Abstract
Background
Red and processed meat, recognized carcinogens, are risk factors for colorectal neoplasia, including polyps, the precursor for colorectal cancer. The mechanism is unclear. One possible explanation is the mutagenic activity of these foods, perhaps due to generation during cooking [e.g., heterocyclic amine (HCA) intake]. Few studies have evaluated meat intake and sessile serrated lesion (SSL) risk, a recently recognized precursor, and no study has evaluated meat cooking methods and meat-derived mutagens with SSL risk.
Objective
We evaluated intakes of meat, meat cooking methods, and inferred meat mutagens with SSL risk and in comparison to risk of other polyps.
Methods
Meat, well-done meat, and inferred meat mutagen intakes were evaluated. Polytomous logistic regression models were used to estimate ORs and 95% CIs among cases (556 hyperplastic polyp, 1753 adenoma, and 208 SSL) and controls (3804) in the large colonoscopy-based, case-control study, the Tennessee Colorectal Polyp Study.
Results
The highest quartile intakes of red meat (OR: 2.38; 95% CI: 1.44, 3.93), processed meat (OR: 2.03; 95% CI: 1.30, 3.17), well-done red meat (OR: 2.19; 95% CI: 1.34, 3.60), and the HCA 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQX; OR: 2.48; 95% CI: 1.49, 4.16) were associated with increased risk of SSLs in comparison to the lowest quartile intake.
Conclusions
High intakes of red and processed meats are strongly and especially associated with SSL risk and part of the association may be due to HCA intake. Future studies should evaluate other mechanism(s) and the potential for primary prevention.
Keywords: sessile serrated lesion, colorectal, adenoma, meat, heterocyclic amines, polycyclic aromatic hydrocarbons, etiology
Introduction
Colorectal cancer (CRC) is the third most common cancer in the United States and the second leading cause of cancer-related mortality (1). The primary pathway to CRC is the conventional adenoma (AD)–carcinoma sequence (also known as the chromosomal instability pathway) (1, 2). There are other colorectal polyps in addition to AD. Hyperplastic polyps (HPs) are a common polyp type that has been presumed to be benign with little malignant potential. Within the past 2 decades, a third type of colorectal polyp, the sessile serrated lesion (SSL), has been recognized. A diagnostic consensus for SSL was not reached until 2010 (1, 3-6). SSLs represent an alternative pathway to carcinogenesis, which may account for 20–30% of CRCs (7, 8), particularly microsatellite instable cancer (9).
Recent studies suggest that SSLs are overrepresented in interval cancers, which may represent the potential for more rapid conversion to malignancy or may reflect their difficulty to locate on colonoscopy, which may lead to missed lesions (10). Thus, it is valuable to understand how the development of these polyps might be impacted by modifiable lifestyle factors. However, given the recency of the SSL diagnosis consensus, there are few epidemiologic studies that have evaluated risk factors for SSLs.
The WHO has classified red meat as a class 2A carcinogen, indicating that it probably causes cancer, and processed meat as a group 1 carcinogen, indicating that it is known to cause cancer. A recent expert panel report also supported a role for red and processed meats in CRC risk (11). Meat-derived mutagens may account for red and processed meat carcinogenicity. Heterocyclic amines (HCAs) and polycyclic aromatic hydrocarbons (PAHs) are generated during high-temperature cooking of meat whose reactive metabolites may cause DNA damage. Their mutagenic activity varies based on many factors including the method of cooking, cooking time, and temperature (12). HCAs become capable of damaging DNA when they are activated by specific enzymes in the body; the bioactivity of these enzymes differs among people and may contribute to cancer risks associated with exposure to HCAs (13, 14). One potential mechanism that could explain the mutagenicity of HCAs and PAHs is the formation of DNA adducts, which increase with the intake of dietary HCAs and PAHs (12, 15).
There are few studies that have evaluated the association of meat intake and inferred meat-derived mutagens with SSLs, including our previous publication evaluating total red meat intake in which we observed a strong association between red meat intake and SSLs (16). However, to our best knowledge, no previous study has evaluated the risk of SSLs with processed meat intake, meat doneness levels, or inferred meat-mutagen intake. In this study, we conduct this analysis and compare risks for SSLs with risks for AD and HP.
Methods
Study design and population
Data from the case-control Tennessee Colorectal Polyp Study (TCPS) was used in this analysis. Details of the study design have been previously published (16). In brief, the TCPS was conducted from February 2003 to October 2010 and included 7621 participants recruited from colonoscopy clinics at Vanderbilt University Medical Center and the Tennessee Valley Veterans Affairs Medical Center (Supplemental Figure 1). The primary outcome of the TCPS was risk of colorectal polyps. The study was approved for human subjects research by both institutions, and all study participants provided written informed consent. Individuals were excluded if they had a personal history of adenoma, cancer, inflammatory bowel disease, or familial CRC syndrome. The majority of participants (90.5%) were recruited prior to colonoscopy before their examination findings were known.
Patient and public involvement
During the multiyear fielding of this study, patient advocates were included in the study team and participated in review of study procedures, progress, and research findings.
Meat and inferred meat-mutagen intake amounts
Of the 7621 participants providing an informed written consent and participating in ≥1 component of the study, 6462 (84.3%) completed an interviewer-administered telephone questionnaire that solicited information about family history, health history, and lifestyle factors, including a detailed assessment of meat intake and meat doneness using an at-home booklet with pictures of cooked meats, which has been previously described in detail (Supplemental Figure 1) (17). Intakes of other dietary factors were collected from a self-administered FFQ developed to capture dietary intake in the southeastern United States (18).
