Skip to main content
Download PDF
Download ePub

Adherence to a low-fat dietary pattern reduces head and neck cancer risk: evidence from the PLCO trial

Nutrition Journal volume 23, Article number: 125 (2024) Cite this article

Abstract

Purpose

Low-fat dietary (LFD) pattern refers to a dietary structure with reduced fat intake. The aim was to investigate the association between LFD pattern and risk of head and neck cancer (HNC).

Methods

Data were derived from the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial. LFD score was used to assess adherence to an LFD pattern, with higher scores indicating greater adherence. Cox regression was used to evaluate the association between LFD score and risk of HNC and its subtypes. To visualize the trend in risk of HNC and its subtypes with changing LFD score, restricted cubic spline plots were utilized. A series of subgroup analyses were conducted to identify potential confounders. Sensitivity analyses were performed to assess the robustness of the results.

Results

Among 98,459 participants of PLCO trial, 268 cases with HNC were identified during an average of 8.8 years of follow-up. In the fully adjusted model, participants in the highest compared with the lowest quartiles of LFD score had a lower risk of HNC (HR Q4 vs. Q1: 0.60; 95% CI: 0.40–0.90; P for trend = 0.026) and larynx cancer (HRQ4 vs. Q1: 0.46; 95% CI: 0.22–0.96; P for trend = 0.039). The restricted cubic spline plots demonstrated a linear dose-response relationship between the LFD score and the risk of HNC and its subtypes (all P for nonlinearity > 0.05). The primary association remained robust in the sensitivity analysis.

Conclusion

Our findings suggest that adherence to an LFD pattern may lower the risk of HNC in the US population.

Peer Review reports

Introduction

Head and neck cancer (HNC), including cancers of the oral cavity, pharynx, and larynx, is the sixth most common cancer worldwide [1]. In 2018, there were an estimated 890,000 new cases and 450,000 deaths from HNC globally [1]. Known risk factors for HNC include tobacco use, alcohol consumption, human papillomavirus (HPV) infection, and genetic factors [2]. However, these established factors do not fully account for the observed patterns in HNC occurrence [3]. Therefore, identifying novel modifiable risk factors is important for the prevention of this malignancy. Dietary pattern has emerged as a potential risk factor for HNC in recent years [4,5,6,7,8,9]. Dietary patterns take into account the synergistic interactions between different food components and can provide more rational dietary guidance compared to individual foods or nutrients. Current studies have evaluated the associations between different dietary patterns, such as Western and prudent/healthy patterns, and HNC risk, with inconsistent results [10,11,12].

The occurrence and development of cancer are inseparable from metabolism. Changes in glucose metabolism are critical to the growth and progression of cancer [13], and mainly involve four aspects: tricarboxylic acid cycle, glycolysis, gluconeogenesis, and glycogen synthesis [14]. Accumulated studies have described deregulated levels of enzymes and molecules related to lipid metabolism in HNC, e.g., elevated fatty acids [15], fatty acid binding protein and fatty acid synthase [16, 17]. Unhealthy dietary patterns disrupt lipid and glucose metabolism and play an important role in HNC. Consumption of fried foods, high-fat and processed meats, and sweets has been associated with an increased risk of throat cancer [8, 18, 19]. So, is it possible to choose a dietary pattern to reduce the risk of head and neck squamous cell carcinoma?

The low-fat dietary (LFD) pattern refers to a dietary structure with reduced fat intake, where a greater proportion of energy intake is derived from carbohydrates and protein. It is commonly used for weight control, reducing cardiovascular disease risk, and dietary management of diabetes. However, obesity and high intake of animal fat are associated with HNC. This pattern represents a prudent dietary choice that may protect against HNC by modulating metabolism, oxidative stress, and insulin resistance [20]. Only one case-control study has examined the LFD pattern scores in association with HNC risk, and reported no significant relationship [21]. However, no prospective cohort studies have specifically investigated the LFD pattern in relation to HNC risk.

In the present study, we aimed to prospectively examine the association between LFD score, incorporating intakes of carbohydrates, fat, and protein, and the risk of HNC. We conducted this prospective study utilizing data from the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial.

Methods and methods

Study population and exclusion criteria

To determine whether screening tests could reduce colorectal, lung, prostate and ovarian cancer-related mortality, the National Cancer Institute (NCI) of USA conducted a randomized controlled trial named Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial. The design and methodology of this trial have been previously reported [22]. They recruited close to 155,000 volunteers aged 55–74 years from 10 centers between November 1993 and July 2001. Cancer diagnosis data up to December 2009 were collected. Demographic and medical history data for all participants were collected via baseline questionnaire (BQ). In addition, dietary data were collected using dietary history questionnaire (DHQ) and supplemental questionnaires (SQX). DHQ collects dietary information from participants based on the 137-item Food Frequency Questionnaire (FFQ). Multiple studies have confirmed that FFQ is a good nutritional assessment model [23, 24]. SQX is used to supplement some data not collected by BQ. The institutional review boards at each PLCO study center and the NCI provided ethical approval for the study protocols. Informed consent in writing was collected from all participants prior to their enrollment. For this analysis, we used the NCI-approved public dataset (project ID: PLCO-1335). Therefore, inclusion of human participants did not require ethical approval.

Exclusion criteria was as follows: (1) Failure to return BQ; (2) Participants with invalid DHQ (including the missing DHQ response items exceed 8 or more, extreme caloric intake assessed by DHQ (first and last percentiles) and DHQ completion date not available before death); (3) Participants with a cancer diagnosis before completing the DHQ; (4) Participants with an occurred outcome events between randomization to DHQ completion (who developed HNC, died, or were lost to follow-up in the period); (5) Participants with potentially unreliable dietary intake (including females with “<600 or > 3500 kcal/day” and males with “<800 or > 4200 kcal/day”) [25].

