Abstract
Tumor necrosis factor-α (TNF) and toll-like receptors (TLR) are important mediators of inflammation. We examined 10 of these genes with respect to breast cancer risk and mortality in a genetically admixed population of Hispanic/Native American (NA) (2111 cases, 2597 controls) and non-Hispanic white (NHW) (1481 cases, 1585 controls) women. Additionally, we explored if diet and lifestyle factors modified associations with these genes. Overall, these genes (collectively) were associated with breast cancer risk among women with >70% NA ancestry (PARTP = 0.0008), with TLR1 rs7696175 being the primary risk contributor (OR 1.77, 95% CI 1.25, 2.51). Overall, TLR1 rs7696175 (HR 1.40, 95% CI 1.03, 1.91; Padj=0.032), TLR4 rs5030728 (HR 1.96, 95% CI 1.30, 2.95; Padj=0.014), and TNFRSF1A rs4149578 (HR 2.71, 95% CI 1.28, 5.76; Padj=0.029) were associated with increased breast cancer mortality. We observed several statistically significant interactions after adjustment for multiple comparisons, including interactions between our dietary oxidative balance score and CD40LG and TNFSF1A; between cigarette smoking and TLR1, TLR4, and TNF; between body mass index (BMI) among pre-menopausal women and TRAF2; and between regular use of aspirin/non-steroidal anti-inflammatory drugs and TLR3 and TRA2. In conclusion, our findings support a contributing role of certain TNF-α and TLR genes in both breast cancer risk and survival, particularly among women with higher NA ancestry. Diet and lifestyle factors appear to be important mediators of the breast cancer risk associated with these genes.
Keywords: Breast cancer, TLR, TNF, TRAIL, TRAF2, survival, cigarettes, oxidative stress
Introduction
Tumor necrosis factor-α (TNF), a pro-inflammatory cytokine, stimulates cell proliferation and induces cell differentiation and is thought to be one of the most important promoters of inflammation. Additionally, TNF is a modulator of insulin resistance, especially among individuals who are obese or have chronic inflammation conditions; TNF has been reported to inhibit insulin-induced glucose uptake by targeting components of the insulin-signaling cascade [1–5]. TNF mediates cell survival and apoptosis through TNF receptors by activating at least two major signaling pathways, NFκB and the p38 mitogen-activated protein (MAP) kinase pathway. Tumor necrosis factor receptor superfamily member 1A (TNFRSF1A or TNFR1) is a major receptor for TNF-alpha that activates NFκB, mediates apoptosis, and functions as a regulator of inflammation. TNF receptor-associated factor 2 (TRAF2) is a member of the TRAF protein family that interacts with TNF receptors. TRAF2 is required for TNF activation of mitogen activated protein kinase 8 (MAPK8 alias JNK1) as well as NFκB and therefore is thought to influence the apoptotic effects of TNF. TNFSF10 (TRAIL) protein expression has been elevated in adriamycin-treated breast cells [6]. This protein preferentially induces apoptosis in transformed and tumor cells. CD40LG, also known as TNFSF5 and TRAP, is involved in TNF-signaling pathway and related cytokine activity. Toll-like receptors (TLR) also are mediators of inflammation and potentially important modulators of cancer risk through their involvement in the NFκB-signaling pathway [7,8]. TLR4 specifically has been linked to breast cancer [9] and to colon tumor progression and metastatic potential [10,11]. TRAIL has been designated CD253 (cluster of differentiation 253); TLR2 has been designated as CD282; and TLR3 has been designated as CD283.
In this study we examine genetic variation in TLR and TNF-related genes as they relate to breast cancer risk and survival. TNF rs1800629 has been associated with breast cancer risk in a small case-control study of Mexican women [12], suggesting that this gene and possibly its related pathway are important for breast cancer risk in Latina women. We evaluate associations by genetic ancestry since breast cancer incidence rates differ between non-Hispanic white (NHW), Hispanic, and Native American (NA) women living in the Southwestern United States [13]. We also evaluate associations by lifestyle factors that are associated with inflammation and insulin and could therefore modify risk associated with these genes and pathway. Factors we evaluate include dietary oxidative balance score (DOBS) [14], body mass index (BMI), regular cigarette smoking, use of aspirin or other non-steroidal anti-inflammatory drugs (NSAIDs), and having been diagnosed with diabetes. Given the association of these genes with apoptosis and metastatic potential, we evaluate their association with breast cancer mortality.
Methods
The Breast Cancer Health Disparities Study includes participants from three population-based case-control studies [13], the 4-Corners Breast Cancer Study (4-CBCS) [15], the Mexico Breast Cancer Study (MBCS)[16], and the San Francisco Bay Area Breast Cancer Study (SFBCS) [17,18], who completed an in-person interview and who had a blood or mouthwash sample available for DNA extraction. Information on exposures was collected up to the referent year, defined as the calendar year before diagnosis for cases or before selection into the study for controls. 4-CBCS participants were between 25 and 79 years; MBCS participants were between 28 and 74 years; and SFBCS participants were between 35 to 79 years. All participants signed informed written consent prior to participation and each study was approved by their Institutional Review Board for Human Subjects.