Participants were asked about their usual intake in the past 12 mo of 11 meats (hamburgers or cheeseburgers from fast food, hamburgers or cheeseburgers not from fast food, beef steaks, pork chops or ham steaks, bacon, sausage, hotdogs or franks, chicken, fish, meat gravies made with drippings, and short ribs or spareribs). For each meat, participants were asked to report their usual intake frequency (never or once a month or less, 2–3 times/mo, once a week, 2–3 times/wk, 4–6 times/wk, once a day, or ≥2 times/d) and serving size (small, medium, or large based on specified weights) and the frequency (never, about one-quarter of the time, about one-half of the time, about three-quarters of the time, and always) of ≤5 different cooking methods (oven-broiled or oven-baked, grilled or barbecued, pan-fried, deep fried, or all other ways). In addition, participants were asked to select their preference for meat doneness level for 7 different types of meat or preparations of meat (hamburger, cheeseburger, or beef patties, beef steaks, pork chops, bacon, grilled chicken, pan-fried chicken, and pan-fried or grilled fish) using a series of 3 photographs in their at-home booklet of each meat cooked to increasing doneness levels. Intakes of meat were converted to inferred meat-derived mutagens based on the amount of meat intake and the usual level of meat well-doneness using the software CHARRED created by the National Cancer Institute as previously described (19). This analysis examined the following specific mutagens: 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline (MeIQX), 2-amino-3,4,8-trimethylimidazo[4,5-f]-quinoxaline (DiMeIQX), 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), benzo[α]pyrene (BaP), and total mutagenic activity.
Case and control definitions
Individuals were classified into polyp cases or controls based on the findings at the colonoscopy. All colonoscopies were performed as part of routine care using standard clinical practice. Individuals were classified as a control if they had no polyps at the colonoscopy during a complete colonoscopy with confirmed visualization of the cecum (n = 3804). Individuals with polyps were classified into case groups based on the histopathological diagnosis of all retrieved polyps. Diagnoses of all polyps were standardized by a study pathologist as previously described (16). AD cases had ≥1 tubular, tubulovillous, or villous AD without a synchronous SSL. SSL cases had ≥1 SSL. HP cases had ≥1 HP without any other synchronous polyp.
Statistical analysis
The current analysis is based on a total of 6361 eligible participants (86.0%) who completed the telephone interview meat questionnaire, including 212 SSL cases, 559 HP cases, 1787 AD cases, and 3803 polyp-free controls (Supplemental Figure 1). Characteristics were compared between cases and controls using Mantel-Haenszel chi-square testing for categorical variables and general linear models for continuous variables with adjustment for age (5-y categories) and sex, where appropriate. Quantiles were derived for meat and inferred meat-mutagen intake amounts using the distribution among controls. SSL cases were compared with controls, HP cases, and AD cases. The associations between intake and the risks of HPs, ADs, and SSLs were evaluated using multinomial logistic regression models to estimate ORs and 95% CIs. All models were adjusted for sex (male/female), age (years), recruitment site [academic, Veterans Affairs (VA)], race (white, nonwhite), indication for colonoscopy (screening, family history, diagnostic, other), regular alcohol consumption status (never, former, current), cigarette smoking status (never, former, current), regular exercise in the past 10 y (no/yes), BMI (kg/m2), educational attainment (high school or less, some college, college graduate, graduate/professional school), current regular use of nonsteroidal anti-inflammatory drugs (NSAIDs; no/yes), and total energy intake (kilocalories per day derived from an FFQ). For individuals who did not provide an FFQ (n = 916 including 31 SSL cases), total energy intake was imputed by assigning the group mean within age (5-y categories), sex (male/female), and study site (academic/VA) strata. An ɑ of 0.0167 (0.05/3 comparison groups) was used. All statistical analyses were completed using Statistical Analysis System (SAS Enterprise 7.15; SAS Institute).
Ethics approval
Written informed consent was obtained from all study participants, and the study protocol was approved by the institutional review board at each study site.
Results
Demographic characteristics of each of the 4 groups examined (no-polyp controls, ADs, HPs, and SSLs) are shown in Table 1. In comparison to controls, SSL cases were more likely to be male (P = 0.016) and have a current or past smoking history (P < 0.001), a family history of CRC (P = 0.008), and a lower use of NSAIDs (P = 0.007). In comparison to AD cases, SSL cases were younger (P = 0.023) and more likely to be female (P = 0.012).
TABLE 1.
