genetic association datasets for pancreatic cancer risk

For evaluation of associations with pancreatic cancer risk, we used data from four GWASs conducted in the Pancreatic Cancer Cohort Consortium (PanScan) and the Pancreatic Cancer Case–Control Consortium (PanC4; ref. 7). Detailed information on quality control and imputation has been provided elsewhere (7). Data from approximately 9,269 cases and 12,530 controls of European ancestry were used. Only variants with imputation quality of r2 ≥ 0.3 were retained.

MR analysis

We performed separate MR analyses for each type of PUFA. On the basis of power estimation (https://shiny.cnsgenomics.com/mRnd/), the minimal detectable ORs per SD of genetically predicted PUFA levels at 80% power and alpha of 0.05 ranged from 1.08 to 1.13 for linoleic acid, 1.06 to 1.21 for arachidonic acid, 1.13 to 1.15 for adrenic acid, 1.17 to 1.27 for GLA, and 1.08 to 1.11 for DGLA. We created a weighted genetic score (wGRS) to represent the genetically estimated PUFA level using information from published GWASs of PUFA plasma levels and data from PanScan/PanC4 GWASs. For each subject, a wGRS was created as the weighted sum of the number of association alleles at each locus multiplied by the point estimate for the association with plasma PUFA level:

formula

where |{\beta _i}$| is the regression coefficient of the ith SNP for the PUFA and |SN{P_i}$| is the dosage of the association alleles (0, 1, 2) of the ith SNP. All association alleles were converted to correspond to increased PUFA levels. We used logistic regression models to assess the associations between wGRS and pancreatic cancer risk. Besides unadjusted analyses, we performed analyses adjusting for age, sex, and the top principal components. We stratified the data by age (<50 years, 50–70 years, and ≥70 years) and sex. We also used the summary statistics of the PanScan/PanC4 GWASs to estimate the MR associations using the fixed effects inverse variance-weighted approach (4).

The instruments used for each of the n-6 PUFA are included in Table 1. There was no evidence of association (at P < 0.01) between any of the wGRS and common pancreatic cancer risk factors. We did not detect any statistically significant associations between genetically predicted plasma n-6 PUFA levels and pancreatic cancer risk (Table 2). The estimates for adrenic acid and GLA had wide confidence intervals, consistent with the estimated lower power for these. The associations remained virtually unchanged regardless of covariate adjustment, analyzing individual-level versus summary statistics data, or within strata of age or sex.

Table 1.

Genetic instruments for plasma phospholipid levels of n-6 PUFAs (% of total fatty acids) that were genome-wide significant (P < 5 × 10−8) in previous GWASs.

