Dietary composition can influence systemic inflammation; higher levels of circulating inflammatory biomarkers are associated with increased risk of breast and other cancers. A total of 438 overweight/obese, healthy, postmenopausal women were randomized to a caloric-restriction diet (goal: 10% weight-loss), aerobic-exercise (225 min/week moderate-to-vigorous activity), combined diet+exercise, or control. Dietary inflammatory index (DII) and energy-adjusted (E-DII) scores were derived from food frequency questionnaires (FFQ) and could be calculated for 365 participants with complete FFQs at baseline and 12 months. Changes from baseline to 12 months in E-DII scores in the intervention arms versus controls were analyzed using generalized estimating equations, adjusted for confounders. We examined associations between changes in previously measured biomarkers and E-DII at 12 months. Participants randomized to diet and diet+exercise arms had greater reductions in E-DII (−104.4% and −84.4%), versus controls (−34.8%, both P < 0.001). Weight change had a more marked effect than E-DII change on biomarkers at 12-months; associations between E-DII and biomarker changes were reduced after adjustment by weight change. Changes in E-DII at 12 months, adjusted for weight change, were negatively associated with changes in ghrelin [r = −0.19; P = 0.05 (diet), r = −0.29; P = 0.02 (diet+exercise)], and positively with VEGF [r = 0.22; P = 0.03 (diet+exercise)], and red blood cell counts [r = 0.30; P = 0.004 (exercise)]. C-reactive protein (CRP) and IL6 levels were not associated with E-DII changes at 12 months. In conclusion, a behavior change of low-calorie, low-fat diet significantly reduces dietary inflammatory potential, modulating biomarkers that are associated with tumorigenesis, such as VEGF, but not CRP or IL6.

Prevention Relevance:

Diets high in saturated fats and low in fruit and vegetable intake are associated with increased inflammation, which increases cancer risk. This study showed that changes in diet quality had effects on factors associated with cancer; however, the majority of beneficial effects were associated with weight loss rather than diet quality.

High body weight is associated with increased risk of a variety of cancers (1). The exact mechanisms linking obesity and cancer have yet to be elucidated, but some hormones and peptides are elevated in obese versus normal-weight persons, such as estrogens and insulin, and decrease with weight loss. Research indicates that excessive accumulation of adipose tissue creates a protumorigenic environment. This is characterized by inflammation, macrophage invasion, and increased angiogenesis, required for increased adipose tissue expansion, driven by factors such as VEGF (2). Emerging data from animal models demonstrate the role of immune mediators such as TNFα and adipokines that couple nutritional status to adaptive responses, and how dysregulation of these systems in obesity, for example, leads to metabolic diseases and chronic inflammation (3). In turn, increased levels of circulating inflammatory and other biomarkers are associated with increased risk of breast and other cancers (4, 5).

Dietary patterns have been shown to be associated with markers of systemic inflammation in observational studies (6). Dietary composition can influence systemic inflammation, with high glycemic index diets, diets high in saturated fats and red meat, and low in overall fruit and vegetables associated with increased inflammation (7). “Immunometabolism,” a process where dietary components or metabolic processes such as free fatty acids, insulin, and glucose bind to receptors and trigger downstream pathways, such as JNK and NF-κB, may lead to dysregulation of a variety of immune mediators (3).

Shivappa and colleagues developed a literature-derived, population-based dietary inflammatory index (DII), which distinguishes dietary patterns on a continuum from maximally anti-inflammatory to maximally proinflammatory (8). A subsequent meta-analysis of 24 published studies demonstrated that individuals in the highest (most pro-inflammatory) versus lowest (most anti-inflammatory) DII categories had a 25% increased risk of overall cancer [RR = 1.25; 95% confidence interval (CI), 1.16–1.35], and a 67% increased risk of cancer mortality (RR = 1.67; 95% CI, 1.13–2.48; ref. 9). After stratification for cancer type, higher DII was associated with a 12% higher risk of breast cancer (RR = 1.12; 95% CI, 1.03–1.22), a 33% higher risk of colorectal cancer (RR = 1.33; 95% CI, 1.22–1.46), and a 30% higher risk of lung cancer (RR = 1.30; 95% CI, 1.13–1.50; ref. 9).

The Nutrition and Exercise in Women (NEW) study was a 12-month randomized controlled trial (RCT), which investigated the effects of dietary weight loss alone and in combination with physical activity, compared with controls (no intervention) on biomarkers related to breast cancer risk in 438 overweight postmenopausal women (10). The dietary weight loss program focused on calorie and fat reduction and weight-loss goals; however, we have not previously examined specific inflammation-related dietary patterns in trial participants and their effects on study biomarkers related to cancer. The NEW study has been extensively characterized in terms of biomarker profiles, dietary data, and anthropometrics, and we have previously demonstrated a statistically significant 49% and 28% mean reduction in C-reactive protein (CRP) and IL6 levels, respectively, in women who lost at least 5% of their baseline weight (10).

Here, we investigated baseline correlations between E-DII scores and circulating biomarkers previously measured in the study, including: (i) inflammation-related biomarkers (CRP, IL6, TNFα) and serum amyloid protein A (SAA); (ii) sex hormones (estradiol, estrone, testosterone, androstenedione); metabolic-related hormones and proteins (insulin, glucose, HOMA); obesity- and satiety-related hormones (ghrelin, leptin); and angiogenesis markers [pigment epithelium-derived factor (PEDF), VEGF, plasminogen activator inhibitor type-1 (PAI-1)]. These biomarkers were measured as part of the parent trial as they are associated with mechanistic cancer pathways such as angiogenesis or insulin signaling, breast cell growth modulation via sex-steroid hormones, or inflammatory potential. Leptin and ghrelin, which are associated with obesity, also are associated with inflammatory states. Increased levels of leptin, considered a link between the neuroendocrine and the immune system, promote glucose and fatty acid oxidation (11, 12). Ghrelin also has effects on both the innate and acquired immune response: in vitro, it inhibits proinflammatory responses and NF-κB activation (13), and downstream effects of inhibition of ghrelin binding to its receptor in knockout mice models include reduced expression of TNFα, IL6, and IL1β (14). We also assessed changes in the E-DII score between baseline (pre-randomization) and 12 months by intervention arm. Finally, we determined the degree to which the change in biomarkers could be attributed to changes in E-DII independently of weight loss.

This study is ancillary to the NEW (www.clinicaltrials.gov NCT00470119) study, a 12-month RCT testing the effects of caloric restriction and/or exercise on circulating hormones and other outcomes. The study was carried out in the Fred Hutchinson Cancer Research Cancer (FHCRC; Seattle, WA). It was performed with the approval of the FHCRC Institutional Review Board, in accordance with an assurance filed with and approved by the U.S. Department of Health and Human Services. The study was conducted in accordance with the U.S. Federal Policy for the Protection of Human Subjects or the “Common Rule.” Investigators obtained written informed consent from each participant.

The study is described in detail elsewhere (15). Briefly, postmenopausal, healthy overweight or obese (BMI > 25 kg/m2), sedentary women ages 50 to 75 years, not taking hormonal therapy, were recruited through media and mass mailings. Exclusion criteria included: >100 min/week of moderate physical activity; diagnosed diabetes or other serious medical condition(s); postmenopausal hormone use; consumption of >2 alcoholic drinks/day; current smoking; participation in another structured weight loss program; and contraindication to participation (e.g., abnormal exercise tolerance test, inability to attend sessions). Eligible women were randomized to one of four arms: (i) reduced-calorie dietary modification (“diet,” N = 118); (ii) moderate-to-vigorous intensity aerobic exercise (“exercise,” N = 117); (iii) combined diet and exercise (“diet+exercise”, N = 117); or (iv) control (no intervention; N = 87). Randomization was stratified by BMI (≥ or <30 kg/m2) and race/ethnicity. Investigators and laboratory staff were blinded to randomization arm.