Characteristics of the study participants in the Tennessee Colorectal Polyp Study1
| P for SSL vs. other groups2,3 | ||||||||
|---|---|---|---|---|---|---|---|---|
| Characteristics | No-polyp controls (n = 3803) | HPs (n = 559) | ADs (n = 1787) | SSLs (n = 212) | P-heterogeneity2 | Controls | HP | AD |
| Age, y | 57.2 (56.9, 57.4) | 56.8 (56.2, 57.4) | 59.0 (58.6, 59.3) | 57.2 (56.7, 58.8) | <0.001 | 0.307 | 0.124 | 0.023 |
| Sex, % F | 44.8 | 36.1 | 28.0 | 36.3 | <0.001 | 0.016 | 0.962 | 0.012 |
| Race, % white | 89.4 | 91.1 | 88.9 | 91.1 | 0.226 | |||
| Study site, % | <0.001 | 0.071 | 0.489 | 0.450 | ||||
| Academic medical center | 74.9 | 66.7 | 66.8 | 69.3 | ||||
| Veterans Affairs medical center | 25.1 | 33.3 | 33.2 | 30.7 | ||||
| Educational attainment, % | <0.001 | 0.056 | 0.548 | 0.555 | ||||
| High school or less | 23.3 | 29.3 | 28.5 | 24.4 | ||||
| Some college | 28.4 | 28.3 | 28.3 | 27.3 | ||||
| College graduate | 21.3 | 21.8 | 22.1 | 27.6 | ||||
| Graduate/professional school | 26.9 | 20.6 | 21.1 | 21.1 | ||||
| Indication for colonoscopy, % | 0.172 | |||||||
| Average risk screening | 57.1 | 54.1 | 55.3 | 57.5 | ||||
| Family history | 12.6 | 13.7 | 12.7 | 16.3 | ||||
| Diagnostic/follow-up | 22.5 | 21.6 | 23.7 | 15.7 | ||||
| Other | 7.8 | 10.5 | 8.3 | 10.5 | ||||
| Family history of colorectal cancer, % | 9.0 | 8.7 | 9.3 | 12.0 | 0.006 | 0.008 | 0.125 | 0.217 |
| Regular alcohol intake, % | 0.016 | 0.130 | 0.958 | 0.360 | ||||
| Never | 59.5 | 53.5 | 56.3 | 53.8 | ||||
| Former | 21.6 | 23.7 | 23.9 | 21.8 | ||||
| Current | 18.8 | 22.8 | 19.8 | 24.4 | ||||
| Cigarette smoking status, % | <0.001 | <0.001 | 0.276 | 0.076 | ||||
| Never | 53.2 | 31.7 | 42.8 | 37.0 | ||||
| Former | 34.2 | 38.1 | 33.3 | 33.2 | ||||
| Current | 12.6 | 30.1 | 23.9 | 29.8 | ||||
| Regular physical activity in the past 10 y, % | 58.3 | 53.2 | 52.3 | 52.6 | <0.001 | 0.184 | 0.942 | 0.700 |
| Current use of NSAIDs, % | 51.7 | 50.9 | 45.7 | 44.7 | <0.001 | 0.007 | 0.066 | 0.204 |
| BMI, kg/m2 | 28.0 | 28.8 | 28.7 | 28.8 | <0.001 | 0.058 | 0.861 | 0.852 |
| Total energy intake, kcal/d | 2048 (2011, 2086) | 2173 (2084, 2261) | 2162 (2108, 2215) | 2031 (1891, 2172) | <0.001 | 0.818 | 0.090 | 0.082 |
1Values are least-square means of log transformed data (lower confidence limit, upper confidence limit) or frequencies standardized to age (5-y categories) and sex distribution of controls, with the exception of age, sex, race, educational attainment, study site, and indication for colonoscopy which are presented as n (%) for categorical data and least-square means (lower confidence limit, upper confidence limit) for continuous data. Statistical significance was set at P < 0.0167. AD, conventional adenoma; HP, hyperplastic polyp; NSAID, nonsteroidal anti-inflammatory drug; SSL, sessile serrated lesion.
2 P values adjusted for age (5-y categories) and sex for all comparisons excluding age, sex, and study site.
3 P values for comparisons of SSL cases vs. controls, HP cases, and AD cases where all P-heterogeneity < 0.05.
Evaluation of meat type and SSL risk
Table 2 presents results for the associations of meat intake with the risk of SSLs. A nearly 2-fold increased risk of SSLs in comparison to controls was observed for total meat intake for the highest versus the lowest intake quartile. This association appeared to be particularly strong for and limited to red meat (OR: 2.38; 95% CI: 1.44, 3.93; P-trend < 0.001) and processed meat (OR: 2.03; 95% CI: 1.30, 3.17; P-trend = 0.001), with no association observed for intake of total white meat. SSLs were also associated with red meat intake in comparison to ADs (P-trend = 0.012) and suggestively associated with risk in comparison to HPs (P-trend = 0.028), both of which polyp types were also associated with red meat intake in comparison to controls (data not shown). The association with processed meat was only observed for SSLs and not for the other polyp cases versus controls (data not shown).
TABLE 2.
Associations between intake of meat types and SSL risk: the Tennessee Colorectal Polyp Study1
| SSLs vs. polyp-free controls | SSLs vs. other polyp cases | ||||||
|---|---|---|---|---|---|---|---|
| SSL | SSL vs. hyperplastic polyp only | SSL vs. conventional adenoma | |||||
| Meat intake amount, g/d (min, max) | Controls, n | n | OR (95% CI)2 | n | OR (95% CI)2 | n | OR (95% CI)2 |
| Total meat | |||||||
| Q1: (0, <59.4) | 951 | 35 | 1.00 (ref) | 101 | 1.00 (ref) | 394 | 1.00 (ref) |
| Q2: (59.4, <90.9) | 951 | 44 | 1.43 (0.88, 2.34) | 116 | 1.22 (0.7, 2.12) | 339 | 1.67 (1.01, 2.76) |
| Q3: (90.9, <138.5) | 951 | 61 | 1.81 (1.13, 2.90) | 158 | 1.29 (0.76, 2.18) | 459 | 1.76 (1.09, 2.85) |
| Q4: ≥138.5 | 950 | 72 | 2.20 (1.36, 3.56) | 184 | 1.51 (0.88, 2.59) | 595 | 1.96 (1.20, 3.21) |
| P-trend | <0.001 | 0.145 | 0.011 | ||||
| White meat | |||||||
