The association between dietary inflammatory index and risk of ...

2 downloads 0 Views 217KB Size Report
Dec 31, 2014 - In the current study, we utilized this novel index in the Women's Health Initiative to prospectively evaluate its association with risk of CRC in ...
Cancer Causes Control (2015) 26:399–408 DOI 10.1007/s10552-014-0515-y

ORIGINAL PAPER

The association between dietary inflammatory index and risk of colorectal cancer among postmenopausal women: results from the Women’s Health Initiative Fred K. Tabung • Susan E. Steck • Yunsheng Ma • Angela D. Liese • Jiajia Zhang • Bette Caan • Lifang Hou • Karen C. Johnson • Yasmin Mossavar-Rahmani Nitin Shivappa • Jean Wactawski-Wende • Judith K. Ockene • James R. Hebert



Received: 22 September 2014 / Accepted: 18 December 2014 / Published online: 31 December 2014 Ó Springer International Publishing Switzerland 2014

Abstract Purpose Inflammation is a process central to carcinogenesis and in particular to colorectal cancer (CRC). Previously, we developed a dietary inflammatory index (DII) from extensive literature review to assess the inflammatory potential of diet. In the current study, we utilized this novel index in the Women’s Health Initiative to prospectively evaluate its association with risk of CRC in postmenopausal women. Methods The DII was calculated from baseline food frequency questionnaires administered to 152,536 women aged 50–79 years without CRC at baseline between 1993 and 1998 and followed through 30 September 2010. Incident CRC cases were ascertained through a central physician adjudication process. Multiple covariate-adjusted Cox proportional hazards regression models were used to

estimate hazard ratios (HR) and 95 % confidence intervals (95 % CI) for colorectal, colon (proximal/distal locations), and rectal cancer risk, by DII quintiles (Q). Results During an average 11.3 years of follow-up, a total of 1,920 cases of CRC (1,559 colon and 361 rectal) were identified. Higher DII scores (representing a more pro-inflammatory diet) were associated with an increased incidence of CRC (HRQ5–Q1 1.22; 95 % CI 1.05, 1.43; ptrend = 0.02) and colon cancer, specifically proximal colon cancer (HRQ5–Q1 1.35; 95 % CI 1.05, 1.67; ptrend = 0.01) but not distal colon cancer (HRQ5–Q1 0.84; 95 % CI 0.61, 1.18; ptrend = 0.63) or rectal cancer (HRQ5–Q1 1.20; 95 % CI 0.84, 1.72; ptrend = 0.65). Conclusion Consumption of pro-inflammatory diets is associated with an increased risk of CRC, especially cancers located in the proximal colon. The absence of a significant association for distal colon cancer and rectal cancer

F. K. Tabung  S. E. Steck (&)  N. Shivappa  J. R. Hebert Cancer Prevention and Control Program, University of South Carolina, 915 Greene Street, Discovery I Building, Columbia, SC 29208, USA e-mail: [email protected]

B. Caan Division of Research, Kaiser Permanente, Oakland, CA, USA

F. K. Tabung  S. E. Steck  A. D. Liese  J. Zhang  N. Shivappa  J. R. Hebert Department of Epidemiology and Biostatistics, University of South Carolina, Columbia, SC, USA S. E. Steck  A. D. Liese  J. R. Hebert Center for Research in Nutrition and Health Disparities, University of South Carolina, Columbia, SC, USA Y. Ma  J. K. Ockene Division of Preventive and Behavioral Medicine, Department of Medicine, University of Massachusetts Medical School, Worcester, MA, USA

L. Hou Department of Preventive Medicine and The Robert H. Lurie Comprehensive Cancer Center, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA K. C. Johnson Department of Preventive Medicine, University of Tennessee Health Science Center, Memphis, TN, USA Y. Mossavar-Rahmani Department of Epidemiology and Population Health, Albert Einstein College of Medicine, Bronx, NY, USA J. Wactawski-Wende Department of Epidemiology and Environmental Health, University at Buffalo, The State University of New York, Buffalo, NY, USA

123

400

may be due to the small number of incident cases for these sites. Interventions that may reduce the inflammatory potential of the diet are warranted to test our findings, thus providing more information for colon cancer prevention. Keywords Dietary inflammatory index  Colorectal cancer  Women’s Health Initiative

Cancer Causes Control (2015) 26:399–408

with the lowest DII quintile [22] as well as in the Bellvitge CRC case–control study where the odds ratio for the fourth quartile compared with the first was 1.65 (95 % CI 1.05–2.60) [23]. Our objective in the current study was to examine whether pro-inflammatory diets, as measured by the DII, are associated with increased risk of CRC in the WHI, a larger, more racially and geographically diverse population of postmenopausal women in the USA.

Introduction Inflammation is a process central to carcinogenesis and other chronic diseases, and there is evidence that diet modulates inflammation [1–4]. Chronic inflammatory conditions are associated with cancer risk; for example, patients with inflammatory bowel disease have an increased risk of developing colorectal cancer (CRC) [5, 6]. Moreover, several studies have shown a reduced risk of colon cancer with use of aspirin or other anti-inflammatory agents [7–9]. Specific components of the diet have been shown to be associated with lower levels of inflammation, e.g., fruits and vegetables, omega-3 polyunsaturated fatty acids (PUFAs), fiber, and moderate alcohol intake [10, 11]. Such components of diet are generally known to have a much wider safety margin with prudent use than do pharmaceutical agents [12]. In contrast, dietary components such as saturated fatty acids (SFA), high-glycemic index foods, and a high x-6/x-3 PUFA ratio are associated with increased levels of inflammation [4, 13–15]. Given that nutrients or foods are not consumed in isolation, a protective or deleterious effect of diet will likely include a combination of these dietary factors [16]. Dietary pattern analysis can provide an approach to examine the relationship between diet and the risk of chronic diseases that produces more intuitively appealing results that may be more predictive of disease risk than are individual foods or nutrients [3, 10, 17–19]. The dietary inflammatory index (DII) was developed [20] and construct validated [21] to assess the overall quality of diet with regard to its inflammatory potential. We previously found that food frequency questionnaire (FFQ)-derived DII scores were significantly associated with inflammatory biomarkers, where higher DII scores (representing more pro-inflammatory diets) were positively associated with interleukin-6 (IL-6), tumor necrosis factor alpha receptor 2 (TNFa-R2), and high-sensitivity C-reactive protein (hs-CRP) in a subset of women in the Women’s Health Initiative (WHI) (unpublished results: Tabung, Steck, Zhang, Ma, Liese, Agalliu et al. 2014). In addition, we observed that a higher DII score was associated with increased risk of CRC in the Iowa Women’s Health Study, with a hazard ratio (HR) of 1.20 [95 % confidence interval (CI) 1.01–1.43] comparing the highest

