Abstract
Findings on the associations of dietary/tissue levels of omega-6 polyunsaturated fatty acids (n-6 PUFAs) with the risk of colorectal cancer (CRC) are conflicting. We conducted a dose-response meta-analysis to assess the associations of dietary/tissue levels of n-6 PUFAs [total, linoleic acid (LA), and arachidonic acid (AA)] with CRC risk in adults. Twenty prospective cohort studies with a total sample size of 787,490 participants were included. Comparing extreme intake levels of LA revealed the summary relative risks (RR) of 1.15 (95% confidence interval (CI): 1.05–1.27) for CRC, and 1.30 (95% CI: 1.00–1.68) for rectal cancer, indicating a significant positive association for LA. However, neither total n-6 PUFAs nor AA were associated with cancers. A significant positive association was also found between a 1 gr/day increase in dietary LA intake and risk of colon cancer (RR: 1.01, 95% CI: 1.00–1.02). There were no significant associations between tissue levels of total n-6 PUFAs (RR: 0.94, 95% CI: 0.75–1.19), LA (RR: 0.93, 95% CI: 0.61–1.41), and AA (RR: 0.97, 95% CI: 0.70–1.33) and CRC risk. In conclusion, these findings suggest that dietary intake, but not tissue levels, of LA was associated with an increased risk of colorectal, colon, and rectal cancers. (PROSPERO registration: CRD42024516584).
Subject terms: Gastrointestinal cancer, Epidemiology
Introduction
Colorectal cancer (CRC) is a prevalent gastrointestinal cancer worldwide [1], with its incidence increasing at an alarming rate and predicted to reach 2.5 million new cases by 2035. CRC is the fourth leading cause of cancer-related deaths worldwide, accounting for more than 900,000 deaths annually [1, 2]. This cancer also imposes a significant economic burden on healthcare systems, highlighting the need for effective preventive strategies.
It is well established that lifestyle factors, including both genetic and environmental influences, play an important role in the etiology of CRC [3]. Of different environmental factors, diet has always been of particular interest. Emerging evidence from epidemiological studies indicates that adherence to a Western-style or high-fat diet is associated with an increased risk of CRC [4], while adherence to Mediterranean or DASH diet is associated with a reduced risk [5, 6]. These dietary patterns have been characterized by differing notably in their content of various fatty acids, particularly omega-6 polyunsaturated fatty acids (n-6 PUFA). However, it remains unclear whether the increased risk of CRC is directly attributable to the effects of these fatty acids or other dietary factors also contribute. Vegetable oils like sunflower, safflower, soybean, corn, and canola oils, nuts, and seeds are other dietary sources of these fatty acids, particularly linoleic acid (LA) [7]. Overall, due to the role of n-6 PUFAs in inflammatory responses, they may increase the risk of some cancers [8, 9]. However, the findings of two systematic reviews reveal that increasing dietary intake of LA, the most abundant n-6 PUFA, does not have a significant effect on inflammatory markers [10, 11]. On the other hand, it has been shown that the conversion rate of dietary LA to AA in humans is low [12]. Overall, the role of dietary LA on inflammation is still unclear. In addition, findings from observational studies on the link between n-6 PUFAs and CRC risk are inconsistent [13–32], with some studies indicated a significant positive association between dietary and tissue biomarkers of n-6 PUFA and risk of CRC [17, 20, 22, 23, 29], while other studies did not report any significant association [13, 15, 20, 22, 24, 27, 28, 31] and even an inverse association [14].
A recent meta-analysis (Lu et al. 2023) summarized available findings on the association between dietary/tissue biomarkers of n-6 PUFA and CRC [33], revealing that the n-6/n-3 PUFA ratio is related to a higher risk of CRC. However, several eligible studies were not included in that meta-analysis [23, 32], and it did not assess the dose-response relationship between dietary n-6 PUFA and CRC risk, focusing only on comparisons between the highest and lowest intake levels. A dose-response analysis provides additional insights into the association between dietary n-6 PUFA and CRC risk, recognizing that fat intake does vary substantially across different populations. Accordingly, we have now conducted a comprehensive systematic review and dose-response meta-analysis of all existing prospective cohort studies to provide an improved understanding of association(s) between dietary/tissue biomarkers of n-6 PUFAs and risk of CRC in adults.
Methods
This systematic review and meta-analysis was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses protocol [34].
Search strategy
We conducted a systematic search using online databases, including PubMed, Scopus, and ISI Web of Science until March 5, 2024, to identify prospective cohort studies that examined the associations between dietary intake and tissue biomarkers of n-6 PUFAs (total, LA, and AA) and risk of CRC. Supplementary Table 1 presented the medical subject heading terms (MeSH) and non-MeSH terms used in the search strategy. In addition to the mentioned databases, we conducted a web-based search in Google Scholar using a combination of “omega-6 fatty acid” and “colorectal cancer” terms. In this search engine, we screened the first 500 papers ranked based on relevancy. In the systematic search, no restrictions were considered for publication time or the language of articles. We also searched the reference lists of the included articles and recent reviews to ensure comprehensive inclusion.
Inclusion criteria
We included prospective studies (e.g., prospective cohort, nested case-control, case-cohort studies), those that recruited adults (≥ 18 years), and reported relative risk (RR) estimates, including hazard ratio (HR), risk ratio (RR), and odds ratio (OR), with 95% confidence interval (CI) for the associations of dietary/tissue biomarkers of n-6 PUFAs (total, LA, AA) with colorectal, colon, or rectal cancers risk or presenting required data for the calculation of these effect sizes. If findings from one dataset were published in more than one article, we selected the one with the greatest number of cases or longer follow-up duration.
Exclusion criteria
We excluded letters, comments, reviews, meta-analyses, animal studies, abstracts, citations, and those studies with insufficient data. We also did not include studies that were conducted on children and adolescents, recruited critically ill participants, had a retrospective design, and considered genetically predicted dietary of n-6 PUFAs. In addition, studies that assessed dietary intake of fatty acids, without considering n-6 PUFAs (total, LA, and AA), were excluded.
Data extraction
Study selection and data extraction were conducted by two independent investigators (NA and NE) and any disagreement between them was resolved by discussion with a third researcher (OS). For the meta-analysis, we extracted any reported relative risk estimates, including RR, HR, and OR, along with 95% CI for the associations of dietary intake and tissue biomarkers of n-6 PUFA (total, LA, and AA) with colorectal, colon, and rectal cancer risk. In the case of studies that reported several risk estimates, we selected the one with the most adjustments. In addition to the risk estimates, we extracted additional information on the first author’s name, publication year, cohort name, sample size, number of cases, gender and age of participants, study location, follow-up duration (years), methods used to assess dietary or tissue levels of n-6 PUFA and colorectal cancer diagnosis, and confounding variables adjusted in the statistical analysis. For studies that reported their findings by gender, we considered that study as two separate studies. For the studies that reported main analyses for total population as well as stratified analyses for gender or any other variables, we used the effect size for total population in our primary analysis.
Quality assessment
The quality of studies included in the current meta-analysis was evaluated using the Newcastle Ottawa Scale (NOS) [35]. According to this scale, a maximum of 9 points would be given to each study according to the following parameters: 4 points for selection of participants, 2 points for comparability, and 3 points for the assessment of outcomes. Since the median score of included studies in the current meta-analysis was 7, we considered studies with a score of ≥7 as high-quality studies.
Statistical analysis
We included the relative risks (RRs, HRs, and ORs) of colorectal, colon, and rectal cancers, reported for the comparison between the highest and lowest intakes/tissue levels of n-6 PUFA (total, LA, and AA), into the meta-analysis. Since the relative risks are non-normally distributed variables, we included the natural log form (and its standard error) of these risk estimates into the statistical analyses. To calculate the summary relative risk, a random-effects model was used to take between-study heterogeneity into account. To assess heterogeneity among the included studies, both Q-statistic and I2 values were used. For the I2 statistic, values of >50% were considered as significant heterogeneity between studies. In addition to the main analysis, subgroup analyses were conducted based on gender, study location, and study quality. To assess publication bias, we used Egger’s regression asymmetry test [36]. For the significant publication bias, the trim-and-fill method was used to detect the effect of probable missing studies on the overall relative risk [37]. Furthermore, sensitivity analysis was conducted to evaluate the dependency of overall effect size on one study. In this analysis, each study was excluded to assess the influence of that study on the overall estimate.