Interventions

The dietary intervention (diet) was a modification of the Diabetes Prevention Program (DPP) and Look AHEAD (Action for Health in Diabetes) with goals of 1,200 to 2,000 kcal/day, <30% daily calories from fat, 10% weight-loss by 6-months, and weight maintenance thereafter. Although specific foods or dietary patterns were not prescribed, participants learned how to buy and prepare lower-calorie and lower-fat foods and beverages, and how to increase fruit and vegetable intake (16). Participants had at least two individual meetings with a dietician followed by weekly group meetings for 6 months; thereafter, they attended monthly, with biweekly phone/email contact. Intervention adherence was defined by percent of in-person nutrition session attendance. The exercise intervention goal was 45 minutes of moderate-to-vigorous [≥4 metabolic equivalents (MET)] intensity exercise at a target heart rate of 70% to 85% observed maximum, 5 days/week by week 7. Participants attended three facility-based supervised sessions/week and exercised 2 days/week at home. They recorded exercise mode, duration, peak heartrate, and perceived exertion at each session. Activities of ≥4 METs (17) counted towards the prescribed target. Women in this arm were asked not to change their diet. Women randomized to combined diet+exercise arm received both the diet and the exercise interventions. Controls were asked not to change their diet or exercise habits, and were offered four weight-loss classes and 8 weeks of exercise training at study completion.

Study measures

All study measures were obtained at baseline and 12 months by trained personnel blinded to participants' randomization status. Height and weight were measured from which body mass index (BMI, kg/m2) was calculated. Body composition (fat mass and percent body fat) was measured by dual-energy X-ray absorptiometry (DXA) whole-body scanner (GE Lunar). Cardiorespiratory fitness (VO2max) was assessed using a maximal graded treadmill test according to a modified branching protocol (18). Dose and frequency of NSAID medications and use of vitamin D and calcium supplements were abstracted from participant questionnaires that recorded intake at baseline and 12 months.

Participants completed a 120-item food frequency questionnaire (FFQ) that assessed usual diet (19), and questionnaires that collected information on demographics, medical history, medication use, and physical activity patterns. Analytes measured in fasting blood samples included adipokines, sex steroid hormones, biomarkers of inflammation and angiogenesis, lipids, and blood counts, were measured as described previously (10, 20–22).

The development and validation of the population-based DII have been described previously (8, 23). The DII was developed to calculate the overall inflammatory potential of an individual's diet. It is based on 45 pro- and anti-inflammatory food parameters which either increased or decreased circulating biomarkers of inflammation (CRP, IL1β, IL4, IL6, IL10, and TNFα). These parameters were identified and scored from review and ranking of 1,943 articles focused on diet-related inflammation. The index is calculated using a scoring algorithm; a higher score indicates a pro-inflammatory diet, and a lower score indicates an anti-inflammatory diet (8). The maximally pro-inflammatory diet score is +7.98, and the maximally anti-inflammatory DII score is −8.87 (8, 23). The effective range in most studies is approximately −5.5 to +5.5 (24).

In this study, 17 of the 45 food components were unavailable in the FFQ-derived data, and were omitted from the DII/E-DII calculation. Food components included were: alcohol (g), caffeine (g), carbohydrate (g), cholesterol (mg), total fat (g), fiber (g), iron (mg), magnesium (mg), mono-unsaturated fatty acids (g), Omega 3 (g), protein (g), poly-unsaturated fatty acids (g), saturated fat (g), Selenium (μg), Thiamin (mg), Trans Fat (g), Vitamins A (RE), B-6, B-12, C (mg), D (μg), and E (mg), Zinc (mg). Food components not included as they could not be calculated from the FFQ were eugenol, garlic, ginger, onion, omega-6, saffron, turmeric, green tea, black tea, falan-3-ol, flavones, flavonols, flavonones, anthocyanins, isoflavones, pepper, thyme, oregano, and rosemary. These parameters are missing from approximately two-thirds of studies that have used the DII, and Shivappa and colleagues have successfully validated the association of DII with inflammation when calculated without these parameters (23, 25). Only intake from foods, and not from supplements, was used in the DII calculations. For these analyses, energy-adjusted DII (E-DII) scores, which are based on the intake of each food component as expressed per 1,000 kcal consumed, was used in this analysis (26). The E-DII uses the same computational strategy, but utilized an energy-adjusted comparative global database.

A total of 438 participants were randomized, and 426 (97.3%) completed an FFQ at baseline. Of these, 365 (85.7%) completed the FFQ at 12 months. Participants missing FFQ data at either timepoint were excluded from the analysis (N = 73; 12 control, 20 diet, 22 exercise, and 19 diet+exercise).

Statistical analyses

Descriptive data are presented as geometric means and 95% CI. Partial Pearson correlation coefficients were calculated between baseline biomarker measures and E-DII scores. We also examined baseline associations between biomarkers and tertiles of E-DII. Mean changes in E-DII scores from baseline to 12 months, stratified by arm, were computed. The intervention effects on these E-DII score variables were examined on the basis of the assigned treatment at randomization, regardless of adherence or study retention (i.e., intent-to-treat). Mean 12-month changes in each intervention arm were compared with controls using the generalized estimating equations (GEE) modification of linear regression to account for intra-individual correlation over time. Intervention effects are presented as both absolute and relative change. The model was adjusted for age, baseline BMI category (<30 kg/m2, >30 kg/m2) race/ethnicity, and use of NSAIDs. We also compared differences in changes from baseline to 12 months in intervention arms vs. controls in users versus non-users of NSAIDs, calcium or vitamin D at baseline, adjusted for age, baseline BMI, race/ethnicity (White, Black, other), and medication (statins, NSAIDS) use.

Finally, we examined associations between changes in analytes previously measured in NEW and change in E-DII scores between baseline and 12 months, using partial Pearson coefficients, stratified by study arm. We tested the partial correlation between weight change at 12 months and biomarkers, adjusted for E-DII, followed by partial correlations between change in E-DII and biomarkers, adjusted for weight change.

All statistical tests were two sided. Statistical analyses were performed using SAS software (version 8.2; SAS Institute Inc.).

Participants were a mean 58 years of age, with a mean BMI of 30.7 kg/m2; the majority were non-Hispanic White (82.7%; Table 1). Mean baseline E-DII scores averaged −2.03 across arms (−1.76, controls; −1.86 diet; −2.41, exercise; and −2.05, diet+exercise). E-DII correlated statistically significantly at baseline with IL6 (r = 0.20, P < 0.0001, Table 2) with highest levels in the third tertile of E-DII (Ptrend across E-DII tertiles P = 0.002). No other statistically significant associations with baseline E-DII scores were observed.

Table 1.

Baseline characteristics.