Data Harmonization
Data were harmonized across all study centers and questionnaires as previously described [13]. Women were classified as either pre-menopausal or post-menopausal based on responses to questions on menstrual history. Pre-menopausal women were those who reported still having periods during the referent year (defined as the year before diagnosis for cases or before selection into the study for controls). Post-menopausal women were those who reported either a natural menopause or if they reported taking hormone therapy (HT) and were still having periods or were at or above the 95th percentile of age for those who reported having a natural menopause (i.e., ≥ 12 months since their last period). Women in 4-CBCS and SFBCS were asked to self-identify their race/ethnicity and were classified as non-Hispanic white (NHW), Hispanic, Native American (NA) or a combination of these groups. Women in MBCS were not asked their race or ethnicity.
Lifestyle variables included BMI calculated as self-reported weight (kg) during the referent year divided by measured height squared (m2) and categorized as normal (<25 kg/m2), overweight (25–29.9 kg/m2), and obese (≥30 kg/m2). Cigarette smoking was evaluated as current, former, or never a regular smoker, where regular was defined as having smoked one or more cigarettes for six months or longer in 4-CBCS and SFBCS (data available for a subset of subjects only) or having smoked 100 or more cigarettes in MCBCS. A dietary oxidative balance score (DOBS) that included nutrients with anti- or pro-oxidative balance properties was developed as previously reported [14]. Dietary information was collected via a computerized validated diet history questionnaire in 4-CBCS [19,20], a 104-item semi-quantitative Food Frequency Questionnaire (FFQ) in MBCS [21], and a modified version of the Block Food Frequency Questionnaire in SFBCS [22]. Alcohol consumption was based on long-term use; consumption during the referent year was used for a subset of SFBCS women without information on long-term use. Regular use of aspirin or NSAIDS defined as three or more times a week for at least one month was available for the 4-CBCS only. A history of diabetes was defined as ever being told by a health care provider that you had diabetes or high blood sugar (available only for a subset of SFBCS participants).
Genetic Data
DNA was extracted from either whole blood (n=7287) or mouthwash (n=634) samples. Whole Genome Amplification (WGA) was applied to the mouthwash-derived DNA samples prior to genotyping. A tagSNP approach was used to capture variation across the entire candidate genes. Genes were selected based on the literature available at the time the platform was developed that indicated a potential effect on inflammation. TagSNPs were selected using the following parameters: linkage disequilibrium (LD) blocks were defined using a Caucasian LD map in concordance with the custom-made GoldenGate chemistry array and an r2=0.8; minor allele frequency (MAF) >0.1; range= −1500 bps from the initiation codon to +1500 bps from the termination codon; and 1 SNP/LD bin. Additionally, 104 Ancestry Informative Markers (AIMs) were used to distinguish European and NA ancestry [13]. All markers were genotyped using a multiplexed bead array assay format based on GoldenGate chemistry (Illumina, San Diego, California). A genotyping call rate of 99.93% was attained (99.65% for WGA samples). We included 132 blinded internal replicates representing 1.6% of the sample set. The duplicate concordance rate was 99.996% as determined by 193,297 matching genotypes among sample pairs. In the current analysis we evaluated tagSNPs for CD40LG alias TNFSF5 and TRAP (3 SNPs), TLR1 (1 SNP), TLR2 (4 SNPs), TLR3 (4 SNPs), TLR4 (8 SNPs), TNF (2 SNPs on Illumina and 1 taqman), TNFRSF1A (4 SNPs), TNFRSF11A (25 SNPs), TNFSF10 (12 SNPs), and TRAF2 (4 SNPs). Online Supplement 1 provides a description of these genes and SNPs.
Tumor Characteristics and Survival
Data on estrogen receptor (ER) and progesterone receptor (PR) tumor status and survival were available for cases from 4-CBCS and SFBCS only. Cancer registries in Utah, Colorado, Arizona, New Mexico, and California provided information on stage at diagnosis, months of survival after diagnosis, cause of death, and ER and PR status. Surveillance Epidemiology and End Results (SEER) disease stage was categorized as local, regional, or distant.
Statistical Methods
Genetic ancestry estimation
The program STRUCTURE was used to estimate individual ancestry for each study participant assuming two founding populations [23,24]. A three-founding population model was assessed but did not fit the population structure. Participants were classified by level of percent NA ancestry (≤28%, >28–70%, and >70%), based on the distribution of genetic ancestry in the control population [13].
