Abstract
Objectives
Dietary Approaches to Stop Hypertension (DASH) pattern has been found to aid in the reduction of obesity, oxidative stress, and chronic inflammation, which are all strongly linked to the development of head and neck cancer (HNC). Nevertheless, no epidemiological studies have investigated the association between this dietary pattern and HNC risk. This study was conducted with the purpose of bridging this gap in knowledge.
Design
A prospective cohort study involving 98,459 American adults aged 55 years and older.
Setting and Participants
Data were drawn from the Prostate, Lung, Colorectal, and Ovarian (PLCO) Trial. In the present study, participants with dependable energy intake data who furnished baseline and dietary history information were identified as the study population.
Methods
Diet was assessed by food frequency questionnaires and the DASH score was calculated to assess each participant's adherence to DASH eating pattern. Cox proportional hazards models were used to calculate multivariable adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for the occurrence of HNC. To visualize the variation in cancer risk for HNC and its subtypes across the entire spectrum of DASH scores, restricted cubic spline plots were utilized. Additionally, a series of predefined subgroup analyses were performed to identify potential effect modifiers, and several sensitivity analyses were conducted to assess the stability of the findings.
Results
During a follow-up period of 871,879.6 person-years, 268 cases of HNC were identified, comprising 161 cases pertaining to oral cavity and pharynx cancers, as well as 96 cases of larynx cancer. In the fully adjusted model, adherence to the DASH diet was associated with a remarkable 57% reduction in the risk of HNC when comparing extreme quartiles (HR quartile 4 vs 1: 0.43; 95% CI: 0.28, 0.66; P for trend < 0.001). The restricted cubic spline plots demonstrated a linear dose-response relationship between the DASH score and the risk of HNC as well as its subtypes. Subgroup analysis revealed that the protective effect of the DASH diet against HNC was particularly pronounced in individuals with lower daily energy intake. The primary association remained robust in the sensitivity analysis.
Conclusions
In American middle-aged and older population, adherence to the DASH diet may help prevent HNC, particularly for individuals with lower daily energy intake.
Key words: Dietary Approaches to Stop Hypertension, head and neck cancer, epidemiology, cohort study, cancer prevention
Abbreviations
- BMI
body mass index
- BQ
baseline questionnaire
- CI
confidence interval
- DASH
dietary approaches to stop hypertension
- DHQ
dietary history questionnaire
- HPV
human papillomavirus
- HR
hazard ratio
- HNC
head and neck cancer
- NCI
national cancer institute
- PLCO
prostate, lung, colorectal, and ovarian
- SD
standard deviation
Introduction
Globally, head and neck cancer (HNC) ranks as the seventh most prevalent form of malignancy, with roughly 930,000 new cases and 460,000 fatalities in 2020, accounting for 4.9% of cancer diagnoses and 4.7% of cancer-related deaths, respectively (1). Despite recent advancements in treatment, HNC's prognosis remains dismal due to late detection and high recurrence rates (2). Additionally, treatment for HNC frequently causes structural, functional, and aesthetic deficits that significantly impact patients' quality of life (3). Hence, HNC prevention assumes greater importance for the public than clinical treatment, and identifying potential risk or protective factors is critical to its prevention.
Currently, established risk factors for HNC include obesity, tobacco exposure, alcohol consumption, heredity, and human papillomavirus (HPV) infection, which increases the risk of oropharyngeal cancer (4, 5). Meanwhile, a growing body of research has focused on the relationship between diet and HNC risk. Previous studies have reported that the intake of fruits, grains, and dietary fiber was associated with a reduced risk of HNC (6, 7, 8), while processed meat consumption may increase the risk (9). Nowadays, research on dietary patterns is becoming increasingly popular because dietary patterns research examines the diet as a whole, considering the complex interaction effects among components, and could provide more comprehensive dietary guidance to the public. For instance, there is evidence suggesting that adherence to a Western-style dietary pattern, characterized by frequent consumption of red meat, added sugars, and fried foods, has been positively associated with HNC incidence (10). Conversely, adhering to the Mediterranean diet has been shown to potentially lower the risk of developing HNC (11). It is of significant importance to continue identifying dietary patterns associated with the development of HNC, as this information can guide nutritional strategies for preventing HNC on a public health scale.
Dietary Approaches to Stop Hypertension (DASH) pattern is a dietary pattern that promotes the consumption of fruits, nuts and legumes, vegetables, grains, low-fat dairy products, while restricting the intake of sodium, sugar-sweetened beverages, and red and processed meat. It is composed of several elements believed to possess potential protective properties against HNC, such as fruits and grains, while simultaneously curtailing the intake of harmful components such as red and processed meat. Furthermore, previous evidence indicated that the DASH diet was associated with the reduction of long-term weight gain, oxidative stress, and chronic inflammation (12, 13). As previously mentioned, obesity is a significant risk factor for HNC, and oxidative damage and chronic inflammation are commonly shared pathways for various cancers, including HNC (14, 15). In actuality, the DASH diet has been found to be beneficial in preventing a variety of cancers, including but not limited to basal cell cancer (16), gastric cancer (17), and breast cancer (18). Nevertheless, the association between DASH diet and the risk of HNC remains unclear. To bridge this knowledge gap, we conducted this prospective study using data from the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial.
Methods
Study population
The raw data used in our study was sourced exclusively from the PLCO trial, which has been exhaustively detailed in the original study (19). This is a population-based prospective cohort study that aimed to investigate whether certain prespecified examination methods, such as PSA, chest X-ray, sigmoidoscopy, transvaginal ultrasound, and others, could enhance the prognosis of PLCO cancers. Between 1993 and 2001, ten screening centers located across the United States invited eligible participants to enroll in the trial. Exclusions applied to individuals with a history of PLCO cancers, those who had recently undergone PLCO cancer screening, or were undergoing cancer-related treatment (Supplementary Figure 1). Ultimately, the trial enrolled 154,887 participants aged 55–74 years who provided informed consent. Upon enrollment, the randomization procedure was initiated, whereby each participant was randomly allocated to either the control or intervention group. Participants in the control group received usual care, while those in the intervention group underwent a regular prespecified screening program for PLCO cancers. After randomization, several questionnaires were administered to obtain information on each participant. The Baseline Questionnaire (BQ) was used to gather voluntary reports of baseline risk factors, such as demographics and medical history. The Diet History Questionnaire (DHQ) comprised inquiries concerning the consumption of 137 food items and supplements over the previous year, including daily frequency, serving count, and other relevant details. Raw responses were processed into analysis-ready variables in terms of gram intake, pyramid servings, etc. Specifically, nutrient and food group intake quantities were computed by aligning the questionnaire responses with National Nutrient Database for Standard Reference of the USDA. This database furnished nutrient profiles for each food item based on national dietary data. Individual-level mean daily nutrient intakes were ascertained by multiplying the frequency weight of the chosen frequency category by the nutrient content of the selected portion size. The validity of the DHQ for assessing nutrient intake among middle-aged and older American adults has been confirmed in previous studies (20, 21). Following the information collection process, a long-term regular follow-up was conducted, during which data on cancer diagnoses were collected until 2009, and mortality data were collected until 2018 in the PLCO trial.
In accordance with our research objectives, certain participants were excluded from the PLCO trial (Supplementary Figure 1). These included those with incomplete BQ (n=4,918), those who submitted incomplete or invalid DHQ (n=38,462), those with a prior history of cancer before randomization (n=9,684), those who experienced an outcome event between randomization and DHQ completion (n=68), and male participants with daily calorie intake under 800 or over 4200 kcal, female participants with daily calorie intake under 600 or over 3500 kcal (n=3,296) (22). For the purpose of this study, the duration of follow-up time spanned from the DHQ completion to the emergence of any of the outcome events (Supplementary Figure 2). An outcome event was defined as the occurrence of HNC, death, loss to follow-up, or reaching the follow-up endpoint (December 31, 2009), whichever came first. Additionally, our research was conducted under a license granted by the National Cancer Institute (NCI, approval number: PLCO-1233).