ControlDietExerciseDiet+ExAll
(N = 75)(N = 98)(N = 94)(N = 98)(N = 365)
VariableMeanSDMeanSDMeanSDMeanSDMeanSD
Age (years) 57.26 4.33 58.07 6.09 58.63 5.20 58.10 4.51 58.06 5.12 
BMI (kg/m230.50 3.80 30.62 3.81 30.53 3.80 30.97 4.53 30.67 4.00 
Waist circumference (cm) 94.63 10.33 93.51 9.07 95.01 10.50 92.82 9.47 93.94 9.82 
VO2max (kg/mL/min)) 23.15 4.13 22.72 3.93 22.39 4.20 23.76 4.27 23.00 4.15 
Usual physical activity (min/wk) 22.79 41.96 35.53 47.19 40.79 46.03 33.53 44.07 33.73 45.26 
Total calories (kcal/day) 1958 684.2 1857 612.7 1996 557.0 1884 588.5 1921 608.2 
DII (adjusted for total calorie intake)a −1.76 1.88 −1.86 1.85 −2.41 1.94 −2.05 2.15 −2.03 1.97 
 N % N % N % N % N % 
Race/ethnicity 
Non-Latina White 62 82.7 88 89.8 80 85.1 85 86.7 315 86.3 
African American 8.0 5.1 11 11.7 3.1 25 6.8 
Latina 4.0 1.0 2.1 5.1 11 3.0 
Other 5.3 4.1 1.1 5.1 14 3.8 
Education 
College graduate/above 50 66.7 63 64.3 57 60.6 71 72.4 241 66.0 
Statin usersb 16 21.3 9.2 17 18.1 25 25.5 67 18.4 
NSAIDS usersb 20 26.7 40 40.8 28 29.8 39 39.8 127 34.8 
Smoker (ever) 25 33.3 43 43.9 37 39.4 39 39.8 144 39.5 
ControlDietExerciseDiet+ExAll
(N = 75)(N = 98)(N = 94)(N = 98)(N = 365)
VariableMeanSDMeanSDMeanSDMeanSDMeanSD
Age (years) 57.26 4.33 58.07 6.09 58.63 5.20 58.10 4.51 58.06 5.12 
BMI (kg/m230.50 3.80 30.62 3.81 30.53 3.80 30.97 4.53 30.67 4.00 
Waist circumference (cm) 94.63 10.33 93.51 9.07 95.01 10.50 92.82 9.47 93.94 9.82 
VO2max (kg/mL/min)) 23.15 4.13 22.72 3.93 22.39 4.20 23.76 4.27 23.00 4.15 
Usual physical activity (min/wk) 22.79 41.96 35.53 47.19 40.79 46.03 33.53 44.07 33.73 45.26 
Total calories (kcal/day) 1958 684.2 1857 612.7 1996 557.0 1884 588.5 1921 608.2 
DII (adjusted for total calorie intake)a −1.76 1.88 −1.86 1.85 −2.41 1.94 −2.05 2.15 −2.03 1.97 
 N % N % N % N % N % 
Race/ethnicity 
Non-Latina White 62 82.7 88 89.8 80 85.1 85 86.7 315 86.3 
African American 8.0 5.1 11 11.7 3.1 25 6.8 
Latina 4.0 1.0 2.1 5.1 11 3.0 
Other 5.3 4.1 1.1 5.1 14 3.8 
Education 
College graduate/above 50 66.7 63 64.3 57 60.6 71 72.4 241 66.0 
Statin usersb 16 21.3 9.2 17 18.1 25 25.5 67 18.4 
NSAIDS usersb 20 26.7 40 40.8 28 29.8 39 39.8 127 34.8 
Smoker (ever) 25 33.3 43 43.9 37 39.4 39 39.8 144 39.5 

aExpressed per 1,000 kcal consumed.

bBaseline use.

Table 2.

Pearson correlation between baseline DII and NEW study covariates.

DII score N = 365
VariableRhoP
Anthropometrics 
 Percent body-fat 0.05 0.30 
 BMI 0.12 0.03 
 Weight 0.08 0.12 
Inflammation cytokines and biomarkers 
 CRP 0.07 0.17 
 SAA 0.12 0.02 
 TNFα 0.09 0.11 
 IL6 0.20 <0.0001 
 IL10 0.07 0.20 
Angiogenesis biomarkers 
 PAI-1 0.01 0.95 
 PEDF 0.14 0.009 
 VEGF 0.01 0.95 
Adipokines, insulin resistance, C-peptide 
 Adiponectin −0.01 0.88 
 Ghrelin −0.07 0.18 
 Leptin 0.12 0.02 
 Homeostatic model assessment of insulin resistance (HOMA-IR) 0.10 0.06 
Sex steroid hormones 
 Androstenedione −0.02 0.71 
 Estrone 0.01 0.95 
 Estradiol 0.09 0.10 
 Testosterone −0.03 0.52 
 Sex steroid hormone binding globulin (SHBG) −0.01 0.83 
Blood cell counts and indices 
 Mean corpuscular volume (MCV) −0.14 0.007 
 Lymphocytes −0.04 0.40 
 White blood cell count 0.11 0.04 
 Red blood cell count 0.14 0.01 
 Platelets 0.04 0.47 
DII score N = 365
VariableRhoP
Anthropometrics 
 Percent body-fat 0.05 0.30 
 BMI 0.12 0.03 
 Weight 0.08 0.12 
Inflammation cytokines and biomarkers 
 CRP 0.07 0.17 
 SAA 0.12 0.02 
 TNFα 0.09 0.11 
 IL6 0.20 <0.0001 
 IL10 0.07 0.20 
Angiogenesis biomarkers 
 PAI-1 0.01 0.95 
 PEDF 0.14 0.009 
 VEGF 0.01 0.95 
Adipokines, insulin resistance, C-peptide 
 Adiponectin −0.01 0.88 
 Ghrelin −0.07 0.18 
 Leptin 0.12 0.02 
 Homeostatic model assessment of insulin resistance (HOMA-IR) 0.10 0.06 
Sex steroid hormones 
 Androstenedione −0.02 0.71 
 Estrone 0.01 0.95 
 Estradiol 0.09 0.10 
 Testosterone −0.03 0.52 
 Sex steroid hormone binding globulin (SHBG) −0.01 0.83 
Blood cell counts and indices 
 Mean corpuscular volume (MCV) −0.14 0.007 
 Lymphocytes −0.04 0.40 
 White blood cell count 0.11 0.04 
 Red blood cell count 0.14 0.01 
 Platelets 0.04 0.47 

At 12 months E-DII scores dropped (i.e., became more anti-inflammatory) in all arms; however, the magnitude of change was statistically significantly larger only in the diet and diet+exercise arms (−104.4% and −84.4% respectively, both P < 0.001) compared with the control arm (−34.8%). Participants in the exercise arm decreased their scores by −21.6% (P = 0.68 compared with controls; Table 3). There were no statistically significant differences between changes in E-DII scores across arms, stratified by either NSAID use (Table 4), vitamin D, or calcium supplementation (Supplementary Tables S1 and S2); all tests for interaction P > 0.25.

Table 3.

12 months changes in DII by study arm.

Baseline12 MonthChange in E-DII scoresP value
ArmNMean (SD)NMean (SD)Value%P1P2
E-DII Control 75 −1.76 (1.88) 75 −2.37 (1.82) −0.61 −34.8 — — 
 Diet 98 −1.86 (1.85) 98 −3.80 (1.18) −1.94 −104.4 <0.001 <0.001 
 Exercise 94 −2.41 (1.94) 94 −2.93 (1.79) −0.52 −21.6 0.68 0.68 
 Diet+Exercise 98 −2.05 (2.15) 98 −3.78 (1.51) −1.73 −84.4 <0.001 <0.001 
Baseline12 MonthChange in E-DII scoresP value
ArmNMean (SD)NMean (SD)Value%P1P2
E-DII Control 75 −1.76 (1.88) 75 −2.37 (1.82) −0.61 −34.8 — — 
 Diet 98 −1.86 (1.85) 98 −3.80 (1.18) −1.94 −104.4 <0.001 <0.001 
 Exercise 94 −2.41 (1.94) 94 −2.93 (1.79) −0.52 −21.6 0.68 0.68 
 Diet+Exercise 98 −2.05 (2.15) 98 −3.78 (1.51) −1.73 −84.4 <0.001 <0.001 
Baseline12 MonthChange in E-DII scoresP value
ArmNMean (SD)NMean (SD)Value%P1P2
BMI Control 75 30.50 (3.80) 75 30.25 (4.02) −0.26 −0.84 — — 
 Diet 98 30.62 (3.81) 98 27.54 (4.20) −3.07 −10.0 <0.001 <0.001 
 Exercise 94 30.53 (3.80) 94 29.65 (3.97) −0.88 −2.90 0.015 0.015 
 Diet+exercise 98 30.97 (4.53) 98 27.32 (4.56) −3.65 −11.8 <0.001 <0.001 
Baseline12 MonthChange in E-DII scoresP value
ArmNMean (SD)NMean (SD)Value%P1P2
BMI Control 75 30.50 (3.80) 75 30.25 (4.02) −0.26 −0.84 — — 
 Diet 98 30.62 (3.81) 98 27.54 (4.20) −3.07 −10.0 <0.001 <0.001 
 Exercise 94 30.53 (3.80) 94 29.65 (3.97) −0.88 −2.90 0.015 0.015 
 Diet+exercise 98 30.97 (4.53) 98 27.32 (4.56) −3.65 −11.8 <0.001 <0.001 
Baseline12 MonthChange in E-DII scoresP value
ArmNMean (SD)NMean (SD)Value%P1P2
Percent body-fat Control 75 47.87 (4.35) 75 47.52 (4.87) −0.35 −0.74 — — 
 Diet 98 47.39 (4.20) 98 42.40 (6.58) −4.99 −10.5 <0.001 <0.001 
 Exercise 94 47.83 (4.13) 94 45.98 (5.13) −1.85 −3.86 <0.001 <0.001 
 Diet+exercise 98 47.85 (4.76) 98 41.22 (7.35) −6.63 −13.9 <0.001 <0.001 
Baseline12 MonthChange in E-DII scoresP value
ArmNMean (SD)NMean (SD)Value%P1P2
Percent body-fat Control 75 47.87 (4.35) 75 47.52 (4.87) −0.35 −0.74 — — 
 Diet 98 47.39 (4.20) 98 42.40 (6.58) −4.99 −10.5 <0.001 <0.001 
 Exercise 94 47.83 (4.13) 94 45.98 (5.13) −1.85 −3.86 <0.001 <0.001 
 Diet+exercise 98 47.85 (4.76) 98 41.22 (7.35) −6.63 −13.9 <0.001 <0.001 

P1 is the P value from GEE model comparing the 12 months changes in E-DII between the intervention arms versus controls, unadjusted.