SNP Associations
Genes and SNPs were assessed for their association with breast cancer risk overall, by strata of genetic ancestry, and by menopausal status in the whole population and by ER/PR status for the 4-CBCS and SFBCS. All statistical analyses were performed using SAS version 9.3 (SAS Institute, Cary, NC). Logistic regression models were used to estimate odds ratios (OR) and 95% confidence intervals (CI) for breast cancer risk associated with SNPs. Confounding variables adjusted in these analyses were study, BMI in the referent year, and parity as a categorical variables and age (five-year categories) and genetic ancestry as continuous variables. A p value of <0.05 was considered statistically significant, although results are presented for those where the unadjusted p values was <0.05 and the multiple comparison adjusted p value was <0.15. Associations at this level are presented since group sample sizes vary and these associations could be relevant for replication in other populations. Associations with SNPs were assessed assuming a co-dominant model. Based on the initial assessment, SNPs that appeared to have a dominant or recessive mode of inheritance were evaluated with those inheritance models in subsequent analyses. For stratified analyses, the p value was based on the Wald chi-square test comparing the homozygote rare to the homozygote common when presenting the co-dominant model. Tests for interactions were evaluated using Wald one degree of freedom (1-df) tests. The multinomial p value reported for ER/PR status using the glogit link in the logistic procedure excludes controls. Adjustments for multiple comparisons within the gene used the step-down Bonferroni correction, taking into account the degree of correlation of the SNPs within genes using the SNP spectral decomposition method proposed by Nyholt [25] and modified by Li and Ji [26].
Interactions
We assessed gene by environment interactions among environmental and lifestyle factors that could influence candidate genes given their potential involvement in inflammation, including BMI (separately for pre- and post-menopausal women), smoking (current, former, or never smokers), dietary oxidative balance score, and regular use of aspirin/NSAID (for 4-CBCS participants only). The dietary oxidative balance score (DOBS) was based on each individual’s ranking of anti-oxidants (vitamin C, vitamin E, beta carotene (data for beta carotene were not available for MBCS), folic acid, and dietary fiber) and pro-oxidants (alcohol). Nutrients were evaluated as nutrient per 1000 calories and quartiles of intake and the DOBS were based on study-specific distributions. Alcohol consumption was classified into three levels: the top 25th percentile of consumption, all other drinkers, and non-drinkers. In creating the DOBS, participants were assigned values of zero for low levels (first quartile) of exposure to anti-oxidants or high exposure to pro-oxidants (fourth quartile), one for intermediate levels (second and third quartiles) of exposure, and two for high levels (fourth quartile) of exposure to anti-oxidants and low exposure (first quartile) to pro-oxidants.
Survival Analysis
Survival months were calculated based on month and year of diagnosis and month and year of death or last contact. Survival updates were received in the winter of 2013 which included complete survival surveillance through December of 2012. Associations between SNPs and breast cancer-specific mortality among cases with a first primary invasive breast cancer were evaluated using Cox proportional hazards models to obtain multivariate hazard ratios (HR) and 95% confidence intervals (CI) among all women and by genetic ancestry strata. Since survival data were not available for MBCS, the upper two ancestry strata were combined to evaluate survival by genetic ancestry. Individuals were censored when they died of causes other than breast cancer or were lost to follow-up. We present Wald p values for all women and by ancestry strata based on the comparison between the homozygote rare and common genotypes using models adjusted for age, study, genetic ancestry, BMI, and SEER stage. Interactions between genetic variants and genetic ancestry with survival were assessed using p values from 1-df Wald chi-square tests.
ARTP Analysis
We used the adaptive rank truncated product (ARTP) method that utilizes a highly efficient permutation algorithm to determine the significance of association of each gene and of all genes combined with breast cancer risk by genetic ancestry and by ER/PR status. Case/control status was permuted 10,000 times within R version 3.0.2 (R Foundation for Statistical Computing, Vienna, Austria) and p values based on 1-df Wald chi-square tests were generated from logistic regression models. We also assessed associations with mortality using the ARTP method, permuting vital status and survival months together. Likelihood-ratio test p values were calculated from Cox proportional hazard models. We controlled the logistic and Cox models using the adjustment variables previously stated. We report both pathway and gene p values based on the ARTP method (PARTP) [27,28]. Since ARTP has not been developed to incorporate lifestyle factors when evaluating interactions, results for interactions were adjusted for multiple comparisons as described above.
Results
The majority of women were Hispanic/NA, post-menopausal, had ER+/PR+ tumors, and were diagnosed with local stage disease (Table 1). Among NHW women 21.4% had died, compared to 19.8% of Hispanic/NA women; 47.6% of deaths among NHW women were from breast cancer, compared to 55.9% of deaths among Hispanic/NA women. Among NHW women, 44.4 to 45.9% had a BMI of <25 kg/m2, compared to 17.6 to 23.5% among Hispanic/NA women.
Table 1.