Assessment of the DASH score
The DASH score was employed to assess each participant's adherence to the DASH diet. The calculation method of the DASH score was proposed by Dr. Fung et al. in 2008 and has been widely used in nutritional investigations (23, 24, 25). For each DASH component, participants were grouped into quintiles based on their consumption levels obtained from the DHQ. The scoring criteria for the beneficial components, including fruits, nuts and legumes, vegetables, grains, and low-fat dairy products, ranged from 1 (for participants in the lowest quintile) to 5 points (for those in the highest quintile). Conversely, for the detrimental components such as sodium, sugar-sweetened beverages, red and processed meat, the scores ranged from 1 (for those in the highest quintile) to 5 points (for those in the lowest quintile). Supplementary Table 1 presents the specific criteria used to determine the score for each component. The DASH score was derived by adding up the scores for all eight components, resulting in a range of 8 to 40. A higher DASH score indicated better adherence to the DASH diet.
Determination of HNC cases
An annual report was issued to each participant, requesting information on any cancer diagnosis. If a diagnosis was reported, the date of diagnosis, location, type of cancer, medical facility where the diagnosis was made, and contact information of the physician who made the diagnosis were requested. In the case of non-response, the research team made efforts to contact participants by mail or phone to confirm their cancer diagnosis and vital status. For each HNC case, researchers endeavored to verify the diagnosis and obtain details by reaching out to the diagnostic units or physicians. Furthermore, death certificates provided by relatives, friends, and physicians were utilized as sources of diagnostic information. Our analysis considered the following types of HNC: oral cavity (lip, tongue, salivary gland, floor of mouth), pharynx (nasopharynx, oropharynx, hypopharynx), nasal cavity and ear, and larynx.
Assessment of covariates
In order to carry out a comprehensive investigation into the relationship between the DASH diet and HNC risk, potential confounding factors were taken into account in subsequent analyses. These factors included gender, race, educational level, body mass index (BMI, calculated as weight divided by height squared in kg/m2), cigar and pipe smoking status, pack-years of cigarette smoking (number of packs per day multiplied by years smoked), history of hypertension, and family history of HNC, all of which were obtained from the BQ. Age, alcohol drinking intensity (g/day), and daily calorie intake (kcal/day) were obtained from the DHQ.
Statistical analysis
In epidemiological research, it is common to perform missing data imputation for participants with missing covariates, rather than simply excluding them (26). In our analysis, all missing covariate proportions were below 5%. Categorical variables were imputed using the mode, while continuous variables were imputed using the median (27). Supplementary Table 2 displays the distribution characteristics of covariates with missing values before and after imputation. All subsequent analyses were conducted on the full dataset after data imputation.
All participants were categorized into quartiles based on their respective DASH scores. The follow-up person-years for each quartile were calculated separately. The incidence rate was obtained by dividing HNC cases by the total follow-up person-years and multiplying by 1000, expressed as the number of cases per 1000 person-years. To investigate the potential association between the DASH score and the risk of HNC, three proportional hazards models were employed: the unadjusted model, model 1 which adjusted for demographic factors including age (as a continuous variable), gender (male vs female), and race (white vs non-white), and the fully adjusted model 2 which was considered the most comprehensive model in our analysis. Model 2 not only adjusted for demographic factors but also included additional adjustments for educational level (college below vs college/postgraduate), BMI (as a continuous variable), cigar smoking status (never vs current/former), pipe smoking status (never vs current/former), pack-years of cigarette smoking (never vs > 0 and ≤ 20 vs > 20), alcohol drinking intensity (g/day, never vs > 0 and ≤ 5 vs > 5 and ≤ 10 vs >10 and ≤ 20 vs > 20 and ≤ 30 vs > 30) (28), daily calorie intake (kcal/day, as a continuous variable), hypertension history (no vs yes), and family history of HNC (no vs yes/possible). It should be emphasized that the identification of adjusted covariates in the multivariate analysis was based on existing investigations on the risk of HNC (28, 29). The lowest quartile with the lowest DASH scores was defined as the reference group, and hazard ratios (HRs) and their corresponding 95% confidence intervals (CIs) were calculated for each of the other quartiles and compared to the reference group. To assess the trend between HNC risk and DASH score, we assigned the median DASH score of each quartile to all participants in the group and treated them as continuous variables. The trend test was conducted to assess the presence of a statistically significant trend between the risk of HNC and the DASH score, and P values were calculated accordingly. Simultaneously, we investigated the association between the DASH score and the risk of the two primary subtypes of HNC (oral cavity and pharynx, larynx) using the aforementioned method.
We utilized restricted cubic spline plots to depict the variability in cancer risk for HNC and its subtypes across the DASH score range. It is noteworthy that in this analysis, HRs were adjusted for the same confounding factors as in fully adjusted model 2. P values for nonlinearity were calculated to assess whether a non-linear dose-response relationship existed between the corresponding cancer risk and the DASH score. Furthermore, we examined the relationship between the intake of each DASH component (as a continuous variable) and the risk of HNC to strengthen our findings and identify components that may potentially contribute to the established association between the DASH diet and HNC risk.
A series of pre-determined subgroup analyses were conducted by stratifying for various factors including age (years, ≤ 65 vs > 65), gender (male vs female), BMI (kg/m2, ≤ 25 vs > 25), pack-years of cigarette smoking (never vs > 0 and ≤ 20 vs > 20), history of hypertension (no vs yes), and daily calorie intake (≤ median vs > median). P values for likelihood ratio tests were calculated to assess the significance of the interaction between the DASH score and each stratification factor. Several sensitivity analyses were conducted to confirm the robustness of our results. Specifically, we performed a repeat analysis after separately excluding the following groups of participants: (i) those with missing data, to eliminate any potential impact of data imputation on the results, (ii) those with a family history of HNC, as they may be more genetically susceptible to HNC, (iii) those with oropharyngeal cancer, as HPV infection may have an impact on their development, which were not adjusted for in our multivariate analysis, and (iv) HNC cases observed within the first 2 or 4 years of follow-up, in order to eliminate possible reverse causality.
The statistical analyses were conducted using R software (version 4.2.1). Statistical significance was defined as a two-tailed P value <0.05.
Results
Participant characteristics
A total of 98,459 participants were screened for eligibility in our study. The mean DASH score was 24.00 with a standard deviation of 4.62. Participants were stratified into quartiles based on their DASH scores: quartile 1 included scores ranging from 8–21 (n=29,523), quartile 2 included scores ranging from 22–24 (n=23,433), quartile 3 included scores ranging from 25–27 (n=22,564), and quartile 4 included scores ranging from 28–40 (n=22,939). The detailed baseline characteristics of the study population are displayed in Table 1. Participants were assigned to higher quartiles to indicate greater adherence to the DASH diet. As expected, the daily intake of beneficial components of the DASH diet increased with higher levels of DASH adherence, whereas the intake of sodium, sugar-sweetened beverages, red and processed meats exhibited an opposite trend. Demographically, participants in quartile 4 with the highest DASH scores (28–40) tended to be older (mean age 66.38 years) compared to those in quartile 1 with the lowest scores (8–21), who had a mean age of 64.51 years. The proportion of females was substantially higher in quartile 4 (66.36%) relative to quartile 1 (37.95%). Educational attainment was also greater among quartile 4, with 64.24% holding at least college education versus 51.91% of those in quartile 1. Behaviorally, the percentage of never smokers was higher in quartile 4 for cigarettes (56.77% vs 39.14%), cigars (92.13% vs 84.02%), and pipes (90.54% vs 82.68%). A greater proportion of quartile 4 consumed no alcohol (30.43% vs 25.26%). Mean BMI was lower among quartile 4 (26.14 kg/m2) compared to quartile 1 (27.95 kg/m2), as was average daily caloric intake (1690 kcal vs 1792 kcal).