For the linear dose-response analysis, we used the generalized least squares trend (glst command in STATA) estimation method [38, 39]. Firstly, study specific slopes were estimated, and then, these slopes were combined to obtain an overall average slope. Using a random-effects model, the study specific slopes were combined. In the glst method, outcome distribution, the total number of participants, and the effect sizes with the variance estimates for ≥3 quantitative categories of exposure were required. For each study, we assigned the median or mean amount of n-6 PUFA (total, LA, and AA) in each category to the corresponding effect size. For studies that reported the intake of n-6 PUFAs (total, LA, and AA) as the percent of energy (%E), we converted them to gr/day. For studies that reported n-6 PUFA as ranges, the midpoint in each category was estimated. When the highest and lowest categories were open ended, we assumed the length of the open-ended interval to be the same as that of the adjacent interval. Also, a possible non-linear dose-response association was examined using restricted cubic splines with 3 knots at centiles of 10%, 50%, and 90% of the distribution. The correlation within each set of provided risk estimates was accounted for and the study specific estimates were combined by using a linear mixed effects meta-analysis. This method estimates the study specific slopes and combines them to obtain an overall average slope in a single stage. The significance for non-linearity was calculated by null hypothesis testing, in which we considered the coefficient of the second spline to be equal to zero. Statistical analyses were performed using STATA version 14.0. For all tests, including Cochran’s Q test, P < 0.05 was considered statistically significant.
Result
Literature search
4005 articles were retrieved from the initial search (Fig. 1). Following the exclusion of duplicate papers (n = 1238) and those that did not meet the inclusion criteria (n = 2737), 30 potentially relevant full-text articles were identified. Of the 30 papers, a further 7 articles were excluded. These studies used a case-control design (n = 4) [40–43], reporting an overall risk estimate of gastrointestinal cancer, but not CRC (n = 1) [44], and reporting colorectal adenoma as an outcome (n = 2) [45, 46]. Five duplicate articles were identified, of which two were related to the Netherlands cohort study [31, 47], three to the Nurses’ Health Study and Health Professionals Follow-up Study [30, 48, 49]. As these articles evaluated similar exposure variables, we included only the one with higher quality, with the greatest number of cases, or higher follow-up period [30, 31], and excluded others [47–49]. Two other duplicate articles were also identified, and all were included, given that different outcome variables (CRC, colon cancer, and rectal cancer) were investigated [13, 19]. Accordingly, 20 prospective cohort studies were included in the current systematic review and meta-analysis [13–32], of which fourteen evaluated the association between n-6 PUFA and CRC (10 on dietary n-6 PUFA intake [13, 15, 18, 21, 22, 29, 30], 3 on tissue levels of n-6 PUFA [16, 20, 23], and one on both [17]), eleven assessed the link between LA and CRC (7 on dietary LA intake [22, 24, 26, 27, 29, 30, 32], 3 on tissue levels of LA [14, 20, 23], and one on both [17]), nine assessed the relation between AA and CRC (5 on dietary AA intake [22, 24, 26, 29, 30], 3 on tissue levels of AA [14, 20, 23], and one on both [17]). In addition, three studies investigated the relationship between dietary or tissue levels of n-6 PUFA and colon or rectal cancers without considering CRC in relation to dietary or tissue levels of n-6 PUFA [19, 28, 31].
Fig. 1.
Flow diagram of study selection.
Characteristics of included studies
The characteristics of included studies, published between 1999 and 2021, are summarized in Supplementary Table 2. The number of participants ranged from 460 to 134,017, totaling 787,490, with an age range between 27 and 84 years. Duration of the follow-up ranged from 6 to 26 years, with a total of 10,694 CRC, 5417 colon cancer, and 2533 rectal cancer cases were recorded. Three studies recruited men only [16, 26, 27], five were conducted solely on women [22, 24, 25, 29, 32], and the remaining studies enrolled both genders [23, 28, 30, 31], of which, only four reported sex-stratified effect sizes [15, 20, 28, 30]. Five studies were from the United States (US) [15, 16, 22, 25, 30], one from Australia [17], and seven from European populations [18, 21, 23, 27, 29, 31, 32], and Asian [13, 14, 19, 20, 24, 26, 28] countries. In addition to the assessment of dietary intake at the study baseline, 5 studies repeated this assessment during the follow-up period [15, 24, 26, 28, 30].
Of sixteen studies on dietary intake of n-6 PUFAs, fifteen used food frequency questionnaire [13, 15, 17, 19, 21, 22, 24–32] and one used food diaries [18] for dietary assessment. Two studies collected dietary data through a face-to-face interview [19, 26], and others used self-reported data in their analysis [13, 15, 17, 21, 22, 24, 25, 27–32]. In all studies on tissue levels of n-6 PUFAs, fatty acids were measured in blood [14, 16, 17, 20, 23], and all used chromatography methods to measure n-6 fatty acids concentrations. CRC and its subtypes were determined using data from medical records or cancer registries in thirteen studies [13, 14, 24, 31, 32], self-reported data in five [15, 16, 22, 25, 30], and both medical records and self-reported data in two studies [23, 26]. In the majority of studies, relative risk estimates were adjusted for some important confounders including family history of CRC (n = 7), age (n = 17), body mass index (n = 17), smoking (n = 17), alcohol consumption (n = 17), physical activity (n = 14), energy intake (n = 16), and other dietary variables, including processed meat, fiber and calcium (n = 8). The NOS scores of the studies ranged from 6 to 9, with a median of 7 Sixteen studies had a score of ≥7, and were, accordingly, considered high-quality studies [13, 14, 32] (Supplementary Tables 3 and 4).
Findings from the systematic review
None of the included studies revealed a significant association between dietary intake or tissue levels of n-6 PUFAs and the risk of CRC and colon cancer. Of eight studies, investigating the association between dietary intake of n-6 PUFAs and risk of rectal cancer, two reported a positive association [17, 29]. Of studies assessing dietary intake of LA, one study showed a positive association with CRC [29] and two indicated a positive association with rectal cancer [17, 29]. No significant association was found for colon cancer. In contrast, one study revealed a significant inverse association between dietary intake of LA and risk of CRC and colon cancer [14]. No other significant association was seen for tissue levels of LA and AA with risk of CRC, colon, and rectal cancer.
Meta-analysis on n-6 PUFAs and risk of CRC
Dietary n-6 PUFA
Sixteen studies evaluated the association between dietary intake of n-6 PUFAs and risk of CRC [13, 15, 21, 22, 28–30]. Among 738,604 participants included in these studies, 9152 cases of CRC, 3797 cases of colon cancer, and 1676 cases of rectal cancer were recorded during the follow-up period. Comparing the highest categories of dietary n-6 PUFAs intake with the lowest, RR for CRC, was 1.05 (95% CI: 0.94–1.17, P = 0.38, I2 = 48.9%; P = 0.02 for heterogeneity), indicating no significant association between dietary n-6 PUFAs intake and CRC risk (Table 1 and Supplementary Fig. 1). Such non-significant association was seen for colon (RR: 1.02, 95% CI: 0.91–1.15, P = 0.71, I2 = 0.0%; P = 0.88 for heterogeneity) and rectal (RR: 1.19, 95% CI: 0.93–1.52, P = 0.18, I2 = 47.6%; P = 0.06 for heterogeneity) cancers. Fifteen studies were eligible for the linear dose-response analysis of dietary n-6 PUFAs intake and risk of CRC [13, 15, 21, 22, 24, 25, 28–30]. There was no significant linear association between a 1 gr/day increase in dietary n-6 PUFAs intake and the risk of colorectal (RR: 1.00, 95% CI: 0.99–1.01, P = 0.97, I2 = 41.4%; P = 0.07 for heterogeneity), colon (RR: 1.00, 95% CI: 0.99–1.01, P = 0.88, I2 = 0%; P = 1.00 for heterogeneity), and rectal (RR: 1.01, 95% CI: 0.98–1.03, P = 0.51, I2 = 61.3%; P = 0.02 for heterogeneity) cancers (Table 1 and Supplementary Fig. 2). Eleven studies were included in the non-linear dose-response analysis [13, 15, 18, 19, 21, 22, 24, 25, 28–30], revealing no evidence of a non-linear association for colorectal (P = 0.39 for non-linearity), colon (P = 0.70 for non-linearity), and rectal (P = 0.64 for non-linearity) cancers in relation to dietary n-6 PUFAs intake (Fig. 2).