Description of study population by self-reported race/ethnicity
| U.S. non-Hispanic White | U. S. Hispanic/Native American or Mexican | |||||||
|---|---|---|---|---|---|---|---|---|
| Controls | Cases | Controls | Cases | |||||
| N | % | N | % | N | % | N | % | |
| Total | 1585 | 37.9 | 1481 | 41.2 | 2597 | 62.1 | 2111 | 58.8 |
| Study | ||||||||
| 4-CBCS | 1321 | 83.3 | 1227 | 82.8 | 723 | 27.8 | 597 | 28.3 |
| MBCS | 0 | 0 | 0 | 0 | 994 | 38.3 | 816 | 38.7 |
| SFBCS | 264 | 16.7 | 254 | 17.2 | 880 | 33.9 | 698 | 33.1 |
| Age (years) | ||||||||
| <40 | 116 | 7.3 | 89 | 6 | 311 | 12 | 200 | 9.5 |
| 40–49 | 408 | 25.7 | 409 | 27.6 | 831 | 32 | 713 | 33.8 |
| 50–59 | 409 | 25.8 | 413 | 27.9 | 756 | 29.1 | 617 | 29.2 |
| 60–69 | 349 | 22 | 361 | 24.4 | 526 | 20.3 | 430 | 20.4 |
| >70 | 303 | 19.1 | 209 | 14.1 | 173 | 6.7 | 151 | 7.2 |
| Mean | 56.6 | 56 | 52.3 | 52.7 | ||||
| Menopausal Status | ||||||||
| Pre-menopausal | 494 | 31.5 | 489 | 33.5 | 1027 | 40.7 | 836 | 40.9 |
| Post-menopausal | 1075 | 68.5 | 970 | 66.5 | 1499 | 59.3 | 1210 | 59.1 |
| Estimated Percent Native American Ancestry | ||||||||
| 0–28 | 1577 | 99.5 | 1472 | 99.4 | 278 | 10.7 | 275 | 13 |
| 29–70 | 7 | 0.4 | 7 | 0.5 | 1686 | 64.9 | 1393 | 66 |
| 71–100 | 1 | 0.1 | 2 | 0.1 | 633 | 24.4 | 443 | 21 |
| ER/PR Status2 | ||||||||
| ER+/PR+ | NA1 | 695 | 68.2 | NA1 | 605 | 61.9 | ||
| ER+/PR- | NA1 | 121 | 11.9 | NA1 | 115 | 11.8 | ||
| ER−/PR+ | NA1 | 15 | 1.5 | NA1 | 28 | 2.9 | ||
| ER−/PR− | NA1 | 188 | 18.4 | NA1 | 229 | 23.4 | ||
| Vital Status2,3 | ||||||||
| Deceased | NA1 | 255 | 21.4 | NA1 | 229 | 19.8 | ||
| Alive | NA1 | 935 | 78.6 | NA1 | 929 | 80.2 | ||
| Cause of Death2,3 | ||||||||
| Breast Cancer | NA1 | 121 | 47.5 | NA1 | 128 | 55.9 | ||
| Other | NA1 | 134 | 52.5 | NA1 | 101 | 44.1 | ||
| SEER Summary Stage2,3 | ||||||||
| Local | NA1 | 831 | 71 | NA1 | 650 | 59.6 | ||
| Regional | NA1 | 325 | 27.8 | NA1 | 432 | 39.6 | ||
| Distant | NA1 | 15 | 1.3 | NA1 | 9 | 0.8 | ||
| Smoking Status4 | ||||||||
| Never | 794 | 60.3 | 688 | 56.1 | 1616 | 72.1 | 1298 | 70.1 |
| Former | 360 | 27.3 | 386 | 31.5 | 347 | 15.5 | 322 | 17.4 |
| Current | 163 | 12.4 | 152 | 12.4 | 278 | 12.4 | 231 | 12.5 |
| BMI (kg/m2) | ||||||||
| <25 | 699 | 44.4 | 678 | 45.9 | 453 | 17.6 | 492 | 23.5 |
| 25–29.9 | 465 | 29.5 | 433 | 29.3 | 951 | 36.9 | 768 | 36.7 |
| ≥30 | 412 | 26.1 | 367 | 24.8 | 1172 | 45.5 | 832 | 39.8 |
| NSAID use5 | ||||||||
| No | 708 | 53.7 | 670 | 54.7 | 446 | 61.7 | 395 | 66.2 |
| Yes | 610 | 46.3 | 554 | 45.3 | 277 | 38.3 | 202 | 33.8 |
| Dietary Oxidative Balance Score6 [mean (SD)] | ||||||||
| 4-CBCS | 6.3 (2.7) | 6.3 (2.6) | 6.7 (2.5) | 6.5 (2.6) | ||||
| MBCS | NA1 | NA1 | 5.9 (2.0) | 5.7 (2.0) | ||||
| SFBCS | 5.6 (2.6) | 5.7 (2.6) | 6.9 (2.5) | 6.1 (2.5) | ||||
Data not applicable (NA)
Data unavailable from Mexico Breast Cancer Study (MBCS).
Includes first primary invasive breast cancer cases from the 4-Corners Breast Cancer Study (4-CBCS) and San Francisco Bay Area Breast Cancer Study (SFBCS) only
Data available for a subset of SFBCS participants only
Data only available for the 4-CBCS
Dietary Oxidative Balance Score (DOBS) includes alcohol (pro-oxidant), vitamin C, vitamin E, beta carotene (data not available for MBCS), folic acid, and dietary fiber (anti-oxidants).