Table 1.
Baseline characteristics of study population according to quartiles of DASH scores a
| Characteristics | Overall | Quartiles of DASH scores | P value | |||
|---|---|---|---|---|---|---|
| Quartile 1 (8–21) | Quartile 2 (22–24) | Quartile 3 (25–27) | Quartile 4 (28–40) | |||
| Number of participants | 98459 | 29523 | 23433 | 22564 | 22939 | |
| DASH score | 24.00±4.62 | 18.56±2.28 | 23.03±0.81 | 25.95±0.82 | 30.08±1.96 | 0.000b |
| Age | 65.52±5.73 | 64.51±5.57 | 65.51±5.71 | 65.98±5.69 | 66.38±5.79 | 0.000b |
| Gender | 0.000c | |||||
| Male | 47218 (47.96%) | 18320 (62.05%) | 11607 (49.53%) | 9574 (42.43%) | 7717 (33.64%%) | |
| Female | 51241 (52.04%) | 11203 (37.95%) | 11826 (50.47%) | 12990 (57.57%) | 15222 (66.36%) | |
| Race | <0.001c | |||||
| White | 91221 (92.65%) | 27219 (92.20%) | 21774 (92.92%) | 21069 (93.37%) | 21159 (92.24%) | |
| Non-white | 7238 (7.35%) | 2304 (7.80%) | 1659 (7.08%) | 1495 (6.63%) | 1780 (7.76%) | |
| Education level | <0.001c | |||||
| College below | 41535 (42.19%) | 14199 (48.09%) | 10144 (43.29%) | 8990 (39.84%) | 8202 (35.76%) | |
| College/Postgraduate | 56924 (57.81%) | 15324 (51.91%) | 13289 (56.71%) | 13574 (60.16%) | 14737 (64.24%) | |
| Body mass index (kg/m2) | 27.20±4.79 | 27.95±4.81 | 27.45±4.78 | 27.05±4.73 | 26.14±4.61 | 0.000b |
| Pack-years of cigarette smoking | 0.000c | |||||
| Never (0) | 47213 (47.95%) | 11556 (39.14%) | 10930 (46.64%) | 11705 (51.87%) | 13022 (56.77%) | |
| >0 and ≤20 | 20644 (20.97%) | 5700 (19.31%) | 4989 (21.29%) | 4864 (21.56%) | 5091 (22.19%) | |
| >20 | 30602 (31.08%) | 12267 (41.55%) | 7514 (32.07%) | 5995 (26.57%) | 4826 (21.04%) | |
| Cigar smoking | <0.001c | |||||
| Never | 86569 (87.92%) | 24806 (84.02%) | 20486 (87.42%) | 20144 (89.27%) | 21133 (92.13%) | |
| Current/Former | 11890 (12.08%) | 4717 (15.98%) | 2947 (12.58%) | 2420 (10.73%) | 1806 (7.87%) | |
| Pipe smoking | <0.001c | |||||
| Never | 84827 (86.15%) | 24411 (82.68%) | 19986 (85.29%) | 19661 (87.13%) | 20769 (90.54%) | |
| Current/Former | 13632 (13.85%) | 5112 (17.32%) | 3447 (14.71%) | 2903 (12.87%) | 2170 (9.46%) | |
| Alcohol drinking intensity (g/day) | <0.001c | |||||
| Never (0) | 26681 (27.10%) | 7457 (25.26%) | 6114 (26.09%) | 6130 (27.17%) | 6980 (30.43%) | |
| >0 and ≤5 | 38524 (39.13%) | 10943 (37.07%) | 9110 (38.88%) | 9030 (40.02%) | 9441 (41.16%) | |
| >5 and ≤10 | 9666 (9.82%) | 2951 (10.00%) | 2395 (10.22%) | 2167 (9.60%) | 2153 (9.39%) | |
| >10 and ≤20 | 9537 (9.69%) | 2900 (9.82%) | 2308 (9.85%) | 2278 (10.10%) | 2051 (8.94%) | |
| >20 and ≤30 | 7370 (7.49%) | 2325 (7.88%) | 1806 (7.71%) | 1723 (7.64%) | 1516 (6.61%) | |
| >30 | 6681 (6.79%) | 2947 (9.98%) | 1700 (7.25%) | 1236 (5.48%) | 798 (3.48%) | |
| History of hypertension | <0.001c | |||||
| No | 66641 (67.68%) | 19587 (66.34%) | 15651 (66.79%) | 15250 (67.59%) | 16153 (70.42%) | |
| Yes | 31818 (32.32%) | 9936 (33.66%) | 7782 (33.21%) | 7314 (32.41%) | 6786 (29.58%) | |
| Family history of HNC | <0.001c | |||||
| No | 94516 (96.00%) | 28148 (95.34%) | 22563 (96.29%) | 21700 (96.17%) | 22105 (96.36%) | |
| Yes/Possible | 3943 (4.00%) | 1375 (4.66%) | 870 (3.71%) | 864 (3.83%) | 834 (3.64%) | |
| Energy intake from diet (kcal/day) | 1728.71±658.04 | 1792.50±685.67 | 1709.94±679.96 | 1704.04±658.02 | 1690.05±589.63 | <0.001b |
| DASH components intake | ||||||
| Fruits (g/day) | 275.24±213.29 | 172.80±161.71 | 254.35±194.61 | 310.29±207.58 | 393.94±226.47 | 0.000b |
| Nuts and legumes (g/day) | 20.57±26.06 | 12.08±15.84 | 17.42±20.40 | 21.89±24.46 | 33.40±36.18 | 0.000b |
| Vegetables (g/day) | 284.83±181.87 | 210.68±138.99 | 261.72±159.58 | 304.33±176.62 | 384.70±206.03 | 0.000b |
| Grains (g/day) | 61.53±59.68 | 33.96±38.55 | 54.11±50.11 | 70.08±59.14 | 97.17±71.17 | 0.000b |
| Low-fat dairy (g/day) | 137.18±222.00 | 47.98±131.73 | 108.89±198.87 | 170.04±235.35 | 248.57±264.19 | 0.000b |
| Sodium from diet (mg/day) | 2728.47±1126.48 | 2788.05±1135.77 | 2708.86±1166.99 | 2712.09±1148.16 | 2687.92±1044.88 | <0.001b |
| Sugared beverages (g/day) | 398.08±463.51 | 535.76±578.37 | 401.38±442.44 | 344.63±389.73 | 270.07±314.11 | 0.000b |
| Red/processed meats (g/day) | 12.26±14.62 | 19.59±18.49 | 12.86±13.24 | 9.48±11.18 | 4.96±6.72 | 0.000b |
a. Values are means (standard deviation) for continuous variables and percentages for categorical variables. b. Group comparisons of continuous variables utilized analysis of variance (ANOVA). c. Categorical variables employed chi-squared tests to assess differences across quartiles..