123

Methods Participants The WHI is a large clinical investigation of strategies for the prevention and control of some of the most common causes of morbidity and mortality among postmenopausal women. The design of the WHI has been described in detail elsewhere [24]. Briefly, the WHI enrolled a total of 161,808 postmenopausal women 50–79 years old, in 40 clinical centers across the USA between 1993 and 1998. The women were enrolled into either the clinical trials (CT) that included 68,132 women or the observational study (OS) that included 93,676 women. There were three overlapping components of the CT, including the dietary modification trial (DMT), hormone therapy trial (HT), which included an estrogen-plus-progestin study of women with a uterus and the estrogen-alone study of women without a uterus, and the calcium and vitamin D trial (CaD). Women who proved to be ineligible for, or who were unwilling to enroll in, the CT were invited to be part of the prospective cohort of women in the OS [24]. Women of racial/ethnic minority groups represented 17.1 % of the overall sample. Exclusion criteria for both the OS and the CT included any medical condition associated with a predicted survival of less than 3 years, alcoholism, other drug dependency, mental illness (e.g., major depressive disorder), dementia, active participation in another intervention trial and not likely to live in the area for at least 3 years. Demographic information and dietary data were obtained by self-report using standardized questionnaires. Certified staff drew blood samples from participants and performed physical exam and measurements, including blood pressure, height, and weight, at the baseline clinic visit. Women were further excluded from the DMT if their FFQ-assessed diets had \32 % energy from fat [25]. The WHI protocol was approved by the institutional review boards at the Clinical Coordinating Center (CCC) at the Fred Hutchinson Cancer Research Center (Seattle, WA) and at each of the 40 clinical centers [26].

Cancer Causes Control (2015) 26:399–408

401

Diet assessment

Outcomes ascertainment

As part of baseline enrollment screening for the WHI, all participants completed a standardized 122-item FFQ developed for the WHI to estimate average daily nutrient intake over the previous 3-month period [25]. FFQs were mailed to participants who completed and returned them to the Clinical Coordinating Centers. FFQ data were considered complete if all adjustment questions, all summary questions, 90 % of the foods, and at least one-half of every food group section was complete [27]. The nutrient database, linked to the University of Minnesota Nutrition Data System for Research (NDSR) [28], is based on the US Department of Agriculture Standard Reference Releases and manufacturer information. This FFQ has demonstrated good comparability to 24-h dietary recall interviews and food records in the WHI [25]. When compared to biomarkers such as doubly labeled water and urinary nitrogen, the FFQ was shown to underestimate energy intake by 27–32 % and protein intake by 10–15 % [29, 30].

The WHI outcomes ascertainment and adjudication methods have been previously described [31]. Briefly, participants (or next-of-kin) self-reported cancer diagnoses on questionnaires annually in the OS or semiannually in the CT through 2005 and annually in both the OS and the CT, thereafter. CRC events reported were verified by centrally trained physician adjudicators after review of medical records and pathology reports. The outcome for these analyses was CRC, including cancers of the colon and rectum (including rectum and rectosigmoid). Proximal colon cancers were defined as cancers of the cecum, ascending colon, right colon, hepatic flexure of colon, and transverse colon (ICD = C18.0, C18.2–18.4), and distal colon cancers were defined as cancers of the splenic flexure of colon, descending colon, left colon, and sigmoid colon (ICD = C18.5–18.7). Separate analyses also were conducted considering stage of CRC at diagnosis (localized, regional, and distant).

Description of the dietary inflammatory index (DII)

Covariates

Details of the development [20] and construct validation [21] of the DII have been previously described. Briefly, an extensive literature search was performed to obtain peerreviewed journal articles that examined the association between six well-known inflammatory biomarkers (IL-1b, IL-4, IL-6, IL-10, TNFa, and CRP) and 45 specific foods and nutrients (components of the DII). Literature-derived inflammatory effect scores for each of the DII components were standardized to a representative global diet database, constructed based on 11 datasets from diverse populations in different parts of the world. Overall DII scores for each individual participant represent the sum of each of the DII components in relation to the comparison global diet database [20]. The DII score characterizes an individual’s diet on a continuum from maximally anti-inflammatory to maximally pro-inflammatory, with a higher DII score indicating a more pro-inflammatory diet and a lower (more negative) DII score indicating a more anti-inflammatory diet. In the WHI FFQ, 32 of the 45 original DII components were available for inclusion in the overall DII score (see [20] for list of 45 DII components). Components such as ginger, turmeric, garlic, oregano, hot pepper, rosemary, eugenol, saffron, flavan-3-ol, flavones, flavonols, flavonones, and anthocyanidins that are included in the original DII calculation [20] were not included in the current study because they were not available from the WHI FFQ. The absence of these components is likely to have a minimal impact on overall DII scores because most of the missing food items are likely consumed in small quantities in this population.

Covariates included in the models were as follows: total energy intake (kcal/day); age (years); body mass index [BMI = weight (kg)/height (m)2] categorized into normal weight (\25 kg/m2), overweight (25 to \ 30 kg/m2), and obese (C30 kg/m2); race groups, European American (EA), African–American (AA), Hispanic (HP), and Asian or Pacific Islander (A/PI); educational levels were categorized into less than high school, some high school/GED, at least some college/graduate education; smoking status was categorized into current, past, and never; physical activity (PA) was categorized based on current public health recommendations [32], as meeting or not meeting PA recommendations (C150 min/week of moderate intensity PA or C75 min/week of vigorous intensity PA versus \150 min/week of moderate intensity PA or \75 min/ week of vigorous intensity PA, respectively); family history of CRC (yes/no); diabetes (yes/no); hypertension (yes/ no); arthritis (yes/no); history of colonoscopy/sigmoidoscopy (yes/no); history of occult blood tests (yes/no); nonsteroidal anti-inflammatory drug (NSAID) use (yes/no); category and duration of estrogen use and category and duration of combined estrogen and progesterone use both categorized into five groups (none, \5y, 5 to \10y, 10 to \15y, and C15y); DMT arm (intervention, control, not randomized to DMT); HT arm (estrogen-alone intervention, estrogen-alone control, combined estrogen and progesterone intervention, estrogen and progesterone control, not randomized to HT); and CaD arm (intervention, control, not randomized to CaD). Data on potential confounders were collected by self-administered