Few genes and SNPs were significantly associated with breast cancer risk (Table 2, shows those with an adjusted p value of <0.15). TLR1, TLR2, and TNFRSF11A had the strongest association among women with the highest level of NA ancestry. Of the 25 SNPs evaluated for TNFRSF11A, five were associated with breast cancer risk among those with high NA ancestry. Of these, rs7237982 (ORGG 2.34, 95% CI 1.05, 5.21), rs17069845 (ORTC/CC 0.74, 95% CI 0.57, 0.97), and rs8083511 (ORCC 1.74, 95% CI 1.12, 2.70) were significantly associated with breast cancer risk and the ORs were significantly different from those in other ancestry groups prior to adjustment for multiple comparisons (data not shown in table). CD40LG rs1126535 was significantly associated with breast cancer risk among those with low NA ancestry and CD40LG rs5939073 was associated with breast cancer risk among those with intermediate ancestry.
Table 2.
Summary of significant associations between pathway genes and breast cancer risk by percent Native American ancestry
| <=28% Native American Ancestry | >28 – 70% Native American Ancestry | >70% Native American Ancestry | Interaction P-value (raw; adjusted)2 | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
|
| ||||||||||
| Controls | Cases | OR1 (95% CI) | Controls | Cases | OR (95% CI) | Controls | Cases | OR (95% CI) | ||
| CD40LG (rs1126535) | 0.014, 0.036 | |||||||||
| TT | 1232 | 1136 | 1.00 | 516 | 457 | 1.00 | 81 | 68 | 1.00 | |
| TC | 552 | 526 | 1.05 (0.90, 1.21) | 824 | 662 | 0.91 (0.77, 1.07) | 291 | 182 | 0.77 (0.52, 1.13) | |
| CC | 57 | 78 | 1.47 (1.03, 2.10) | 337 | 270 | 0.94 (0.76, 1.16) | 256 | 187 | 0.96 (0.65, 1.41) | |
| P-value (raw; adjusted) 2 | 0.032, 0.083 | 0.576, 0.940 | 0.821, 1.000 | |||||||
| CD40LG (rs5930973) | 0.022, 0.036 | |||||||||
| GG | 1637 | 1560 | 1.00 | 1630 | 1327 | 1.00 | 623 | 433 | 1.00 | |
| GA/AA | 205 | 182 | 0.93 (0.75, 1.15) | 49 | 63 | 1.63 (1.11, 2.40) | 6 | 4 | 0.79 (0.21, 2.94) | |
| P-value (raw; adjusted) | 0.519, 0.855 | 0.014, 0.036 | 0.730, 1.000 | |||||||
| TLR1 (rs7696175) | <.001, <.001 | |||||||||
| CC/CT | 1472 | 1444 | 1.00 | 1462 | 1174 | 1.00 | 550 | 355 | 1.00 | |
| TT | 370 | 298 | 0.82 (0.69, 0.97) | 215 | 216 | 1.25 (1.01, 1.53) | 78 | 82 | 1.77 (1.25, 2.51) | |
| P-value (raw; adjusted) | 0.020, 0.020 | 0.038, 0.038 | 0.001, 0.001 | |||||||
| TLR2 (rs4696483) | 0.012, 0.036 | |||||||||
| CC/CT | 1796 | 1714 | 1.00 | 1637 | 1352 | 1.00 | 617 | 422 | 1.00 | |
| TT | 46 | 27 | 0.61 (0.38, 0.99) | 42 | 38 | 1.13 (0.72, 1.78) | 12 | 15 | 1.83 (0.84, 4.00) | |
| P-value (raw; adjusted) | 0.047, 0.142 | 0.583, 1.000 | 0.131, 0.392 | |||||||
| TNFRSF11A (rs8099222) | 0.227, 1.000 | |||||||||
| GG | 1055 | 994 | 1.00 | 1028 | 877 | 1.00 | 494 | 320 | 1.00 | |
| GA | 681 | 640 | 1.00 (0.87, 1.15) | 546 | 453 | 0.97 (0.83, 1.13) | 131 | 102 | 1.11 (0.82, 1.50) | |
| AA | 97 | 92 | 1.01 (0.75, 1.36) | 103 | 60 | 0.69 (0.49, 0.96) | 4 | 15 | 4.66 (1.51, 14.40) | |
| P-value (raw; adjusted) | 0.970, 1.000 | 0.028, 0.493 | 0.008, 0.133 | |||||||
| TNFRSF11A (rs8089829) | 0.004, 0.070 | |||||||||
| AA | 525 | 532 | 1.00 | 644 | 499 | 1.00 | 324 | 203 | 1.00 | |
| AG | 916 | 826 | 0.88 (0.75, 1.02) | 759 | 653 | 1.11 (0.95, 1.31) | 272 | 189 | 1.06 (0.82, 1.38) | |
| GG | 398 | 380 | 0.94 (0.78, 1.13) | 274 | 236 | 1.10 (0.89, 1.36) | 33 | 44 | 1.92 (1.17, 3.16) | |
| P-value (raw; adjusted) | 0.495, 1.000 | 0.381, 1.000 | 0.010, 0.161 | |||||||
Odds Ratios (OR) and 95% Confidence Intervals (CI) adjusted for age, study, BMI during referent year, parity, and genetic ancestry.