Association between DASH scores and the risk of HNC
During a follow-up period of 871,879.6 person-years, a total of 268 cases of HNC were identified, including 125 cases of oral cavity cancer, 96 cases of larynx cancer, 19 cases of oropharynx cancer, 11 cases of hypopharynx cancer, 11 cases of nasal cavity and ear cancer, 4 cases of nasopharynx cancer and 2 cases of oral cavity and pharynx cancer not otherwise specified (NOS). The overall incidence rate of HNC was 0.31 cases per 1,000 person-years. The mean and standard deviation of the follow-up time were 8.84 and 1.94 years, respectively. In the unadjusted Cox model, participants in the highest quartile of DASH scores exhibited a significantly reduced risk of HNC compared with the lowest quartile (HR quartile 4 vs 1: 0.26; 95% CI: 0.17, 0.40; P for trend < 0.001). After adjusted for fully potential confounding factors, this inverse relationship was still observed (HR quartile 4 vs 1: 0.43; 95% CI: 0.28, 0.66; P for trend < 0.001). Furthermore, a similar inverse relationship was observed between the DASH score and the risk of HNC subtypes (For oral and pharynx cancer, HR quartile 4 vs 1: 0.46; 95% CI: 0.27, 0.77; P for trend= 0.002; For larynx cancer, HR quartile 4 vs 1: 0.37; 95% CI: 0.16, 0.82; P for trend= 0.017) (Table 2). Importantly, when repeated analysis was performed among participants with complete data (n=94,085), a similar association between DASH score and the incidence of HNC and its subtypes was observed (Supplementary Table 3).
Table 2.
Association of DASH scores with the risk of head and neck cancer and its subtypes a
| Quartiles of DASH score | No. of | No. of | Person-years | Hazard | ratio (95% confidence | interval) |
|---|---|---|---|---|---|---|
| Participants | Cases | Unadjusted | Model 1b | Model 2c | ||
| Head and Neck | ||||||
| Quartile 1 (8–21) | 29523 | 126 | 257050.1 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| Quartile 2 (22–24) | 23433 | 65 | 207269.2 | 0.64 (0.47, 0.86) | 0.71 (0.52, 0.95) | 0.78 (0.58, 1.05) |
| Quartile 3 (25–27) | 22564 | 50 | 200506.7 | 0.51 (0.37, 0.70) | 0.60 (0.43, 0.84) | 0.70 (0.50, 0.98) |
| Quartile 4 (28–40) | 22939 | 27 | 207053.6 | 0.26 (0.17, 0.40) | 0.35 (0.23, 0.53) | 0.43 (0.28, 0.66) |
| P for trend | <0.001 | <0.001 | <0.001 | |||
| Oral cavity and Pharynx | ||||||
| Quartile 1 (8–21) | 29477 | 70 | 256831.7 | 1.00 (reference) | 1.000 (reference) | 1.00 (reference) |
| Quartile 2 (22–24) | 23413 | 43 | 207195.2 | 0.76 (0.52, 1.11) | 0.80 (0.55, 1.17) | 0.86 (0.58, 1.26) |
| Quartile 3 (25–27) | 22543 | 29 | 200400.8 | 0.53 (0.34, 0.82) | 0.58 (0.37, 0.90) | 0.65 (0.42, 1.01) |
| Quartile 4 (28–40) | 22919 | 19 | 206926.5 | 0.34 (0.20, 0.56) | 0.39 (0.23, 0.66) | 0.46 (0.27, 0.77) |
| P for trend | <0.001 | <0.001 | 0.002 | |||
| Larynx | ||||||
| Quartile 1 (8–21) | 29466 | 53 | 256843.2 | 1.00 (reference) | 1.00 (reference) | 1.00 (reference) |
| Quartile 2 (22–24) | 23391 | 16 | 207107.0 | 0.37 (0.21, 0.65) | 0.45 (0.26, 0.79) | 0.52 (0.30, 0.92) |
| Quartile 3 (25–27) | 22531 | 20 | 200305.7 | 0.48 (0.29, 0.81) | 0.66 (0.39, 1.11) | 0.83 (0.49, 1.41) |
| Quartile 4 (28–40) | 22899 | 7 | 206767.0 | 0.16 (0.07, 0.36) | 0.27 (0.12, 0.60) | 0.37 (0.16, 0.82) |
| P for trend | <0.001 | <0.001 | 0.017 | |||
a. Hazard ratio was calculated using Cox proportional hazard regression models, P values were calculated from significance testing for the underlying linear trend in Cox models. b. Adjusted for age (years), gender (male, female) and race (white, non-white). c. Adjusted for model 1 plus education level (college below, college/postgraduate), alcohol drinking intensity (0, >0 and ≤5, >5 and ≤10, >10 and ≤20, >20 and ≤30, >30), cigar smoking (never, current/former), pipe smoking (never, current/former), pack-years cigarettes smoking (0, >0 and ≤20, >20), body mass index (continuous), history of hypertension (no, yes), family history of head and neck cancer (no, yes/possible) and energy intake from diet (continuous).
Additional analyses
Restricted cubic spline plots were utilized to depict the variations in the incidence of HNC, oral and pharynx cancer, and larynx cancer across the spectrum of DASH scores. As shown in Figure 1, with increasing DASH scores, the risk of these cancers decreased, showing a linear dose-response manner (all P values for nonlinearity > 0.05). In subgroup analyses, the inverse association with HNC appeared stronger for male (HR quartile 4 vs 1: 0.44; 95% CI: 0.26, 0.73; P for trend= 0.001) versus female (HR quartile 4 vs 1: 0.50; 95% CI: 0.21, 1.17; P for trend= 0.026) (Table 3, P for interaction = 0.015). Adherence to the DASH diet appeared to confer a comparatively lower HNC risk for participants with lower (≤ median) daily energy intake (HR quartile 4 vs 1: 0.28; 95% CI: 0.11, 0.68; P for trend= 0.017) versus participants with higher (> median) daily energy intake (HR quartile 4 vs 1: 0.54; 95% CI: 0.33, 0.88; P for trend= 0.002) (Table 3, P for interaction = 0.008). However, no significant interaction was found when stratifying for age, BMI, pack-years of cigarette smoking, and history of hypertension (Table 3, all P for interaction > 0.05). Furthermore, the primary association between DASH score and the risk of HNC remained robust in a series of sensitivity analyses (Table 4).
Figure 1.
Dose-response analysis on the association of the DASH scores with the risk of head and neck cancer (A), oral and pharynx cancer (B), and larynx cancer (C). The likelihood ratio tests were employed in the analysis.
Table 3.