Table 1.
Summary risk estimates for association between dietary intake (or tissue levels) of omega-6 fatty acids and the risk of colorectal, colon, and rectal cancer in adultsa.
| No of effect sizes | Pooled relative risk (95% CI)b | P | I2 (%)c | P-heterogeneityd | |
|---|---|---|---|---|---|
| Dietary intake of n-6 PUFA | |||||
| The highest versus lowest comparison | |||||
| Colorectal cancer | |||||
| Overall | 13 | 1.05 (0.94–1.17) | 0.38 | 48.9 | 0.02 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 3 | 1.06 (0.85–1.31) | 0.62 | 61.7 | 0.07 |
| Women | 6 | 1.07 (0.88–1.31) | 0.48 | 58.2 | 0.04 |
| Men and women | 4 | 1.03 (0.84–1.26) | 0.78 | 46.0 | 0.14 |
| Study location | |||||
| US | 6 | 1.01 (0.85–1.20) | 0.93 | 59.9 | 0.03 |
| Non-US countries | 7 | 1.10 (0.95–1.27) | 0.21 | 38.5 | 0.14 |
| Study qualitye | |||||
| High quality | 11 | 1.07 (0.95–1.20) | 0.30 | 49.5 | 0.03 |
| Low quality | 2 | 0.97 (0.68–1.40) | 0.88 | 69.4 | 0.07 |
| Colon cancer | |||||
| Overall | 8 | 1.02 (0.91–1.15) | 0.71 | 0 | 0.88 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 3 | 1.06 (0.89–1.26) | 0.50 | 0 | 0.47 |
| Women | 3 | 0.95 (0.78–1.16) | 0.62 | 0 | 0.72 |
| Men and women | 2 | 1.10 (0.79–1.52) | 0.58 | 0 | 0.84 |
| Study location | |||||
| US | 2 | 1.02 (0.85–1.23) | 0.84 | 0 | 0.35 |
| Non-US countries | 6 | 1.03 (0.88–1.20) | 0.75 | 0 | 0.83 |
| Rectal cancer | |||||
| Overall | 8 | 1.19 (0.93–1.52) | 0.18 | 47.6 | 0.06 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 3 | 1.20 (0.94–1.53) | 0.14 | 0 | 0.42 |
| Women | 3 | 1.06 (0.55–2.05) | 0.86 | 75.8 | 0.02 |
| Men and women | 2 | 1.41 (0.82–2.42) | 0.22 | 44.7 | 0.18 |
| Study location | |||||
| US | 2 | 0.98 (0.55–1.75) | 0.96 | 65.2 | 0.09 |
| Non-US countries | 6 | 1.27 (0.96–1.68) | 0.10 | 44.2 | 0.11 |
| Linear dose-response association (per 1 g/day increase) | |||||
| Colorectal cancer | |||||
| Overall | 11 | 1.00 (0.99–1.01) | 0.97 | 41.4 | 0.07 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 2 | 1.00 (0.98–1.02) | 1.00 | 70.7 | 0.07 |
| Women | 6 | 1.01 (0.99–1.02) | 0.45 | 59.4 | 0.03 |
| Men and women | 3 | 1.00 (0.99–1.01) | 0.42 | 0 | 0.59 |
| Study location | |||||
| US | 6 | 1.00 (0.99–1.01) | 0.86 | 53.7 | 0.06 |
| Non-US countries | 5 | 1.00 (0.99–1.01) | 0.87 | 36.1 | 0.18 |
| Study qualitye | |||||
| High quality | 9 | 1.00 (0.99–1.01) | 0.91 | 45.1 | 0.07 |
| Low quality | 2 | 1.00 (0.98–1.02) | 0.91 | 59.7 | 0.12 |
| Colon cancer | |||||
| Overall | 7 | 1.00 (0.99–1.01) | 0.88 | 0 | 1.00 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 2 | 1.00 (0.99–1.01) | 1.00 | 0 | 1.00 |
| Women | 3 | 1.00 (0.99–1.01) | 0.99 | 0 | 0.97 |
| Men and women | 2 | 1.01 (0.98–1.03) | 0.58 | 0 | 0.72 |
| Study location | |||||
| US | 2 | 1.00 (0.99–1.01) | 1.00 | 0 | 1.00 |
| Non-US countries | 5 | 1.00 (0.99–1.02) | 0.72 | 0 | 0.99 |
| Rectal cancer | |||||
| Overall | 7 | 1.01 (0.98–1.03) | 0.51 | 61.3 | 0.02 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 2 | 1.00 (0.98–1.03) | 0.67 | 0 | 0.41 |
| Women | 3 | 1.03 (0.95–1.12) | 0.43 | 80.1 | 0.01 |
| Men and women | 2 | 1.02 (0.97–1.07) | 0.49 | 66.0 | 0.09 |
| Study location | |||||
| US | 2 | 1.00 (0.98–1.02) | 0.84 | 0.70 | 0.32 |
| Non-US countries | 5 | 1.02 (0.98–1.06) | 0.42 | 71.4 | 0.01 |
| Tissue levels of n-6 PUFA | |||||
| The highest versus lowest comparison | |||||
| Colorectal cancer | |||||
| Overall | 5 | 0.94 (0.75–1.19) | 0.62 | 0 | 0.58 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 2 | 0.65 (0.40–1.07) | 0.09 | 0 | 0.86 |
| Women | 1 | 1.15 (0.48–2.75) | 0.75 | 0 | <0.001 |
| Men and women | 2 | 1.03 (0.79–1.35) | 0.82 | 0 | 0.81 |
| Study location | |||||
| US | 1 | 0.63 (0.34–1.17) | 0.14 | 0 | <0.001 |
| Non-US countries | 4 | 1.01 (0.79–1.29) | 0.96 | 0 | 0.81 |
| Study qualitye | |||||
| High quality | 2 | 1.03 (0.79–1.35) | 0.82 | 0 | 0.81 |
| Low quality | 3 | 0.75 (0.49–1.15) | 0.19 | 0 | 0.53 |
aAbbreviation: n-6 PUFA omega-6 polyunsaturated fatty acids, CI confidence interval, US United States.
bObtained from the random effects model.
cInconsistency- the percentage of variation across studies due to heterogeneity.
dObtained from the Q-test.
eStudies with a median score of 7 or more, based on the NOS, were considered as a high-quality study.
Fig. 2. Non-linear dose-response association between dietary omega-6 fatty acids and colorectal cancer in adults.
A Colorectal cancer, B colon cancer, and C rectal cancer. The solid lines indicate the spline model. The dashed lines present the 95% CI. CRC colorectal cancer, CC colon cancer, RC rectal cancer, RR relative risk.
Tissue n-6 PUFAs
Four studies assessed the relationship between tissue levels of n-6 PUFAs and CRC risk [16, 17, 20, 23], with a total of 7453 participants and 1811 cases of CRC. There was no significant association between tissue levels of n-6 PUFAs and risk of CRC (RR: 0.94, 95% CI: 0.75–1.19, P = 0.62, I2 = 0%, P = 0.58 for heterogeneity) (Table 1 and Supplementary Fig. 3). The number of studies for n-6 PUFAs levels and colon/rectal cancer was not sufficient for a meta-analysis. Also, due to the limited number of studies, performing the dose-response analysis was not possible for all outcomes.