Construction and evaluation of LFD score

We calculated LFD score according to previously reported standards [26, 27]. In brief, participants were stratified into 11 tiers (Supplementary Table 1) based on percentage of energy derived from constant macronutrients including fat, carbohydrate and protein consumption. For fat, participants in the lowest tier were assigned 10 points and those in the highest tier were assigned 0 points. For protein and carbohydrate, the order of tiers was reversed. We then summed scores for the three constant macronutrients to calculate LFD score for all participants, ranging from 0 to 30. Thus, higher LFD score indicate better adherence to LFD pattern. Specifically, we extracted the percent energy from the three macronutrients (protein, fat and carbohydrate) using data obtained from the DHQ. The derivation of nutrient values utilized DietCalc software, the USDA National Nutrient Database for Standard Reference, and the Nutrition Data System. The raw responses from DHQ were processed into analyzable variables such as daily food frequency, pyramid servings, and gram intake. The reliability of DHQ to assess participants’ dietary and nutrition data has been reported elsewhere [23, 28].

Diagnosis of HNC

In the PLCO cancer screening trial, diagnosis of HNC was primarily dependent on annual study update questionnaires mailed to living participants. The PLCO trial confirmed diagnoses of HNC through medical record abstraction (MRA) of participants suspected by the trial to have HNC. Note that if the MRA process does not find records indicating cancer diagnosis, even if a source such as a death certificate indicates cancer, the cancer is not considered confirmed. For patients diagnosed with primary HNC, additional information regarding diagnostic procedures, cancer staging, grading, histopathological type, and initial cancer therapy was recorded by the participants’ physicians. We extracted information on diagnosis date, type (e.g., HNC, oral cavity and pharynx cancer, and larynx cancer), and ICD-O-2 codes from the PLCO data. The endpoint was incidence of HNC and its subtypes (oral cavity and pharynx cancer, and larynx cancer).

Dealing with covariates

We obtained participants’ baseline covariates including sex (male, female), education level (less than college, college graduate, postgraduate), trial group (intervention, control), height, race (non-Hispanic White, non-Hispanic Black, Hispanic, and other race/ethnicity), weight, smoking status (never, current or former), history of hypertension (no, yes), aspirin use (no, yes), history of diabetes (no, yes) and family history of HNC (no, yes, possibly) from the BQ. Body mass index (BMI) was calculated by dividing weight (kg) by height squared (m2). Additionally, the DHQ was used to collect participants’ age, alcohol intake (no, yes), intake of dietary cholesterol, pyramid food servings (grain, total fruit, vegetable, dairy), saturated fatty acids, mono-unsaturated fatty acids, poly-unsaturated fatty acids, fish and other seafood, eggs, total dietary fiber from diet, discretionary fat, red processed meat, cakes, cookies, pies, and pastries, sugars and sweets, total lean meat, added sugars, alcohol, energy intake in diet, and consumption of total carbohydrate, fat, and protein. Total minutes of moderate to vigorous physical activity per week was obtained from the SQX as physical activity.

Statistical analysis

Data were assessed for missingness prior to analysis. Missingness ranging from 0 to 26% was observed for covariates in the BQ and DHQ. For variables missing < 5% of data, continuous variables such as BMI were imputed using median replacement, while categorical variables including education, family cancer history, regular aspirin use, and diabetes were imputed using mode substitution. For physical activity data missing > 25% in the Supplementary Questionnaire (SQX), multiple imputation was performed using chained equations to generate five complete datasets. Details of missingness and imputation for each variable are presented in Supplementary Table 2. Logistic regression models were fitted on each imputed dataset and results were pooled using Rubin’s rules. Imputation procedures were performed in R version 4.3.1.

Cox proportional hazards regression models were utilized to calculate hazard ratios (HRs) and 95% confidence intervals (CIs) to investigate potential associations between LFD score and risk of HNC incidence. Models were adjusted for potential confounders and used follow-up time as the time metric. Participants were categorized into quartiles based on LFD score (Quartile 1, Quartile 2, Quartile 3 and Quartile 4), with the lowest scoring quartile (Quartile 1) serving as the reference group. Person-years were quantified according to follow-up time for each quartile. The Schoenfeld residuals method was utilized to evaluate the time-invariance of the LFD score. Trend analysis for HNC risk across increasing LFD score quartiles was conducted using Cox regression models, with median values for each quartile assigned to all participants within that quartile and treated as a continuous variable. To control for potential confounding, two multivariable Cox regression models were utilized in the analyses. Model 1 was adjusted for participants’ age, race, and sex. Model 2 was further adjusted for education, trial group, energy intake from diet, family history of HNC, smoking pack-years, alcohol intake, BMI, aspirin use, physical activity, and diabetes history.

We used restricted cubic spline (RCS) regression with knots located the 10th, 50th, and 90th percentiles to accurately describe the risk of HNC across the entire range of LFD score. It is noted that the number of knots was selected based on the Akaike’s information criterion (AIC) and the Bayesian information criterion (BIC), with the lowest AIC and BIC values indicating the best-fitted model. When AIC results were inconsistent with those from BIC, BIC results were used for knot selection, as BIC is more conservative than AIC [29]. The median of the first quartile of LFD score was set as the reference value. The pnonlinearity was also determined by testing the null hypothesis. If the p value is less than 0.05, it means that the nonlinear effect of the model is significant. Furthermore, the above methodology was employed to assess the association between LFD score and the incidence rate of HNC subtypes, including oral cavity and pharynx cancer and larynx cancer.