| Q1: (0, <24.9) | 971 | 54 | 1.00 (ref) | 142 | 1.00 (ref) | 488 | 1.00 (ref) |
| Q2: (24.9, <45.3) | 931 | 63 | 1.42 (0.95, 2.13) | 144 | 1.30 (0.83, 2.05) | 451 | 1.40 (0.93, 2.12) |
| Q3: (45.3, <85.0) | 1269 | 62 | 1.11 (0.74, 1.66) | 187 | 0.98 (0.62, 1.55) | 525 | 1.22 (0.80, 1.85) |
| Q4: ≥85.0 | 612 | 32 | 1.18 (0.72, 1.92) | 82 | 1.17 (0.67, 2.04) | 313 | 1.07 (0.65, 1.76) |
| P-trend | 0.742 | 0.923 | 0.813 | ||||
| Red meat | |||||||
| Q1: (0, <16.3) | 965 | 33 | 1.00 (ref) | 91 | 1.00 (ref) | 306 | 1.00 (ref) |
| Q2: (16.3, <39.5) | 937 | 39 | 1.14 (0.68, 1.91) | 123 | 0.88 (0.49, 1.56) | 397 | 0.96 (0.56, 1.63) |
| Q3: (39.5, <75.7) | 953 | 56 | 1.69 (1.04, 2.74) | 147 | 1.29 (0.74, 2.24) | 427 | 1.45 (0.88, 2.39) |
| Q4: ≥75.7 | 948 | 84 | 2.38 (1.44, 3.93) | 198 | 1.62 (0.92, 2.85) | 657 | 1.69 (1.01, 2.83) |
| P-trend | <0.001 | 0.028 | 0.012 | ||||
| Processed meat | |||||||
| Q1: (0) | 1070 | 42 | 1.00 (ref) | 118 | 1.00 (ref) | 377 | 1.00 (ref) |
| Q2: (1.0, <8.0) | 834 | 37 | 1.08 (0.66, 1.75) | 116 | 0.84 (0.49, 1.45) | 354 | 1.00 (0.61, 1.65) |
| Q3: (8.0, <22.5) | 944 | 49 | 1.27 (0.81, 2.00) | 148 | 1.04 (0.63, 1.74) | 439 | 1.20 (0.75, 1.91) |
| Q4: ≥22.5 | 946 | 84 | 2.03 (1.30, 3.17) | 176 | 1.72 (1.04, 2.85) | 613 | 1.74 (1.10, 2.74) |
| P-trend | 0.001 | 0.018 | 0.011 | ||||
1max, maximum; min, minimum; NSAID, nonsteroidal anti-inflammatory drug; Q, quartile; ref, reference; SSL, sessile serrated lesion.
2ORs and 95% CIs were derived from multinomial logistic regression models adjusted for age, sex, race, study site, educational attainment, indication for colonoscopy, alcohol intake, smoking history, physical activity, BMI, total energy intake, and NSAID use. Statistical significance was set at P < 0.0167.
Evaluation of well-done meat intake and SSL risk
Table 3 presents results for the associations of well-done meat intake with the risk of SSLs. In comparison to controls, an 80% increased risk of SSLs was observed for total well-done meat intake (OR: 1.83; 95% CI: 1.15, 2.89; P-trend = 0.001) as well as a suggestive 60% increased risk for well-done processed meat (OR: 1.58; 95% CI: 1.08, 2.31; P-trend = 0.019) and a more than double increased risk for well-done red meat intake (OR: 2.19; 95% CI: 1.34, 3.60; P-trend < 0.001) for the highest versus the lowest intake quartile. Well-done red meat intake was also suggestively but not significantly associated with SSLs in comparison to HPs or ADs.
TABLE 3.
Associations between well-done meat intake and SSL risk; the Tennessee Colorectal Polyp Study1
| SSLs vs. polyp-free controls | SSLs vs. other polyp cases | ||||||
|---|---|---|---|---|---|---|---|
| SSL | SSL vs. hyperplastic polyp only | SSL vs. conventional adenoma | |||||
| Well-done meat intake amount, g/d (min, max) | Controls, n | n | OR (95% CI)2 | n | OR (95% CI)2 | n | OR (95% CI)2 |
| Total well-done meat | |||||||
| Q1: (0, <24.90) | 952 | 40 | 1.00 (ref) | 109 | 1.00 (ref) | 342 | 1.00 (ref) |
| Q2: (24.90, <49.31) | 950 | 32 | 0.90 (0.55, 1.49) | 99 | 0.99 (0.56, 1.75) | 388 | 0.82 (0.49, 1.38) |
| Q3: (49.31, <85.20) | 951 | 67 | 1.65 (1.06, 2.59) | 164 | 1.26 (0.76, 2.09) | 455 | 1.39 (0.88, 2.21) |
| Q4: ≥85.20 | 950 | 73 | 1.83 (1.15, 2.89) | 187 | 1.46 (0.87, 2.46) | 602 | 1.43 (0.89, 2.29) |
| P-trend | 0.001 | 0.093 | 0.035 | ||||
| Total well-done white meat | |||||||
| Q1: (0, <4.7) | 948 | 46 | 1.00 (ref) | 143 | 1.00 (ref) | 427 | 1.00 (ref) |
| Q2: (4.7, <14.2) | 948 | 70 | 1.55 (1.03, 2.34) | 112 | 0.94 (0.57, 1.55) | 490 | 1.29 (0.85, 1.96) |
| Q3: (14.2, <29.8) | 972 | 40 | 1.04 (0.66, 1.64) | 162 | 1.4 (0.86, 2.27) | 422 | 0.95 (0.59, 1.51) |
| Q4: ≥29.8 | 924 | 55 | 1.43 (0.92, 2.20) | 140 | 0.59 (0.34, 1.04) | 442 | 1.21 (0.78, 1.89) |
| P-trend | 0.362 | 0.814 | 0.732 | ||||
| Total well-done red meat | |||||||
| Q1: (0, <9.50) | 985 | 33 | 1.00 (ref) | 97 | 1.00 (ref) | 320 | 1.00 (ref) |
| Q2: (9.50, <28.35) | 940 | 40 | 1.24 (0.74, 2.05) | 124 | 0.99 (0.56, 1.75) | 393 | 1.07 (0.64, 1.81) |
| Q3: (28.35, <58.50) | 928 | 60 | 1.81 (1.12, 2.94) | 150 | 1.38 (0.8, 2.38) | 441 | 1.52 (0.92, 2.49) |
| Q4: ≥58.50 | 950 | 79 | 2.19 (1.34, 3.60) | 188 | 1.69 (0.97, 2.96) | 633 | 1.68 (1.01, 2.80) |
| P-trend | <0.001 | 0.025 | 0.018 | ||||
| Total well-done processed meat | |||||||
| Q1: (0) | 1529 | 67 | 1.00 (ref) | 173 | 1.00 (ref) | 572 | 1.00 (ref) |
| Q2: (0.10, <3.90) | 451 | 19 | 0.75 (0.41, 1.35) | 60 | 0.65 (0.34, 1.25) | 197 | 0.68 (0.37, 1.24) |
| Q3: (3.90, <13.43) | 866 | 45 | 1.09 (0.72, 1.63) | 141 | 0.83 (0.53, 1.32) | 407 | 1.02 (0.67,1.56) |
| Q4: ≥13.43 | 948 | 81 | 1.58 (1.08, 2.31) | 184 | 1.28 (0.83, 1.98) | 607 | 1.38 (0.93, 2.04) |
| P-trend | 0.019 | 0.304 | 0.099 | ||||
1max, maximum; min, minimum; NSAID, nonsteroidal anti-inflammatory drug; Q, quartile; ref, reference; SSL, sessile serrated lesion.