P2 is the P value from GEE model comparing the 12 months changes in E-DII between the intervention arms versus controls, adjusted for age, baseline BMI (<30 kg/m2, ≥30 kg/m2), race/ethnicity (White, Black, Other), and medication (statin, NSAIDS) use.

Table 4.

12 months changes in E-DII by study arm and baseline use of NSAIDs.

Baseline NSAID nonusersBaseline NSAID users
Baseline12 MonthP-valueBaseline12 MonthP-valueInteraction P-value
MeanNMeanMeanMean
Study armN(SD)(SD)Change in E-DII (%)P1P2N(SD)N(SD)Change in E-DII (%)P1P2Pi1Pi2
Control 55 −1.53 55 −2.15 −0.62 — — 20 −2.38 20 −2.96 −0.58 — — — — 
  (1.78)  (1.65) (−40.71)    (2.05)  (2.14) (−24.42)     
Diet 58 −1.92 58 −3.98 −2.06 <0.001 <0.001 40 −1.77 40 −3.54 −1.77 0.002 0.002 0.13 0.12 
  (1.93)  (1.13) (−107.2)    (1.75)  (1.22) (−100.0)     
Exercise 66 −2.03 66 −2.53 −0.50 0.66 0.66 28 −3.30 28 −3.86 −0.56 0.94 0.94 0.50 0.66 
  (2.00)  (1.87) (−24.79)    (1.46)  (1.15) (−16.86)     
Diet+exercise 59 −2.02 59 −3.73 −1.71 <0.001 <0.001 39 −2.10 39 −3.85 −1.75 0.002 0.002 0.58 0.66 
  (2.24)  (1.61) (−84.97)    (2.03)  (1.38) (−83.43)     
Baseline NSAID nonusersBaseline NSAID users
Baseline12 MonthP-valueBaseline12 MonthP-valueInteraction P-value
MeanNMeanMeanMean
Study armN(SD)(SD)Change in E-DII (%)P1P2N(SD)N(SD)Change in E-DII (%)P1P2Pi1Pi2
Control 55 −1.53 55 −2.15 −0.62 — — 20 −2.38 20 −2.96 −0.58 — — — — 
  (1.78)  (1.65) (−40.71)    (2.05)  (2.14) (−24.42)     
Diet 58 −1.92 58 −3.98 −2.06 <0.001 <0.001 40 −1.77 40 −3.54 −1.77 0.002 0.002 0.13 0.12 
  (1.93)  (1.13) (−107.2)    (1.75)  (1.22) (−100.0)     
Exercise 66 −2.03 66 −2.53 −0.50 0.66 0.66 28 −3.30 28 −3.86 −0.56 0.94 0.94 0.50 0.66 
  (2.00)  (1.87) (−24.79)    (1.46)  (1.15) (−16.86)     
Diet+exercise 59 −2.02 59 −3.73 −1.71 <0.001 <0.001 39 −2.10 39 −3.85 −1.75 0.002 0.002 0.58 0.66 
  (2.24)  (1.61) (−84.97)    (2.03)  (1.38) (−83.43)     

P1 is the P value from GEE model comparing the 12 months changes in E-DII between the intervention arms versus controls within the categories of NSAID use, unadjusted.

P2 is the P value from GEE model comparing the 12 months changes in E-DII between the intervention arms versus controls within the categories of baseline NSAID use, adjusted for age, baseline BMI (<30 kg/m2, ≥30 kg/m2), race/ethnicity (White, Black, Other).

Pi1 is the P value from GEE model comparing differences in changes from baseline to 12 months in intervention arms versus controls in users versus nonusers of NSAID at baseline, unadjusted.

Pi2 is the P value from GEE model comparing differences in changes from baseline to 12 months in intervention arms versus controls in users versus nonusers of NSAID at baseline, adjusted for age, baseline BMI (<30 kg/m2, ≥30 kg/m2) and race/ethnicity (White, Black, Other).

Table 5.

Partial Pearson correlation between changes in energy adjusted DII scores (E-DII), or in weight and 12 months changes biomarker levels, by study arm.