Meta-analysis on LA and risk of CRC
Dietary LA
Ten studies in 9 articles, with a total sample size of 438,873 participants, 5304 cases of CRC, 3485 cases of colon cancer, and 1468 cases of rectal cancer, were included [17, 22, 24, 26, 27, 29–32], for dietary LA analysis. RR for the risk of CRC, comparing the highest categories of dietary LA intake with the lowest categories, was 1.15 (95% CI: 1.05–1.27, P = 0.003, I2 = 0%, P = 0.44 for heterogeneity), indicating a significant positive association (Table 2 and Supplementary Fig. 4). Such positive association was seen for colon (RR: 1.10, 95% CI: 0.99–1.23, P = 0.09, I2 = 0%, P = 0.78 for heterogeneity) and rectal (RR: 1.30, 95% CI: 1.00–1.68, P = 0.05, I2 = 57.0%, P = 0.03 for heterogeneity) cancers. However, these associations were marginally significant. Nine studies in 8 articles were eligible for the linear dose-response analysis [17, 22, 24, 27, 29–32], and there was no significant linear association between a 1 gr/day increase in dietary LA intake and risk of CRC (RR: 1.01, 95% CI: 0.99–1.03, P = 0.12, I2 = 42.1%; P = 0.11 for heterogeneity) and rectal cancer (RR: 1.03, 95% CI: 1.00–1.06, P = 0.09, I2 = 76.0; P = 0.001 for heterogeneity). In contrast, a 1 gr/day increase in dietary LA intake was associated with a 1% increased risk of colon cancer (RR: 1.01, 95% CI: 1.00–1.02, P = 0.02, I2 = 0%; P = 0.60 for heterogeneity) (Table 2 and Supplementary Fig. 5). Eight studies in 7 articles were included in the non-linear dose-response analysis [22, 24, 27, 29–32], revealing no evidence of non-linearity for the associations between dietary LA intake and risk of colorectal (P = 0.88 for non-linearity), colon (P = 0.44 for non-linearity) and rectal (P = 0.41 for non-linearity) cancers (Fig. 3).
Table 2.
Summary risk estimates for association between dietary intake (or tissue levels) of linoleic acid and the risk of colorectal, colon, and rectal cancer in adultsa.
| No of effect sizes | Pooled relative risk (95% CI)b | P | I2 (%)c | P-heterogeneityd | |
|---|---|---|---|---|---|
| Dietary intake of LA | |||||
| The highest versus lowest comparison | |||||
| Colorectal cancer | |||||
| Overall | 9 | 1.15 (1.05–1.27) | 0.003 | 0 | 0.44 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 3 | 1.20 (1.03–1.38) | 0.02 | 0 | 0.93 |
| Women | 5 | 1.11 (0.94–1.32) | 0.22 | 26.9 | 0.24 |
| Men and women | 1 | 1.41 (0.99–2.00) | 0.05 | 0 | <0.001 |
| Study location | |||||
| US | 3 | 1.10 (0.90–1.34) | 0.36 | 43.1 | 0.17 |
| Non-US countries | 6 | 1.23 (1.08–1.40) | 0.002 | 0 | 0.77 |
| Colon cancer | |||||
| Overall | 7 | 1.10 (0.99–1.23) | 0.09 | 0 | 0.78 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 2 | 1.15 (0.95–1.38) | 0.16 | 0 | 0.82 |
| Women | 3 | 1.01 (0.84–1.20) | 0.95 | 0 | 0.73 |
| Men and women | 2 | 1.23 (0.97–1.56) | 0.08 | 0 | 0.52 |
| Study location | |||||
| US | 2 | 1.09 (0.92–1.29) | 0.34 | 0 | 0.47 |
| Non-US countries | 5 | 1.12 (0.96–1.30) | 0.15 | 0 | 0.62 |
| Study qualitye | |||||
| High quality | 6 | 1.07 (0.95–1.21) | 0.26 | 0 | 0.88 |
| Low quality | 1 | 1.31 (0.97–1.77) | 0.08 | 0 | <0.001 |
| Rectal cancer | |||||
| Overall | 7 | 1.30 (1.00–1.68) | 0.05 | 57.0 | 0.03 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 2 | 1.29 (0.98–1.70) | 0.07 | 0 | 0.72 |
| Women | 3 | 1.30 (0.66–2.55) | 0.45 | 81.1 | 0.01 |
| Men and women | 2 | 1.36 (0.79–2.34) | 0.27 | 66.2 | 0.09 |
| Study location | |||||
| US | 2 | 0.92 (0.55–1.56) | 0.77 | 63.6 | 0.10 |
| Non-US countries | 5 | 1.48 (1.17–1.86) | 0.001 | 24.2 | 0.26 |
| Study qualitye | |||||
| High quality | 6 | 1.35 (1.00–1.82) | 0.05 | 61.2 | 0.02 |
| Low quality | 1 | 1.03 (0.66–1.61) | 0.90 | 0 | <0.001 |
| Linear dose-response association (per 1 g/day increase) | |||||
| Colorectal cancer | |||||
| Overall | 7 | 1.01 (0.99–1.03) | 0.12 | 42.1 | 0.11 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 2 | 1.01 (0.99–1.03) | 0.44 | 44.9 | 0.18 |
| Women | 5 | 1.02 (0.99–1.05) | 0.16 | 52.8 | 0.08 |
| Study location | |||||
| US | 3 | 1.01 (0.99–1.02) | 0.51 | 46.1 | 0.16 |
| Non-US countries | 4 | 1.02 (0.99–1.05) | 0.12 | 30.6 | 0.23 |
| Colon cancer | |||||
| Overall | 6 | 1.01 (1.00–1.02) | 0.02 | 0 | 0.60 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 1 | 1.01 (0.99–1.02) | 0.19 | 0 | <0.001 |
| Women | 3 | 1.00 (0.97–1.03) | 0.86 | 35.8 | 0.21 |
| Men and women | 2 | 1.01 (1.00–1.02) | 0.04 | 0 | 1.00 |
| Study location | |||||
| US | 2 | 1.01 (1.00–1.02) | 0.10 | 0 | 1.00 |
| Non-US countries | 4 | 1.01 (0.99–1.02) | 0.41 | 16.2 | 0.31 |
| Study qualitye | |||||
| High quality | 5 | 1.01 (1.00–1.02) | 0.18 | 0 | 0.48 |
| Low quality | 1 | 1.01 (1.00–1.02) | 0.05 | 0 | <0.001 |
| Rectal cancer | |||||
| Overall | 6 | 1.03 (1.00–1.06) | 0.09 | 76.0 | 0.001 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 1 | 1.01 (0.98–1.04) | 0.51 | 0 | <0.001 |
| Women | 3 | 1.09 (0.97–1.22) | 0.15 | 86.5 | 0.001 |
| Men and women | 2 | 1.02 (0.98–1.07) | 0.36 | 83.3 | 0.01 |
| Study location | |||||
| US | 2 | 1.00 (0.98–1.02) | 0.99 | 0 | 0.36 |
| Non-US countries | 4 | 1.06 (1.00–1.12) | 0.04 | 83.4 | <0.001 |
| Study qualitye | |||||
| High quality | 5 | 1.04 (1.00–1.09) | 0.07 | 78.3 | 0.001 |
| Low quality | 1 | 1.00 (0.98–1.02) | 1.00 | 0 | <0.001 |
| Tissue levels of LA | |||||
| The highest versus lowest comparison | |||||
| Colorectal cancer | |||||
| Overall | 5 | 0.93 (0.61–1.41) | 0.74 | 69.0 | 0.01 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 1 | 0.57 (0.24–1.37) | 0.21 | 0 | <0.001 |
| Women | 1 | 1.88 (0.78–4.53) | 0.16 | 0 | <0.001 |
| Men and women | 3 | 0.90 (0.55–1.46) | 0.67 | 78.4 | 0.01 |
| Study qualitye | |||||
| High quality | 3 | 0.90 (0.55–1.46) | 0.67 | 78.4 | 0.01 |
| Low quality | 2 | 1.03 (0.32–3.33) | 0.96 | 71.9 | 0.06 |
aAbbreviation: LA linoleic acid, CI confidence interval, US United States.
bObtained from the random effects model.
cInconsistency- the percentage of variation across studies due to heterogeneity.
dObtained from the Q-test.
eStudies with a median score of 7 or more, based on the NOS, were considered as a high-quality study.
Fig. 3. Non-linear dose-response association between dietary linoleic acid and colorectal cancer in adults.