123

402

questionnaires on demographics, medical history, and lifestyle factors [24]. Statistical analyses Data from both components (OS and CT) of the WHI were utilized. Women who reported previous CRC at baseline or who were missing previous CRC status at baseline were excluded (n = 2,387), as well as women with implausible reported total energy intake values (B600 or C5,000 kcal/day, n = 4,688) or extreme BMI values (B15 or C50 kg/m2, n = 2,142). Frequencies and percentages were computed to describe the distribution of covariates across quintiles of the DII. Cox proportional hazard (PH) regression models were used to calculate HRs, 95 % CIs and linear trends for risk of CRC, colon and rectal cancers, by DII quintiles and with adjustment for multiple covariates. Models also were constructed separately for proximal colon cancer and distal colon cancer, as well as for CRC stage at diagnosis. The PH assumption was assessed for each covariate using Martingale-based residuals. Smoking status and CaD arm violated the PH assumption; therefore, all Cox models were stratified by these two covariates. The lowest DII quintile (representing the most anti-inflammatory diet) was the referent for all models. Potential effect modification of the association between the DII and CRC by age group, educational level, smoking status, NSAID use, waist-to-hip ratio, waist circumference, race/ethnicity, and BMI was investigated by stratifying the Cox PH models by levels of the potential effect modifier. Significant effect modification was considered at a p value of 0.10 for the DII 9 covariate interaction term. Potential confounders that changed HRs by [10 % were retained in the final model. Tests of linear trend between colorectal cancer incidence and increments of DII score adjusted for covariates were computed by assigning the median value of each quintile to each participant in the quintile, and this variable was entered into models as ordinal values. In sensitivity analyses, CRC cases that occurred within three years from baseline were excluded to reduce the likelihood that baseline diet may have changed recently due to the presence of subclinical disease. Statistical analyses were conducted using SAS version 9.3Ò (SAS Institute, Cary, NC). All tests were two sided.

Results Table 1 presents the distribution of participants’ characteristics across quintiles of the DII. Participants with higher DII scores (representing a more pro-inflammatory diet) consisted of a higher proportion of women who were

123

Cancer Causes Control (2015) 26:399–408

overweight or obese, not meeting PA guidelines, with lower educational attainment, and current smokers. During an average 11.3 years of follow-up, a total of 1,920 cases of CRC (1,559 colon and 361 rectal) were identified. In the main analysis, consumption of more proinflammatory diets was associated with an increased risk of CRC, comparing the highest with the lowest DII quintile (HR 1.22; 95 % CI 1.05, 1.43; ptrend = 0.02) and colon cancer (HR 1.23; 95 % CI 1.03, 1.46; ptrend = 0.02). The HR for rectal cancer also was elevated in the fifth quintile but was not statistically significant (HR 1.20; 95 % CI 0.84, 1.72; ptrend = 0.65). There was statistically significantly higher risk of proximal colon cancer, but not distal colon cancer (Table 2), for women in the highest DII quintile. The magnitude of risk estimates increased when CRC cases that developed within 3 years from baseline were excluded. For example, the HRs were 1.30 (95 % CI 1.09, 1.56; ptrend = 0.008) for CRC and 1.36 (95 % CI 1.11, 1.66; ptrend = 0.003) for colon cancer comparing the highest with the lowest DII quintile (Table 3). The DII did not appear to be differentially associated with disease stage. Analysis in strata of potential effect modifiers showed differences in the association between DII and CRC in categories of NSAID use, where nonusers of NSAIDs were at increased risk of CRC (HRQ5 vs Q1 1.31, 95 % CI 1.05, 1.65), while risk was not increased among regular users of NSAIDs (HRQ5 vs Q1 1.11, 95 % CI 0.89, 1.38). No other variables (age group, education, smoking, BMI, waist circumference, waist-to-hip ratio, or physical activity) were found to modify the association between the DII and CRC (data not shown).

Discussion In this large prospective examination of the association between the DII and CRC risk, more extreme pro-inflammatory diets (i.e., the highest quintile of intake) were associated with increased risk of CRC. The effects were most apparent for cancers located in the proximal colon. We found no substantial association between the DII and distal colon cancer or rectal cancer, though analyses were likely limited by the small number of incident cancer cases at these sites. Our findings are similar to previous results obtained from the Iowa Women’s Health Study, in which the highest quintile of DII was associated with 20 % increased risk of CRC among postmenopausal women [22]. Our results also support studies of overall diet quality and CRC risk [33–36]. This is expected given that compliance with diets based on dietary recommendations is consistent with healthful culinary traditions [e.g., macrobiotic, dietary approaches to stop hypertension (DASH), and Mediterranean meal

Cancer Causes Control (2015) 26:399–408

403

Table 1 Participant characteristics (n, %) by quintiles of the dietary inflammatory index (DII) at baseline; Women’s Health Initiative, 1993–1998 Characteristic

Q1 (-7.055, \-3.136) (healthiest) n = 30,508

Q2 (-3.136, \-1.995) n = 30,507

Q3 (-1.995, \-0.300) n = 30,507

Q4 (-0.300, \1.953) n = 30,506

Q5 (1.953, 5.636) (least healthy) n = 30,508

Age categories, year 50–59

9,242 (30.3)

9,276 (30.4)

9,960 (32.6)

10,628 (34.8)

11,584 (38.0)

60–69

14,147 (46.4)

14,024 (46.0)

13,672 (44.8)

13,607 (44.6)

13,194 (43.2)

7,119 (23.3)

7,207 (23.6)

6,875 (22.6)

6,271 (20.6)

5,730 (18.8)

Normal (\25)

12,973 (42.5)

11,378 (37.3)

10,635 (34.9)

9,795 (32.2)

9,095 (29.8)

Overweight (25 to \ 30)

10,368 (34.0)

10,638 (34.9)

10,710 (35.1)

10,780 (35.3)

10,752 (35.2)

7,167 (23.5)

8,491 (27.8)

9,162 (30.0)

9,931 (32.5)

10,661 (35.0)

70? Body mass index (kg/m2)

Obesity (C30) Race/ethnicity American Indian or Alaskan Native

87 (0.3)

89 (0.3)

124 (0.4)

141 (0.5)

189 (0.6)

Asian or Pacific Islander

1,332 (4.4)

625 (2.1)

658 (2.2)

783 (2.6)

522 (1.7)

African–American

1,358 (4.5)

1,783 (5.8)

2,313 (7.6)

2,867 (9.4)

638 (2.1)

785 (2.6)

1,140 (3.7)

1,319 (4.3)

1,918 (6.3)

25,860 (84.8) 333 (1.1)

24,978 (82.0) 345 (1.1)

22,792 (74.7) 359 (1.2)

Hispanic/Latino European American Other Missing

26,661 (87.4) 366 (1.1)

26,840 (88.0) 320 (1.0)

66 (0.2)

65 (0.2)

79 (0.2)

73 (0.2)

4,644 (15.2)

84 (0.3)

Physical activity (PA) Not meeting PA recommendations

13,049 (42.8)

16,102 (52.8)

17,326 (56.8)

17,999 (59.0)