Correlation with weight change (WC) or E-DII change
WC (adjusted for DII change)E-DII (adjusted for weight change)
BiomarkerRhoPRhoP
Inflammatory biomarkers 
CRP (mg/L) Control 0.02 0.86 0.20 0.09 
 Diet 0.34 0.001 −0.05 0.61 
 Exercise −0.02 0.86 −0.02 0.83 
 Diet+exercise 0.23 0.03 0.05 0.61 
SAA (mg/L) Control 0.29 0.01 0.10 0.41 
 Diet 0.26 0.009 −0.08 0.42 
 Exercise −0.07 0.48 −0.06 0.56 
 Diet+exercise 0.12 0.26 −0.06 0.51 
TNFα (pg/mL) Control −0.01 0.93 −0.03 0.79 
 Diet 0.25 0.02 0.06 0.58 
 Exercise −0.11 0.32 −0.08 0.48 
 Diet+exercise 0.07 0.50 0.08 0.44 
IL6 (pg/mL) Control −0.15 0.21 0.15 0.12 
 Diet 0.19 0.06 −0.02 0.83 
 Exercise 0.10 0.37 0.22 0.04 
 Diet+exercise 0.23 0.03 0.09 0.38 
IL10 (pg/mL) Control 0.10 0.43 −0.12 0.32 
 Diet 0.11 0.28 −0.11 0.30 
 Exercise −0.09 0.41 0.05 0.68 
 Diet+exercise 0.03 0.81 0.01 0.92 
Biomarkers of angiogenesis 
PAI-1 (ng/mL) Control 0.17 0.16 −0.01 0.94 
 Diet 0.35 0.001 0.05 0.67 
 Exercise 0.23 0.03 −0.01 0.92 
 Diet+exercise 0.33 0.001 0.02 0.87 
PEDF (μg/mL) Control 0.12 0.31 0.40 <0.0001 
 Diet 0.38 <0.0001 −0.09 0.40 
 Exercise 0.40 <0.0001 −0.09 0.41 
 Diet+exercise 0.53 <0.0001 0.17 0.10 
VEGF (pg/mL) Control 0.17 0.15 0.08 0.50 
 Diet 0.43 <0.0001 0.02 0.86 
 Exercise 0.08 0.46 0.01 0.99 
 Diet+exercise 0.33 0.001 0.22 0.03 
Adipokines and markers of insulin resistance, and appetite regulation 
C-Peptide Control 0.28 0.02 0.09 0.43 
 Diet 0.27 0.008 0.03 0.81 
 Exercise 0.32 0.002 −0.07 0.51 
 Diet+exercise 0.44 <0.0001 0.04 0.74 
Adiponectin (μg/mL) Control −0.20 0.09 −0.19 0.10 
 Diet −0.31 0.002 0.11 0.29 
 Exercise 0.04 0.69 −0.01 0.93 
 Diet+exercise −0.20 0.05 0.04 0.70 
Ghrelin (pg/mL) Control −0.04 0.75 −0.03 0.84 
 Diet −0.24 0.02 −0.19 0.05 
 Exercise −0.38 <0.0001 0.01 0.92 
 Diet+exercise −0.18 0.08 −0.23 0.02 
Leptin (ng/mL) Control 0.34 0.003 −0.03 0.82 
 Diet 0.65 <0.0001 −0.03 0.79 
 Exercise 0.64 <0.0001 0.14 0.19 
 Diet+exercise 0.73 <0.0001 0.13 0.22 
HOMA-IR Control 0.27 0.02 0.13 0.27 
 Diet 0.15 0.14 0.09 0.41 
 Exercise 0.21 0.05 −0.09 0.40 
 Diet+exercise 0.28 0.005 0.04 0.72 
Low- and high-density lipoproteins 
LDL Control −0.16 0.17 0.14 0.25 
 Diet 0.14 0.18 −0.02 0.84 
 Exercise 0.15 0.16 0.07 0.52 
 Diet+exercise 0.22 0.03 −0.01 0.93 
HDL Control −0.04 0.74 0.07 0.58 
 Diet −0.14 0.17 0.15 0.16 
 Exercise −0.13 0.23 −0.01 0.94 
 Diet+exercise −0.26 0.01 0.11 0.30 
Sex steroid hormones 
Androstenedione (ng/dL) Control 0.12 0.33 −0.03 0.79 
 Diet −0.13 0.22 0.13 0.20 
 Exercise 0.06 0.58 −0.04 0.73 
 Diet+exercise −0.05 0.66 −0.05 0.62 
Estrone (pg/mL) Control −0.03 0.79 0.07 0.58 
 Diet 0.15 0.16 0.18 0.09 
 Exercise 0.20 0.06 −0.19 0.08 
 Diet+exercise 0.08 0.44 −0.02 0.83 
Estradiol (pg/mL) Control 0.03 0.80 0.01 0.99 
 Diet 0.07 0.51 0.15 0.16 
 Exercise 0.17 0.11 −0.17 0.10 
 Diet+exercise 0.30 0.003 0.01 0.97 
Testosterone (ng/dL) Control 0.01 0.92 −0.02 0.87 
 Diet −0.04 0.71 0.03 0.77 
 Exercise 0.21 0.04 −0.08 0.46 
 Diet+exercise 0.03 0.74 0.01 0.93 
Cell counts 
Lymphocytes (counts/μL) Control 0.01 0.98 −0.19 0.11 
 Diet −0.26 0.01 0.09 0.40 
 Exercise −0.09 0.39 −0.11 0.31 
 Diet+exercise 0.06 0.56 −0.03 0.74 
Red blood cells (counts/μL) Control 0.15 0.21 0.17 0.16 
 Diet 0.35 0.001 −0.11 0.27 
 Exercise 0.20 0.06 0.30 0.004 
 Diet+exercise 0.07 0.47 −0.11 0.29 
Platelets Control 0.12 0.33 0.11 0.37 
 Diet 0.30 0.003 −0.01 0.96 
 Exercise 0.03 0.82 −0.03 0.80 
 Diet+exercise 0.38 <0.0001 0.10 0.33 
Correlation with weight change (WC) or E-DII change
WC (adjusted for DII change)E-DII (adjusted for weight change)
BiomarkerRhoPRhoP
Inflammatory biomarkers 
CRP (mg/L) Control 0.02 0.86 0.20 0.09 
 Diet 0.34 0.001 −0.05 0.61 
 Exercise −0.02 0.86 −0.02 0.83 
 Diet+exercise 0.23 0.03 0.05 0.61 
SAA (mg/L) Control 0.29 0.01 0.10 0.41 
 Diet 0.26 0.009 −0.08 0.42 
 Exercise −0.07 0.48 −0.06 0.56 
 Diet+exercise 0.12 0.26 −0.06 0.51 
TNFα (pg/mL) Control −0.01 0.93 −0.03 0.79 
 Diet 0.25 0.02 0.06 0.58 
 Exercise −0.11 0.32 −0.08 0.48 
 Diet+exercise 0.07 0.50 0.08 0.44 
IL6 (pg/mL) Control −0.15 0.21 0.15 0.12 
 Diet 0.19 0.06 −0.02 0.83 
 Exercise 0.10 0.37 0.22 0.04 
 Diet+exercise 0.23 0.03 0.09 0.38 
IL10 (pg/mL) Control 0.10 0.43 −0.12 0.32 
 Diet 0.11 0.28 −0.11 0.30 
 Exercise −0.09 0.41 0.05 0.68 
 Diet+exercise 0.03 0.81 0.01 0.92 
Biomarkers of angiogenesis 
PAI-1 (ng/mL) Control 0.17 0.16 −0.01 0.94 
 Diet 0.35 0.001 0.05 0.67 
 Exercise 0.23 0.03 −0.01 0.92 
 Diet+exercise 0.33 0.001 0.02 0.87 
PEDF (μg/mL) Control 0.12 0.31 0.40 <0.0001 
 Diet 0.38 <0.0001 −0.09 0.40 
 Exercise 0.40 <0.0001 −0.09 0.41 
 Diet+exercise 0.53 <0.0001 0.17 0.10 
VEGF (pg/mL) Control 0.17 0.15 0.08 0.50 
 Diet 0.43 <0.0001 0.02 0.86 
 Exercise 0.08 0.46 0.01 0.99 
 Diet+exercise 0.33 0.001 0.22 0.03 
Adipokines and markers of insulin resistance, and appetite regulation 
C-Peptide Control 0.28 0.02 0.09 0.43 
 Diet 0.27 0.008 0.03 0.81 
 Exercise 0.32 0.002 −0.07 0.51 
 Diet+exercise 0.44 <0.0001 0.04 0.74 
Adiponectin (μg/mL) Control −0.20 0.09 −0.19 0.10 
 Diet −0.31 0.002 0.11 0.29 
 Exercise 0.04 0.69 −0.01 0.93 
 Diet+exercise −0.20 0.05 0.04 0.70 
Ghrelin (pg/mL) Control −0.04 0.75 −0.03 0.84 
 Diet −0.24 0.02 −0.19 0.05 
 Exercise −0.38 <0.0001 0.01 0.92 
 Diet+exercise −0.18 0.08 −0.23 0.02 
Leptin (ng/mL) Control 0.34 0.003 −0.03 0.82 
 Diet 0.65 <0.0001 −0.03 0.79 
 Exercise 0.64 <0.0001 0.14 0.19 
 Diet+exercise 0.73 <0.0001 0.13 0.22 
HOMA-IR Control 0.27 0.02 0.13 0.27 
 Diet 0.15 0.14 0.09 0.41 
 Exercise 0.21 0.05 −0.09 0.40 
 Diet+exercise 0.28 0.005 0.04 0.72 
Low- and high-density lipoproteins 
LDL Control −0.16 0.17 0.14 0.25 
 Diet 0.14 0.18 −0.02 0.84 
 Exercise 0.15 0.16 0.07 0.52 
 Diet+exercise 0.22 0.03 −0.01 0.93 
HDL Control −0.04 0.74 0.07 0.58 
 Diet −0.14 0.17 0.15 0.16 
 Exercise −0.13 0.23 −0.01 0.94 
 Diet+exercise −0.26 0.01 0.11 0.30 
Sex steroid hormones 
Androstenedione (ng/dL) Control 0.12 0.33 −0.03 0.79 
 Diet −0.13 0.22 0.13 0.20 
 Exercise 0.06 0.58 −0.04 0.73 
 Diet+exercise −0.05 0.66 −0.05 0.62 
Estrone (pg/mL) Control −0.03 0.79 0.07 0.58 
 Diet 0.15 0.16 0.18 0.09 
 Exercise 0.20 0.06 −0.19 0.08 
 Diet+exercise 0.08 0.44 −0.02 0.83 
Estradiol (pg/mL) Control 0.03 0.80 0.01 0.99 
 Diet 0.07 0.51 0.15 0.16 
 Exercise 0.17 0.11 −0.17 0.10 
 Diet+exercise 0.30 0.003 0.01 0.97 
Testosterone (ng/dL) Control 0.01 0.92 −0.02 0.87 
 Diet −0.04 0.71 0.03 0.77 
 Exercise 0.21 0.04 −0.08 0.46 
 Diet+exercise 0.03 0.74 0.01 0.93 
Cell counts 
Lymphocytes (counts/μL) Control 0.01 0.98 −0.19 0.11 
 Diet −0.26 0.01 0.09 0.40 
 Exercise −0.09 0.39 −0.11 0.31 
 Diet+exercise 0.06 0.56 −0.03 0.74 
Red blood cells (counts/μL) Control 0.15 0.21 0.17 0.16 
 Diet 0.35 0.001 −0.11 0.27 
 Exercise 0.20 0.06 0.30 0.004 
 Diet+exercise 0.07 0.47 −0.11 0.29 
Platelets Control 0.12 0.33 0.11 0.37 
 Diet 0.30 0.003 −0.01 0.96 
 Exercise 0.03 0.82 −0.03 0.80 
 Diet+exercise 0.38 <0.0001 0.10 0.33 

aWeight change (adjusted): partial correlation with weight change adjusted for E-DII density change; E-DII (adjusted): partial correlation with E-DII density change adjusted for weight change.