Subgroup analyses on the association of DASH scores with the risk of head and neck cancer a
| Subgroup variable | Participants/cases | Person-years | Hazard Ratio (95% Confidence Interval) by DASH Scores b | P trend | P interaction | |||
|---|---|---|---|---|---|---|---|---|
| Quartile 1 (8–21) | Quartile 2 (22–24) | Quartile 3 (25–27) | Quartile 4 (28–40) | |||||
| Age (years) | 0.290 | |||||||
| ≤65 | 51302/121 | 455497.2 | 1.00 (reference) | 0.66 (0.42, 1.06) | 0.56 (0.32, 0.98) | 0.56 (0.30, 1.06) | 0.014 | |
| >65 | 47157/147 | 416382.4 | 1.00 (reference) | 0.90 (0.60, 1.35) | 0.82 (0.53, 1.26) | 0.38 (0.21, 0.67) | 0.002 | |
| Gender | 0.015 | |||||||
| Male | 47218/213 | 413369.4 | 1.00 (reference) | 0.63 (0.44, 0.90) | 0.79 (0.55, 1.14) | 0.44 (0.26, 0.73) | 0.001 | |
| Female | 51241/55 | 458510.2 | 1.00 (reference) | 1.50 (0.78, 2.89) | 0.50 (0.21, 1.18) | 0.50 (0.21, 1.17) | 0.026 | |
| Body mass index (kg/m2) | 0.201 | |||||||
| ≤25 | 33582/91 | 300270.0 | 1.00 (reference) | 0.62 (0.36, 1.06) | 0.53 (0.30, 0.96) | 0.30 (0.14, 0.62) | <0.001 | |
| >25 | 64877/177 | 571609.6 | 1.00 (reference) | 0.88 (0.61, 1.27) | 0.82 (0.55, 1.24) | 0.57 (0.33, 0.96) | 0.039 | |
| Pack-years of cigarette smoking | 0.222 | |||||||
| Never (0) | 47213/62 | 424378.8 | 1.00 (reference) | 1.07 (0.57, 1.98) | 0.86 (0.44, 1.68) | 0.27 (0.10, 0.74) | 0.015 | |
| >0 and ≤20 | 20644/45 | 184035.8 | 1.00 (reference) | 0.67 (0.32, 1.41) | 0.47 (0.19, 1.12) | 0.49 (0.20, 1.20) | 0.058 | |
| >20 | 30602/161 | 263465.0 | 1.00 (reference) | 0.73 (0.49, 1.09) | 0.76 (0.49, 1.18) | 0.56 (0.32, 0.99) | 0.025 | |
| History of hypertension | 0.502 | |||||||
| No | 66641/183 | 594960.3 | 1.00 (reference) | 0.86 (0.59, 1.24) | 0.82 (0.54, 1.23) | 0.59 (0.36, 0.97) | 0.038 | |
| Yes | 31818/85 | 276919.3 | 1.00 (reference) | 0.65 (0.38, 1.10) | 0.52 (0.29, 0.96) | 0.20 (0.08, 0.50) | <0.001 | |
| Energy intake from diet (kcal/day) | 0.008 | |||||||
| ≤median | 49230/96 | 436424.3 | 1.00 (reference) | 1.23 (0.75, 2.00) | 1.03 (0.60, 1.76) | 0.28 (0.11, 0.68) | 0.017 | |
| >median | 49229/172 | 435455.3 | 1.00 (reference) | 0.59 (0.40, 0.88) | 0.57 (0.36, 0.88) | 0.54 (0.33, 0.88) | 0.002 | |
a. Hazard ratio was calculated using Cox proportional hazard regression models, P trend was calculated from significance testing for the underlying linear trend in Cox models, P interaction for likelihood ratio tests was calculated from significance testing of interaction terms in Cox models. b. Hazard ratios were adjusted for age (years), gender (male, female), race (white, non-white), education level (college below, college/postgraduate), alcohol drinking intensity (0, >0 and ≤5, >5 and ≤10, >10 and ≤20, >20 and ≤30, >30), cigar smoking (never, current/former), pipe smoking (never, current/former), pack-years cigarettes smoking (0, >0 and ≤20, >20), body mass index (continuous), history of hypertension (no, yes), family history of head and neck cancer (no, yes/possible) and energy intake from diet (continuous).
Table 4.
Sensitivity analyses on the association of DASH scores with the risk of head and neck cancer a
| No. of Participants | No. of Cases | HR Quartile 4 vs. Quartile 1 (95% confidence interval)b | P trend | |
|---|---|---|---|---|
| Categories | ||||
| Excluded cases observed within the first 2 years of follow-up | 98407 | 216 | 0.47 (0.29, 0.75) | 0.002 |
| Excluded cases observed within the first 4 years of follow-up | 98341 | 150 | 0.47 (0.27, 0.83) | 0.011 |
| Excluded participants with family history of head and neck cancer c | 94516 | 251 | 0.46 (0.30, 0.71) | <0.001 |
| Exclude participants with missing data | 94085 | 254 | 0.42 (0.27, 0.66) | <0.001 |
| Excluded participants with oropharyngeal cancer | 98440 | 249 | 0.42 (0.27, 0.66) | <0.001 |
| Excluded participants with nasopharynx cancer | 98455 | 264 | 0.44 (0.29, 0.68) | <0.001 |
a. Hazard ratio was calculated using Cox proportional hazard regression models, P trend was calculated from significance testing for the underlying linear trend in Cox models. b. Hazard ratios were adjusted for age (years), gender (male, female), race (white, non-white), education level (college below, college/postgraduate), alcohol drinking intensity (0, >0 and ≤5, >5 and ≤10, >10 and ≤20, >20 and ≤30, >30), cigar smoking (never, current/former), pipe smoking (never, current/former), pack-years cigarettes smoking (0, >0 and ≤20, >20), body mass index (continuous), history of hypertension (no, yes), family history of head and neck cancer (no, yes/possible) and energy intake from diet (continuous). c. Hazard ratio was not adjusted for family history of head and neck cancer.
Individual components and the risk of HNC
An exploration into the association between the consumption of each DASH component and the susceptibility to HNC was also undertaken. According to Supplementary Table 4, a reduced risk of HNC was found to be associated with higher intake of beneficial components of fruits (HR quartile 4 vs 1: 0.40; 95% CI: 0.27, 0.60; P for trend < 0.001) and grains (HR quartile 4 vs 1: 0.53; 95% CI: 0.38, 0.75; P for trend < 0.001). For red and processed meat, participants in the highest quartile of consumption demonstrated a 74% increased risk of HNC compared to those in the lowest quartile, despite the lack of statistical significance observed in trend tests (HR quartile 4 vs 1: 1.74; 95% CI: 1.11, 2.72; P for trend= 0.207). No significant associations were observed between the intake of the remaining components and the risk of HNC.
Discussion
In this large prospective cohort analysis utilizing data from the PLCO trial, we discovered that adherence to the DASH diet significantly reduced the risk of HNC and its subtypes. The restricted cubic spline plots showed a consistent decline in the risk of HNC and its subtypes with increasing DASH scores, which followed a linear dose-response manner. Subgroup analysis showed that this inverse relationship was more evident among participants with lower daily calorie intake. Our results were robust, as a series of sensitivity analyses revealed no significant changes in the primary association.
The DASH diet was originally formulated and examined in a clinical trial financed by the National Institutes of Health. The findings of the trial revealed that individuals who closely adhered to the DASH diet demonstrated a substantial reduction in blood pressure in comparison to those in the control group (30). Despite its name being linked with hypertension, the DASH diet has been found to provide numerous benefits, including weight loss, enhanced insulin sensitivity, and regulated metabolism (31). Additionally, this dietary pattern has been found to be effective in preventing various types of cancer. A nested case-control study revealed that the DASH diet was linked to a 68% reduction in basal cell carcinoma risk when comparing extreme quintiles (16). Another hospital-based case-control study demonstrated that, after adjusting for potential confounders, a high level of adherence to the DASH diet was associated with a 54% decrease in gastric cancer risk (17). The DASH diet has also been shown to have a protective effect against highly aggressive prostate cancer (32). A meta-analysis revealed a significant inverse relationship between adherence to the DASH diet and breast cancer risk (18). To our knowledge, no published research has explored the association between the DASH diet and the risk of HNC. Our study contributes to filling this knowledge gap by demonstrating a significant reduction in the risk of HNC and its subtypes among individuals most adherent to the DASH diet, indicating a protective effect of the DASH diet against HNC. Additionally, the observed linear dose-response relationship implies that the protective effect increases progressively as the DASH score increases. Therefore, our findings underscore the significance of adhering to the DASH diet to the highest degree in order to efficiently prevent HNC.
In analyses of individual components, we found that increased consumption of fruits and grains was linked to a reduced risk of HNC, indicating that they may be potential contributors to the observed inverse association between the DASH diet and HNC risk. Notably, the protective effects of fruits and grains against HNC have been consistently reported in previous studies (28), providing further support to our findings. Nevertheless, it is still not advisable to blindly increase the consumption of fruits and grains, because the DASH diet is a comprehensive eating pattern composed of a variety of foods. An excessive emphasis on increasing the intake of specific food items may inevitably result in a decrease in the consumption of other components. This imbalance could ultimately lead to a reduction in the DASH score, which may not be beneficial for the prevention of HNC. Furthermore, it is important to recognize that the interactions and synergies among different components are not captured in analyses focusing on individual components alone (25). Therefore, adhering to the DASH diet in a comprehensive manner, rather than solely emphasizing the increase in specific components, is more likely to contribute to effective prevention of HNC.