A Colorectal cancer, B colon cancer, and C rectal cancer. The solid lines indicate the spline model. The dashed lines present the 95% CI. LA linoleic acid, CRC colorectal cancer, CC colon cancer, RC rectal cancer, RR relative risk.
Tissue LA
Five studies in 4 papers assessed the relationship between tissue levels of LA and CRC risk [14, 17, 20, 23], with a total of 7693 participants and recorded 1983 cases of CRC, 211 cases of colon cancer, and 139 cases of rectal cancer. Combining the results from these articles, we found no significant association between tissue levels of LA and CRC risk (RR: 0.93, 95% CI: 0.61–1.41, P = 0.74, I2 = 69.0%, P = 0.01 for heterogeneity) (Table 2 and Supplementary Fig. 6). Due to the limited number of studies, we were not able to perform a meta-analysis for colon/rectal cancer and also the dose-response analysis.
Meta-analysis on AA and risk of CRC
Dietary AA
Seven studies, with a total sample size of 346,743 participants and 4659 CRC cases, 2746 colon cancer cases, and 1149 rectal cancer cases, evaluated the association between dietary AA intake and CRC risk [17, 22, 24, 26, 29, 30]. Comparing the highest and lowest categories of dietary AA intake, RR for CRC was 0.97 (95% CI: 0.86–1.10, P = 0.66, I2 = 25.7%; P = 0.23 for heterogeneity), indicating a non-significant association between dietary AA intake and CRC risk (Table 3 and Supplementary Fig. 7). Such non-significant association was also observed for colon (RR: 0.97, 95% CI: 0.84–1.10, P = 0.61, I2 = 0.90%; P = 0.40 for heterogeneity) and rectal (RR: 0.89, 95% CI: 0.74–1.08, P = 0.24, I2 = 0%; P = 0.74 for heterogeneity) cancers. Six studies were eligible for the linear dose-response analysis [17, 22, 24, 29, 30], and there was no linear association between a 100 mg/day increase in dietary AA intake and the risk of colorectal (RR: 1.02, 95% CI: 0.94–1.11, P = 0.59, I2 = 53.3%; P = 0.07 for heterogeneity), colon (RR: 1.01, 95% CI: 0.94–1.09, P = 0.76, I2 = 23.5%; P = 0.27 for heterogeneity), and rectal (RR: 0.98, 95% CI: 0.93–1.04, P = 0.52, I2 = 0%; P = 0.80 for heterogeneity) cancers (Table 3 and Supplementary Fig. 8). The non-linear dose-response also showed no evidence of non-linearity for colorectal (P = 0.09 for non-linearity), colon cancer (P = 0.19 for non-linearity), and rectal (P = 0.76 for non-linearity) cancers in relation to dietary AA intake [22, 24, 29, 30] (Fig. 4).
Table 3.
Summary risk estimates for association between dietary intake (or tissue levels) of arachidonic acid and the risk of colorectal, colon, and rectal cancer in adultsa.
| No of effect sizes | Pooled relative risk (95% CI)b | P | I2 (%)c | P-heterogeneityd | |
|---|---|---|---|---|---|
| Dietary intake of AA | |||||
| The highest versus lowest comparison | |||||
| Colorectal cancer | |||||
| Overall | 7 | 0.97 (0.86–1.10) | 0.66 | 25.7 | 0.23 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 2 | 0.97 (0.81–1.15) | 0.69 | 24.5 | 0.25 |
| Women | 4 | 0.96 (0.76–1.20) | 0.71 | 49.7 | 0.11 |
| Men and women | 1 | 1.12 (0.79–1.58) | 0.52 | 0 | <0.001 |
| Study location | |||||
| US | 3 | 0.97 (0.84–1.13) | 0.71 | 0 | 0.57 |
| Non-US countries | 4 | 1.00 (0.80–1.25) | 0.98 | 56.8 | 0.07 |
| Colon cancer | |||||
| Overall | 5 | 0.97 (0.84–1.10) | 0.61 | 0.90 | 0.40 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 2 | 1.05 (0.82–1.36) | 0.68 | 46.7 | 0.17 |
| Women | 2 | 0.86 (0.69–1.07) | 0.17 | 0 | 0.62 |
| Men and women | 1 | 0.98 (0.66–1.45) | 0.92 | 0 | <0.001 |
| Study location | |||||
| US | 2 | 1.04 (0.77–1.40) | 0.82 | 59.2 | 0.12 |
| Non-US countries | 3 | 0.91 (0.75–1.09) | 0.30 | 0 | 0.71 |
| Rectal cancer | |||||
| Overall | 5 | 0.89 (0.74–1.08) | 0.24 | 0 | 0.74 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 2 | 0.83 (0.64–1.08) | 0.17 | 0 | 0.90 |
| Women | 2 | 0.86 (0.61–1.22) | 0.41 | 0 | 0.60 |
| Men and women | 1 | 1.18 (0.74–1.88) | 0.49 | 0 | <0.001 |
| Study location | |||||
| US | 2 | 0.87 (0.62–1.23) | 0.43 | 0 | 0.65 |
| Non-US countries | 3 | 0.90 (0.71–1.14) | 0.38 | 0 | 0.42 |
| Linear dose-response association (per 100 mg/day increase) | |||||
| Colorectal cancer | |||||
| Overall | 5 | 1.02 (0.94–1.11) | 0.59 | 53.3 | 0.07 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 1 | 1.10 (0.94–1.29) | 0.25 | 0 | <0.001 |
| Women | 4 | 1.01 (0.92–1.11) | 0.84 | 56.1 | 0.08 |
| Study location | |||||
| US | 3 | 1.02 (0.94–1.11) | 0.57 | 0 | 0.37 |
| Non-US countries | 2 | 1.19 (0.76–1.86) | 0.45 | 82.7 | 0.02 |
| Colon cancer | |||||
| Overall | 4 | 1.01 (0.94–1.09) | 0.76 | 23.5 | 0.27 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 1 | 1.21 (0.98–1.49) | 0.08 | 0 | <0.001 |
| Women | 2 | 0.98 (0.94–1.03) | 0.48 | 0 | 0.56 |
| Men and women | 1 | 0.95 (0.43–2.08) | 0.90 | 0 | <0.001 |
| Study location | |||||
| US | 2 | 1.09 (0.92–1.28) | 0.31 | 46.5 | 0.17 |
| Non-US countries | 2 | 0.98 (0.94–1.03) | 0.38 | 0 | 0.94 |
| Rectal cancer | |||||
| Overall | 4 | 0.98 (0.93–1.04) | 0.52 | 0 | 0.80 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 1 | 0.91 (0.65–1.27) | 0.58 | 0 | <0.001 |
| Women | 2 | 0.98 (0.93–1.04) | 0.55 | 0 | 0.60 |
| Men and women | 1 | 1.38 (0.54–3.50) | 0.50 | 0 | <0.001 |
| Study location | |||||
| US | 2 | 1.00 (0.82–1.22) | 0.98 | 0 | 0.50 |
| Non-US countries | 2 | 0.98 (0.93–1.04) | 0.51 | 0 | 0.47 |
| Tissue levels of AA | |||||
| The highest versus lowest comparison | |||||
| Colorectal cancer | |||||
| Overall | 5 | 0.97 (0.70–1.33) | 0.83 | 45.8 | 0.12 |
| Subgroup analysis | |||||
| Sex | |||||
| Men | 1 | 1.16 (0.49–2.75) | 0.74 | 0 | <0.001 |
| Women | 1 | 0.65 (0.30–1.42) | 0.28 | 0 | <0.001 |
| Men and women | 3 | 1.01 (0.66–1.55) | 0.96 | 68.1 | 0.04 |
| Study qualitye | |||||
| High quality | 3 | 1.01 (0.66–1.55) | 0.96 | 68.1 | 0.04 |
| Low quality | 2 | 0.84 (0.47–1.51) | 0.57 | 0 | 0.33 |
| Colon cancer | |||||
| Overall | 2 | 1.42 (1.00–2.03) | 0.05 | 0 | 0.80 |
| Rectal cancer | |||||
| Overall | 2 | 1.35 (0.83–2.18) | 0.23 | 0 | 0.84 |
aAbbreviation: AA arachidonic acid, CI confidence interval, US United States.
bObtained from the random effects model.
cInconsistency- the percentage of variation across studies due to heterogeneity.
dObtained from the Q-test.
eStudies with a median score of 7 or more, based on the NOS, were considered as a high-quality study.