ChrSNPGRCh37/hg19 positionAlleleaEAFβSEP% VEb per allele% VE per IVcF-statistic per IVd
Linoleic acid (18:2n6) 
 10 rs10740118 65101207 G/C 0.56 0.2484 0.0431 8.08 × 10−9 0.2–0.7 9.4–25.1 452–1461 
 11 rs174547 61570783 C/T 0.32 1.4737 0.0417 4.98 × 10−274 7.6–18.1   
 11 rs2727270 61603237 T/C 0.44 0.69 0.07 2.60 × 10−21 0.5–2.4   
 16 rs16966952 15135943 A/G 0.31 0.3512 0.0439 1.23 × 10−15 0.5–2.5   
 16 rs2280018 15150833 A/C 0.38 0.38 0.05 3.60 × 10−14 0.6–1.4   
Arachidonic acid (20:4n6) 
 11 rs174547 61570783 T/C 0.68 1.6909 0.0253 3.30 × 10−971 3.7–37.6 4.1–44 311–5708 
 11 rs102275 61557803 T/C 0.68 2.49 0.1 6.60 × 10−147 0.3–5.8   
 16 rs16966952 15135943 G/A 0.69 0.1989 0.0314 2.43 × 10−10 0.1–0.6   
Adrenic acid (22:4,n6) 
 11 rs174547 61570783 T/C 0.67 0.0483 0.0019 6.26 × 10−140 7.8–10.9 7.8–10.9 1844–2667 
GLA (18:3,n6) 
 11 rs174547 61570783 T/C 0.67 0.0156 0.0009 2.29 × 10−72 2.2–4.6 2.5–6.4 186–497 
 16 rs16966952 15135943 G/A 0.69 0.0061 0.0009 5.05 × 10−11 0.3–1.8   
 11 rs10899123e 75501207 C/G 0.91 0.0055 0.0014 9.97 × 10−5 NA   
DGLA (20:3,n6) 
 11 rs174547 61570783 C/T 0.33 0.3550 0.0136 2.63 × 10−151 8.7–11.1 13.5–26.3 850–1944 
 11 rs968567 61595564 T/C 0.16 0.29 0.02 1.30 × 10−42 1.4–7.9   
 16 rs16966952 15135943 G/A 0.69 0.2204 0.013 7.55 × 10−65 2.0–4.5   
 16 rs2280018 15150833 C/A 0.61 0.16 0.02 4.50 × 10−25 1.4–2.8   
ChrSNPGRCh37/hg19 positionAlleleaEAFβSEP% VEb per allele% VE per IVcF-statistic per IVd
Linoleic acid (18:2n6) 
 10 rs10740118 65101207 G/C 0.56 0.2484 0.0431 8.08 × 10−9 0.2–0.7 9.4–25.1 452–1461 
 11 rs174547 61570783 C/T 0.32 1.4737 0.0417 4.98 × 10−274 7.6–18.1   
 11 rs2727270 61603237 T/C 0.44 0.69 0.07 2.60 × 10−21 0.5–2.4   
 16 rs16966952 15135943 A/G 0.31 0.3512 0.0439 1.23 × 10−15 0.5–2.5   
 16 rs2280018 15150833 A/C 0.38 0.38 0.05 3.60 × 10−14 0.6–1.4   
Arachidonic acid (20:4n6) 
 11 rs174547 61570783 T/C 0.68 1.6909 0.0253 3.30 × 10−971 3.7–37.6 4.1–44 311–5708 
 11 rs102275 61557803 T/C 0.68 2.49 0.1 6.60 × 10−147 0.3–5.8   
 16 rs16966952 15135943 G/A 0.69 0.1989 0.0314 2.43 × 10−10 0.1–0.6   
Adrenic acid (22:4,n6) 
 11 rs174547 61570783 T/C 0.67 0.0483 0.0019 6.26 × 10−140 7.8–10.9 7.8–10.9 1844–2667 
GLA (18:3,n6) 
 11 rs174547 61570783 T/C 0.67 0.0156 0.0009 2.29 × 10−72 2.2–4.6 2.5–6.4 186–497 
 16 rs16966952 15135943 G/A 0.69 0.0061 0.0009 5.05 × 10−11 0.3–1.8   
 11 rs10899123e 75501207 C/G 0.91 0.0055 0.0014 9.97 × 10−5 NA   
DGLA (20:3,n6) 
 11 rs174547 61570783 C/T 0.33 0.3550 0.0136 2.63 × 10−151 8.7–11.1 13.5–26.3 850–1944 
 11 rs968567 61595564 T/C 0.16 0.29 0.02 1.30 × 10−42 1.4–7.9   
 16 rs16966952 15135943 G/A 0.69 0.2204 0.013 7.55 × 10−65 2.0–4.5   
 16 rs2280018 15150833 C/A 0.61 0.16 0.02 4.50 × 10−25 1.4–2.8   

Abbreviations: Chr, chromosome; EAF, effect allele frequency; IV, instrumental variable; VE, variation explained.

aThe first listed allele represents the effect allele associated with an increased level of corresponding PUFA; the second allele represents the alternative allele.

b% VE = [2 × β2 × EAF × (1 − EAF)/var(PUFA)] × 100, unless indicated in article, such as in Guan and colleagues (6).

c% VE per IV = sum of the %VE per allele for each SNP included in the IV.

dF-statistic is a measure of the strength of the genetic instrument and is calculated as follows: [R2 × (n − 1 − k)]/[(1 − R2) × k], where R2 = % VE, n = sample size, k = total number of instrumental variables.

eGenetic variant, rs10899123, showed an association at 5 × 10−8 < P < 0.05 in the CHARGE studies (Guan and colleagues, 2014; ref. 6), although it showed a GWAS significant association in the study by Hu and colleagues (2016; ref. 9). It was included in the genetic instrument in sensitivity analyses while it did not significantly change the association. The analyses excluding it in the instrument are reported in Table 2.