Meeting PA recommendations

16,601 (54.4)

13,321 (43.7)

11,817 (38.7)

10,819 (35.5)

8,378 (27.5)

856 (2.8)

1,081 (3.5)

1,367 (4.5)

1,693 (5.5)

1,805 (5.9)

738 (2.4)

1,141 (3.7)

1,472 (4.8)

1,723 (5.6)

2,579 (8.4)

Missing

20,324 (66.6)

Educational level \High school Some high school/GED

13,922 (45.6)

16,358 (53.6)

16,968 (55.6)

17,166 (56.4)

18,755 (61.5)

At least some years of college

15,642 (51.3)

12,784 (42.0)

11,875 (39.0)

11,385 (37.2)

8,916 (29.2)

Missing

206 (0.7)

224 (0.7)

192 (0.6)

232 (0.8)

258 (0.9)

Smoking status Never

15,128 (49.6)

15,415 (50.5)

15,618 (51.2)

15,414 (50.5)

15,234 (49.9)

Past

13,911 (45.6)

13,094 (43.0)

12,462 (40.8)

12,489 (40.9)

11,571 (37.9)

1,101 (3.6) 368 (1.2)

1,680 (5.5) 318 (1.0)

2,068 (6.8) 359 (1.2)

2,224 (7.3) 379 (1.2)

3,323 (10.9) 380 (1.3)

No

26,590 (87.2)

26,418 (86.6)

26,525 (86.9)

26,471 (86.8)

26,444 (86.7)

Yes

2,369 (7.8)

2,525 (8.3)

2,412 (7.9)

2,389 (7.8)

2,286 (7.5)

Missing

1,549 (5.0)

1,564 (5.1)

1,570 (5.2)

1,646 (5.4)

1,778 (5.8)

No

12,988 (42.6)

12,302 (40.3)

13,090 (42.9)

13,803 (45.3)

14,597 (47.8)

Yes

17,520 (57.4)

18,205 (59.7)

17,417 (57.1)

16,703 (54.7)

15,911 (52.2)

Current Missing Family history of colorectal cancer

NSAID use

plans] that tend to be anti-inflammatory [37]. For example, Miller et al. [33] examined the association of four indices developed to capture the DASH dietary pattern and risk of CRC. Increased compliance to the DASH diet was consistently associated with reduced risk of CRC across all four DASH indices, though the extent of the predicted

reduced risk depended on the method used to develop the DASH index. Another study using the DASH diet index also observed a reduced risk of CRC with higher index scores, and similar to the current study results, found a significantly reduced risk of colon but not rectal cancer [34]. In another study, Reedy et al. [35] examined the

123

404

Cancer Causes Control (2015) 26:399–408

Table 2 Hazard ratios and 95 % confidence intervals for colorectal cancer risk across quintiles of the dietary inflammatory index, Women’s Health Initiative, 1993–2010 Q1 (-7.055, \-3.136) (healthiest) Referent

Q2 (-3.136, \-1.995)

Q3 (-1.995, \-0.300)

Q4 (-0.300, \1.953)

Q5 (1.953, 5.636) (least healthy)

HR (95 % CI)a

HR (95 % CI)

HR (95 % CI)

HR (95 % CI)

ptrend

Colorectal cancer Age-adjusted model

1.00

1.10 (0.96, 1.26)

1.06 (0.93, 1.22)

1.11 (0.97, 1.27)

1.38 (1.21, 1.57)* \0.0001

Multivariable-adjusted modelb

1.00

1.05 (0.91, 1.21)

0.98 (0.84, 1.13)

1.02 (0.88, 1.19)

1.22 (1.05, 1.43)*

Colorectal cancer cases, 1,920

365 (19.0 %)

388 (20.2 %)

359 (18.7 %)

373 (19.4 %)

435 (22.7 %)

Age-adjusted model

1.00

1.08 (0.93, 1.26)

1.04 (0.89, 1.21)

1.10 (0.95, 1.28)

1.35 (1.17, 1.56)* \0.0001

Multivariable-adjusted model

1.00

1.05 (0.89, 1.23)

0.98 (0.83, 1.15)

1.07 (0.91, 1.26)

1.23 (1.03, 1.46)*

Colon cancer cases, 1,559

299 (19.2 %)

314 (20.1 %)

288 (18.5 %)

312 (20.0 %)

346 (22.2 %)

Age-adjusted model Multivariable-adjusted model

1.00 1.00

1.18 (0.98, 1.41) 1.16 (0.96, 1.41)

1.00 (0.82, 1.21) 0.98 (0.79, 1.20)

1.15 (0.96, 1.38) 1.15 (0.94, 1.41)

1.35 (1.13, 1.62)* 1.35 (1.09, 1.67)*

Proximal colon cancer cases, 1,034d

193 (18.7 %)

221 (21.4 %)

181 (17.5 %)

210 (20.3 %)

229 (22.2 %)

1.00

0.90 (0.68, 1.20)

1.06 (0.80, 1.39)

1.02 (0.77, 1.35)

1.21 (0.93, 1.59)

0.08

Multivariable-adjusted model

1.00

0.80 (0.58, 1.09)

0.91 (0.67, 1.23)

0.90 (0.67, 1.22)

0.84 (0.61, 1.18)

0.63

Distal colon cancer cases, 428d

90 (21.0 %)

76 (17.7 %)

88 (20.6 %)

88 (20.6 %)

86 (20.1 %)

0.02

Colon cancer 0.02

Proximal colon (C18.0–18.4)c 0.002 0.01

Distal colon (C18.5–18.7)c Age-adjusted model

Rectal cancere Age-adjusted model

1.00

1.19 (0.87, 1.63)

1.17 (0.86, 1.61)

1.15 (0.84, 1.58)

1.48 (1.10, 2.01)*

0.02

Multivariable-adjusted model

1.00

1.07 (0.76, 1.50)

0.98 (0.70, 1.39)

0.84 (0.58, 1.20)

1.20 (0.84, 1.72)

0.65

Rectal cancer cases, 361

66 (18.3 %)

74 (20.5 %)

71 (19.7 %)

61 (16.9 %)

89 (24.6 %)