P values in table for SNPs are unadjusted and adjusted for multiple comparisons using the step down Bonferroni correction. ARTPp for <=28% NA ancestry = 0.267 for CD40LG, 0.019 for TLR1, 0.205 for TLR2, and 0.93 for TNFRSF11A; for >28–70% NA ancestry = 0.0498 for CD40LG, 0.037 for TLR1, 0.822 for TLR2 and 0.348 for TNFRSF11A; for >70% NA ancestry CD40LG = 0.762, TLR1 = 0.0008, TLR2= 0.446, and TNFRSF11A = 0.176. Pathway p value only significant for >70% NA = 0.015.
No differences in risk were identified by menopausal status (data not shown), and only two significant associations were identified by ER/PR phenotype (Table 3). TLR3 rs5743305 was associated with ER−/PR+ tumors and TNFRSF1A rs4149578 was significantly associated with ER−/PR− tumors. Both of these genes had statistically significant ARTP p values of 0.011 and 0.023, respectively.
Table 3.
Associations between TLR3 and TNFRSF1 and breast cancer risk, by ER/PR tumor phenotype.
| Controls | ER+/PR+ | ER+/PR− | ER−/PR+ | ER−/PR− | Multinomial P-values (raw; adjusted)2,3 | |||||
|---|---|---|---|---|---|---|---|---|---|---|
|
| ||||||||||
| N | N | OR1 (95% CI) | N | OR (95% CI) | N | OR (95% CI) | N | OR (95% CI) | ||
| TLR3 (rs5743305)4 | 0.011, 0.040 | |||||||||
| TT | 1321 | 528 | 1.00 | 101 | 1.00 | 10 | 1.00 | 157 | 1.00 | |
| TA | 1415 | 597 | 1.04 (0.91, 1.20) | 106 | 0.97 (0.73, 1.29) | 21 | 1.98 (0.93, 4.23) | 201 | 1.20 (0.96, 1.50) | |
| AA | 430 | 172 | 0.97 (0.79, 1.19) | 28 | 0.82 (0.53, 1.27) | 12 | 3.70 (1.58, 8.67) | 57 | 1.11 (0.80, 1.53) | |
| P-value (raw; adjusted) 2 | 0.774, 1.000 | 0.387, 1.000 | 0.003,0.010 | 0.534, 1.000 | ||||||
| TNFRSF1A (rs4149578)4 | 0.018, 0.053 | |||||||||
| GG | 2487 | 1000 | 1.00 | 184 | 1.00 | 33 | 1.00 | 350 | 1.00 | |
| GA/AA | 678 | 295 | 1.09 (0.93, 1.27) | 50 | 0.99 (0.72, 1.38) | 10 | 1.13 (0.55, 2.30) | 64 | 0.67 (0.51, 0.89) | |
| P-value (raw; adjusted) | 0.284, 0.851 | 0.971, 1.000 | 0.747, 1.000 | 0.006, 0.017 | ||||||
Odds Ratios (OR) and 95% Confidence Intervals (CI) adjusted for age, study, BMI during referent year, parity, and genetic ancestry.
P values for SNPs in table are unadjusted and adjusted for multiple comparisons using the step down Bonferroni correction.
Multinomial p values exclude controls.
ARTP p value for TLR3 for ER−/PR+ tumors was 0.011; ARTP p value for ER−/PR− tumors for TNFRSF1A was 0.023.
Two of the three SNPs analyzed in CD40LG and one of four SNPs in TNFRSF1A significantly interacted with DOBS (Table 4). CD40LG homozygote variant was associated with increased risk of breast cancer only among those with low DOBS. For the homozygote common genotype of TNFRSF1A rs4149570, breast cancer risk decreased with increasing DOBS. TLR1, TLR4 (1 of 4 SNPs) and TNF (1 of 4 SNPs) interacted with cigarette smoking. TLR1 homozygote rare genotype significantly increased risk only among never-smokers; TLR4 rs111536898 rare allele decreased risk among never-smokers; TNF rs1800630 rare allele increased risk among current smokers only. Three of four SNPs in TRAF2 were associated with BMI among pre-menopausal women only. For these SNPs, the rare allele was associated with a reduced risk of breast cancer among obese women. Two SNPs of TLR3 interacted with aspirin/NSAIDs with the greatest effect among regular users. TRAF2 rs4880073 also interacted with aspirin/NSAID use with the AA genotype reducing breast cancer risk among non-regular users. After adjustment for multiple comparisons, no significant interactions between having diabetes and any SNPs were observed.