The reduced risk of HNC with the DASH diet may be attributed to the following mechanisms: (i) The DASH diet limits the consumption of red and processed meat, which are rich in carcinogens such as aromatic amines and nitrites (33). Moreover, the diet encourages an increase in fiber intake, which has been proven to bind with carcinogens, thereby reducing their contact with the epithelial cells of the oral cavity, pharynx, and larynx (8). (ii) Adherence to the DASH diet has been observed to be associated with a lower BMI (12). Considering obesity is a recognized risk factor for HNC, it is plausible that the protective effect of the DASH diet against HNC may be partly mediated through its ability to promote weight loss (4). (iii) Previous studies have demonstrated that the DASH diet has the potential to alleviate systemic chronic inflammation (34), which has been associated with the development of various types of malignancies, including HNC (35). (iv) There is strong evidence linking oxidative stress to the development of HNC. The DASH diet, which is rich in antioxidants such as vitamin C, vitamin E, flavonoids, and beta-carotene, has been found to provide substantial antioxidant activity (36). Studies have indicated that individuals who adhere to the DASH diet exhibit lower levels of oxidative stress and experience less oxidative damage (37), indicating the potential of this dietary pattern in reducing the risk of HNC.
Our subgroup analysis indicated that the protective effect of the DASH diet against HNC was more prominent in participants with lower daily energy intake. One possible explanation is that the DASH diet counteracted the detrimental effects of an excessively calorie-restricted diet. Individuals with reduced dietary energy intake often exhibit a tendency to curtail their food consumption, resulting in lower intake of essential nutrients and an increased risk of nutrient deficiencies, which are associated with an increased risk of developing HNC (38, 39). The DASH diet, as a comprehensive dietary pattern, provides adequate amounts of essential nutrients, including antioxidant substances, which may contribute to its protective effect against HNC. Furthermore, calorie-restricted diets are widely recognized to facilitate weight loss, anti-inflammatory responses, and anti-oxidative damage, which are in line with the health-promoting effects of the DASH diet (40). Previous investigations have demonstrated that the combination of the DASH diet and a calorie-restricted diet can produce a synergistic effect, leading to more notable advantages such as weight loss, decreased levels of blood lipids, improved insulin sensitivity, and, possibly, enhancement of the anticancer potential (41, 42).
Our study boasts notable strengths. Most significantly, it is the first to establish that adherence to the DASH diet can play a preventative role against HNC, particularly for individuals with lower daily energy intake. Additionally, our use of dietary pattern surveys can provide the public with more comprehensive dietary guidance. In addition, the population-based prospective study design, the sufficiently long follow-up time, and the robust study results presented in sensitivity analysis all contribute to the reliability of our findings.
However, several limitations in our study must be acknowledged. Firstly, the dietary information collected through the DHQ relied on self-reported data, which could be subject to non-differential bias. In addition, this memory-based approach for assessing dietary intake over the past year is subject to recall bias, with participants potentially underreporting or selectively forgetting their dietary behaviors over the lengthy recall period. This may lead to misclassification of dietary exposures, thus biasing the observed association between diet and disease risk. Secondly, the study population consisted of Americans aged 55–74, and thus the findings may not be generalizable to other populations. Thirdly, while we adjusted for a range of confounding factors in the multivariate analysis, some residual confounders, such as HPV infection were not considered due to unavailability of data. HPV infection is an established risk factor for oropharyngeal cancer (43), but in our study, there were only 19 oropharyngeal cases included in the HNC cases, which may have a limited impact on our results. Furthermore, when we excluded oropharyngeal cases in the sensitivity analysis, the findings were largely unchanged. Fourthly, our study only collected dietary information at baseline and did not consider changes in participants' dietary habits over time, which may have introduced non-differential bias. Nonetheless, in nutritional epidemiology, it is commonly accepted that adults' dietary habits remain relatively stable in the short term (44). Furthermore, previous research has suggested that using baseline dietary data tends to yield weaker associations compared to using cumulative averages (45).
Conclusions
In American middle-aged and older population, adherence to the DASH diet may aid in the prevention of head and neck cancer, particularly for individuals with lower daily energy intake. Further studies encompassing a larger population are required to validate the findings.
Acknowledgments
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-1233.
Contributor Information
Ling Xiang, Email: 306359@hospital.cqmu.edu.cn.
Linglong Peng, Email: penglinglong_cqmu@cqmu.edu.cn.
Authors' contributions:
ZZ, methodology, investigation, software, and writing - original draft; YW, LX and LP, conceptualization, supervision, validation, and writing - review & editing; MY, project administration, data curation, and visualization; HG, formal analysis and resources. All authors read and approved the final manuscript.
Funding:
This work was supported by Natural Science Foundation Project of Chongqing, Chongqing Science and Technology Commission, China [cstc2021jcyjmsxmX0153 (Linglong Peng)], and [CSTB2022NSCQ-MSX1005 (Haitao Gu)].
Availability of data and materials:
The data that support the findings of this study are available from the National Cancer Institute but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of the National Cancer Institute.
Ethics approval and consent to participate:
The PLCO trial received approval from the Institutional Review Board of the NCI and was conducted in accordance with the principles outlined in the Declaration of Helsinki. Prior to their participation in the trial, all participants provided informed consent for inclusion.
Competing interests:
All authors declare that they have no competing interests.
Electronic Supplementary Material
Supplementary material is available in the online version of this article at https://doi.org/10.1007/s12603-023-2009-7.
Supplementary material, approximately 28.4 MB.