Stratified analyses were conducted to evaluate potential effect modification by age (≤ 65, > 65 years), sex, BMI (≤ 30, > 30 kg/m2), smoking (never, former, current), alcohol intake (yes/no), and diabetes history (yes/no) on the relationship between LFD score and HNC risk. Multiplicative interaction was assessed using likelihood ratio tests comparing Cox models with and without cross-product interaction terms.

Sensitivity analyses were performed to evaluate the robustness of findings: (1) Repeated analysis in participants with non-missing data; (2) Excluded participants with family history of HNC; (3) Excluded participants with a history of diabetes. and (4) Excluded cases observed within the first 2 and 4 years of follow-up to address the concern of reverse causality.

Baseline characteristics of the study population are presented as mean ± standard deviation for continuous variables and number (percentage) for categorical variables. All statistical tests were two-sided and performed using R version 4.3.1 (R Foundation for Statistical Computing, Vienna, Austria); P values < 0.05 were considered statistically significant.

Results

Screening of the study population

Our aim was to analyze the association between LFD patterns and risk of HNC. Therefore, based on the five exclusion criteria mentioned in the methods, the numbers of participants excluded for each criterion were as follows: (1) n = 4,918; (2) n = 38,462; (3) n = 9,684; (4) n = 68; (5) n = 3,296. Finally, our study included 98,459 participants (Fig. 1). In Cox proportional hazards regression analysis, follow-up time was defined as the interval between completion of the DHQ at baseline and HNC diagnosis, death, loss to follow-up, or end of follow-up on December 31, 2009, whichever came first (Fig. 2)

Fig. 1
figure 1

The flow chart of identifying eligible participants. PLCO, Prostate, Lung, Colorectal, and Ovarian; BQ, baseline questionnaire; DHQ, diet history questionnaire

Full size image
Fig. 2
figure 2

The timeline and follow-up scheme of our study

Full size image

Characteristics of the study population

Baseline characteristics of the 98,459 study participants across quartiles of LFD score are presented in table 1. Quartile 1 (n = 26,715) had scores of 0–10. Quartile 2 (n = 26,151) had scores of 11–15. Quartile 3 (n = 24,764) had scores of 16–20. Quartile 4 (n = 20,829) had scores of 21–30. The mean (standard deviation) LFD score was 15.00 (6.27), with higher scores indicating greater adherence to the LFD dietary pattern. Compared to participants in the lowest LFD quartile, those in the highest quartile were more likely to be older (66.19 ± 5.79 vs. 65.52 ± 5.73), female (64.60% vs. 44.71%), other race/ethnicity (4.89% vs. 2.84%), college-educated (41.47% vs. 32.85%), control group (50.74% vs. 47.91%), physically active (140.27 ± 114.12 vs109.42 ± 102.98), no drinker (33.66% vs. 22.55%), never smokers (56.87% vs. 40.13%), with a lower BMI (26.57 ± 4.74 vs. 27.58 ± 4.86), lower energy intake from diet (1502.45 ± 529.56 vs. 1936.85 ± 722.85) and no family history of HNC (96.49% vs. 95.53%) (all P < 0.001). Intakes of saturated fatty acids (12.15 ± 5.31 vs. 27.39 ± 12.51), mono-unsaturated fatty acids (14.47 ± 6.21 vs. 32.39 ± 13.82), and poly-unsaturated fatty acids (9.41 ± 4.04 vs. 18.94 ± 8.32) were lower among participants in the highest versus lowest LFD quartile (all P < 0.001). In terms of energy sources, percent of energy from fat (23.59 ± 3.97 vs. 39.44 ± 6.15) were lower, while percent of energy from carbohydrates (61.10 ± 5.89 vs. 43.36 ± 6.86), and percent of energy from protein (16.98 ± 2.39 vs. 14.34 ± 2.72) were higher among participants in the highest versus lowest LFD quartile (all P < 0.001). Intakes of saturated fatty acids (12.15 ± 5.31 vs. 27.39 ± 12.51), mono-unsaturated fatty acids (14.47 ± 6.21 vs. 32.39 ± 13.82), and poly-unsaturated fatty acids (9.41 ± 4.04 vs. 18.94 ± 8.32) were lower among participants in the highest versus lowest LFD quartile (all P < 0.001). Intakes of vegetables (4.24 ± 2.58 vs. 3.72 ± 2.01), fruits (3.62 ± 2.19 vs. 1.89 ± 1.35), dairy foods (1.69 ± 1.25 vs. 1.21 ± 0.97), fish and other seafood (0.51 ± 0.60 vs. 0.43 ± 0.44), eggs (0.83 ± 4.52 vs. 0.20 ± 2.07), total dietary fiber from diet (20.28 ± 8.94 vs. 16.41 ± 7.24) were higher, while intakes of grain (4.50 ± 2.21 vs. 4.70 ± 2.32), discretionary fat (27.88 ± 12.03 vs. 69.02 ± 28.65), red processed meat (6.33 ± 8.30 vs. 17.74 ± 18.66), cakes, cookies, pies, and pastries (17.81 ± 18.34 vs. 32.40 ± 33.78), sugars and sweets (19.89 ± 18.35 vs. 28.51 ± 27.41), total lean meat (5.90 ± 3.35 vs. 8.17 ± 4.60), added sugars (10.52 ± 6.41 vs. 12.98 ± 8.59), alcohol (0.25 ± 0.47 vs. 1.06 ± 2.11), and cholesterol from diet (139.49 ± 72.63 vs. 269.60 ± 152.36) were lower among participants in the highest versus lowest LFD quartile (all P < 0.001) (table 1)

Table 1 Baseline characteristics of study population according to quartiles of LFD scores
Full size table