Fig. 4. Non-linear dose-response association between dietary arachidonic acid and colorectal cancer in adults.
A Colorectal cancer, B colon cancer, and C rectal cancer. The solid lines indicate the spline model. The dashed lines present the 95% CI. AA arachidonic acid, CRC colorectal cancer, CC colon cancer, RC rectal cancer, RR relative risk.
Tissue AA
Five studies assessed the relationship between tissue levels of AA and CRC risk [14, 17, 20, 23], with a total of 7693 participants, of which 1983 cases of CRC, 881 cases of colon cancer, and cases of 538 rectal cancer were recorded. Combining the results from these studies revealed no significant association between tissue levels of AA and risks of colorectal (RR: 0.97, 95% CI: 0.70–1.33, P = 0.83, I2 = 45.8%, P = 0.12 for heterogeneity) and rectal (RR: 1.35, 95% CI: 0.83–2.18, P = 0.23, I2 = 0%, P = 0.84 for heterogeneity) cancers (Table 3 and Supplementary Fig. 9). However, we found a marginally significant positive association between tissue levels of AA and colon cancer risk (RR: 1.42, 95% CI: 1.00–2.03, P = 0.05, I2 = 0%, P = 0.80 for heterogeneity). Due to the limited number of studies, we did not perform a dose-response analysis.
Subgroup and sensitivity analyses, and publication bias
Tables 1, 2, and 3 show findings from the subgroup analyses of dietary intake/tissue levels of omega-6 (total, LA, and AA) and risk of CRC and its subtypes. According to these analyses, no significant association was found between dietary intake/tissue levels of n-6 PUFAs and the risk of CRC, colon and rectal cancer in all subgroups (Table 1). However, a significant positive association was seen between dietary LA intake (for the highest versus lowest comparison) and CRC risk in men and non-US population. For rectal cancer, a significant positive association was seen with dietary LA intake in subgroups of non-US populations and high-quality studies (Table 2). For dietary intake/tissue levels of AA, we found no significant association in any subgroups (Table 3). The sensitivity analysis showed that the exclusion of any single study from the analysis did not change the pooled effect sizes significantly. According to the visual inspection of funnel plots and both Begg’s test and Egger’s regression, there was no significant publication bias in the associations evaluated (P > 0.10).
Discussion
This review is, to our knowledge, the first dose-response meta-analysis exploring the associations between dietary/tissue biomarkers of n-6 PUFAs and risk of CRC. Our analysis revealed that higher intake of LA was associated with an increased risk of colorectal and rectal cancers. Also, in the dose-response analysis, each 1 gr/day increase in dietary LA intake was associated with a 1% higher risk of colon cancer. There was no evidence of any associations with tissue levels of LA, or dietary intake/tissue levels of n-6 PUFAs and AA.
CRC is one of the most common gastrointestinal cancers [1], with dietary factors playing an important role in its pathogenesis [50]. Of different dietary factors, n-6 PUFAs have received much attention due to their roles in inflammatory responses through the production of inflammatory prostaglandins [12]. However, findings on the association between dietary intake and tissue levels of n-6 PUFAs and CRC risk are conflicting [14, 17, 20, 31]. In this meta-analysis, we found a significant positive association between dietary LA intake and colorectal, colon, and rectal cancers. Contrary to our finding, a recent meta-analysis, conducted by Lu et al., showed no significant association between dietary LA intake and risk of CRC [33]. While this could be due to missing out some eligible studies [23, 32], Lu et al. also combined the risk estimates from cohort studies with those from case-control studies. However, our meta-analysis included only prospective studies, avoiding recall or selection biases, which could be the subject of concern in case-control studies. Similar to our findings, experimental studies have shown that high-LA and high-glucose diets increase the levels of advanced glycation end products (AGE) and the receptor of advanced glycation end products (RAGE), which are associated with CRC progression [51]. High-fat corn oil was also shown to promote colon tumorigenesis by up-regulating the cyclooxygenase-2 expression [52].
In the current meta-analysis, we found no significant association between tissue levels of LA and CRC risk. Nevertheless, a positive association was seen for dietary LA intake. The disparity might be due to LA changes during food cooking or processing. On the other hand, oils high in LA, when exposed to food processing methods such as frying and high-heat cooking, can undergo oxidation, leading to the formation of harmful compounds like lipid peroxides and aldehydes [53], which may be associated with an increased risk of CRC [54]. Moreover, a higher intake of LA, which is commonly found in vegetable oils, can lead to increased energy consumption and contribute to obesity [55], a known risk factor for CRC [56, 57]. Despite this, evidence suggests anti-cancer properties of LA levels in tissues or blood [14], indicating that dietary LA might be associated with CRC independently of tissue LA levels. Similar to our findings, Lu et al. also reported that tissue levels of LA were not associated with CRC risk [33]. Also, it should be noted that the amount of LA in foods is low and therefore the estimation of its intake might be affected by measurement error. In addition, the positive association between dietary LA and CRC might be due to the effect of confounding variables such as other dietary factors rather than LA. Therefore, our findings on the relation between LA and CRC should be considered with caution, warranting further studies.
We found that dietary intakes of total n-6 PUFAs and AA were not associated with risk of colorectal, colon, and rectal cancers. Similarly, Lu et al. also showed no significant association between dietary intakes of total n-6 PUFAs and AA and risk of colorectal, colon, and rectal cancers [33]. The observed disparity between dietary intake of n-6 PUFAs and LA might be explained by the different effects of n-6 PUFAs (LA and long-chain n-6 PUFAs) on cancer incidence. Therefore, considering the different effects of n-6 PUFAs on cancer etiology, the cumulative effects of these fatty acids on cancer incidence might be null. For dietary intake of AA, the observed null connection may be due to AA metabolites that are produced by different pathways. Lipoxin A4 (LXA4) is produced by the lipoxygenase pathway and is considered as a tumor growth suppressor due to its anti-angiogenic properties [58]. However, PGE2 produced via the cyclooxygenase pathway plays an important role in the development of CRC [59]. Therefore, a combination of AA metabolites with different health benefits might justify the null association between dietary AA intake and risk of CRC.
Our findings should be interpreted with caution due to several limitations, most of which are common to observational studies and meta-analyses. These include being unable to perform a dose-response meta-analysis due to the limited number of studies in some associations. In addition, since the current meta-analysis was conducted on observational studies, causality cannot be established. Also, the existence of measurement and reporting errors in the estimation of food and nutrient intake is inevitable in observational studies. Moreover, previous studies have not examined the influence of processing and cooking methods or the sources of n-6 PUFAs on the association between these fatty acids and CRC risk. The source of n-6 PUFAs, whether it is derived from plant-based foods or from fast foods and snacks, may have different effects. Therefore, future studies should assess the influence of different food sources of n-6 PUFAs on CRC risk. In addition, differences in n-6 PUFAs intake among geographical regions could have affected the highest and lowest levels of n-6 PUFAs intake and the results from these comparisons. To control these differences, we conducted the subgroup and dose-response analyses.
In conclusion, we found that dietary intake of LA is associated with an increased risk of colorectal, colon, and rectal cancers. However, no significant association was found for total n-6 PUFAs and AA, either in the highest versus lowest comparison or in the dose-response meta-analysis. Also, in the dose-response analysis, each 1 gr/day increase in dietary LA intake (equal to 3.67 grams of sunflower oil, 1.6 grams of walnut oil, or 2.35 grams of pumpkin oil [60]) was associated with a 1% higher risk of colon cancer. Although the increased risk of 1% with a 1 gr/day more intake of LA might be clinically unimportant, this risk might be large in the higher dosages of LA intake. Among the studies included in the current meta-analysis, the range of LA intake was between 0 and 20 gr/day. Therefore, higher intakes of LA can provide higher risk of colon cancer. In terms of tissue levels of n-6 PUFAs, LA, and AA, no significant association was found with CRC risk. Future studies, particularly well-designed prospective cohort studies, should assess the influence of dietary/tissue levels of n-6 PUFAs on the risk of CRC mortality. Further studies are also needed to investigate whether specific foods rich in n-6 PUFAs are differentially associated with CRC risk.