2ORs and 95% CIs were derived from multinomial logistic regression models adjusted for age, sex, race, study site, educational attainment, indication for colonoscopy, alcohol intake, smoking history, physical activity, BMI, total energy intake, and NSAID use. Statistical significance was set at P < 0.0167.
Evaluation of associations between meat mutagens and polyp risk
The associations of specific inferred meat mutagens with the risk of SSLs are presented in Table 4. An increased risk of SSLs was observed for MeIQX intake. Specifically, MeIQX was associated with a strong 2.5-fold increased risk of SSLs in comparison to controls (OR: 2.48; 95% CI: 1.49, 4.16; P-trend = 0.004) and a ≥2-fold increased risk in comparison to HP or AD cases, although the latter findings were not statistically significant. There were no associations observed for intakes of PhIP, DiMeIQX, or BaP and SSL risk.
TABLE 4.
Associations between intake of inferred meat-derived mutagens and SSL risk: the Tennessee Colorectal Polyp Study1
| SSLs vs. polyp-free controls | SSLs vs. other polyp cases | ||||||
|---|---|---|---|---|---|---|---|
| SSL | SSL vs. hyperplastic polyp only | SSL vs. conventional adenoma | |||||
| Inferred daily intake amount, ng/d (min, max) | Controls, n | n | OR (95% CI)2 | n | OR (95% CI)2 | n | OR (95% CI)2 |
| MeIQX intake | |||||||
| Q1: (0, <12.2) | 930 | 27 | 1.00 (ref) | 103 | 1.00 (ref) | 313 | 1.00 (ref) |
| Q2: (12.2, <32.9) | 930 | 59 | 2.17 (1.32, 3.58) | 119 | 2.06 (1.17, 3.60) | 379 | 1.98 (1.18, 3.31) |
| Q3: (32.9, <70.3) | 930 | 46 | 1.69 (1.00, 2.84) | 155 | 1.40 (0.79, 2.49) | 452 | 1.41 (0.83, 2.40) |
| Q4: ≥70.3 | 929 | 77 | 2.48 (1.49, 4.16) | 173 | 2.22 (1.25, 3.93) | 597 | 1.97 (1.16, 3.33) |
| P-trend | 0.004 | 0.044 | 0.075 | ||||
| PhIP intake | |||||||
| Q1: (0, <73.25) | 930 | 40 | 1.00 (ref) | 112 | 1.00 (ref) | 356 | 1.00 (ref) |
| Q2: (73.25, <169.38) | 930 | 56 | 1.30 (0.83, 2.02) | 116 | 1.38 (0.83, 2.30) | 385 | 1.27 (0.81, 2.01) |
| Q3: (169.38, <336.20) | 930 | 55 | 1.34 (0.86, 2.09) | 152 | 1.19 (0.72, 1.96) | 480 | 1.08 (0.69, 1.70) |
| Q4: ≥336.20 | 929 | 58 | 1.39 (0.89, 2.18) | 170 | 1.17 (0.71, 1.93) | 520 | 1.12 (0.71, 1.78) |
| P-trend | 0.173 | 0.793 | 0.870 | ||||
| DiMeIQX intake | |||||||
| Q1: (0, <0.83) | 931 | 45 | 1.00 (ref) | 112 | 1.00 (ref) | 364 | 1.00 (ref) |
| Q2: (0.83, <2.74) | 930 | 47 | 0.98 (0.63, 1.53) | 122 | 1.01 (0.61, 1.67) | 394 | 0.97 (0.61, 1.53) |
| Q3: (2.74, <5.96) | 929 | 47 | 0.96 (0.62, 1.50) | 142 | 0.90 (0.54, 1.48) | 421 | 0.93 (0.59, 1.47) |
| Q4: ≥5.96 | 929 | 70 | 1.38 (0.91, 2.10) | 174 | 1.18 (0.73, 1.91) | 562 | 1.12 (0.73, 1.72) |
| P-trend | 0.131 | 0.538 | 0.608 | ||||
| BaP intake | |||||||
| Q1: (0, <9.10) | 930 | 45 | 1.00 (ref) | 109 | 1.00 (ref) | 343 | 1.00 (ref) |
| Q2: (9.10, <31.54) | 930 | 54 | 1.23 (0.79, 1.90) | 133 | 1.2 (0.73, 1.98) | 420 | 1.12 (0.72, 1.77) |
| Q3: (31.54, <79.05) | 930 | 50 | 1.05 (0.66, 1.65) | 136 | 1.02 (0.61, 1.7) | 466 | 0.87 (0.55, 1.39) |
| Q4: ≥79.05 | 929 | 60 | 1.30 (0.83, 2.03) | 172 | 1.14 (0.69, 1.88) | 512 | 1.09 (0.69, 1.73) |
| P-trend | 0.382 | 0.820 | 0.972 | ||||
| Meat-mutagen intake | |||||||
| Q1: (0, <3645) | 930 | 39 | 1.00 (ref) | 98 | 1.00 (ref) | 333 | 1.00 (ref) |
| Q2: (3645, <7234) | 930 | 51 | 1.36 (0.86, 2.14) | 140 | 1.03 (0.62, 1.73) | 403 | 1.18 (0.74, 1.88) |
| Q3: (7234, <13,019) | 930 | 53 | 1.33 (0.84, 2.10) | 142 | 1.06 (0.63, 1.78) | 466 | 1.03 (0.65, 1.65) |
| Q4: ≥13,019 | 929 | 66 | 1.65 (1.05, 2.58) | 170 | 1.25 (0.75, 2.07) | 539 | 1.27 (0.80, 2.01) |
| P-trend | 0.039 | 0.366 | 0.418 | ||||
1BaP, benzo[α]pyrene; DiMeIQX, 2-amino-3,4,8-trimethylimidazo[4,5-f]-quinoxaline; max, maximum; MeIQX, 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline; min, minimum; NSAID, nonsteroidal anti-inflammatory drug; PhIP, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine; Q, quartile; ref, reference; SSL, sessile serrated lesion.
2ORs and 95% CIs were derived from multinomial logistic regression models adjusted for age, sex, race, study site, educational attainment, indication for colonoscopy, alcohol intake, smoking history, physical activity, BMI, total energy intake, and NSAID use. Statistical significance was set at P < 0.0167.
Discussion
In this large case-control study, higher intakes of total meat, red meat, processed meat, well-done meat, well-done red meat, and well-done processed meat were associated with an increased risk of SSLs. For nearly all meat intake categories associated with increased risk of SSL, the risks of SSLs were suggestively or significantly increased in comparison to HPs and ADs. Inferred MeIQX was associated with a strong increased risk of SSLs in comparison to controls, HPs, or ADs. The association was null for the other inferred meat mutagens investigated.