Table 2.

Associations between 1 SD increase in PUFA-specific wGRSs and pancreatic cancer risk in PanScan and PanC4 studiesa.

Linoleic acidArachidonic acidDGLA
SubgroupCases/controlsOR (95% CI)POR (95% CI)POR (95% CI)P
Overallb 9,269/12,530 0.99 (0.97–1.01) 0.30 1.01 (1.00–1.02) 0.37 0.94 (0.87–1.01) 0.10 
Overallc 9,206/12,525 1.00 (0.98–1.02) 0.89 1.00 (0.99–1.01) 0.53 0.94 (0.87–1.02) 0.13 
Data sourcec 
PanScan1 1,746/1,812 1.03 (0.97–1.08) 0.31 0.99 (0.97–1.01) 0.35 0.97 (0.81–1.16) 0.72 
PanScan2 1,768/1,841 0.99 (0.93–1.04) 0.59 1.01 (0.99–1.03) 0.47 0.97 (0.82–1.15) 0.73 
PanScan3 1,528/5,080 0.99 (0.93–1.05) 0.66 1.01 (0.98–1.03) 0.56 0.91 (0.76–1.10) 0.34 
PanC4 4,164/3,792 1.00 (0.96–1.03) 0.91 1.01 (0.99–1.02) 0.53 0.94 (0.83–1.06) 0.30 
Overalld 9,040/12,496 1.00 (0.97–1.03) 0.95 1.00 (0.99–1.02) 0.73 0.95 (0.87–1.03) 0.21 
Agee 
>70 3,494/3,385 1.02 (0.98–1.06) 0.42 1.00 (0.98–1.01) 0.68 1.02 (0.90–1.17) 0.73 
50–70 3,917/6,916 0.99 (0.96–1.03) 0.74 1.01 (0.99–1.02) 0.55 0.91 (0.81–1.02) 0.11 
≤50 1,795/2,224 0.98 (0.93–1.03) 0.42 1.01 (0.99–1.04) 0.32 0.90 (0.75–1.07) 0.24 
Sexf 
Male 4,985/7,801 1.00 (0.97–1.03) 0.94 1.00 (0.99–1.02) 0.66 0.97 (0.87–1.08) 0.55 
Female 4,221/4,225 1.00 (0.97–1.04) 0.99 1.00 (0.99–1.02) 0.71 0.99 (0.95–1.04) 0.77 
Linoleic acidArachidonic acidDGLA
SubgroupCases/controlsOR (95% CI)POR (95% CI)POR (95% CI)P
Overallb 9,269/12,530 0.99 (0.97–1.01) 0.30 1.01 (1.00–1.02) 0.37 0.94 (0.87–1.01) 0.10 
Overallc 9,206/12,525 1.00 (0.98–1.02) 0.89 1.00 (0.99–1.01) 0.53 0.94 (0.87–1.02) 0.13 
Data sourcec 
PanScan1 1,746/1,812 1.03 (0.97–1.08) 0.31 0.99 (0.97–1.01) 0.35 0.97 (0.81–1.16) 0.72 
PanScan2 1,768/1,841 0.99 (0.93–1.04) 0.59 1.01 (0.99–1.03) 0.47 0.97 (0.82–1.15) 0.73 
PanScan3 1,528/5,080 0.99 (0.93–1.05) 0.66 1.01 (0.98–1.03) 0.56 0.91 (0.76–1.10) 0.34 
PanC4 4,164/3,792 1.00 (0.96–1.03) 0.91 1.01 (0.99–1.02) 0.53 0.94 (0.83–1.06) 0.30 
Overalld 9,040/12,496 1.00 (0.97–1.03) 0.95 1.00 (0.99–1.02) 0.73 0.95 (0.87–1.03) 0.21 
Agee 
>70 3,494/3,385 1.02 (0.98–1.06) 0.42 1.00 (0.98–1.01) 0.68 1.02 (0.90–1.17) 0.73 
50–70 3,917/6,916 0.99 (0.96–1.03) 0.74 1.01 (0.99–1.02) 0.55 0.91 (0.81–1.02) 0.11 
≤50 1,795/2,224 0.98 (0.93–1.03) 0.42 1.01 (0.99–1.04) 0.32 0.90 (0.75–1.07) 0.24 
Sexf 
Male 4,985/7,801 1.00 (0.97–1.03) 0.94 1.00 (0.99–1.02) 0.66 0.97 (0.87–1.08) 0.55 
Female 4,221/4,225 1.00 (0.97–1.04) 0.99 1.00 (0.99–1.02) 0.71 0.99 (0.95–1.04) 0.77 