References
- 1.Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA: a cancer journal for clinicians. 2021;71(3):209–249. doi: 10.3322/caac.21660. PubMed PMID: 33538338. [DOI] [PubMed] [Google Scholar]
- 2.Giraldi L, Leoncini E, Pastorino R, Wünsch-Filho V, de Carvalho M, Lopez R, et al. Alcohol and cigarette consumption predict mortality in patients with head and neck cancer: a pooled analysis within the International Head and Neck Cancer Epidemiology (INHANCE) Consortium. Annals of oncology: official journal of the European Society for Medical Oncology. 2017;28(11):2843–2851. doi: 10.1093/annonc/mdx486. 10.1093/annonc/mdx486 PubMed PMID: 28945835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Partington EJ, Moore LS, Kahmke R, Warram JM, Carroll W, Rosenthal EL, et al. Laser-Assisted Indocyanine Green Dye Angiography for Postoperative Fistulas After Salvage Laryngectomy. JAMA otolaryngology— head & neck surgery. 2017;143(8):775–781. doi: 10.1001/jamaoto.2017.0187. 10.1001/jamaoto.2017.0187 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Wang K, Yu XH, Tang YJ, Tang YL, Liang XH. Obesity: An emerging driver of head and neck cancer. Life sciences. 2019;233:116687. doi: 10.1016/j.lfs.2019.116687. 10.1016/j.lfs.2019.116687 PubMed PMID: 31348948. [DOI] [PubMed] [Google Scholar]
- 5.Chow LQM. Head and Neck Cancer. The New England journal of medicine. 2020;382(1):60–72. doi: 10.1056/NEJMra1715715. 10.1056/NEJMra1715715 PubMed PMID: 31893516. [DOI] [PubMed] [Google Scholar]
- 6.Maasland DH, van den Brandt PA, Kremer B, Goldbohm RA, Schouten LJ. Consumption of vegetables and fruits and risk of subtypes of head-neck cancer in the Netherlands Cohort Study. International journal of cancer. 2015;136(5):E396–E409. doi: 10.1002/ijc.29219. 10.1002/ijc.29219 PubMed PMID: 25220255. [DOI] [PubMed] [Google Scholar]
- 7.Lam TK, Cross AJ, Freedman N, Park Y, Hollenbeck AR, Schatzkin A, et al. Dietary fiber and grain consumption in relation to head and neck cancer in the NIH-AARP Diet and Health Study. Cancer causes & control: CCC. 2011;22(10):1405–1414. doi: 10.1007/s10552-011-9813-9. 10.1007/s10552-011-9813-9 PubMed PMID: 21785948. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kawakita D, Lee YA, Turati F, Parpinel M, Decarli A, Serraino D, et al. Dietary fiber intake and head and neck cancer risk: A pooled analysis in the International Head and Neck Cancer Epidemiology consortium. International journal of cancer. 2017;141(9):1811–1821. doi: 10.1002/ijc.30886. 10.1002/ijc.30886 PubMed PMID: 28710831; PMCID 5797849. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Perloy A, Maasland DHE, van den Brandt PA, Kremer B, Schouten LJ. Intake of meat and fish and risk of head-neck cancer subtypes in the Netherlands Cohort Study. Cancer causes & control: CCC. 2017;28(6):647–656. doi: 10.1007/s10552-017-0892-0. 10.1007/s10552-017-0892-0 PubMed PMID: 28382514. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Dalmartello M, Decarli A, Ferraroni M, Bravi F, Serraino D, Garavello W, et al. Dietary patterns and oral and pharyngeal cancer using latent class analysis. International journal of cancer. 2020;147(3):719–727. doi: 10.1002/ijc.32769. 10.1002/ijc.32769 PubMed PMID: 31677269. [DOI] [PubMed] [Google Scholar]
- 11.Saka-Herrán C, Pereira-Riveros T, Jané-Salas E, López-López J. Association between the Mediterranean Diet and Vitamin C and the Risk of Head and Neck Cancer. Nutrients. 2023;15(13) doi: 10.3390/nu15132846. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mozaffarian D. Dietary and Policy Priorities for Cardiovascular Disease, Diabetes, and Obesity: A Comprehensive Review. Circulation. 2016;133(2):187–225. doi: 10.1161/CIRCULATIONAHA.115.018585. 10.1161/CIRCULATIONAHA.115.018585 PubMed PMID: 26746178; PMCID 4814348. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Aleksandrova K, Koelman L, Rodrigues CE. Dietary patterns and biomarkers of oxidative stress and inflammation: A systematic review of observational and intervention studies. Redox biology. 2021;42:101869. doi: 10.1016/j.redox.2021.101869. 10.1016/j.redox.2021.101869 PubMed PMID: 33541846; PMCID 8113044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Ambatipudi S, Langdon R, Richmond RC, Suderman M, Koestler DC, Kelsey KT, et al. DNA methylation derived systemic inflammation indices are associated with head and neck cancer development and survival. Oral oncology. 2018;85:87–94. doi: 10.1016/j.oraloncology.2018.08.021. 10.1016/j.oraloncology.2018.08.021 PubMed PMID: 30220325; PMCID 6156796. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Dequanter D, Dok R, Nuyts S. Basal oxidative stress ratio of head and neck squamous cell carcinomas correlates with nodal metastatic spread in patients under therapy. OncoTargets and therapy. 2017;10:259–263. doi: 10.2147/OTT.S118980. 10.2147/OTT.S118980 PubMed PMID: 28123307; PMCID 5234556. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Leone A, Martínez-González M, Martin-Gorgojo A, Sánchez-Bayona R, De Amicis R, Bertoli S, et al. Mediterranean diet, Dietary Approaches to Stop Hypertension, and Pro-vegetarian dietary pattern in relation to the risk of basal cell carcinoma: a nested case-control study within the Seguimiento Universidad de Navarra (SUN) cohort. The American journal of clinical nutrition. 2020;112(2):364–372. doi: 10.1093/ajcn/nqaa127. 10.1093/ajcn/nqaa127 PubMed PMID: 32492135. [DOI] [PubMed] [Google Scholar]
- 17.Toorang F, Sasanfar B, Hadji M, Esmaillzadeh A, Zendehdel K. Adherence to “dietary approaches to stop hypertension” eating plan in relation to gastric cancer. Nutrition journal. 2020;19(1):40. doi: 10.1186/s12937-020-00560-w. 10.1186/s12937-020-00560-w PubMed PMID: 32393262; PMCID 7216586. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Shu L, Huang YQ, Zhang XY, Zheng PF, Zhu Q, Zhou JY. Adherence to the Dietary Approaches to Stop Hypertension diet reduces the risk of breast cancer: A systematic review and meta-analysis. Frontiers in nutrition. 2022;9:1032654. doi: 10.3389/fnut.2022.1032654. 10.3389/fnut.2022.1032654 PubMed PMID: 36698472. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Andriole GL, Crawford ED, Grubb RL, 3rd, Buys SS, Chia D, Church TR, et al. Mortality results from a randomized prostate-cancer screening trial. The New England journal of medicine. 2009;360(13):1310–1319. doi: 10.1056/NEJMoa0810696. 10.1056/NEJMoa0810696 PubMed PMID: 19297565; PMCID 2944770. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Subar AF, Thompson FE, Kipnis V, Midthune D, Hurwitz P, McNutt S, et al. Comparative validation of the Block, Willett, and National Cancer Institute food frequency questionnaires: the Eating at America's Table Study. American journal of epidemiology. 2001;154(12):1089–1099. doi: 10.1093/aje/154.12.1089. 10.1093/aje/154.12.1089 PubMed PMID: 11744511. [DOI] [PubMed] [Google Scholar]
- 21.Prentice RL, Mossavar-Rahmani Y, Huang Y, Van Horn L, Beresford SA, Caan B, et al. Evaluation and comparison of food records, recalls, and frequencies for energy and protein assessment by using recovery biomarkers. American journal of epidemiology. 2011;174(5):591–603. doi: 10.1093/aje/kwr140. 10.1093/aje/kwr140 PubMed PMID: 21765003; PMCID 3202154. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Tabung FK, Liu L, Wang W, Fung TT, Wu K, Smith-Warner SA, et al. Association of Dietary Inflammatory Potential With Colorectal Cancer Risk in Men and Women. JAMA oncology. 2018;4(3):366–373. doi: 10.1001/jamaoncol.2017.4844. 10.1001/jamaoncol.2017.4844 PubMed PMID: 29346484; PMCID 5844836. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Fung TT, Chiuve SE, McCullough ML, Rexrode KM, Logroscino G, Hu FB. Adherence to a DASH-style diet and risk of coronary heart disease and stroke in women. Archives of internal medicine. 2008;168(7):713–720. doi: 10.1001/archinte.168.7.713. 10.1001/archinte.168.7.713 PubMed PMID: 18413553. [DOI] [PubMed] [Google Scholar]