SSLs are a type of serrated polyp (SP) along with HPs and traditional serrated adenomas. Due to the lack of a consensus definition until 2010, many historical studies that have evaluated SP risk have not been able to differentiate SSLs from HPs and may also have misclassification of SP subtype (8). Thus, although there are a few studies, including our own (16), which have evaluated meat intake (20-24), well-done meat intake (17, 25, 26), and inferred meat-mutagen intake (25, 27) with risks of SPs or HPs, there is no other study that has evaluated all of these factors in relation to SSL risk. A recent study evaluated red meat intake with risk of nonadvanced and advanced adenomas, which included some SSLs (20). They confirmed our previous finding of an association between red meat intake and polyp risk. In a recent meta-analysis of risk factors for SPs, there was no observed association between processed meat intake and SP risk (28); however, they were unable to evaluate SSL risk. Future analyses need to make the distinction between HPs and SSLs as risk factors may differ between these polyp types.
Unlike previous studies that found a strong association between inferred meat-mutagen exposure and AD or SP risk (29), with the exception of MeIQX, the associations with SSL risk in this study were null across all other inferred meat mutagens. MeIQX is the mutagen most closely derived from red meat (30). The mutagenicity of MeIQX has been demonstrated in animal studies (29). One potential mechanism that could explain this is the formation of DNA adducts (15), which increase with the intake of dietary HCAs and PAHs.
In this study, the overall intakes of red and processed meats were slightly more strongly associated with SSL risk than well-done intakes of these meats, suggesting that it may not be well-doneness per se that is responsible for the increased risk. This observation could be the result of a limited sample size of individuals who consume well-done meats, or it could be because of other mechanisms unrelated to PAHs and HCAs. Meat also contains N-nitroso compounds and heme iron (15). The mechanism for each of these components is different so it is critical to evaluate the evidence for each component separately in future studies. Other possible mechanisms of meat carcinogenicity include effects of bile acids on inflammation or cancer stem cells (31, 32).
Larger studies of SSLs could help confirm or counter the null findings in this study. Specifically, this study might have been underpowered to observe weak associations between the other HCAs and PAHs. However, the observed and strong statistically significant associations despite a limited sample size of SSLs relative to the other polyp types suggest that SSL risk is significantly increased by red meat intake and MeIQX exposure may be one of the mechanisms for this association. Additional studies are needed to evaluate other possible mechanisms.
Recall bias is always a concern in a case-control study; however, most study participants completed the telephone questionnaire within 13 d of their colonoscopy, including before they may have received their final report, and polyps are, in general, considered a benign diagnosis. In addition, most participants with an SSL would have been told that they have an HP as there was not a standardized diagnosis of this polyp type during the period of the study. We cannot preclude the possibility that some SSLs were missed during the colonoscopy, resulting in misclassification of the case status. However, SSLs are a rare polyp and missed SSLs are very rare; therefore, the potential case misclassification is unlikely to explain the observed associations. We did not include intakes of all meat types (e.g., lamb, goat, and turkey) and so may have underestimated intakes of red meat and poultry for some participants. There could have been errors in the HCA-exposure measurement; however, any misclassification is likely to be nondifferential, which tends to bias results towards the null. Our study is strengthened by the standardized pathological review of all polyps removed from study participants, the extensive information collected on meat intake and cooking preferences, and a population with a wide range of characteristics and behaviors.