Abbreviation: CI, confidence interval.

aResults for adrenic acid and GLA not shown; their associations were not significant, with relatively wide CIs.

bORs and 95% CIs estimated using individual-level data without adjustment, and represent 1 SD increase in each PUFA-specific wGRS.

cORs and 95% CIs estimated using individual-level data with adjustment of age (under 50, 50–60, 60–70, 70–80, and above 80), sex, and 10 or seven principal components for PanScan and PanC4 data, respectively, and represent 1 SD increase in each PUFA-specific wGRS.

dORs and 95% CIs estimated using summary statistics data.

eORs and 95% CIs estimated using individual-level data with adjustment of age, sex, and 10 or seven principal components for PanScan and PanC4 data, respectively, and represent 1 SD increase in each PUFA-specific wGRS.

fORs and 95% CIs estimated using individual-level data with adjustment of age (under 50, 50–60, 60–70, 70–80, and above 80) and 10 or seven principal components for PanScan and PanC4 data, respectively, and represent 1 SD increase in each PUFA-specific wGRS.

We did not observe significant associations between genetically predicted n-6 PUFA levels and pancreatic cancer risk in the PanScan/PanC4 subjects. As the proportion of variance of n-6 PUFA that can be explained by the summed association magnitudes of these GWAS-identified loci is relatively high, there is reasonable statistical power to detect any meaningful associations. PUFA has been reported to be associated with colorectal cancer risk on the basis of MR analysis (8). Most dietary sources of n-6 PUFA are consumed infrequently. A limitation of this study is that there is no information for the genetic variants associated with total n-6 PUFA levels in previous literature, and thus, we could not determine the association between total n-6 PUFA levels and pancreatic cancer risk using genetic instruments. Alternative designs of a direct assessment of dietary sources and measurement of PUFA levels in blood at repeat timepoints can better characterize the relationship between PUFA and pancreatic cancer risk. Further studies are also needed to investigate the potential relationships in subjects of other populations.

P. Haycock reports grants from Cancer Research UK during the conduct of the study. M. Du reports grants from NCI (P30CA008748), Geoffrey Beene Foundation, Arnold and Arlene Goldstein Family Foundation, and Society of Memorial Sloan Kettering Cancer Center during the conduct of the study. G.G. Giles reports grants from National Health and Medical Research Council (paid to institution, Cancer Council Victoria) during the conduct of the study. L. Le Marchand reports grants from NCI during the conduct of the study. R.E. Neale reports grants from National Health and Medical Research Council during the conduct of the study. J.E. Buring reports grants from NIH during the conduct of the study. C.S. Fuchs reports personal fees from Agios, Amylin Pharmaceuticals, Bain Capital, CytomX Therapeutics, Daiichi Sankyo, Eli Lilly, Entrinsic Health, EvolveImmune Therapeutics, Genentech, Merck, Taiho, and Unum Therapeutics outside the submitted work; serves as a director for CytomX Therapeutics and owns unexercised stock options for CytomX Therapeutics and Entrinsic Health; is a cofounder of EvolveImmune Therapeutics and has equity in this private company; and has provided expert testimony for Amylin Pharmaceuticals and Eli Lilly. I-M. Lee reports grants from NIH during the conduct of the study. R.L. Milne reports grants from National Health and Medical Research Council during the conduct of the study. A.L. Oberg reports grants from NCI (P50 CA102701) during the conduct of the study. I.M. Thompson Jr reports grants from NCI, NIH (several grants in support of conduct and administration of SELECT and PCPT studies) during the conduct of the study. J. Wactawski-Wende reports grants from NIH/National Heart, Lung, and Blood Institute (funding for WHI) during the conduct of the study. A. Zeleniuch-Jacqotte reports grants from NIH/NCI during the conduct of the study. A.P. Klein reports grants from NCI during the conduct of the study. X.-O. Shu reports grants and personal fees from NCI (grant review) during the conduct of the study. L. Wu reports grants from NCI during the conduct of the study. No potential conflicts of interest were disclosed by the other authors.