* Statistically significant a Hazard ratio and 95 % confidence interval b

All multivariable models are adjusted for age, total energy intake, body mass index, race/ethnicity, physical activity, educational level, smoking status, family history of colorectal cancer, hypertension, diabetes, arthritis, history of colonoscopy, history of occult blood tests, NSAID use, category and duration of estrogen use, category and duration of estrogen and progesterone use, dietary modification trial arm, hormone therapy trial arm, and calcium and vitamin D arm

c

ICD-O-2 codes used to define location of colon cancer include C18.0 (cecum), C18.2 (ascending colon, right colon), C18.3 (hepatic flexure of colon), C18.4 (transverse colon), C18.5 (splenic flexure of colon), C18.6 (descending colon, left colon), and C18.7 (sigmoid colon)

d

Proximal and distal colon cancer cases do not add up to the total number of colon cancer cases because of missing ICD codes and exclusion of codes C18.8 and C18.9 for large intestine Not Otherwise Specified

e

Rectal cancer include all rectum and rectosigmoid cases

association of four different dietary indices and CRC risk and found that higher scores on all four indices were associated with a decreased CRC risk in men, but only the Healthy Eating Index 2005 was associated with decreased risk in women. In contrast to our study findings, the WHI DMT’s low-fat dietary pattern did not reduce risk of CRC in postmenopausal women after 8.1 years of follow-up [38]. The targeted dietary intervention would have been expected to have anti-inflammatory effects by reducing dietary fat and increasing consumption of vegetables, fruits, and whole grains [38]. The absence of an intervention effect on CRC risk could be due to an insufficiently large difference in dietary intake between the intervention and comparison groups and shorter follow-up

123

duration compared with current investigation [38]. Alternatively, the fact that the annualized incidence rates of colon polyps or adenomas (self-report) were lower in the intervention group than in the comparison group (2.16 vs 2.35 %, respectively; HR 0.91; 95 % CI 0.87–0.95) suggests that the intervention may have slowed progression to CRC or that more than 8.1 years of follow-up was needed to detect clinically apparent CRC [38]. In any event, these findings are mitigated somewhat in the current analyses by comparing dietary intake across all participants, regardless of intervention status, thus allowing for comparisons of participants across a broader range of intake. As with any observational study, however, the role of unmeasured or residual confounding cannot be entirely ruled out.

Cancer Causes Control (2015) 26:399–408

405

Table 3 Hazard ratios and 95 % confidence intervals for colorectal cancer risk across quintiles of the dietary inflammatory index, excluding colorectal cancer cases which developed within 3 years from baseline, Women’s Health Initiative, 1993–2010 Q1 (-7.055, \-3.139) (healthiest) Referent

Q2 (-3.139, \-1.997)

Q3 (-1.997, \-0.306)

Q4 (-0.306, \1.949)

ptrend

HR (95 % CI)

Q5 (1.949, 5.636) (least healthy) HR (95 % CI)

a

HR (95 % CI)

Age-adjusted modeld

1.00

1.10 (0.94, 1.28)

Multivariable-adjusted modele

1.00

1.07 (0.91, 1.27)

1.07 (0.92, 1.25)

1.10 (0.94, 1.29)

1.42 (1.23, 1.65)*

\0.0001

1.00 (0.84, 1.19)

1.04 (0.87, 1.23)

1.30 (1.09, 1.56)*

Colorectal cancer cases, 1,462

274 (18.7 %)

0.008

296 (20.2 %)

273 (18.7 %)

282 (19.3 %)

337 (23.1 %)

Age-adjusted modeld Multivariable-adjusted modele

1.00

1.12 (0.94, 1.33)

1.05 (0.89, 1.25)

1.13 (0.95, 1.34)

1.46 (1.24, 1.72)*

\0.0001

1.00

1.11 (0.92, 1.33)

1.00 (0.82, 1.21)

1.11 (0.92, 1.33)

1.36 (1.11, 1.66)*

Colon cancer cases, 1,202

0.003

222 (18.5 %)

246 (20.5 %)

218 (18.1 %)

239 (19.9 %)

277 (23.0 %)

Age-adjusted model Multivariable-adjusted model

1.00 1.00

1.20 (0.96, 1.48) 1.18 (0.95, 1.46)

0.96 (0.76, 1.21) 0.93 (0.74, 1.18)

1.20 (0.97, 1.50) 1.15 (0.91, 1.44)

1.45 (1.17, 1.80)* 1.38 (1.09, 1.75)*

Proximal colon cancer cases, 827d

153 (18.5 %)

179 (21.6 %)

139 (16.8 %)

168 (20.3 %)

188 (22.7 %)

1.00

0.91 (0.63, 1.33)

1.17 (0.82, 1.67)

1.01 (0.76, 1.58)

1.23 (0.86, 1.75)

0.17

Multivariable-adjusted model

1.00

0.86 (0.59, 1.26)

1.08 (0.75, 1.55)

0.97 (0.67, 1.41)

1.03 (0.69, 1.54)

0.71

Distal colon cancer cases, 296d

58 (19.6 %)

52 (17.6 %)

65 (22.0 %)

59 (19.9 %)

62 (20.9 %)

HR (95 %bCI)

Colorectal cancer

Colon cancer

Proximal colon (C18.0–18.4)c 0.0007 0.01

Distal colon (C18.5–18.7)c Age-adjusted model

Rectal cancere Age-adjusted modeld

1.00

1.00 (0.69, 1.43)

1.15 (0.81, 1.64)

1.00 (0.69, 1.44)

1.27 (0.89, 1.80)*

0.22

Multivariable-adjusted modele

1.00

0.93 (0.63, 1.38)

0.99 (0.67, 1.47)

0.77 (0.51, 1.16)

1.08 (0.71, 1.65)

0.96

Rectal cancer cases, 260

52 (20.0 %)

50 (19.2 %)

55 (21.5 %)

43 (16.5 %)

60 (23.1 %)

* Statistically significant Hazard ratio and associated 95 % confidence interval

a

b

All multivariable models are adjusted for total energy intake, age, body mass index, race/ethnicity, physical activity, educational level, smoking status, family history of colorectal cancer, hypertension, diabetes, arthritis, history of colonoscopy, history of occult blood tests, NSAID use, category and duration of estrogen use, category and duration of estrogen and progesterone use, dietary modification trial arm, hormone therapy trial arm, and calcium and vitamin D trial arm

c

ICD-O-2 codes used to define location of colon cancer include C18.0 for cecum, C18.2, ascending colon, right colon, C18.3, hepatic flexure of colon, C18.4, transverse colon, C18.5, splenic flexure of colon, C18.6, descending colon, left colon, and C18.7, sigmoid colon

d

Proximal and distal colon cancer cases do not add up to the total number of colon cancer cases because of missing ICD codes and codes C18.8 and C18.9 for large intestine Not Otherwise Specified are not included

e

Rectal cancer include all rectum and rectosigmoid cases

The difference in results by anatomic site, with significant associations in the proximal colon but not in the distal colon or rectum, is similar to other studies and supports the idea that CRC is a heterogeneous group of diseases. Studies have found differences in risk factors for colon and rectal cancer; for example, obesity—a state of low-grade chronic systemic inflammation—is associated with increased risk of colon cancer [39], but not rectal cancer [40]. Biologically, rectal and distal colon tumors have been found to share similar mutational frequencies and other characteristics which are different from those observed in tumors of the proximal colon [41, 42] and may thus be influenced by different mechanisms of carcinogenesis [43].