Association between LFD score and risk of HNC

Over a mean (standard deviation) follow-up time of 8.84 (1.94) years, representing 869,807.9 person-years, 268 incident HNC cases were ascertained, including 161 cases of oral cavity and pharynx cancer, and 96 cases of larynx cancer. In the unadjusted Cox model, participants in the highest versus lowest quartile of LFD score had a 60% lower hazard of HNC (HRQ4 vs. Q1 = 0.40, 95% CI: 0.27–0.59; Ptrend < 0.001). In multivariate Cox proportional hazards models adjusting for age, sex, race, education, trial group, energy intake from diet, family history, smoking, alcohol intake, aspirin use, physical activity, BMI, and diabetes, participants in the highest versus lowest quartile of LFD score had a 40% lower hazard of HNC (HRQ4 vs. Q1 = 0.60, 95% CI: 0.40–0.90; Ptrend = 0.026). Furthermore, a similar inverse relationship was observed between the LFD score and the risk of HNC subtype (for larynx cancer, HRQ4 vs. Q1: 0.46; 95% CI: 0.22–0.96; Ptrend= 0.039) (table 2). We employed RCS model to illustrate the fluctuations in the incidence of HNC, oral and pharynx cancer, as well as larynx cancer across the entire range of LFD score. As depicted in Fig. 3, as LFD score increased, there was a corresponding decrease in the risk of these cancers, displaying a linear dose-response relationship (all P for nonlinearity > 0.05)

Table 2 Association of LFD scores with the risk of HNC and its subtypes
Full size table
Fig. 3
figure 3

Dose–response analyses on the association of LFD scores with the risk of HNC (A), oral and pharynx cancer (B), and larynx cancer (C). HRs were adjusted for age(years), sex (male, female) and race (non-Hispanic White, non-Hispanic Black, Hispanic, and other race/ethnicity), educational level (college below, college graduate, postgraduate), BMI (kg/m2), family history of HNC (no, yes, possibly), smoker (never, current or former), drinker (no, yes), aspirin use (no, yes), history of hypertension (no, yes), history of diabetes (no, yes), and physical activity (min/week)

Full size image

  • In addition, we also added the results for models with the LFD score run as a continuous variable. In the unadjusted Cox model, LFD score was associated with clinically significant HNC (HR = 0.96, 95% CI: 0.94–0.97; P < 0.001). In multivariate Cox proportional hazards models adjusting for age, sex, race, education, trial group, energy intake from diet, family history, smoking, alcohol intake, aspirin use, history of hypertension, physical activity, BMI, and diabetes, LFD score was still associated with clinically significant HNC (HR = 0.98, 95% CI: 0.96-1.00; P = 0.028) (Supplementary Table 3). Association of intake of PUFA, MUFA, or SFA with the risk of HNC was also analyzed, indicating non-significant result (Supplementary Table 4).

Subgroup and sensitivity analyses

In subgroup analyses, the inverse association between LFD score and HNC risk did not differ significantly by age, sex, trial group, energy intake from diet, BMI, smoking status, alcohol intake, or history of hypertension or diabetes (all P for interaction > 0.05) (table 3). Subgroup analyses demonstrated that the inverse association was stronger among participants with no aspirin use (Pinteraction <0.001). In sensitivity analyses, results remained unchanged restricted to participants excluding those with a history of diabetes or family history of HNC at baseline (Ptrend <0.05). The associations remained similar when we further excluded cases observed within the first 2 years or 4 years of follow-up although the p-value did not reach statistical significance (table 4). This suggests that the inverse association between LFD score and HNC risk is well robust. Non-significant results are associated with a smaller number of cases remaining after exclusion

Table 3 Subgroup analyses on the association of LFD scores with the risk of HNC
Full size table
Table 4 Sensitivity analyses on the association of LFD scores with the risk of HNC
Full size table

Discussion

In this large prospective cohort study, we found that adherence to an LFD pattern was significantly associated with a reduced risk of HNC. Restricted cubic spline analysis showed linear dose-response relationships between LFD score and risks of total HNC as well as its subtypes. A series of sensitivity analyses yielded consistent results without significant changes after multiple adjustments

Over the past few decades, substantial epidemiologic evidence has accumulated to suggest that adherence to an LFD pattern may help prevent obesity, type 2 diabetes, cardiovascular disease, and certain types of cancer [30]. Early ecological studies revealed associations between national fat consumption and mortality rates from these diseases [31, 32]. This stimulated extensive research on the health effects of reducing dietary fat intake. Several landmark randomized controlled trials including the women’s Health Initiative did not find protective effects of LFD on cancer or cardiovascular outcomes [33, 34]. However, most of these trials had poor long-term compliance, limiting their ability to truly test LFD. The role of dietary fat intake in cancer development has been investigated, but findings on LFD and cancer risk have been inconsistent. Some results showing no association between greater adherence to an LFD pattern and risk of cancer align with the totality of evidence from both observational studies and clinical trials. For example, pooled analyses of prospective cohort studies have found no significant associations between lower fat intake and risk of prostate cancers [35]. Randomized trials including the women’s Health Initiative also demonstrated no effects of an LFD intervention on incidence of breast and colorectal cancers over 7–8 years of follow-up [33, 34]. The LFD patterns emphasized in previous studies focused on reducing dietary fat and increasing intakes of fruits, vegetables, and grains. Recent studies have also reported different conclusions. LFD may reduce the risk of colorectal cancer [36]. LFD scores were negatively associated with risks of liver and lung cancer [27, 37]. No studies have yet examined the association between LFD and HNC risk. Established high-risk factors for HNC include alcohol drinking and cigarette smoking. Some studies have also examined associations between dietary patterns and HNC risk. Mediterranean dietary patterns may reduce the incidence of HNC in US populations [4, 38]. Moderate egg intake (< 4 eggs per week) and daily vitamin intake help lower HNC risk [4]. Consumption of healthful diets rich in fruits and vegetables correlates with reduced HNC risk [39]. The dietary pattern scores constructed in this study considered the energy proportion of three macronutrients and the low-fat contribution, overcoming the limitations of traditional methods for evaluating LFD patterns. Our results demonstrate that after adjusting for potential confounders, LFD score was inversely associated with HNC risk in a nonlinear dose-response manner. To our knowledge, this is the first prospective study with up to 98,456 participants demonstrating an inverse association between LFD patterns and HNC risk