Supplementary information
Author contributions
NA contributed to the literature search. NA and NE contributed to data extraction and data analysis. NA and OS drafted the manuscript which was critically revised for important intellectual content by all authors. MA contributed to the manuscript editing and obtained funding. JAS and GA contributed to the manuscript editing. OS supervised the study. All authors have read and approved the final manuscript.
Funding
The study was financially supported by Abadan University of Medical Sciences, Abadan, Iran (code: 1746). The funder had no role in the design and conduct of the study; collection of the data; preparation, review, or approval of the manuscript; or the decision to submit the manuscript for publication.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Competing interests
The authors declare no competing interests.
Ethics approval and consent to participate
All analyses were based on previously published studies; thus, no ethical approval is required.
Footnotes
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Masoomeh Asadi, Email: Asadi.masoomeh@ymail.com.
Omid Sadeghi, Email: omidsadeghi69@yahoo.com.
Supplementary information
The online version contains supplementary material available at 10.1038/s41387-025-00367-w.
References
- 1.Dekker E, Tanis PJ, Vleugels JLA, Kasi PM, Wallace MB. Colorectal cancer. Lancet. 2019;394:1467–80. [DOI] [PubMed] [Google Scholar]
- 2.Arnold M, Sierra MS, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global patterns and trends in colorectal cancer incidence and mortality. Gut. 2017;66:683–91. [DOI] [PubMed] [Google Scholar]
- 3.Ionescu VA, Gheorghe G, Bacalbasa N, Chiotoroiu AL, Diaconu C. Colorectal cancer: from risk factors to oncogenesis. Medicina (Kaunas). 2023;59:1646–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Mehta, Song RS, Nishihara M, Drew DA R, Wu K, Qian ZR, et al. Dietary patterns and risk of colorectal cancer: analysis by tumor location and molecular subtypes. Gastroenterology. 2017;152:1944–53.e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bamia C, Lagiou P, Buckland G, Grioni S, Agnoli C, Taylor AJ, et al. Mediterranean diet and colorectal cancer risk: results from a European cohort. Eur J Epidemiol. 2013;28:317–28. [DOI] [PubMed] [Google Scholar]
- 6.Fung TT, Hu FB, Wu K, Chiuve SE, Fuchs CS, Giovannucci E. The mediterranean and dietary approaches to stop hypertension (DASH) diets and colorectal cancer. Am J Clin Nutr. 2010;92:1429–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Spector AA, Kim HY. Discovery of essential fatty acids. J Lipid Res. 2015;56:11–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Smith WL, DeWitt DL, Garavito RM. Cyclooxygenases: structural, cellular, and molecular biology. Annu Rev Biochem. 2000;69:145–82. [DOI] [PubMed] [Google Scholar]
- 9.Romagnolo DF, Donovan MG, Doetschman TC, Selmin OI. n-6 linoleic acid induces epigenetics alterations associated with colonic inflammation and cancer. Nutrients. 2019;11:171–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Su H, Liu R, Chang M, Huang J, Wang X. Dietary linoleic acid intake and blood inflammatory markers: a systematic review and meta-analysis of randomized controlled trials. Food Funct. 2017;8:3091–103. [DOI] [PubMed] [Google Scholar]
- 11.Johnson GH, Fritsche K. Effect of dietary linoleic acid on markers of inflammation in healthy persons: a systematic review of randomized controlled trials. J Acad Nutr Diet. 2012;112:1029–41, 41.e1-15. [DOI] [PubMed] [Google Scholar]
- 12.Kumar NG, Contaifer D, Madurantakam P, Carbone S, Price ET, Tassell BV, et al. Dietary bioactive fatty acids as modulators of immune function: implications on human health. Nutrients. 2019;11:2974. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Butler LM, Wang R, Koh WP, Stern MC, Yuan JM, Yu MC. Marine n-3 and saturated fatty acids in relation to risk of colorectal cancer in Singapore Chinese: a prospective study. Int J Cancer. 2009;124:678–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Butler LM, Yuan JM, Huang JY, Su J, Wang R, Koh WP, et al. Plasma fatty acids and risk of colon and rectal cancers in the Singapore Chinese Health Study. NPJ Precis Oncol. 2017;1:38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Daniel CR, McCullough ML, Patel RC, Jacobs EJ, Flanders WD, Thun MJ, et al. Dietary intake of ω-6 and ω-3 fatty acids and risk of colorectal cancer in a prospective cohort of US men and women. Cancer Epidemiol Biomarkers Prev. 2009;18:516–25. [DOI] [PubMed] [Google Scholar]
- 16.Hall MN, Campos H, Li H, Sesso HD, Stampfer MJ, Willett WC, et al. Blood levels of long-chain polyunsaturated fatty acids, aspirin, and the risk of colorectal cancer. Cancer Epidemiol Biomarkers Prev. 2007;16:314–21. [DOI] [PubMed] [Google Scholar]
- 17.Hodge AM, Williamson EJ, Bassett JK, MacInnis RJ, Giles GG, English DR. Dietary and biomarker estimates of fatty acids and risk of colorectal cancer. Int J Cancer. 2015;137:1224–34. [DOI] [PubMed] [Google Scholar]
- 18.Key TJ, Appleby PN, Masset G, Brunner EJ, Cade JE, Greenwood DC, et al. Vitamins, minerals, essential fatty acids and colorectal cancer risk in the United Kingdom Dietary Cohort Consortium. Int J Cancer. 2012;131:E320–5. [DOI] [PubMed] [Google Scholar]
- 19.Koh WP, Yuan JM, van den Berg D, Lee HP, Yu MC. Interaction between cyclooxygenase-2 gene polymorphism and dietary n-6 polyunsaturated fatty acids on colon cancer risk: the Singapore Chinese Health Study. Br J Cancer. 2004;90:1760–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kojima M, Wakai K, Tokudome S, Suzuki K, Tamakoshi K, Watanabe Y, et al. Serum levels of polyunsaturated fatty acids and risk of colorectal cancer: a prospective study. Am J Epidemiol. 2005;161:462–71. [DOI] [PubMed] [Google Scholar]
- 21.Kraja B, Muka T, Ruiter R, de Keyser CE, Hofman A, Franco OH, et al. Dietary fiber intake modifies the positive association between n–3 PUFA intake and colorectal cancer risk in a caucasian population. J Nutr. 2015;145:1709–16. [DOI] [PubMed] [Google Scholar]
- 22.Lin J, Zhang SM, Cook NR, Lee IM, Buring JE. Dietary fat and fatty acids and risk of colorectal cancer in women. Am J Epidemiol. 2004;160:1011–22. [DOI] [PubMed] [Google Scholar]
- 23.Linseisen J, Grundmann N, Zoller D, Kühn T, Jansen EH, Chajès V, et al. Red blood cell fatty acids and risk of colorectal cancer in the European prospective investigation into cancer and nutrition (EPIC). Cancer Epidemiol Biomarkers Prev. 2021;30:874–85. [DOI] [PubMed] [Google Scholar]
- 24.Murff HJ, Shu XO, Li H, Dai Q, Kallianpur A, Yang G, et al. A prospective study of dietary polyunsaturated fatty acids and colorectal cancer risk in Chinese women. Cancer Epidemiol Biomarkers Prev. 2009;18:2283–91. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Navarro SL, Neuhouser ML, Cheng TYD, Tinker LF, Shikany JM, Snetselaar L, et al. The interaction between dietary fiber and fat and risk of colorectal cancer in the women’s health initiative. Nutrients. 2016;8:779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Nguyen S, Li H, Yu D, Cai H, Gao J, Gao Y, et al. Dietary fatty acids and colorectal cancer risk in men: a report from the Shanghai Men’s Health Study and a meta-analysis. Int J Cancer. 2021;148:77–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Pietinen P, Malila N, Virtanen M, Hartman TJ, Tangrea JA, Albanes D, et al. Diet and risk of colorectal cancer in a cohort of Finnish men. Cancer Causes Control. 1999;10:387–96. [DOI] [PubMed] [Google Scholar]