The authors assume full responsibility for analyses and interpretation of these data.

D.H. Ghoneim: Formal analysis, writing–review and editing. J. Zhu: Formal analysis, investigation, writing–original draft. W. Zheng: Methodology, writing–review and editing. J. Long: Writing–review and editing. H.J. Murff: Writing–review and editing. F. Ye: Writing–review and editing. V.W. Setiawan: Writing–review and editing. L.R. Wilkens: Writing–review and editing. N.K. Khankari: Resources, writing–review and editing. P. Haycock: Methodology, writing–review and editing. S.O. Antwi: Writing–review and editing. Y. Yang: Writing–review and editing. A.A. Arslan: Resources, data curation. L.E. Beane Freeman: Resources, data curation. P.M. Bracci: Resources, data curation. F. Canzian: Resources, data curation. M. Du: Resources, data curation. S. Gallinger: Resources, data curation. G.G. Giles: Resources, data curation. P.J. Goodman: Resources, data curation. C. Kooperberg: Resources, data curation. L. Le Marchand: Resources, data curation, writing–review and editing. R.E. Neale: Resources, data curation, writing–review and editing. G. Scelo: Resources, data curation. K. Visvanathan: Resources, data curation. E. White: Resources, data curation. D. Albanes: Resources, data curation. P. Amiano: Resources, data curation. G. Andreotti: Resources, data curation. A. Babic: Resources, data curation. W.R. Bamlet: Data curation. S.I. Berndt: Resources, data curation. L.K. Brais: Resources, data curation. P. Brennan: Resources, data curation. B. Bueno-de-Mesquita: Resources, data curation. J.E. Buring: Resources, data curation. P.T. Campbell: Resources, data curation. K.G. Rabe: Data curation. S.J. Chanock: Resources, data curation. P. Duggal: Resources, data curation. C.S. Fuchs: Resources, data curation. J.M. Gaziano: Resources, data curation. M.G. Goggins: Resources, data curation. T. Hackert: Resources, data curation. M.M. Hassan: Resources, data curation. K.J. Helzlsouer: Resources, data curation. E.A. Holly: Resources, data curation. R.N. Hoover: Resources, data curation. V. Katzke: Resources, data curation. R.C. Kurtz: Resources, data curation. I-M. Lee: Resources, data curation. N. Malats: Resources, data curation. R.L. Milne: Resources, data curation, writing–review and editing. N. Murphy: Resources, data curation. A.L. Oberg: Data curation. M. Porta: Resources, data curation. N. Rothman: Resources, data curation. H.D. Sesso: Resources, data curation. D.T. Silverman: Resources, data curation. I.M. Thompson Jr: Resources, data curation. J. Wactawski-Wende: Resources, data curation. X. Wang: Data curation. N. Wentzensen: Resources, data curation. H. Yu: Resources, writing–review and editing. A. Zeleniuch-Jacqotte: Resources, data curation. K. Yu: Data curation. B.M. Wolpin: Resources, data curation. E.J. Jacobs: Resources, data curation. E.J. Duell: Resources, data curation. H.A. Risch: Resources, data curation, writing–review and editing. G.M. Petersen: Resources, data curation. L.T. Amundadottir: Data curation. P. Kraft: Resources, data curation. A.P. Klein: Resources, data curation. R.Z. Stolzenberg-Solomon: Resources, data curation. X.-O. Shu: Conceptualization, resources, supervision, investigation, writing–review and editing. L. Wu: Conceptualization, resources, data curation, formal analysis, supervision, funding acquisition, investigation, methodology, writing–original draft, project administration.