Several potential mechanisms may be involved in the association between LFD scores and risk of HNC. LFD patterns may inhibit carcinogenesis in the head and neck region by regulating hormone levels, improving insulin sensitivity, controlling body weight, reducing intake of detrimental fats, and increasing intake of beneficial foods. Fat intake is associated with changes in hormone levels. High-fat diets may increase levels of estrogen and testosterone, and higher endogenous estrogen is linked to increased risk of HNC [40]. LFD may exert anticancer effects by decreasing hormone levels. High-fat diets can lead to insulin resistance, while activation of insulin and IGF-1 signaling pathways is implicated in cancer development [41]. LFD may suppress tumor development by improving insulin sensitivity and lowering IGF-1 levels. Excess fat intake is associated with obesity, which is a risk factor for HNC [42]. LFD patterns can control body weight and thus lower cancer risk. Different types of fats have varying mechanisms of action. Saturated fats provoke inflammatory responses while trans fats cause oxidative stress [43]. LFD can restrict intake of these detrimental fats. LFDs are rich in beneficial components like dietary fiber and antioxidant nutrients which may have anticancer effects [44].

Strengths of the present analysis include the prospective design, large sample size, long follow-up duration, ability to control for smoking and alcohol as major risk factors, and assessment of overall dietary patterns rather than individual nutrients or foods. Most importantly, this is the first study to demonstrate that adherence to an LFD pattern is associated with reduced risks of total HNC as well as its subtypes, suggesting a potential preventive effect against HNC

This study also has some limitations. First, although we considered a comprehensive range of confounding factors, and sensitivity analysis also showed that the results are stable, residual confounding factors cannot be completely ruled out. Moreover, due to the unavailability of data, we were unable to conduct analyses for certain confounding factors such as HPV infection. There is currently little evidence on whether the association between diet and HNC risk changes in people with different HPV infection status. One study showed that fruit consumption was associated with a reduced HNSCC risk among HPV-16-seronegative individuals but an increased HNSCC risk among the HPV-16-seropositive individuals [45]. The mechanism is currently unclear. It is also possible that the LFD score is associated with HNC risk differently in people with different HPV status, which needs to be explored in more clinical studies in the future. Second, as with other nutritional epidemiology studies, measurement error in dietary assessment remains a concern despite the survey questionnaires used in the cohort being validated and showing good validity. Third, we were unable to evaluate the impact of potential dietary changes over time since only baseline diet information was comprehensive in the PLCO trial, and subsequent supplementary diet questionnaires had missing diet data. Forth, in our study, over 90% of participants were non-hispanic White; over 60% had educational degree of some college or less; and around half were aspirin users or ever smokers. Therefore, our results may not be applicable to other populations. Finally, participants in this study were U.S. adults aged 55 to 74, so the results need to be applied with caution to other populations

Conclusion

In summary, we established an association between higher LFD score and reduced risk of HNC and larynx cancer. Thus, our findings support potential benefits of adhering to LFD dietary patterns as a HNC and larynx cancer prevention strategy

Data availability

No datasets were generated or analysed during the current study.

Abbreviations

NCI:

National cancer institute

PLCO:

Prostate lung colorectal and ovarian

BQ:

Baseline questionnaire

CIs:

Confidence intervals

DHQ:

Dietary history questionnaire

SQX:

Supplemental questionnaire

HRs:

Hazard ratios

LFD:

Low-fat diet

BMI:

Body mass index

HNC:

Head and neck cancer

References

  1. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global Cancer statistics 2020: GLOBOCAN estimates of incidence and Mortality Worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71:209–49.

    Article  PubMed  Google Scholar 

  2. Hashim D, Genden E, Posner M, Hashibe M, Boffetta P. Head and neck cancer prevention: from primary prevention to impact of clinicians on reducing burden. Ann Oncol. 2019;30:744–56.

    Article  PubMed Central  CAS  Google Scholar 

  3. Fiolet T, Srour B, Sellem L, Kesse-Guyot E, Alles B, Mejean C, Deschasaux M, Fassier P, Latino-Martel P, Beslay M, Hercberg S, Lavalette C, Monteiro CA, Julia C, Touvier M. Consumption of ultra-processed foods and cancer risk: results from NutriNet-Sante prospective cohort. BMJ. 2018;360:k322.

    Article  PubMed Central  Google Scholar 

  4. Saka-Herran C, Pereira-Riveros T, Jane-Salas E, Lopez-Lopez J. Association between the Mediterranean Diet and Vitamin C and the Risk of Head and Neck Cancer, Nutrients, 15 (2023).

  5. Giraldi L, Panic N, Cadoni G, Boccia S, Leoncini E. Association between Mediterranean diet and head and neck cancer: results of a large case-control study in Italy. Eur J Cancer Prev. 2017;26:418–23.

    Article  Google Scholar 

  6. Kawakita D, Lee YA, Turati F, Parpinel M, Decarli A, Serraino D, Matsuo K, Olshan AF, Zevallos JP, Winn DM, Moysich K, Zhang ZF, Morgenstern H, Levi F, Kelsey K, McClean M, Bosetti C, Garavello W, Schantz S, Yu GP, Boffetta P, Chuang SC, Hashibe M, Ferraroni M, La Vecchia C, Edefonti V. Dietary fiber intake and head and neck cancer risk: a pooled analysis in the International Head and Neck Cancer Epidemiology consortium. Int J Cancer. 2017;141:1811–21.