We examined associations between changes in weight loss, adjusted for changes in DII at 12 months and circulating biomarkers measured in the parent study, using partial Pearson correlations (Table 5 and Supplementary Table S3). In general, weight change had a more marked effect than DII change on changes in biomarkers, and associations of DII change with biomarker changes are reduced after adjustment by weight change. Twelve-month E-DII–adjusted changes in weight loss correlated statistically significantly with changes in the inflammatory biomarkers CRP, SAA, IL6, and TNFα; with biomarkers of angiogenesis, PAI-1, PEDF, and VEGF; with c-peptide and HOMA-IR scores; with the adipokines adiponectin, ghrelin; with the sex steroid hormones estradiol and testosterone and with HDL, LDL, and lymphocyte, red blood cell, and platelet counts (Table 5).

In contrast, statistically significant associations between changes in E-DII adjusted for weight change at 12 months were seen only for VEGF (r = 0.22, P = 0.03, diet+exercise arm) and for ghrelin [diet arm (r = −0.19, P = 0.05), diet+exercise arm (r = −0.23, P = 0.02)], and red blood cell count [exercise arm (r = 0.30, P = 0.004)].

Dietary effects on risk of cancer may occur through patterns of eating (such as those measured with the DII and Health Eating Index), in addition to specific dietary factors (27). In addition, consumption of energy-dense foods contributes to the likelihood of increased adiposity which, in turn, is linked to chronic inflammation and increased risk of cancer (5, 28, 29).

In this study in all participants combined, baseline E-DII scores were statistically significantly associated with circulating levels of IL6, but not with CRP. In contrast, in 2,567 postmenopausal women from the Women's Health Initiative, highest versus lowest DII quintiles predicted concentrations of IL6, CRP, and TNFα receptor 2. Higher scores statistically significantly predicted higher IL6 and TNFα Receptor 2 concentrations, and were associated with an increased overall inflammatory biomarker score. However, although higher DII scores were associated with elevated CRP concentrations, this was not statistically significant (30). We previously demonstrated that participants randomized to the diet and diet+exercise arms, and those who lost at least 5% of their baseline weight, had significantly reduced levels of CRP and IL6 compared with controls at 12 months (21). However, in this study, after adjustment for weight change, 12-month changes in E-DII were not associated with changes in either IL6 or CRP. We observed similar outcomes for changes in E-DII and HOMA, adipokines, and sex steroid hormones. In general, weight change had a more marked effect than DII change on the biomarker changes, and associations of DII change with biomarker changes are reduced after adjustment by weight change; after adjustment for weight change E-DII was associated only with ghrelin and VEGF levels at 12 months. This suggest that weight loss rather than changes in inflammatory potential of participants' diets accounted for reductions of these biomarkers in the diet and diet+exercise arms.

Changes in E-DII scores were negatively associated with changes in ghrelin in the diet+exercise arm, that is, more anti-inflammatory diets correlate with higher ghrelin levels. Ghrelin is produced by the gastrointestinal tract and stimulates appetite, and circulating levels are reduced in obesity (31). Ghrelin inhibits the expression of the pro-inflammatory IL1β, IL6, and TNFα, by binding to growth hormone secretagogue receptor (GHS-R; ref. 32). Ghrelin activates anabolic pathways leading to weight gain, and although it adaptively responds to changes in body weight, it also responds to short-term regulatory stimuli such as acute fasting and feeding (33). Weight-loss diets result in compensatory metabolic changes, including increased levels of circulating ghrelin, and decreased post-prandial ghrelin suppression (34). However, to our knowledge, there are no other reports of anti-inflammatory dietary patterns resulting in increases in ghrelin, after adjustment for weight loss. Although the role of ghrelin in carcinogenesis is unclear, it appears to have anti-inflammatory and anti-proliferative properties (35), and a case–control study found that ghrelin expression was associated with better breast cancer-specific survival (36).

We previously demonstrated that fasting total ghrelin increased in both diet only and diet+exercise arms compared with controls; greater weight loss was associated with increased ghrelin concentrations, regardless of intervention arm (37). However, ghrelin appears to be regulated by macronutrients in the diet, with higher levels of dietary fats and carbohydrates influencing ghrelin levels (34, 38); high-fat diet-fed mice developed ghrelin resistance (39). A study in 408 New Zealand European, Māori, and Pacific women aged 16 to 45 years, found that women with higher scores for the ‘refined and processed’ pattern had lower levels of circulating ghrelin, compared with those with lower scores (40).

In the NEW study, weight loss was statistically significantly associated with reductions of angiogenic markers—VEGF, PAI-1, and PEDF (22)—in the diet and diet+exercise arms. This reduction can be attributed to loss of fat mass with a combination of a drop in hypoxia-induced factor-1 (HIF-1) and lower requirements for adipose tissue vascularity overall (41). However, the associations observed in the present analyses between weight change adjusted changes in E-DII and VEGF in the diet+exercise arm suggests that dietary changes may have an additional contribution in the reported reductions in these analytes.

There have been few studies examining DII/E-DII scores in the context of weight change. In a small uncontrolled study of 61 overweight/obese individuals enrolled in a year-long dietary weight-loss program, the DII score was positively correlated with weight regain (r = 0.29, P = 0.05) after correction for physical activity (42). A long-term prospective study of 7,027 adults with a BMI <25 kg/m2 were followed for a median of 8.1 years and the DII was calculated from a FFQ at baseline and end-of-study. At the end of the study, 20.4% of the cohort were overweight/obese. The study reported that the hazard ratio for developing obesity was 1.32 (95% CI, 1.02–1.60) in the highest DII quartile compared with the lowest (Ptrend = 0.004; ref. 43).

We also demonstrated that participants randomized to either the diet or diet+exercise arms in a year-long behavioral weight loss RCT, experienced statistically significantly reductions in DII scores at 12 months, compared with participants in the control arm. This reduction is expected, as dietary intervention participants were encouraged to reduce saturated fat intake and increase servings of fresh fruit and vegetables. Participants in the exercise arm had E-DII scores similar to those of the control arm at 12 months. While participants in the exercise and control arm were instructed not to make changes in their diet, at 12 months all participants—even those in the control arm—reduced their scores (i.e., their diets became more anti-inflammatory). This might be explained by the fact that these participants were motivated to join the NEW study to improve their lifestyle through diet, exercise or both, and study enrollment by itself may have motivated participants to make changes to their diet, attributable to behavioral “spillover” (44, 45). At baseline, all participants, on average, had negative E-DII scores indicating a more anti-inflammatory diet. The range of values reported here for DII (−6.00 to +3.92) is narrower than that seen in other studies, for example, −5.36 to +6.25 in Spanish men and women (46) and −7.30 to +5.78 in the Women's Health Initiative study in postmenopausal women (47).