The authors are indebted to the research team and participants of the PanScan and PanC4 consortia participating studies for their contributions to this study. This study was supported by NCI grants K99 CA218892 and R00 CA218892. The Multiethnic Cohort was supported by grant U01 CA164973. The Women's Health Initiative program was funded by the National Heart, Lung, and Blood Institute, NIH, and U.S. Department of Health and Human Services through contracts HHSN268201600018C, HHSN268201600001C, HHSN268201600002C, HHSN268201600003C, and HHSN268201600004C. The Connecticut Pancreas Cancer Study was supported, in part, by NCI-NIH grant 5R01CA098870 (to H.A. Risch). The cooperation of 30 Connecticut hospitals, including Stamford Hospital, in allowing patient access, is gratefully acknowledged. This study was approved by the State of Connecticut Department of Public Health Human Investigation Committee. Certain data used in this study were obtained from the Connecticut Tumor Registry in the Connecticut Department of Public Health. A detailed list of acknowledgments for other PanScan/PanC4 participating studies is included elsewhere (Klein and colleagues; ref. 7).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Gong
Z
,
Holly
EA
,
Bracci
PM
. 
Survival in population-based pancreatic cancer patients: San Francisco Bay area, 1995–1999
.
Am J Epidemiol
2011
;
174
:
1373
81
.
2.
Funahashi
H
,
Satake
M
,
Hasan
S
,
Sawai
H
,
Newman
RA
,
Reber
HA
, et al
Opposing effects of n-6 and n-3 polyunsaturated fatty acids on pancreatic cancer growth
.
Pancreas
2008
;
36
:
353
62
.
3.
Qin
B
,
Xun
P
,
He
K
. 
Fish or long-chain (n-3) PUFA intake is not associated with pancreatic cancer risk in a meta-analysis and systematic review
.
J Nutr
2012
;
142
:
1067
73
.
4.
Burgess
S
,
Butterworth
A
,
Thompson
SG
. 
Mendelian randomization analysis with multiple genetic variants using summarized data
.
Genet Epidemiol
2013
;
37
:
658
65
.
5.
Lemaitre
RN
,
Tanaka
T
,
Tang
W
,
Manichaikul
A
,
Foy
M
,
Kabagambe
EK
, et al
Genetic loci associated with plasma phospholipid n-3 fatty acids: a meta-analysis of genome-wide association studies from the CHARGE Consortium
.
PLoS Genet
2011
;
7
:
e1002193
.
6.
Guan
W
,
Steffen
BT
,
Lemaitre
RN
,
Wu
JHY
,
Tanaka
T
,
Manichaikul
A
, et al
Genome-wide association study of plasma N6 polyunsaturated fatty acids within the Cohorts for Heart and Aging Research in Genomic Epidemiology Consortium
.
Circ Cardiovasc Genet
2014
;
7
:
321
31
.
7.
Klein
AP
,
Wolpin
BM
,
Risch
HA
,
Stolzenberg-Solomon
RZ
,
Mocci
E
,
Zhang
M
, et al
Genome-wide meta-analysis identifies five new susceptibility loci for pancreatic cancer
.
Nat Commun
2018
;
9
:
556
.
8.
Khankari
NK
,
Banbury
BL
,
Borges
MC
,
Haycock
P
,
Albanes
D
,
Arndt
V
, et al
Mendelian randomization of circulating polyunsaturated fatty acids and colorectal cancer risk
.
Cancer Epidemiol Biomarkers Prev
2020
;
29
:
860
70
.
9.
Hu
Y
,
Li
H
,
Lu
L
,
Manichaikul
A
,
Zhu
J
,
Chen
YD
, et al
Genome-wide meta-analyses identify novel loci associated with n-3 and n-6 polyunsaturated fatty acid levels in Chinese and European-ancestry populations
.
Hum Mol Genet
2016
;
25
:
1215
24
.