- 28.Sasazuki S, Inoue M, Iwasaki M, Sawada N, Shimazu T, Yamaji T, et al. Intake of n-3 and n-6 polyunsaturated fatty acids and development of colorectal cancer by subsite: Japan Public Health Center-based prospective study. Int J Cancer. 2011;129:1718–29. [DOI] [PubMed] [Google Scholar]
- 29.Shin A, Cho S, Sandin S, Lof M, Oh MY, Weiderpass E. Omega-3 and -6 fatty acid intake and colorectal cancer risk in Swedish Women’s Lifestyle and Health Cohort. Cancer Res Treat. 2020;52:848–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Song M, Chan AT, Fuchs CS, Ogino S, Hu FB, Mozaffarian D, et al. Dietary intake of fish, ω-3 and ω-6 fatty acids and risk of colorectal cancer: a prospective study in US men and women. Int J Cancer. 2014;135:2413–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Brink M, Weijenberg MP, De Goeij AF, Schouten LJ, Koedijk FD, Roemen GM, et al. Fat and K-ras mutations in sporadic colorectal cancer in The Netherlands Cohort Study. Carcinogenesis. 2004;25:1619–28. [DOI] [PubMed] [Google Scholar]
- 32.Terry P, Bergkvist L, Holmberg L, Wolk A. No association between fat and fatty acids intake and risk of colorectal cancer. Cancer Epidemiol Biomarkers Prev. 2001;10:913–4. [PubMed] [Google Scholar]
- 33.Lu Y, Li D, Wang L, Zhang H, Jiang F, Zhang R, et al. Comprehensive investigation on associations between dietary intake and blood levels of fatty acids and colorectal cancer risk. Nutrients. 2023;15:730. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. Int J Surg. 2021;88:105906. [DOI] [PubMed] [Google Scholar]
- 35.Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol. 2010;25:603–5. [DOI] [PubMed] [Google Scholar]
- 36.van Enst WA, Ochodo E, Scholten RJ, Hooft L, Leeflang MM. Investigation of publication bias in meta-analyses of diagnostic test accuracy: a meta-epidemiological study. BMC Med Res Methodol. 2014;14:70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Lin L. Hybrid test for publication bias in meta-analysis. Stat Methods Med Res. 2020;29:2881–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Greenland S, Longnecker MP. Methods for trend estimation from summarized dose-response data, with applications to meta-analysis. Am J Epidemiol. 1992;135:1301–9. [DOI] [PubMed] [Google Scholar]
- 39.Orsini N, Bellocco R, Greenland S. Generalized least squares for trend estimation of summarized dose–response data. stata journal. 2006;6:40–57. [Google Scholar]
- 40.Shao S, Kao TC, Eckhaus J, Bourgeois J, Perera K, Zhu K. The association of percentage energy from fat and colon cancer risk among members of the US military. Eur J Cancer Prev. 2015;24:188–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Tu K, Ma T, Zhou R, Xu L, Fang Y, Zhang C. Association between dietary fatty acid patterns and colorectal cancer risk: a large-scale case-control study in China. Nutrients. 2022;14:4375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Williams CD, Satia JA, Adair LS, Stevens J, Galanko J, Keku TO, et al. Associations of red meat, fat, and protein intake with distal colorectal cancer risk. Nutr Cancer. 2010;62:701–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Wu Q, Shi D, Dong T, Zhang Z, Ou Q, Fang Y, et al. Serum saturated fatty acids including very long-chain saturated fatty acids and colorectal cancer risk among Chinese Population. Nutrients. 2023;15:1917. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Sellem L, Srour B, Gueraud F, Pierre F, Kesse-Guyot E, Fiolet T, et al. Saturated, mono- and polyunsaturated fatty acid intake and cancer risk: results from the French prospective cohort NutriNet-Sante. Eur J Nutr. 2019;58:1515–27. [DOI] [PubMed] [Google Scholar]
- 45.Cottet V, Collin M, Gross AS, Boutron-Ruault MC, Morois S, Clavel-Chapelon F, et al. Erythrocyte membrane phospholipid fatty acid concentrations and risk of colorectal adenomas: a case-control nested in the French E3N-EPIC cohort study. Cancer Epidemiol Biomarkers Prev. 2013;22:1417–27. [DOI] [PubMed] [Google Scholar]
- 46.Oh K, Willett WC, Fuchs CS, Giovannucci E. Dietary marine n-3 fatty acids in relation to risk of distal colorectal adenoma in women. Cancer Epidemiol Biomarkers Prev. 2005;14:835–41. [DOI] [PubMed] [Google Scholar]
- 47.Weijenberg MP, Luchtenborg M, de Goeij AF, Brink M, van Muijen GN, de Bruïne AP, et al. Dietary fat and risk of colon and rectal cancer with aberrant MLH1 expression, APC or KRAS genes. Cancer Causes Control. 2007;18:865–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.Willett WC, Stampfer MJ, Colditz GA, Rosner BA, Speizer FE. Relation of meat, fat, and fiber intake to the risk of colon cancer in a prospective study among women. N Engl J Med. 1990;323:1664–72. [DOI] [PubMed] [Google Scholar]
- 49.Wan Y, Wu K, Wang L, Yin K, Song M, Giovannucci EL, et al. Dietary fat and fatty acids in relation to risk of colorectal cancer. Eur J Nutr. 2022;61:1863–73. [DOI] [PubMed] [Google Scholar]
- 50.Kato I, Majumdar AP, Land SJ, Barnholtz-Sloan JS, Severson RK. Dietary fatty acids, luminal modifiers, and risk of colorectal cancer. Int J Cancer. 2010;127:942–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Shimomoto T, Luo Y, Ohmori H, Chihara Y, Fujii K, Sasahira T, et al. Advanced glycation end products (AGE) induce the receptor for AGE in the colonic mucosa of azoxymethane-injected Fischer 344 rats fed with a high-linoleic acid and high-glucose diet. J Gastroenterol. 2012;47:1073–83. [DOI] [PubMed] [Google Scholar]
- 52.Singh J, Hamid R, Reddy BS. Dietary fat and colon cancer: modulation of cyclooxygenase-2 by types and amount of dietary fat during the postinitiation stage of colon carcinogenesis. Cancer Res. 1997;57:3465–70. [PubMed] [Google Scholar]
- 53.Choe E, Min DB. Mechanisms and factors for edible oil oxidation. Compr Rev Food Sci. 2006;5:169–86. [Google Scholar]
- 54.Guo C, Li X, Wang R, Yu J, Ye M, Mao L, et al. Association between oxidative DNA damage and risk of colorectal cancer: sensitive determination of urinary 8-Hydroxy-2’-deoxyguanosine by UPLC-MS/MS analysis. Sci Rep. 2016;6:32581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Naughton SS, Mathai ML, Hryciw DH, McAinch AJ. Linoleic acid and the pathogenesis of obesity. Prostaglandins Other Lipid Mediat. 2016;125:90–9. [DOI] [PubMed] [Google Scholar]
- 56.Miranda BCJ, Tustumi F, Nakamura ET, Shimanoe VH, Kikawa D, Waisberg J. Obesity and colorectal cancer: a narrative review. Medicina (Kaunas). 2024;60:1218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Ma Y, Yang Y, Wang F, Zhang P, Shi S, Zou Y, et al. Obesity and risk of colorectal cancer: a systematic review of prospective studies. PLoS ONE. 2013;8:e53916. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Chen Y, Hao H, He S, Cai L, Li Y, Hu S, et al. Lipoxin A4 and its analogue suppress the tumor growth of transplanted H22 in mice: the role of antiangiogenesis. Mol Cancer Ther. 2010;9:2164–74. [DOI] [PubMed] [Google Scholar]
- 59.Rifkin SB, Shrubsole MJ, Cai Q, Smalley WE, Ness RM, Swift LL, et al. PUFA levels in erythrocyte membrane phospholipids are differentially associated with colorectal adenoma risk. Br J Nutr. 2017;117:1615–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 60.Ivanova S, Marinova G, Batchvarov V. Comparison of fatty acid composition of various types of edible oils. Bulg J Agric Sci. 2016;22:5–15. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.