Meat intake and well-done meat intake are factors that can be modified in the diet; therefore, particularly for SSLs as well as other polyp types, strategies such as reducing red and processed-meat intakes may be important for preventing CRC. Future studies are needed to confirm or refute the findings in this study. In addition, a better understanding of all of the mechanisms by which meat intake may affect SSL risk, such as heme and N-nitroso compounds, is needed. Finally, future studies should also evaluate interindividual differences in metabolism, such as HCA-metabolizing enzymes and heme or N-nitroso–metabolizing enzymes, which may affect susceptibility at equivalent exposure levels.
Supplementary Material
Acknowledgments
The authors’ responsibilities were as follows—RMN, WZ, HJM, MJS: designed the research; TS, WES, RMN, WZ, HJM, MJS: conducted the research; TS, MJS: provided essential materials; DM, MJS: analyzed data or performed statistical analysis; MJS: had primary responsibility for final content; and all authors: contributed to writing, review, and/or revision of the manuscript and read and approved the final manuscript. The authors report no conflicts of interest.
Notes
This study was supported by NIH grants P50CA950103, R01CA97386, and K07CA122451. Surveys and sample collection and processing for this study were conducted by the Survey and Biospecimen Shared Resource, which is supported in part by P30CA68485. The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the National Cancer Institute or the NIH. A portion of the participants were studied as the result of resources and the use of facilities at the VA Tennessee Valley Healthcare System.
Supplemental Figure 1 is available from the “Supplementary data” link in the online posting of the article and from the same link in the online table of contents at https://academic.oup.com/ajcn/.
Data described in the manuscript, code book, and analytic code will not be made available because approval of these activities was not obtained in the informed consent form.
Abbreviations used: AD, conventional adenoma; BaP, benzo[α]pyrene; CRC, colorectal cancer; DiMeIQX, 2-amino-3,4,8-trimethylimidazo[4,5-f]-quinoxaline; HCA, heterocyclic amine; HP, hyperplastic polyp; MeIQX, 2-amino-3,8-dimethylimidazo[4,5-f]quinoxaline; NSAID, nonsteroidal anti-inflammatory drug; PAH, polycyclic aromatic hydrocarbon; PhIP, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine; SSL, sessile serrated lesion; TCPS, Tennessee Colorectal Polyp Study; VA, Veterans Affairs.
References
- 1. Leggett B, Whitehall V. Role of the serrated pathway in colorectal cancer pathogenesis. Gastroenterology. 2010;138:2088–100. [DOI] [PubMed] [Google Scholar]
- 2. O'Brien MJ, Yang S, Mack C, Xu H, Huang CS, Mulcahy E, Amorosino M, Farraye FA. Comparison of microsatellite instability, CpG island methylation phenotype, BRAF and KRAS status in serrated polyps and traditional adenomas indicates separate pathways to distinct colorectal carcinoma end points. Am J Surg Pathol. 2006;30:1491–501. [DOI] [PubMed] [Google Scholar]
- 3. East JE, Saunders BP, Jass JR. Sporadic and syndromic hyperplastic polyps and serrated adenomas of the colon: classification, molecular genetics, natural history, and clinical management. Gastroenterol Clin North Am. 2008;37:25–46. [DOI] [PubMed] [Google Scholar]
- 4. Kim SW, Cha JM, Lee JI, Joo KR, Shin HP, Kim GY, Lim SJ. A significant number of sessile serrated adenomas might not be accurately diagnosed in daily practice. Gut Liver. 2010;4:498–502. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Lash RH, Genta RM, Schuler CM. Sessile serrated adenomas: prevalence of dysplasia and carcinoma in 2139 patients. J Clin Pathol. 2010;63:681–6. [DOI] [PubMed] [Google Scholar]
- 6. Longacre TA, Fenoglio-Preiser CM. Mixed hyperplastic adenomatous polyps/serrated adenomas: a distinct form of colorectal neoplasia. Am J Surg Pathol. 1990;14:524–37. [DOI] [PubMed] [Google Scholar]
- 7. Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell. 1990;61:759–67. [DOI] [PubMed] [Google Scholar]
- 8. Snover DC. Update on the serrated pathway to colorectal carcinoma. Hum Pathol. 2011;42:1–10. [DOI] [PubMed] [Google Scholar]
- 9. Sawhney MS, Farrar WD, Gudiseva S, Nelson DB, Lederle FA, Rector TS, Bond JH. Microsatellite instability in interval colon cancers. Gastroenterology. 2006;131:1700–5. [DOI] [PubMed] [Google Scholar]
- 10. Obuch JC, Pigott CM, Ahnen DJ. Sessile serrated polyps: detection, eradication, and prevention of the evil twin. Curr Treat Options Gastroenterol. 2015;13:156–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Bouvard V, Loomis D, Guyton KZ, Grosse Y, Ghissassi FE, Benbrahim-Tallaa L, Guha N, Mattock H, Straif K. Carcinogenicity of consumption of red and processed meat. Lancet Oncol. 2015;16:1599–600. [DOI] [PubMed] [Google Scholar]