    Article  PubMed Central  CAS  Google Scholar 

  7. Li WQ, Park Y, Wu JW, Goldstein AM, Taylor PR, Hollenbeck AR, Freedman ND, Abnet CC. Index-based dietary patterns and risk of head and neck cancer in a large prospective study. Am J Clin Nutr. 2014;99:559–66.

    Article  PubMed Central  CAS  Google Scholar 

  8. Bradshaw PT, Siega-Riz AM, Campbell M, Weissler MC, Funkhouser WK, Olshan AF. Associations between dietary patterns and head and neck cancer: the Carolina head and neck cancer epidemiology study. Am J Epidemiol. 2012;175:1225–33.

    Article  PubMed Central  Google Scholar 

  9. Edefonti V, Bravi F, La Vecchia C, Randi G, Ferraroni M, Garavello W, Franceschi S, Talamini R, Boffetta P, Decarli A. Nutrient-based dietary patterns and the risk of oral and pharyngeal cancer. Oral Oncol. 2010;46:343–8.

    Article  Google Scholar 

  10. Hu FB. Dietary pattern analysis: a new direction in nutritional epidemiology. Curr Opin Lipidol. 2002;13:3–9.

    Article  CAS  Google Scholar 

  11. Mentella MC, Scaldaferri F, Ricci C, Gasbarrini A, Miggiano GAD. Cancer and Mediterranean Diet: A Review, Nutrients, 11 (2019).

  12. Filomeno M, Bosetti C, Garavello W, Levi F, Galeone C, Negri E, La Vecchia C. The role of a Mediterranean diet on the risk of oral and pharyngeal cancer. Br J Cancer. 2014;111:981–6.

    Article  PubMed Central  CAS  Google Scholar 

  13. Fakhri S, Moradi SZ, Farzaei MH, Bishayee A. Modulation of dysregulated cancer metabolism by plant secondary metabolites: a mechanistic review. Semin Cancer Biol. 2022;80:276–305.

    Article  CAS  Google Scholar 

  14. Dauer P, Lengyel E. New roles for Glycogen in Tumor Progression. Trends Cancer. 2019;5:396–9.

    Article  CAS  Google Scholar 

  15. Halczy-Kowalik L, Drozd A, Stachowska E, Drozd R, Zabski T, Domagala W. Fatty acids distribution and content in oral squamous cell carcinoma tissue and its adjacent microenvironment. PLoS ONE. 2019;14:e0218246.

    Article  PubMed Central  CAS  Google Scholar 

  16. Krontiras H, Roye GD, Beenken SE, Myers RB, Mayo MS, Peters GE, Grizzle WE. Fatty acid synthase expression is increased in neoplastic lesions of the oral tongue. Head Neck. 1999;21:325–9.

    Article  CAS  Google Scholar 

  17. Uma RS, Naresh KN, D’Cruz AK, Mulherkar R, Borges AM. Metastasis of squamous cell carcinoma of the oral tongue is associated with down-regulation of epidermal fatty acid binding protein (E-FABP). Oral Oncol. 2007;43:27–32.

    Article  CAS  Google Scholar 

  18. De Vito R, Lee YCA, Parpinel M, Serraino D, Olshan AF, Zevallos JP, Levi F, Zhang ZF, Morgenstern H, Garavello W, Kelsey K, McClean M, Schantz S, Yu GP, Boffetta P, Chuang SC, Hashibe M, La Vecchia C, Parmigiani G, Edefonti V. Shared and Study-specific dietary patterns and Head and Neck Cancer Risk in an International Consortium. Epidemiology. 2019;30:93–102.

    Article  PubMed Central  Google Scholar 

  19. Edefonti V, Hashibe M, Ambrogi F, Parpinel M, Bravi F, Talamini R, Levi F, Yu G, Morgenstern H, Kelsey K, McClean M, Schantz S, Zhang Z, Chuang S, Boffetta P, La Vecchia C, Decarli A. Nutrient-based dietary patterns and the risk of head and neck cancer: a pooled analysis in the International Head and Neck Cancer Epidemiology consortium. Ann Oncol. 2012;23:1869–80.

    Article  CAS  Google Scholar 

  20. Brennan SF, Cantwell MM, Cardwell CR, Velentzis LS, Woodside JV. Dietary patterns and breast cancer risk: a systematic review and meta-analysis. Am J Clin Nutr. 2010;91:1294–302.

    Article  CAS  Google Scholar 

  21. Yang WS, Zeng XF, Liu ZN, Zhao QH, Tan YT, Gao J, Li HL, Xiang YB. Diet and liver cancer risk: a narrative review of epidemiological evidence. Br J Nutr. 2020;124:330–40.

    Article  CAS  Google Scholar 

  22. Zhu CS, Pinsky PF, Kramer BS, Prorok PC, Purdue MP, Berg CD, Gohagan JK. The prostate, lung, colorectal, and ovarian cancer screening trial and its associated research resource. J Natl Cancer Inst. 2013;105:1684–93.

    Article  PubMed Central  Google Scholar 

  23. Subar AF, Thompson FE, Kipnis V, Midthune D, Hurwitz P, McNutt S, McIntosh A, Rosenfeld S. Comparative validation of the Block, Willett, and National Cancer Institute food frequency questionnaires: the eating at America’s table study. Am J Epidemiol. 2001;154:1089–99.

    Article  CAS  Google Scholar 

  24. Thompson FE, Kipnis V, Midthune D, Freedman LS, Carroll RJ, Subar AF, Brown CC, Butcher MS, Mouw T, Leitzmann M, Schatzkin A. Performance of a food-frequency questionnaire in the US NIH-AARP (National Institutes of Health-American Association of Retired Persons) Diet and Health Study. Public Health Nutr. 2008;11:183–95.