Our results appeared to be modified by regular use of NSAIDs, where pro-inflammatory diets were associated with higher risk of CRC only among nonusers of NSAIDs. The adverse effects of a pro-inflammatory diet on inflammation may be potentially masked by the stronger effects of NSAIDs among regular users. The anti-inflammatory effect of NSAIDs on the colonic epithelium could be so strong as to render inconsequential the relative contribution of dietary inflammatory potential, and thus, a DII-CRC association would not be observed among NSAID users, yet would be strong among nonusers of NSAIDs [44–46]. Evidence from experimental models of colon carcinogenesis indicates that cytokines derived from inflammatory cells may drive the uncontrolled proliferation of cancer

123

406

cells, either directly or indirectly [9, 47, 48]. The link between inflammation and colon cancer is further supported by evidence from studies showing a positive association between higher concentrations of inflammatory biomarkers and increased risk of colon cancer [49–51], or a reduced risk of colon cancer with regular use of NSAIDs [7–9]. A pro-inflammatory diet also may be linked to increased colon cancer risk through some component of the metabolic syndrome, especially insulin resistance or glucose intolerance [52, 53]. We previously found that a higher DII score was associated with glucose intolerance among police officers in the Buffalo Cardio-Metabolic Occupational Police Stress study [54]. Glucose intolerance and insulin resistance may lead to CRC through the growth-promoting effects of elevated levels of insulin, glucose, or triglycerides [53]. Strengths of the current study include a large, wellcharacterized population of more than 150,000 women, a long follow-up period, the inclusion of women of diverse race/ethnic groups, and the central adjudication of CRC diagnosis. The use of a novel dietary index to score diet quality based on inflammatory potential supports the evidence linking inflammation and CRC. Limitations include known measurement error in using an FFQ for dietary assessment such as underreporting of energy and protein [29, 30] which may have led to underestimation of the impact of pro-inflammatory diets on CRC in this study. Changes in inflammatory potential of diet over time were not accounted for because only baseline dietary data were used to calculate the DII. It also is important to note that components missing from the FFQ, including ginger, turmeric, garlic, oregano, pepper, rosemary, eugenol, saffron, flavan-3-ol, flavones, flavonols, flavonones, and anthocyanidins, are strongly anti-inflammatory. Even though we showed previously that reasonable predictive ability was retained when replacing 24-h recall-derived DII scores with those derived from a structured questionnaire with fewer DII components [21], there still may be a fall-off in predictive ability in a population that was actively trying to change to a more healthful diet and therefore might be more likely to begin consuming these food items that are not on the FFQ list. Finally, the smaller number of incident cases of distal colon cancer and rectal cancer may have limited our ability to observe an association with the DII in this study.

Cancer Causes Control (2015) 26:399–408

diet and suggests reduction in the inflammatory potential of the diet as a target for future intervention studies aimed at colon cancer prevention. Acknowledgments This project is supported by the Prevent Cancer Foundation Living in Pink grant. The University of South Carolina SPARC grant supported Dr. Fred Tabung, and the National Institutes of Health/National Cancer Institute provided support to Dr. James Hebert (via U54 CA153461 and K05 CA136975). Dr. Yunsheng Ma was partly supported by grant Nos. 1R21 DK083700-01A1 and 1R01HL09457501A1. The National Institutes of Health funded the WHI program through contracts HHSN268201100046C, HHSN268201100001C, HHSN268201100002C, HHSN268201100003C, HHSN268201100004C, and HHSN271201100004C. Conflict of interest All authors declare that they have no conflict of interest.

Appendix: Short list of WHI investigators Program Office (National Heart, Lung, and Blood Institute, Bethesda, Maryland) Jacques Rossouw, Shari Ludlam, Dale Burwen, Joan McGowan, Leslie Ford, and Nancy Geller. Clinical Coordinating Center (Fred Hutchinson Cancer Research Center, Seattle, WA) Garnet Anderson, Ross Prentice, Andrea LaCroix, and Charles Kooperberg. Investigators and Academic Centers (Brigham and Women’s Hospital, Harvard Medical School, Boston, MA) JoAnn E. Manson; (MedStar Health Research Institute/ Howard University, Washington, DC) Barbara V. Howard; (Stanford Prevention Research Center, Stanford, CA) Marcia L. Stefanick; (The Ohio State University, Columbus, OH) Rebecca Jackson; (University of Arizona, Tucson/Phoenix, AZ) Cynthia A. Thomson; (University at Buffalo, Buffalo, NY) Jean Wactawski-Wende; (University of Florida, Gainesville/Jacksonville, FL) Marian Limacher; (University of Iowa, Iowa City/Davenport, IA) Robert Wallace; (University of Pittsburgh, Pittsburgh, PA) Lewis Kuller; and (Wake Forest University School of Medicine, Winston-Salem, NC) Sally Shumaker. Women’s Health Initiative Memory Study (Wake Forest University School of Medicine, Winston-Salem, NC) Sally Shumaker. For a list of all the investigators who have contributed to WHI science, please visit: https://www.whi.org/researchers/ Documents%20%20Write%20a%20Paper/WHI%20Investi gator%20Long%20List.pdf.

Conclusion Consuming a diet with high pro-inflammatory potential is associated with an increased risk of CRC, especially cancer of the proximal colon. This finding strengthens the evidence for a new tool assessing the overall inflammatory capacity of

123

References 1. Villasen˜or A, Ambs A, Ballard-Barbash R, Baumgartner KB, McTiernan A, Ulrich CM, Neuhouser ML (2011) Dietary fiber is

Cancer Causes Control (2015) 26:399–408

2.

3.

4. 5. 6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17. 18.

