Erythrocyte ω-3 polyunsaturated fatty acids are inversely associated with the risk of oral cancer: a case-control study

Oral cancer is a common head and neck malignancy with rising global incidence, particularly in developing countries. Known risk factors include age, sex, smoking, alcohol consumption, HPV infection, and betel nut chewing, yet prior work from the study group suggested about half of oral cancer patients lacked these behavioral exposures, indicating additional etiologic factors. Interest has grown in lipid metabolism and carcinogenesis; erythrocyte membrane fatty acids serve as objective, stable biomarkers of medium- to long-term fatty acid exposure. Among fatty acids, ω-3 PUFAs (ALA, EPA, DPA, DHA) and the ω-3 index (EPA+DHA) have been linked to cardiovascular and neuropsychiatric outcomes and various cancers. However, the role of ω-3 PUFAs in oral cancer remains understudied. This study aimed to assess the association between erythrocyte ω-3 PUFAs and oral cancer risk and to evaluate interactions with smoking and drinking.

Prior studies have associated dietary or circulating ω-3 PUFAs with risks of several cancers (e.g., colorectal, pancreatic, prostate, and breast), though findings have been mixed, with reports of inverse, null, and even positive associations depending on setting and biomarker used. Evidence specific to oral cancer is limited; some head and neck cancer studies based on dietary intake suggested higher ω-3 PUFA intake is associated with reduced risk. The current study addresses the gap by analyzing erythrocyte ω-3 PUFAs, an objective biomarker, in relation to oral cancer risk.

Design: Hospital-based case-control study in Fujian Province, China. Setting: First Affiliated Hospital of Fujian Medical University. Period: September 2010 to January 2019. Participants: 236 histologically confirmed primary oral cancer patients (ages 20–80, Chinese residents in Fujian ≥10 years), excluding recurrent/metastatic cases or those with prior chemo/radiotherapy; 300 cancer-free controls from the hospital health examination center in the same period. Data collection: Structured questionnaire via face-to-face interviews captured demographics and risk factors. Ethics: Informed consent obtained; approved by the Institutional Review Board of Fujian Medical University (2011053; March 10, 2011). Biospecimens: 5 mL fasting EDTA blood; erythrocytes separated (centrifugation 1500 rpm, 10 min, 4°C). Fatty acid measurement: Erythrocytes hemolyzed in hypotonic Tris-HCl (pH 7.4, 4°C, 2 h); membranes pelleted (ultracentrifugation 12,000 rpm, 30 min, 4°C). Lipids extracted with chloroform/methanol (2:1 v/v), dried under N2. Fatty acid methyl esters (FAME) prepared with 14% boron trifluoride ether/methanol (1:3 v/v) at 60°C for 10 h; extracted into hexane, dried, redissolved. Gas chromatography: Agilent 7890B with FID and DB-23 capillary column (60 m × 0.25 mm, 0.15 µm film). Nitrogen carrier gas; split 10:1; injector 250°C; detector 280°C; oven 120°C (5 min), 120→175°C at 10°C/min, then 175→210°C at 5°C/min, then to 230°C. Identification by comparison to standards; results expressed as percent of total fatty acids. Analytical precision (CV): ALA 1.03%, EPA 0.77%, DPA 1.24%, DHA 1.32%. Exposures: Erythrocyte ALA (18:3 n-3), EPA (20:5 n-3), DPA (22:5 n-3), DHA (22:6 n-3); total ω-3 PUFA = ALA+EPA+DPA+DHA; ω-3 index = EPA+DHA. Statistical analysis: Group differences tested by χ² or Fisher’s exact tests for categorical variables; t-test or Wilcoxon rank-sum for continuous variables. Restricted cubic splines assessed non-linear associations between ω-3 PUFAs and oral cancer risk. Unconditional logistic regression modeled ω-3 PUFAs as continuous and as dichotomous (median of controls as cutpoint), estimating crude and adjusted ORs with 95% CIs. Multivariable adjustment included age (continuous), sex, education, BMI (continuous), occupation, oral hygiene, smoking, drinking, and diabetes. Stratified analyses by smoking and drinking assessed multiplicative interactions. Two-sided P<0.05 considered significant. Software: Stata 13.1.