- 24.Mattei J, Bigornia SJ, Sotos-Prieto M, Scott T, Gao X, Tucker KL. The Mediterranean Diet and 2-Year Change in Cognitive Function by Status of Type 2 Diabetes and Glycemic Control. Diabetes care. 2019;42(8):1372–1379. doi: 10.2337/dc19-0130. 10.2337/dc19-0130 PubMed PMID: 31123154; PMCID 6647047. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Ma Y, Yang W, Simon TG, Smith-Warner SA, Fung TT, Sui J, et al. Dietary Patterns and Risk of Hepatocellular Carcinoma Among U.S. Men and Women. Hepatology (Baltimore, Md) 2019;70(2):577–586. doi: 10.1002/hep.30362. 10.1002/hep.30362 PubMed PMID: 30506561. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Zhong GC, Li QJ, Yang PF, Wang YB, Hao FB, Wang K, et al. Low-carbohydrate diets and the risk of pancreatic cancer: a large prospective cohort study. Carcinogenesis. 2021;42(5):724–732. doi: 10.1093/carcin/bgab006. 10.1093/carcin/bgab006 PubMed PMID: 33480980. [DOI] [PubMed] [Google Scholar]
- 27.O'Brien EC, Wu J, Zhao X, Schulte PJ, Fonarow GC, Hernandez AF, et al. Healthcare Resource Availability, Quality of Care, and Acute Ischemic Stroke Outcomes. Journal of the American Heart Association. 2017;6(2) doi: 10.1161/jaha.116.003813. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kawakita D, Lee YA, Gren LH, Buys SS, La Vecchia C, Hashibe M. Fiber intake and the risk of head and neck cancer in the prostate, lung, colorectal and ovarian (PLCO) cohort. International journal of cancer. 2019;145(9):2342–2348. doi: 10.1002/ijc.32162. 10.1002/ijc.32162 PubMed PMID: 30693489; PMCID 7771385. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Kawakita D, Lee YA, Gren LH, Buys SS, La Vecchia C, Hashibe M. The impact of folate intake on the risk of head and neck cancer in the prostate, lung, colorectal, and ovarian cancer screening trial (PLCO) cohort. British journal of cancer. 2018;118(2):299–306. doi: 10.1038/bjc.2017.383. 10.1038/bjc.2017.383 PubMed PMID: 29161239. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Appel LJ, Moore TJ, Obarzanek E, Vollmer WM, Svetkey LP, Sacks FM, et al. A clinical trial of the effects of dietary patterns on blood pressure. DASH Collaborative Research Group. The New England journal of medicine. 1997;336(16):1117–1124. doi: 10.1056/NEJM199704173361601. 10.1056/NEJM199704173361601 PubMed PMID: 9099655. [DOI] [PubMed] [Google Scholar]
- 31.Lari A, Sohouli MH, Fatahi S, Cerqueira HS, Santos HO, Pourrajab B, et al. The effects of the Dietary Approaches to Stop Hypertension (DASH) diet on metabolic risk factors in patients with chronic disease: A systematic review and meta-analysis of randomized controlled trials. Nutrition, metabolism, and cardiovascular diseases: NMCD. 2021;31(10):2766–2778. doi: 10.1016/j.numecd.2021.05.030. 10.1016/j.numecd.2021.05.030 PubMed PMID: 34353704. [DOI] [PubMed] [Google Scholar]
- 32.Schneider L, Su LJ, Arab L, Bensen JT, Farnan L, Fontham ETH, et al. Dietary patterns based on the Mediterranean diet and DASH diet are inversely associated with high aggressive prostate cancer in PCaP. Annals of epidemiology. 2019;29:16–22.e1. doi: 10.1016/j.annepidem.2018.08.012. 10.1016/j.annepidem.2018.08.012 PubMed PMID: 30268488. [DOI] [PubMed] [Google Scholar]
- 33.Mimmler M, Peter S, Kraus A, Stroh S, Nikolova T, Seiwert N, et al. DNA damage response curtails detrimental replication stress and chromosomal instability induced by the dietary carcinogen PhIP. Nucleic acids research. 2016;44(21):10259–10276. doi: 10.1093/nar/gkw791. 10.1093/nar/gkw791 PubMed PMID: 27599846; PMCID 5137439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Frasca D, Diaz A, Romero M, Blomberg BB. Ageing and obesity similarly impair antibody responses. Clinical and experimental immunology. 2017;187(1):64–70. doi: 10.1111/cei.12824. 10.1111/cei.12824 PubMed PMID: 27314456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Li XH, Chang H, Xu BQ, Tao YL, Gao J, Chen C, et al. An inflammatory biomarker-based nomogram to predict prognosis of patients with nasopharyngeal carcinoma: an analysis of a prospective study. Cancer medicine. 2017;6(1):310–319. doi: 10.1002/cam4.947. 10.1002/cam4.947 PubMed PMID: 27860387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Chen C, Ahn EH, Kang SS, Liu X, Alam A, Ye K. Gut dysbiosis contributes to amyloid pathology, associated with C/EBPβ/AEP signaling activation in Alzheimer's disease mouse model. Science advances. 2020;6(31):eaba0466. doi: 10.1126/sciadv.aba0466. 10.1126/sciadv.aba0466 PubMed PMID: 32832679; PMCID 7439296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.McEvoy CT, Hoang T, Sidney S, Steffen LM, Jacobs DR, Jr., Shikany JM, et al. Dietary patterns during adulthood and cognitive performance in midlife: The CARDIA study. Neurology. 2019;92(14):e1589–e1e99. doi: 10.1212/WNL.0000000000007243. 10.1212/WNL.0000000000007243 PubMed PMID: 30842290; PMCID 6448450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Smith LG, Kirk GJD, Jones PJ, Williams AG. The greenhouse gas impacts of converting food production in England and Wales to organic methods. Nature communications. 2019;10(1):4641. doi: 10.1038/s41467-019-12622-7. 10.1038/s41467-019-12622-7 PubMed PMID: 31641128; PMCID 6805889. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Lee BJ, Chan MY, Hsiao HY, Chang CH, Hsu LP, Lin PT. Relationship of Oxidative Stress, Inflammation, and the Risk of Metabolic Syndrome in Patients with Oral Cancer. Oxidative medicine and cellular longevity. 2018;2018:9303094. doi: 10.1155/2018/9303094. 10.1155/2018/9303094 PubMed PMID: 29951168; PMCID 5987344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Welty FK, Alfaddagh A, Elajami TK. Targeting inflammation in metabolic syndrome. Translational research: the journal of laboratory and clinical medicine. 2016;167(1):257–280. doi: 10.1016/j.trsl.2015.06.017. 10.1016/j.trsl.2015.06.017 PubMed PMID: 26207884. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Blumenthal JA, Babyak MA, Sherwood A, Craighead L, Lin PH, Johnson J, et al. Effects of the dietary approaches to stop hypertension diet alone and in combination with exercise and caloric restriction on insulin sensitivity and lipids. Hypertension (Dallas, Tex: 1979) 2010;55(5):1199–1205. doi: 10.1161/HYPERTENSIONAHA.109.149153. 10.1161/HYPERTENSIONAHA.109.149153 PubMed PMID: 20212264. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Perry CA, Van Guilder GP, Kauffman A, Hossain M. A Calorie-Restricted DASH Diet Reduces Body Fat and Maintains Muscle Strength in Obese Older Adults. Nutrients. 2019;12(1) doi: 10.3390/nu12010102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Hayes RB, Ahn J, Fan X, Peters BA, Ma Y, Yang L, et al. Association of Oral Microbiome With Risk for Incident Head and Neck Squamous Cell Cancer. JAMA oncology. 2018;4(3):358–365. doi: 10.1001/jamaoncol.2017.4777. 10.1001/jamaoncol.2017.4777 PubMed PMID: 29327043; PMCID 5885828. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Macdonald HM, New SA, Reid DM. Longitudinal changes in dietary intake in Scottish women around the menopause: changes in dietary pattern result in minor changes in nutrient intake. Public health nutrition. 2005;8(4):409–416. doi: 10.1079/phn2005705. 10.1079/PHN2005705 PubMed PMID: 15975187. [DOI] [PubMed] [Google Scholar]
- 45.Pittas AG, Nelson J, Mitri J, Hillmann W, Garganta C, Nathan DM, et al. Plasma 25-hydroxyvitamin D and progression to diabetes in patients at risk for diabetes: an ancillary analysis in the Diabetes Prevention Program. Diabetes care. 2012;35(3):565–573. doi: 10.2337/dc11-1795. 10.2337/dc11-1795 PubMed PMID: 22323410; PMCID 3322702. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary material, approximately 28.4 MB.
Data Availability Statement
The data that support the findings of this study are available from the National Cancer Institute but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of the National Cancer Institute.