- 12. Alaejos MS, Afonso AM. Factors that affect the content of heterocyclic aromatic amines in foods. Compr Rev Food Sci Food Saf. 2011;10:52–108. [Google Scholar]
- 13. Budhathoki S, Iwasaki M, Yamaji T, Sasazuki S, Takachi R, Sakamoto H, Yoshida T, Tsugane S. Dietary heterocyclic amine intake, NAT2 genetic polymorphism, and colorectal adenoma risk: the colorectal adenoma study in Tokyo. Cancer Epidemiol Biomark Prev. 2015;24:613–20. [DOI] [PubMed] [Google Scholar]
- 14. Domingo JL, Nadal M. Carcinogenicity of consumption of red meat and processed meat: a review of scientific news since the IARC decision. Food Chem Toxicol. 2017;105:256–61. [DOI] [PubMed] [Google Scholar]
- 15. Kruger C, Zhou Y. Red meat and colon cancer: a review of mechanistic evidence for heme in the context of risk assessment methodology. Food Chem Toxicol. 2018;118:131–53. [DOI] [PubMed] [Google Scholar]
- 16. Davenport JR, Su T, Zhao Z, Coleman HG, Smalley WE, Ness RM, Zheng W, Shrubsole MJ. Modifiable lifestyle factors associated with risk of sessile serrated polyps, conventional adenomas and hyperplastic polyps. Gut. 2018;67:456–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Shin A, Shrubsole MJ, Ness RM, Wu H, Sinha R, Smalley WE, Shyr Y, Zheng W. Meat and meat-mutagen intake, doneness preference and the risk of colorectal polyps: the Tennessee Colorectal Polyp Study. Int J Cancer. 2007;121:136–42. [DOI] [PubMed] [Google Scholar]
- 18. Buchowski MS, Schlundt DG, Hargreaves MK, Hankin JH, Signorello LB, Blot WJ. Development of a culturally sensitive food frequency questionnaire for use in the Southern Community Cohort Study. Cell Mol Biol (Noisy-le-grand). 2003;49:1295–304. [PubMed] [Google Scholar]
- 19.National Cancer Institute. Computerized Heterocyclic Amines Resource for Research in Epidemiology of Disease (CHARRED) [Internet]. National Cancer Institute; 1980 [cited 2019 Aug 12]. Available from: https://dceg.cancer.gov/tools/design/charred. [Google Scholar]
- 20. Fliss-Isakov N, Kariv R, Webb M, Ivancovsky D, Margalit D, Zelber-Sagi S. Mediterranean dietary components are inversely associated with advanced colorectal polyps: a case-control study. World J Gastroenterol. 2018;24:2617–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Carr PR, Holleczek B, Stegmaier C, Brenner H, Hoffmeister M. Meat intake and risk of colorectal polyps: results from a large population-based screening study in Germany. Am J Clin Nutr. 2017;105:1453–61. [DOI] [PubMed] [Google Scholar]
- 22. Bailie L, Loughrey MB, Coleman HG. Lifestyle risk factors for serrated colorectal polyps: a systematic review and meta-analysis. Gastroenterology. 2017;152:92–104. [DOI] [PubMed] [Google Scholar]
- 23. Lippi G, Mattiuzzi C, Cervellin G. Meat consumption and cancer risk: a critical review of published meta-analyses. Crit Rev Oncol Hematol. 2016;97:1–14. [DOI] [PubMed] [Google Scholar]
- 24. Carr PR, Walter V, Brenner H, Hoffmeister M. Meat subtypes and their association with colorectal cancer: systematic review and meta-analysis. Int J Cancer. 2016;138:293–302. [DOI] [PubMed] [Google Scholar]
- 25. Wu K, Giovannucci E, Byrne C, Platz EA, Fuchs C, Willett WC, Sinha R. Meat mutagens and risk of distal colon adenoma in a cohort of U.S. men. Cancer Epidemiol Biomark Prev. 2006;15:1120–5. [DOI] [PubMed] [Google Scholar]
- 26. Sinha R, Peters U, Cross AJ, Kulldorff M, Weissfeld JL, Pinsky PF, Rothman N, Hayes RB. Meat, meat cooking methods and preservation, and risk for colorectal adenoma. Cancer Res. 2005;65:8034–41. [DOI] [PubMed] [Google Scholar]
- 27. Fu Z, Shrubsole MJ, Smalley WE, Wu H, Chen Z, Shyr Y, Ness RM, Zheng W. Association of meat intake and meat-derived mutagen exposure with the risk of colorectal polyps by histologic type. Cancer Prev Res. 2011;4:1686–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. He X, Wu K, Ogino S, Giovannucci EL, Chan AT, Song M. Association between risk factors for colorectal cancer and risk of serrated polyps and conventional adenomas. Gastroenterol. 2018;155:355–73..e18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Martinez Góngora V, Matthes KL, Castaño PR, Linseisen J, Rohrmann S. Dietary heterocyclic amine intake and colorectal adenoma risk: a systematic review and meta-analysis. Cancer Epidemiol Prev Biomark. 2019;28:99–109. [DOI] [PubMed] [Google Scholar]
- 30. Zheng W, Lee S-A. Well-done meat intake, heterocyclic amine exposure, and cancer risk. Nutr Cancer. 2009;61:437–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Kundu S, Kumar S, Bajaj A. Cross-talk between bile acids and gastrointestinal tract for progression and development of cancer and its therapeutic implications. IUBMB Life. 2015;67:514–23. [DOI] [PubMed] [Google Scholar]
- 32. Farhana L, Nangia-Makker P, Arbit E, Shango K, Sarkar S, Mahmud H, Hadden T, Yu Y, Majumdar APN. Bile acid: a potential inducer of colon cancer stem cells. Stem Cell Res Ther. 2016;7:181. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.