- Participant characteristics: 236 cases and 300 controls. Cases had a higher proportion of males, smokers, and drinkers, and lower BMI than controls; several factors (sex, age, education, BMI, oral hygiene, smoking, drinking) were associated with oral cancer in univariable analyses. - Erythrocyte ω-3 PUFA levels (median [IQR], % of total fatty acids) were lower in cases than controls: ALA 0.17 (0.13–0.21) vs 0.18 (0.15–0.24); EPA 0.56 (0.39–0.89) vs 0.76 (0.55–0.97); DPA 2.45 (2.14–2.94) vs 2.66 (2.31–3.09); DHA 5.47 (4.51–6.43) vs 5.96 (5.22–6.69); total ω-3 8.72 (7.06–10.18) vs 9.61 (8.19–10.80); ω-3 index 6.17 (5.00–7.20) vs 6.70 (5.96–7.59); all P<0.01. - Restricted cubic splines indicated non-linear inverse associations between ALA, DPA, DHA, total ω-3 PUFAs, ω-3 index and oral cancer risk after multivariable adjustment. - Logistic regression (continuous exposure, multivariable adjusted OR [95% CI]): ALA 0.51 (0.29–0.91); EPA 0.54 (0.37–0.79); DPA 0.23 (0.09–0.57); DHA 0.19 (0.08–0.45); total ω-3 PUFAs 0.12 (0.04–0.32); ω-3 index 0.21 (0.09–0.47). - Logistic regression (dichotomous by control median, multivariable adjusted OR [95% CI]): EPA 0.62 (0.41–0.92); DHA 0.61 (0.41–0.90); total ω-3 0.66 (0.45–0.99); ω-3 index 0.64 (0.43–0.94); consistent inverse associations. - Interactions: Multiplicative interactions between smoking and ALA/EPA/DHA (Pinteraction <0.05) and between drinking and ALA/EPA/DHA/ω-3 index (Pinteraction <0.05). Protective associations were primarily evident among non-smokers and non-drinkers; e.g., EPA ≥0.76% vs <0.76% among non-smokers OR 0.63 (0.39–1.00); among non-drinkers EPA OR 0.57 (0.36–0.88); ALA OR 0.63 (0.41–0.99); DHA OR 0.64 (0.41–0.98).

This study addressed the understudied relationship between erythrocyte ω-3 PUFAs and oral cancer, showing that higher erythrocyte levels of ALA, EPA, DPA, DHA, total ω-3s, and the ω-3 index are associated with lower oral cancer risk, with evidence of non-linear dose–response. These findings align with several studies reporting inverse associations of ω-3 PUFAs with risks of other cancers, though literature is heterogeneous. The observed interactions suggest that smoking and alcohol consumption may attenuate or modify the protective effects of ω-3 PUFAs; smokers also exhibited lower ω-3 levels, potentially reflecting differences in diet or increased oxidative/inflammatory burden. Mechanistically, ω-3 PUFAs may exert anti-inflammatory, anti-oxidative, and anti-proliferative effects relevant to oral carcinogenesis. Overall, the results support a protective role of ω-3 PUFA status against oral cancer, particularly among non-smokers and non-drinkers.

Higher erythrocyte ω-3 PUFA levels are associated with reduced oral cancer risk, with inverse non-linear relationships observed for EPA, DHA, total ω-3 PUFAs, and the ω-3 index. Multiplicative interactions with smoking and drinking indicate that protective effects may be more pronounced in non-smoking and non-drinking populations. Prospective studies are needed to confirm temporality and clarify mechanisms.

- Case-control design limits causal inference and is susceptible to reverse causality; erythrocyte fatty acids were not collected prospectively and may be influenced by disease-related behavioral changes. - Potential residual confounding: despite adjustment for demographics and lifestyle factors, data on medications affecting fatty acid levels were not collected. - Modest sample size, potentially limiting power for stratified and interaction analyses; larger studies are warranted.