Table 4.
Gene and environment interactions associated with breast cancer risk.
| Dietary Oxidative Balance Score (DOBS)1
|
|||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Quartile 1 | Quartile 2 | Quartile 3 | Quartile 4 | Interaction P-value (raw; adjusted)3 | |||||||||
|
| |||||||||||||
| Controls | Cases | OR2 (95% CI) | Controls | Cases | OR (95% CI) | Controls | Cases | OR (95% CI) | Controls | Cases | OR (95% CI) | ||
| CD40LG (rs1126535) | 0.014, 0.029 | ||||||||||||
| TT | 477 | 462 | 1.00 | 461 | 416 | 0.96 (0.80, 1.15) | 475 | 429 | 0.96 (0.80, 1.15) | 402 | 339 | 0.88 (0.72, 1.07) | |
| TC | 359 | 373 | 1.15 (0.95, 1.40) | 373 | 335 | 1.02 (0.83, 1.24) | 523 | 346 | 0.75 (0.62, 0.91) | 376 | 289 | 0.88 (0.72, 1.08) | |
| CC | 129 | 147 | 1.35 (1.02, 1.78) | 119 | 112 | 1.12 (0.83, 1.51) | 213 | 154 | 0.89 (0.69, 1.15) | 156 | 103 | 0.81 (0.60, 1.08) | |
| CD40LG (rs3092936) | 0.011, 0.029 | ||||||||||||
| TT | 620 | 599 | 1.00 | 604 | 566 | 1 (0.85, 1.18) | 693 | 571 | 0.88 (0.75, 1.03) | 545 | 458 | 0.89 (0.75, 1.05) | |
| TC | 276 | 302 | 1.21 (0.99, 1.49) | 289 | 243 | 0.95 (0.77, 1.17) | 403 | 282 | 0.79 (0.65, 0.97) | 313 | 221 | 0.81 (0.65, 1.00) | |
| CC | 68 | 81 | 1.42 (1.00, 2.01) | 60 | 56 | 1.09 (0.74, 1.62) | 117 | 76 | 0.8 (0.58, 1.10) | 78 | 52 | 0.81 (0.55, 1.18) | |
| TNFRSF1A (rs4149570) | 0.011, 0.032 | ||||||||||||
| GG | 399 | 428 | 1.00 | 413 | 376 | 0.87 (0.71, 1.06) | 564 | 439 | 0.75 (0.62, 0.90) | 450 | 309 | 0.65 (0.53, 0.80) | |
| GT | 441 | 436 | 0.90 (0.74, 1.09) | 421 | 386 | 0.85 (0.69, 1.03) | 530 | 385 | 0.68 (0.56, 0.82) | 392 | 335 | 0.8 (0.65, 0.98) | |
| TT | 125 | 116 | 0.81 (0.61, 1.09) | 119 | 102 | 0.78 (0.58, 1.06) | 119 | 105 | 0.81 (0.60, 1.09) | 92 | 88 | 0.86 (0.62, 1.19) | |
| Smoking Status4
|
Interaction P-value | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Never-Smoker | Former Smoker | Current Smoker | ||||||||
| TLR1 (rs7696175) | 0.032, 0.032 | |||||||||
| CC | 895 | 662 | 1.00 | 271 | 271 | 1.26 (1.03, 1.54) | 136 | 124 | 1.16 (0.89, 1.52) | |
| CT | 1130 | 954 | 1.13 (0.99, 1.29) | 323 | 321 | 1.25 (1.04, 1.51) | 225 | 192 | 1.09 (0.87, 1.35) | |
| TT | 368 | 355 | 1.29 (1.08, 1.54) | 111 | 110 | 1.18 (0.88, 1.57) | 75 | 66 | 1.05 (0.74, 1.49) | |
| TLR4 (rs11536898) | 0.012, 0.069 | |||||||||
| CC | 1962 | 1654 | 1.00 | 573 | 550 | 1.06 (0.92, 1.21) | 372 | 313 | 0.93 (0.79, 1.10) | |
| CA/AA | 433 | 317 | 0.83 (0.70, 0.97) | 132 | 152 | 1.22 (0.95, 1.56) | 64 | 69 | 1.15 (0.81, 1.63) | |
| TNF (rs1800630) | 0.005, 0.013 | |||||||||
| CC | 1803 | 1505 | 1.00 | 532 | 508 | 1.07 (0.93, 1.24) | 339 | 269 | 0.88 (0.74, 1.05) | |
| CA/AA | 592 | 466 | 0.93 (0.81, 1.07) | 173 | 194 | 1.21 (0.97, 1.51) | 97 | 112 | 1.31 (0.99, 1.74) | |
| BMI Among Pre-Menopausal Women
|
||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Normal (< 25 kg/m2) | Overweight (25 to <30 kg/m2) | Obese (>= 30 kg/m2) | ||||||||
| TRAF2 (rs2784075)5 | 0.004, 0.008 | |||||||||
| GG | 244 | 244 | 1.00 | 229 | 170 | 0.78 (0.60, 1.03) | 195 | 183 | 1.02 (0.77, 1.34) | |
| GA/AA | 222 | 268 | 1.25 (0.97, 1.62) | 295 | 245 | 0.94 (0.72, 1.22) | 328 | 208 | 0.73 (0.56, 0.95) | |
| TRAF2 (rs7027246) | <.001, 0.002 | |||||||||