19. 20.

associated with circulating concentrations of C-reactive protein in breast cancer survivors: the heal Study. Breast Cancer Res Treat 129:485–494 Esmaillzadeh A, Kimiagar M, Mehrabi Y, Azadbakht L, Hu FB, Willett WC (2007) Dietary patterns and markers of systemic inflammation among iranian women. J Nutr 137:992–998 Chrysohoou C, Panagiotakos DB, Pitsavos C, Das UN, Stefanadis C (2004) Adherence to the Mediterranean diet attenuates inflammation and coagulation process in healthy adults: the Attica Study. J Am Coll Cardiol 44:152–158 Galland L (2010) Diet and inflammation. Nutr Clin Pract 25:634–640 Ullman T, Itzkowitz SH (2011) Intestinal inflammation and cancer. Gastroenterology 140:1807–1816 Ananthakrishnan A, Cheng SC, Cai T, Cagan A, Gainer VS, Szolovits P, Shaw SY, Churchill S, Karlson EW, Murphy SN, Kohane I, Liao KP (2014) Serum inflammatory markers and risk of colorectal cancer in patients with inflammatory bowel diseases. Clin Gastroenterol Hepatol 12(1342–8):e1 Dube´ C, Rostom A, Lewin G, Tsertsvadze A, Barrowman N, Code C, Sampson M, Moher D, U.S. Preventive Services Task Force (2007) The use of aspirin for primary prevention of colorectal cancer: a systematic review prepared for the U.S. Preventive Services Task Force. Ann Intern Med 146:365–375 Rostom A, Dube´ C, Lewin G, Tsertsvadze A, Barrowman N, Code C, Sampson M, Moher D, U.S. Preventive Services Task Force (2007) Nonsteroidal anti-inflammatory drugs and cyclooxygenase-2 inhibitors for primary prevention of colorectal cancer: a systematic review prepared for the U.S. Preventive Services Task Force. Ann Intern Med 146:376–389 Balkwill F, Charles KA, Mantovani A (2005) Smoldering and polarized inflammation in the initiation and promotion of malignant disease. Cancer Cell 7:211–217 Nanri A, Moore MA, Kono S (2007) Impact of C-reactive protein on disease risk and its relation to dietary factors: literature review. Asian Pac J Cancer Prev 8:167–177 Hlebowicz J, Persson M, Gullberg B, Sonestedt E, Wallstro¨m P, Drake I, Nilsson J, Hedblad B, Wirfa¨lt E (2011) Food patterns, inflammation markers and incidence of cardiovascular disease: the Malmo¨ Diet and Cancer study. J Intern Med 270:365–376 Rennie K, Hughes J, Lang R, Jebb SA (2003) Nutritional management of rheumatoid arthritis: a review of the evidence. J Hum Nutr Diet 16:97–109 Raphael W, Sordillo L (2013) Dietary polyunsaturated fatty acids and inflammation: the role of phospholipid biosynthesis. Int J Mol Sci 14:21167–21188 Liu S, Manson JE, Buring JE, Stampfer MJ, Willett WC, Ridker PM (2002) Relation between a diet with a high glycemic load and plasma concentrations of high-sensitivity C-reactive protein in middle-aged women. Am J Clin Nutr 75:492–498 Peairs A, Rankin JW, Lee YW (2011) Effects of acute ingestion of different fats on oxidative stress and inflammation in overweight and obese adults. Nutr J 10:122 World Cancer Research Fund and the American Institute for Cancer Research (2007) Food, nutrition and prevention of cancer: a global perspective. American Institute for Cancer Research, Washington Hu F (2002) Dietary pattern analysis: a new direction in nutritional epidemiology. Curr Opin Lipidol 13:3–9 Randall E, Marshall JR, Graham S, Brasure J (1990) Patterns in food use and their associations with nutrient intakes. Am J Clin Nutr 52:739–745 Kant AK (1996) Indexes of overall diet quality: a review. J Am Diet Assoc 96:785–791 Shivappa N, Steck SE, Hurley TG, Hussey JR, Hebert JR (2013) Designing and developing a literature-derived, population-based dietary inflammatory index. Public Health Nutr 17:1689–1696

407 21. Shivappa N, Steck SE, Hurley TG, Hussey JR, Ma Y, Ockene IS, Tabung FK, Hebert JR (2013) A population-based dietary inflammatory index predicts levels of C-reactive protein in the Seasonal Variation of Blood Cholesterol Study (SEASONS). Public Health Nutrition. 17:1825–1833 22. Shivappa N, Prizment AE, Blair CK, Jacobs DR Jr, Steck SE, Hebert JR (2014) Dietary inflammatory index (DII) and risk of colorectal cancer in Iowa Women’s Health Study. Cancer Epidemiol Biomark Prev 23:2383–2392 23. Zamora-Ros R, Shivappa N, Steck SE et al (2015) Dietary inflammatory index and inflammatory gene interactions in relation to colorectal cancer risk in the Bellvitge colorectal cancer case–control study. Genes Nutr 10:447 24. Women’s Health Initiative Study Group (1998) Design of the Women’s Health Initiative clinical trial and observational study. Control Clin Trials 19:61–109 25. Patterson R, Kristal AR, Tinker LF, Carter RA, Bolton MP, Agurs-Collins T (1999) Measurement characteristics of the Women’s Health Initiative food frequency questionnaire. Ann Epidemiol 9:178–187 26. Ma Y, He´bert JR, Li W, Bertone-Johnson ER, Olendzki B, Pagoto SL, Tinker L, Rosal MC, Ockene IS, Ockene JK, Griffith JA, Liu S (2008) Association between dietary fiber and markers of systemic inflammation in the Women’s Health Initiative Observational Study. Nutrition 24:941–949 27. Patterson RE, Kristal A, Rodabough R et al (2003) Changes in food sources of dietary fat in response to an intensive low-fat dietary intervention: early results from the Women’s Health Initiative. J Am Diet Assoc 103:454–460 28. Nutrition Coordinating Center at the University of Minnesota M, MN (2013) Nutrition Data System for Research (NDSR). University of Minnesota, Minneapolis, MN 29. Neuhouser ML, Tinker L, Shaw PA et al (2008) Use of recovery biomarkers to calibrate nutrient consumption self-reports in the Women’s Health Initiative. Am J Epidemiol 167:1247–1259 30. Prentice RL, Mossavar-Rahmani Y, Huang Y et al (2011) Evaluation and comparison of food records, recalls, and frequencies for energy and protein assessment by using recovery biomarkers. Am J Epidemiol 174:591–603 31. Curb J, McTiernan A, Heckbert SR, Kooperberg C, Stanford J, Nevitt M, Johnson KC, Proulx-Burns L, Pastore L, Criqui M, Daugherty S, WHI Morbidity and Mortality Committee (2003) Outcomes ascertainment and adjudication methods in the Women’s Health Initiative. Ann Epidemiol 13:S122–S128 32. Department of Health and Human Services (2008) Physical activity guidelines for Americans. DHHS, Washington DC 33. Miller P, Cross AJ, Subar AF, Krebs-Smith SM, Park Y, PowellWiley T, Hollenbeck A, Reedy J (2013) Comparison of 4 established DASH diet indexes: examining associations of index scores and colorectal cancer. Am J Clin Nutr 98:794–803 34. Fung T, Hu FB, Wu K, Chiuve SE, Fuchs CS, Giovannucci E (2010) The Mediterranean and Dietary Approaches to Stop Hypertension (DASH) diets and colorectal cancer. Am J Clin Nutr 92:1429–1435 35. Reedy J, Mitrou PN, Krebs-Smith SM, Wirfa¨lt E, Flood A, Kipnis V, Leitzmann M, Mouw T, Hollenbeck A, Schatzkin A, Subar AF (2008) Index-based dietary patterns and risk of colorectal cancer. Am J Epidemiol 168:38–48 36. Romaguera D, Vergnaud AC, Peeters PH, van Gils CH, Chan DS, Ferrari P, Romieu I, Jenab M, Slimani N, Clavel-Chapelon F, Fagherazzi G, Perquier F, Kaaks R, Teucher B, Boeing H, von Ru¨sten A, Tjønneland A, Olsen A, Dahm CC, Overvad K, Quiro´s JR, Gonzalez CA, Sa´nchez MJ, Navarro C, Barricarte A, Dorronsoro M, Khaw KT, Wareham NJ, Crowe FL, Key TJ, Trichopoulou A, Lagiou P, Bamia C, Masala G, Vineis P, Tumino R, Sieri S, Panico S, May AM, Bueno-de-Mesquita HB, Bu¨chner

123

408

37.