    Article  Google Scholar 

  25. Satija A, Bhupathiraju SN, Spiegelman D, Chiuve SE, Manson JE, Willett W, Rexrode KM, Rimm EB, Hu FB. Healthful and unhealthful plant-based diets and the risk of Coronary Heart Disease in U.S. adults. J Am Coll Cardiol. 2017;70:411–22.

    Article  PubMed Central  Google Scholar 

  26. Shan Z, Guo Y, Hu FB, Liu L, Qi Q. Association of Low-Carbohydrate and Low-Fat Diets with Mortality among US adults. JAMA Intern Med. 2020;180:513–23.

    Article  PubMed Central  CAS  Google Scholar 

  27. Peng L, Xiang L, Xu Z, Gu H, Zhu Z, Tang Y, Jiang Y, He H, Wang Y, Zhao X. Association between low-fat diet and liver cancer risk in 98,455 participants: results from a prospective study. Front Nutr. 2022;9:1013643.

    Article  PubMed Central  Google Scholar 

  28. Csizmadi I, Boucher BA, Lo Siou G, Massarelli I, Rondeau I, Garriguet D, Koushik A, Elenko J, Subar AF. Using national dietary intake data to evaluate and adapt the US Diet History Questionnaire: the stepwise tailoring of an FFQ for Canadian use. Public Health Nutr. 2016;19:3247–55.

    Article  PubMed Central  Google Scholar 

  29. Dziak JJ, Coffman DL, Lanza ST, Li R, Jermiin LS. Sensitivity and specificity of information criteria. Brief Bioinform. 2020;21:553–65.

    Article  Google Scholar 

  30. Smith JD, Hou T, Ludwig DS, Rimm EB, Willett W, Hu FB, Mozaffarian D. Changes in intake of protein foods, carbohydrate amount and quality, and long-term weight change: results from 3 prospective cohorts. Am J Clin Nutr. 2015;101:1216–24.

    Article  PubMed Central  CAS  Google Scholar 

  31. Kinlen LJ. Fat and cancer aetiology. Prog Lipid Res. 1986;25:527–31.

    Article  CAS  Google Scholar 

  32. Armstrong BK, Mann JI, Adelstein AM, Eskin F. Commodity consumption and ischemic heart disease mortality, with special reference to dietary practices. J Chronic Dis. 1975;28:455–69.

    Article  CAS  Google Scholar 

  33. Beresford SA, Johnson KC, Ritenbaugh C, Lasser NL, Snetselaar LG, Black HR, Anderson GL, Assaf AR, Bassford T, Bowen D, Brunner RL, Brzyski RG, Caan B, Chlebowski RT, Gass M, Harrigan RC, Hays J, Heber D, Heiss G, Hendrix SL, Howard BV, Hsia J, Hubbell FA, Jackson RD, Kotchen JM, Kuller LH, LaCroix AZ, Lane DS, Langer RD, Lewis CE, Manson JE, Margolis KL, Mossavar-Rahmani Y, Ockene JK, Parker LM, Perri MG, Phillips L, Prentice RL, Robbins J, Rossouw JE, Sarto GE, Stefanick ML, Van Horn L, Vitolins MZ, Wactawski-Wende J, Wallace RB, Whitlock E. Low-fat dietary pattern and risk of colorectal cancer: the Women’s Health Initiative Randomized Controlled Dietary Modification Trial, JAMA, 295 (2006) 643–54.

  34. Prentice RL, Caan B, Chlebowski RT, Patterson R, Kuller LH, Ockene JK, Margolis KL, Limacher MC, Manson JE, Parker LM, Paskett E, Phillips L, Robbins J, Rossouw JE, Sarto GE, Shikany JM, Stefanick ML, Thomson CA, Van Horn L, Vitolins MZ, Wactawski-Wende J, Wallace RB, Wassertheil-Smoller S, Whitlock E, Yano K, Adams-Campbell L, Anderson GL, Assaf AR, Beresford SA, Black HR, Brunner RL, Brzyski RG, Ford L, Gass M, Hays J, Heber D, Heiss G, Hendrix SL, Hsia J, Hubbell FA, Jackson RD, Johnson KC, Kotchen JM, LaCroix AZ, Lane DS, Langer RD, Lasser NL, Henderson MM. Low-fat dietary pattern and risk of invasive breast cancer. Volume JAMA. the Women’s Health Initiative Randomized Controlled Dietary Modification Trial; 2006. pp. 629–42.

  35. Kohler LN, Garcia DO, Harris RB, Oren E, Roe DJ, Jacobs ET. Adherence to Diet and Physical Activity Cancer Prevention guidelines and Cancer outcomes: a systematic review. Cancer Epidemiol Biomarkers Prev. 2016;25:1018–28.

    Article  PubMed Central  Google Scholar 

  36. Sneyd MJ, Cox B. Do low-fat foods alter risk of colorectal cancer from processed meat? Public Health. 2020;183:138–45.

    Article  CAS  Google Scholar 

  37. Zhu Z, Peng L, Zhou H, Gu H, Tang Y, Zhou Z, Xiang L, Wang Y. Low-fat dairy consumption and the risk of lung cancer: a large prospective cohort study. Cancer Med. 2023;12:16558–69.

    Article  PubMed Central  CAS  Google Scholar 

  38. Saraiya V, Bradshaw P, Meyer K, Gammon M, Slade G, Brennan P, Abedi-Ardekani B, Olshan A. The association between diet quality and cancer incidence of the head and neck. Cancer Causes Control. 2020;31:193–202.