Strengths of our study include the use of the validated DII/E-DII, which was specifically designed to assess inflammatory potential of diet while accounting for energy intake differences among individuals; the RCT design; and the very well characterized NEW study. Limitations include the use of FFQs to obtain dietary intake data. Although they are easy to apply to large studies, FFQs are error prone. For example dietary intake over a period of several months is prone to recall bias, participants often have a desire to be more socially acceptable for example when reporting calorically dense foods (48), and variability in seasonal intake contribute to measurement error (49). Our E-DII was missing several components from the original DII developed by Shivappa and colleagues (8). Some of the missing components include spices like turmeric, saffron, garlic, which are not consumed in large quantities in this population, and their omission may not have impacted the results significantly. However missing parameters which are consumed often such as flavonoids and onion could have underestimated associations with diet interventions or weight loss in the present study. Of note, the association between DII and inflammation has also been validated when calculated without several of these parameters. A sensitivity analysis conducted to compare DII scores calculated from 24-hour dietary recalls vs. from 7-day recalls with missing dietary components, found that the DII score generated from the more limited list did not affect associations with levels of CRP (23). Finally, the majority of the study participants were non-Latina Whites enrolled in a RCT with specific eligibility criteria; hence, findings may not be generalizable to other race/ethnic groups.

In summary, a dietary weight loss intervention focused on reducing calories and fat and increasing fruits and vegetables significantly lowered scores of a dietary inflammation index measured with a FFQ in postmenopausal women, Further research is needed to determine whether lowering intake of inflammation-promoting foods and beverages can translate to reducing risk of specific cancers.

N. Shivappa reports other from Connecting Health Innovations outside the submitted work. J.R. Hébert reports other from Connecting Health Innovations LLC outside the submitted work; has a patent for Dietary Inflammatory Index issued, licensed, and with royalties paid from Connecting Health Innovations LLC. C-Y. Wang reports grants from Breast Cancer Research Foundation during the conduct of the study. A. McTiernan reports grants from Breast Cancer Research Foundation during the conduct of the study. No disclosures were reported by the other authors.

C. Duggan: Conceptualization, resources, writing-original draft, writing-review and editing. J.d.D. Tapsoba: Formal analysis, writing-review and editing. N. Shivappa: Formal analysis, writing-review and editing. H.R. Harris: Writing-review and editing. J.R. Hébert: Formal analysis, methodology, writing-review and editing. C.-Y. Wang: Formal analysis, supervision, writing-review and editing. A. McTiernan: Resources, supervision, writing-review and editing.

This work was supported by grants from the NCI at the NIH (P30 CA015704, R01 CA105204-01A1, and U54-CA116847 to A. McTiernan and C. Duggan), and the Breast Cancer Research Foundation (BCRF-16-106, BCRF-17-105, BCRF-18-107, and BCRF-19-107 to A. McTiernan and C. Duggan).