38.

39.

40.

41.

42.

43.

44.

Cancer Causes Control (2015) 26:399–408 FL, Wirfa¨lt E, Manjer J, Johansson I, Hallmans G, Skeie G, Benjaminsen Borch K, Parr CL, Riboli E, Norat T (2012) Is Concordance with World Cancer Research Fund/American Institute for Cancer Research Guidelines for cancer prevention related to subsequent risk of cancer? Results from the EPIC study. Am J Clin Nutr 96:150–163 Steck S, Shivappa N, Tabung FK, Harmon BE, Wirth MD, Hurley TG, Hebert JR (2014) The dietary inflammatory index: a new tool for assessing diet quality based on inflammatory potential. Digest 49:1–9 Beresford S, Johnson KC, Ritenbaugh C, Lasser NL, Snetselaar LG, Black HR, Anderson GL, Assaf AR, Bassford T, Bowen D, Brunner RL, Brzyski RG, Caan B, Chlebowski RT, Gass M, Harrigan RC, Hays J, Heber D, Heiss G, Hendrix SL, Howard BV, Hsia J, Hubbell FA, Jackson RD, Kotchen JM, Kuller LH, LaCroix AZ, Lane DS, Langer RD, Lewis CE, Manson JE, Margolis KL, Mossavar-Rahmani Y, Ockene JK, Parker LM, Perri MG, Phillips L, Prentice RL, Robbins J, Rossouw JE, Sarto GE, Stefanick ML, Van Horn L, Vitolins MZ, Wactawski-Wende J, Wallace RB, Whitlock E (2006) Low-fat dietary pattern and risk of colorectal cancer: the Women’s Health Initiative Randomized Controlled Dietary Modification Trial. J Am Med Assoc 295:643–654 Caan B, Coates AO, Slattery ML, Potter JD, Quesenberry CP Jr, Edwards SM (1998) Body size and the risk of colon cancer in a large case–control study. Int J Obes Relat Metab Disord 22:178–184 Slattery M, Caan BJ, Benson J, Murtaugh M (2003) Energy balance and rectal cancer: an evaluation of energy intake, energy expenditure, and body mass index. Nutr Cancer 46:166–171 Slattery M, Curtin K, Wolff RK, Boucher KM, Sweeney C, Edwards S, Caan BJ, Samowitz W (2009) A comparison of colon and rectal somatic DNA alterations. Dis Colon Rectum 52:1304–1311 Rosenberg R, Maak M, Schuster T, Becker K, Friess H, Gertler R (2010) Does a rectal cancer of the upper third behave more like a colon or a rectal cancer? Dis Colon Rectum 53:761–770 Konishi K, Fujii T, Boku N, Kato S, Koba I, Ohtsu A, Tajiri H, Ochiai A, Yoshida S (1999) Clinicopathological differences between colonic and rectal carcinomas: Are they based on the same mechanism of carcinogenesis? Gut 45:818–821 Tabung FK, Steck SE, Burch JB, Chen CF, Zhang H, Hurley TG, Cavicchia P, Alexander M, Shivappa N, Creek KE, Lloyd SC, Hebert JR (2014) A healthy lifestyle index is associated with

123

45.

46.

47. 48. 49.

50.

51.

52.

53. 54.

reduced risk of colorectal adenomatous polyps among non-users of non-steroidal anti-inflammatory drugs. J Prim Prevent 1–11. Epub ahead of print. doi:10.1007/s10935-014-0372-1 Hauret KG, Bostick RM, Matthews CE, Hussey JR, Fina MF, Geisinger KR, Roufail WM (2004) Physical activity and reduced risk of incident sporadic colorectal adenomas: observational support for mechanisms involving energy balance and inflammation modulation. Am J Epidemiol 159:983–992 Hartman TJ, Yu B, Albert PS, Slattery ML, Paskett E, Kikendall JW, Iber F, Brewer BK, Schatzkin A, Lanza E (2005) Does nonsteroidal anti-inflammatory drug use modify the effect of a low-fat, high-fiber diet on recurrence of colorectal adenomas? Cancer Epidemiol Biomark Prev 14:2359–2365 Monteleone G, Pallone F, Stolfi C (2012) The dual role of inflammation in colon carcinogenesis. Int J Mol Sci 13:11071–11084 Balkwill F (2004) Cancer and the chemokine network. Nat Rev Cancer 4:540–550 Taniguchi K, Karin M (2014) IL-6 and related cytokines as the critical lynchpins between inflammation and cancer. Semin Immunol 26:54–74. Toriola A, Cheng TY, Neuhouser ML, Wener MH, Zheng Y, Brown E, Miller JW, Song X, Beresford SA, Gunter MJ, Caudill MA, Ulrich CM (2013) Biomarkers of inflammation are associated with colorectal cancer risk in women but are not suitable as early detection markers. Int J Cancer 132:2648–2658 Wu J, Cai Q, Li H, Cai H, Gao J, Yang G, Zheng W, Xiang YB, Shu XO (2013) Circulating C-reactive protein and colorectal cancer risk: a report from the shanghai men’s health study. Carcinogenesis 34:2799–2803 Bruce W, Wolever TM, Giacca A (2000) Mechanisms linking diet and colorectal cancer: the possible role of insulin resistance. Nutr Cancer 37:19–26 Giovannucci E (1995) Insulin and colon cancer. Cancer Causes Control 6:164–179 Wirth M, Burch J, Shivappa N, Violanti JM, Burchfiel CM, Fekedulegn D, Andrew ME, Hartley TA, Miller DB, Mnatsakanova A, Charles LE, Steck SE, Hurley TG, Vena JE, He´bert JR (2014) Association of a dietary inflammatory index with inflammatory indices and metabolic syndrome among police officers. J Occup Environ Med 56:986–989