    Article  Google Scholar 

  39. Galvao OP, De Podesta SV, Peres LB, Salaroli M, Cattafesta JRV, De Podesta SLV, von Zeidler JC, de Oliveira LP, Kowalski MK, Ikeda P, Brennan MP, Curado. Consumption of minimally processed foods as protective factors in the genesis of squamous cell carcinoma of the head and neck in Brazil. PLoS ONE. 2019;14:e0220067.

    Article  Google Scholar 

  40. Freedman ND, Park Y, Subar AF, Hollenbeck AR, Leitzmann MF, Schatzkin A, Abnet CC. Fruit and vegetable intake and head and neck cancer risk in a large United States prospective cohort study. Int J Cancer. 2008;122:2330–6.

    Article  CAS  Google Scholar 

  41. Zheng Y, Wu C, Yang J, Zhao Y, Jia H, Xue M, Xu D, Yang F, Fu D, Wang C, Hu B, Zhang Z, Li T, Yan S, Wang X, Nelson PJ, Bruns C, Qin L, Dong Q. Insulin-like growth factor 1-induced enolase 2 deacetylation by HDAC3 promotes metastasis of pancreatic cancer. Signal Transduct Target Ther. 2020;5:53.

    Article  PubMed Central  CAS  Google Scholar 

  42. Gaudet MM, Olshan AF, Chuang SC, Berthiller J, Zhang ZF, Lissowska J, Zaridze D, Winn DM, Wei Q, Talamini R, Szeszenia-Dabrowska N, Sturgis EM, Schwartz SM, Rudnai P, Eluf-Neto J, Muscat J, Morgenstern H, Menezes A, Matos E, Bucur A, Levi F, Lazarus P, La Vecchia C, Koifman S, Kelsey K, Herrero R, Hayes RB, Franceschi S, Wunsch-Filho V, Fernandez L, Fabianova E, Daudt AW, Dal Maso L, Curado MP, Chen C, Castellsague X, Benhamou S, Boffetta P, Brennan P, Hashibe M. Body mass index and risk of head and neck cancer in a pooled analysis of case-control studies in the International Head and Neck Cancer Epidemiology (INHANCE) Consortium. Int J Epidemiol. 2010;39:1091–102.

    Article  PubMed Central  Google Scholar 

  43. Nowak J, Weylandt KH, Habbel P, Wang J, Dignass A, Glickman JN, Kang JX. Colitis-associated colon tumorigenesis is suppressed in transgenic mice rich in endogenous n-3 fatty acids. Carcinogenesis. 2007;28:1991–5.

    Article  CAS  Google Scholar 

  44. Song M, Garrett WS, Chan AT. Nutrients, foods, and colorectal cancer prevention. Gastroenterology. 2015;148:1244–e12601216.

    Article  CAS  Google Scholar 

  45. Meyer MS, Applebaum KM, Furniss CS, Peters ES, Luckett BG, Smith JF, Bryan J, McClean MD, Marsit C, Kelsey KT. Human papillomavirus-16 modifies the association between fruit consumption and head and neck squamous cell carcinoma. Cancer Epidemiol Biomarkers Prev. 2008;17:3419–26.

    Article  PubMed Central  CAS  Google Scholar 

Download references

Acknowledgements

We sincerely appreciate the PLCO study group and PLCO participants. This research has been conducted using the PLCO resource (https://cdas.cancer.gov/plco/) under application number PLCO-1335.

Funding

This work was supported by National Natural Science Foundation of China (Grant Nos. 81860526). This research was funded by China Postdoctoral Science Foundation funded project (Grant Nos. 2021M703683).

Author information

Author notes

  1. Rong Wang and Haoyun Luo contributed equally to this work.

Authors and Affiliations

  1. Department of Radiation Oncology, Zhongshan City People’s Hospital, No.2, Sun Wen East Road, Zhongshan, 528400, Guangdong, China

    Rong Wang & Yijing Ye

  2. Department of Gastrointestinal Surgery, The Second Affiliated Hospital of Chongqing Medical University, Chongqing, China

    Haoyun Luo

  3. Department of Clinical Nutrition, The Second Affiliated Hospital of Chongqing Medical University, No.288 Tianwen Avenue, Nan’an District, Chongqing, 400010, China

    Ling Xiang

  4. Department of Chemotherapy Oncology, Zhongshan City People’s Hospital, No.2, Sun Wen East Road, Zhongshan, 528400, Guangdong, China

    Qijiu Chen

Authors
  1. Rong Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Haoyun Luo
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Yijing Ye
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Ling Xiang
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Qijiu Chen
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

R.W. and L.X. designed the research and applied for the original data, R.W., H.L. and Q.C. performed data collection, statistical analysis, and drafted the original manuscript. Y.Y., and L.X. assisted in statistical methodology. L.X. and Q.C. revised the manuscript.

Corresponding authors

Correspondence to Rong Wang, Ling Xiang or Qijiu Chen.

Ethics declarations

Ethical approval

The study is in line with the Helsinki Declaration. The PLCO Cancer Screening Trial design was approved by the Institutional Review Board of the NCI and each of the screening centers. Participants provided written informed consent prior to participation in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

12937_2024_1026_MOESM1_ESM.docx

Supplementary Material 2: Supplementary table 1. Criteria for determining the LFD scores.Supplementary table 2. Distribution of variables with missing data before and after imputation.

Supplementary Material 2: Supplementary table 4. Association of intake of PUFA, MUFA, or SFA with the risk of HNC.

Supplementary Material 3: Supplementary table 3. Association of LFD scores with the risk of HNC.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Wang, R., Luo, H., Ye, Y. et al. Adherence to a low-fat dietary pattern reduces head and neck cancer risk: evidence from the PLCO trial. Nutr J 23, 125 (2024). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12937-024-01026-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12937-024-01026-z

Keywords

Nutrition Journal

ISSN: 1475-2891

Contact us