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

1.
Keum
N
,
Greenwood
DC
,
Lee
DH
,
Kim
R
,
Aune
D
,
Ju
W
, et al
Adult weight gain and adiposity-related cancers: a dose-response meta-analysis of prospective observational studies
.
J Natl Cancer Inst
2015
;
107
:
djv088
.
2.
Sun
K
,
Kusminski
CM
,
Scherer
PE
. 
Adipose tissue remodeling and obesity
.
J Clin Invest
2011
;
121
:
2094
101
.
3.
Hotamisligil
GS
. 
Inflammation, metaflammation and immunometabolic disorders
.
Nature
2017
;
542
:
177
85
.
4.
Chan
DS
,
Bandera
EV
,
Greenwood
DC
,
Norat
T
. 
Circulating C-reactive protein and breast cancer risk-systematic literature review and meta-analysis of prospective cohort studies
.
Cancer Epidemiol Biomarkers Prev
2015
;
24
:
1439
49
.
5.
Himbert
C
,
Delphan
M
,
Scherer
D
,
Bowers
LW
,
Hursting
S
,
Ulrich
CM
. 
Signals from the adipose microenvironment and the obesity-cancer link—a systematic review
.
Cancer Prev Res
2017
;
10
:
494
506
.
6.
Tabung
FK
,
Giovannucci
EL
,
Giulianini
F
,
Liang
L
,
Chandler
PD
,
Balasubramanian
R
, et al
An empirical dietary inflammatory pattern score is associated with circulating inflammatory biomarkers in a multi-ethnic population of postmenopausal women in the United States
.
J Nutr
2018
;
148
:
771
80
.
7.
Minihane
AM
,
Vinoy
S
,
Russell
WR
,
Baka
A
,
Roche
HM
,
Tuohy
KM
, et al
Low-grade inflammation, diet composition and health: current research evidence and its translation
.
Br J Nutr
2015
;
114
:
999
1012
.
8.
Shivappa
N
,
Steck
SE
,
Hurley
TG
,
Hussey
JR
,
Hebert
JR
. 
Designing and developing a literature-derived, population-based dietary inflammatory index
.
Public Health Nutr
2014
;
17
:
1689
96
.
9.
Fowler
ME
,
Akinyemiju
TF
. 
Meta-analysis of the association between dietary inflammatory index (DII) and cancer outcomes
.
Int J Cancer
2017
;
141
:
2215
27
.
10.
Campbell
KL
,
Foster-Schubert
KE
,
Alfano
CM
,
Wang
CC
,
Wang
CY
,
Duggan
CR
, et al
Reduced-calorie dietary weight loss, exercise, and sex hormones in postmenopausal women: randomized controlled trial
.
J Clin Oncol
2012
;
30
:
2314
26
.
11.
Rivadeneira
DB
,
DePeaux
K
,
Wang
Y
,
Kulkarni
A
,
Tabib
T
,
Menk
AV
, et al
Oncolytic viruses engineered to enforce leptin expression reprogram tumor-infiltrating T cell metabolism and promote tumor clearance
.
Immunity
2019
;
51
:
548
60
.
12.
Abella
V
,
Scotece
M
,
Conde
J
,
Pino
J
,
Gonzalez-Gay
MA
,
Gómez-Reino
JJ
, et al
Leptin in the interplay of inflammation, metabolism and immune system disorders
.
Nat Rev Rheumatol
2017
;
13
:
100
9
.
13.
Li
WG
,
Gavrila
D
,
Liu
X
,
Wang
L
,
Gunnlaugsson
S
,
Stoll
LL
, et al
Ghrelin inhibits proinflammatory responses and nuclear factor-kappaB activation in human endothelial cells
.
Circulation
2004
;
109
:
2221
6
.
14.
Pereira
J
,
da Silva
FC
,
de Moraes-Vieira
PMM
. 
The impact of ghrelin in metabolic diseases: an immune perspective
.
J Diabetes Res
2017
;
2017
:
4527980
.
15.
Foster-Schubert
KE
,
Alfano
CM
,
Duggan
CR
,
Xiao
L
,
Campbell
KL
,
Kong
A
, et al
Effect of diet and exercise, alone or combined, on weight and body composition in overweight-to-obese postmenopausal women
.
Obesity
2012
;
20
:
1628
38
.
16.
Kong
A
,
Beresford
SA
,
Alfano
CM
,
Foster-Schubert
KE
,
Neuhouser
ML
,
Johnson
DB
, et al
Associations between snacking and weight loss and nutrient intake among postmenopausal overweight to obese women in a dietary weight-loss intervention
.
J Am Diet Assoc
2011
;
111
:
1898
903
.
17.
Ainsworth
BE
,
Haskell
WL
,
Whitt
MC
,
Irwin
ML
,
Swartz
AM
,
Strath
SJ
, et al
Compendium of physical activities: an update of activity codes and MET intensities
.
Med Sci Sports Exerc
2000
;
32
:
S498
504
.
18.
Pate
R
,
Blair
S
,
Durstine
J
.
Guidelines for exercise testing and prescription
.
Philadelphia, PA: Lea & Febinger
; 
1991
:
70
72
.
19.
Patterson
RE
,
Kristal
AR
,
Tinker
LF
,
Carter
RA
,
Bolton
MP
,
Agurs-Collins
T
. 
Measurement characteristics of the Women's Health Initiative food frequency questionnaire
.
Ann Epidemiol
1999
;
9
:
178
87
.
20.
Mason
C
,
Foster-Schubert
KE
,
Imayama
I
,
Kong
A
,
Xiao
L
,
Bain
C
, et al
Dietary weight loss and exercise effects on insulin resistance in postmenopausal women
.
Am J Prev Med
2011
;
41
:
366
75
.
21.
Imayama
I
,
Ulrich
CM
,
Alfano
CM
,
Wang
C
,
Xiao
L
,
Wener
MH
, et al
Effects of a caloric restriction weight loss diet and exercise on inflammatory biomarkers in overweight/obese postmenopausal women: a randomized controlled trial.
Cancer Res
2012
;
72
:
2314
26
.
22.
Duggan
C
,
Tapsoba Jde
D
,
Wang
CY
,
McTiernan
A
. 
Dietary weight loss and exercise effects on serum biomarkers of angiogenesis in overweight postmenopausal women: a randomized controlled trial
.
Cancer Res
2016
;
76
:
4226
35
.
23.
Shivappa
N
,
Steck
SE
,
Hurley
TG
,
Hussey
JR
,
Ma
Y
,
Ockene
IS
, et al
A population-based dietary inflammatory index predicts levels of C-reactive protein in the Seasonal Variation of Blood Cholesterol Study (SEASONS)
.
Public Health Nutr
2014
;
17
:
1825
33
.
24.
Hébert
JR
,
Shivappa
N
,
Wirth
MD
,
Hussey
JR
,
Hurley
TG
. 
Perspective: the Dietary Inflammatory Index (DII)-lessons learned, improvements made, and future directions
.
Adv Nutr
2019
;
10
:
185
95
.
25.
Shivappa
N
,
Hébert
JR
,
Rietzschel
ER
,
De Buyzere
ML
,
Langlois
M
,
Debruyne
E
, et al
Associations between dietary inflammatory index and inflammatory markers in the Asklepios Study
.
Br J Nutr
2015
;
113
:
665
71
.
26.
Sokol
A
,
Wirth
MD
,
Manczuk
M
,
Shivappa
N
,
Zatonska
K
,
Hurley
TG
, et al
Association between the dietary inflammatory index, waist-to-hip ratio and metabolic syndrome
.
Nutr Res
2016
;
36
:
1298
303
.
27.
Mayne
ST
,
Playdon
MC
,
Rock
CL
. 
Diet, nutrition, and cancer: past, present and future
.
Nat Rev Clin Oncol
2016
;
13
:
504
.
28.
Arendt
LM
,
McCready
J
,
Keller
PJ
,
Baker
DD
,
Naber
SP
,
Seewaldt
V
, et al
Obesity promotes breast cancer by CCL2-mediated macrophage recruitment and angiogenesis
.
Cancer Res
2013
;
73
:
6080
93
.
29.
Yu
JL
,
Rak
JW
. 
Host microenvironment in breast cancer development: inflammatory and immune cells in tumour angiogenesis and arteriogenesis
.
Breast Cancer Res
2003
;
5
:
83
8
.
30.
Tabung
FK
,
Steck
SE
,
Zhang
J
,
Ma
Y
,
Liese
AD
,
Agalliu
I
, et al
Construct validation of the dietary inflammatory index among postmenopausal women
.
Ann Epidemiol
2015
;
25
:
398
405
.
31.
Kojima
M
,
Hosoda
H
,
Date
Y
,
Nakazato
M
,
Matsuo
H
,
Kangawa
K
. 
Ghrelin is a growth-hormone-releasing acylated peptide from stomach
.
Nature
1999
;
402
:
656
60
.
32.
Dixit
VD
,
Schaffer
EM
,
Pyle
RS
,
Collins
GD
,
Sakthivel
SK
,
Palaniappan
R
, et al
Ghrelin inhibits leptin- and activation-induced proinflammatory cytokine expression by human monocytes and T cells
.
J Clin Invest
2004
;
114
:
57
66
.
33.
Yanagi
S
,
Sato
T
,
Kangawa
K
,
Nakazato
M
. 
The homeostatic force of Ghrelin
.
Cell Metab
2018
;
27
:
786
804
.
34.
Jakubowicz
D
,
Froy
O
,
Wainstein
J
,
Boaz
M
. 
Meal timing and composition influence ghrelin levels, appetite scores and weight loss maintenance in overweight and obese adults
.
Steroids
2012
;
77
:
323
31
.
35.
Baatar
D
,
Patel
K
,
Taub
DD
. 
The effects of ghrelin on inflammation and the immune system
.
Mol Cell Endocrinol
2011
;
340
:
44
58
.
36.
Grönberg
M
,
Ahlin
C
,
Naeser
Y
,
Janson
ET
,
Holmberg
L
,
Fjällskog
ML
. 
Ghrelin is a prognostic marker and a potential therapeutic target in breast cancer
.
PLoS One
2017
;
12
:
e0176059
.
37.
Mason
C
,
Xiao
L
,
Imayama
I
,
Duggan
CR
,
Campbell
KL
,
Kong
A
, et al
The effects of separate and combined dietary weight loss and exercise on fasting ghrelin concentrations in overweight and obese women: a randomized controlled trial
.
Clin Endocrinol
2015
;
82
:
369
76
.
38.
Kong
A
,
Neuhouser
ML
,
Xiao
L
,
Ulrich
CM
,
McTiernan
A
,
Foster-Schubert
KE
. 
Higher habitual intake of dietary fat and carbohydrates are associated with lower leptin and higher ghrelin concentrations in overweight and obese postmenopausal women with elevated insulin levels
.
Nutr Res
2009
;
29
:
768
76
.
39.
Naznin
F
,
Toshinai
K
,
Waise
TMZ
,
Okada
T
,
Sakoda
H
,
Nakazato
M
. 
Restoration of metabolic inflammation-related ghrelin resistance by weight loss
.
J Mol Endocrinol
2018
;
60
:
109
18
.
40.
Jayasinghe
SN
,
Breier
BH
,
McNaughton
SA
,
Russell
AP
,
Della Gatta
PA
,
Mason
S
, et al
Dietary patterns in New Zealand Women: evaluating differences in body composition and metabolic biomarkers
.
Nutrients
2019
;
11
:
1643
.
41.
Forsythe
JA
,
Jiang
BH
,
Iyer
NV
,
Agani
F
,
Leung
SW
,
Koos
RD
, et al
Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1
.
Mol Cell Biol
1996
;
16
:
4604
13
.
42.
Muhammad
HFL
,
Vink
RG
,
Roumans
NJT
,
Arkenbosch
LAJ
,
Mariman
EC
,
van Baak
MA
. 
Dietary intake after weight loss and the risk of weight regain: macronutrient composition and inflammatory properties of the diet
.
Nutrients
2017
;
9
:
1205
.
43.
Ramallal
R
,
Toledo
E
,
Martinez
JA
,
Shivappa
N
,
Hebert
JR
,
Martinez-Gonzalez
MA
, et al
Inflammatory potential of diet, weight gain, and incidence of overweight/obesity: the SUN cohort
.
Obesity
2017
;
25
:
997
1005
.
44.
Pinsky
PF
,
Miller
A
,
Kramer
BS
,
Church
T
,
Reding
D
,
Prorok
P
, et al
Evidence of a healthy volunteer effect in the prostate, lung, colorectal, and ovarian cancer screening trial
.
Am J Epidemiol
2007
;
165
:
874
81
.
45.
Galizzi
MM
,
Whitmarsh
L
. 
How to measure behavioral spillovers: a methodological review and checklist
.
Front Psychol
2019
;
10
:
342
.
46.
Zamora-Ros
R
,
Shivappa
N
,
Steck
SE
,
Canzian
F
,
Landi
S
,
Alonso
MH
, et al
Dietary inflammatory index and inflammatory gene interactions in relation to colorectal cancer risk in the Bellvitge colorectal cancer case-control study
.
Genes Nutr
2015
;
10
:
447
.
47.
Tabung
FK
,
Steck
SE
,
Zhang
J
,
Ma
Y
,
Liese
AD
,
Tylavsky
FA
, et al
Longitudinal changes in the dietary inflammatory index: an assessment of the inflammatory potential of diet over time in postmenopausal women
.
Eur J Clin Nutr
2016
;
70
:
1374
80
.
48.
Hebert
JR
,
Ebbeling
CB
,
Matthews
CE
,
Hurley
TG
,
Ma
Y
,
Druker
S
, et al
Systematic errors in middle-aged women's estimates of energy intake: comparing three self-report measures to total energy expenditure from doubly labeled water
.
Ann Epidemiol
2002
;
12
:
577
86
.
49.
Willett
WC
,
editor
.
Nutritional epidemiology
:
Oxford University Press
:
New York
; 
1998
.

Supplementary data