| GG | 230 | 220 | 1.00 | 222 | 159 | 0.8 (0.60, 1.06) | 183 | 179 | 1.1 (0.83, 1.47) | |
| GA/AA | 235 | 288 | 1.32 (1.02, 1.70) | 300 | 255 | 0.99 (0.76, 1.30) | 337 | 209 | 0.74 (0.57, 0.98) | |
| TRAF2 (rs908831) | 0.018, 0.021 | |||||||||
| AA | 152 | 168 | 1.00 | 143 | 105 | 0.7 (0.50, 0.99) | 124 | 128 | 1.00 (0.71, 1.41) | |
| AG | 227 | 236 | 0.96 (0.72, 1.28) | 261 | 209 | 0.8 (0.59, 1.07) | 263 | 184 | 0.70 (0.52, 0.95) | |
| GG | 87 | 108 | 1.17 (0.81, 1.68) | 120 | 101 | 0.85 (0.59, 1.21) | 136 | 79 | 0.61 (0.42, 0.88) | |
| Aspirin/NSAID Use6
|
Interaction P value | ||||||
|---|---|---|---|---|---|---|---|
| Non-Regular Users | Regular Users | ||||||
|
| |||||||
| Controls | Cases | OR (95% CI) | Controls | Cases | OR (95% CI) | ||
|
| |||||||
| TLR3 (rs11721827) | <.001, 0.003 | ||||||
| AA | 879 | 779 | 1.00 | 631 | 579 | 1.07 (0.92, 1.24) | |
| AC | 239 | 251 | 1.17 (0.96, 1.43) | 226 | 164 | 0.83 (0.66, 1.04) | |
| CC | 24 | 29 | 1.33 (0.76, 2.30) | 26 | 11 | 0.48 (0.23, 0.97) | |
| TLR3 (rs3775291) | 0.014, 0.038 | ||||||
| GG | 552 | 521 | 1.00 | 479 | 358 | 0.82 (0.68, 0.98) | |
| GA | 482 | 438 | 0.98 (0.82, 1.17) | 346 | 324 | 1.04 (0.85, 1.26) | |
| AA | 108 | 101 | 1.00 (0.75, 1.35) | 58 | 71 | 1.36 (0.94, 1.97) | |
| TRAF2 (rs4880073) | 0.016, 0.049 | ||||||
| GG | 330 | 357 | 1.00 | 315 | 243 | 0.72 (0.58, 0.91) | |
| GA | 589 | 518 | 0.81 (0.67, 0.98) | 409 | 375 | 0.87 (0.71, 1.07) | |
| AA | 223 | 185 | 0.76 (0.59, 0.97) | 159 | 136 | 0.81 (0.62, 1.07) | |
Dietary Oxidative Balance Score (DOBS) includes alcohol (pro-oxidant), vitamin C, vitamin E, beta carotene (data not available for MBCS), folic acid, and dietary fiber (anti-oxidants).
Odds Ratios (OR) and 95% Confidence Intervals (CI) adjusted for age, study, BMI during referent year (where appropriate), parity, and genetic ancestry.
P values are unadjusted and adjusted for multiple comparisons using the step down Bonferroni correction.
Data unavailable for subset of SFBCS participants.
In high LD with rs7027246: r2 of 0.79, 0.93, and 0.96 among women with ≤28% NA ancestry, 29–70% NA ancestry, and ≥71% NA ancestry, respectively.
Data available for 4-CBCS participants only.
TLR1 rs7696175, TRL4 rs5030728, TNRFSF1A rs4149578, TNFSF10 rs231985, rs3136597, and rs231983 were associated with breast cancer-specific mortality (Table 5). In all instances the rare genotype was associated with poorer survival. Associations with TLR1, TLR4, TNFRSF1A, and TNFRSF10 rs231985 were slightly stronger among those with greater NA ancestry; however, there were no statistically significant differences in mortality by NA ancestry. Associations with all-cause mortality were similar as those presented for breast cancer-specific mortality (data not shown) with a few exceptions; TLR1 rs7696175, TLR4 rs5030728, and TNFSF10 rs231985 were uniquely associated with breast cancer-specific mortality. On the other hand, TLR4 rs10759932 was strongly associated with overall mortality (ORcc = 1.94, 95% CI 1.14, 3.32) but not with breast cancer-specific mortality (OR=0.93, 95% CI 0.70, 1.24).
Table 5.
Associations between genes and breast cancer-specific mortality by percent Native American Ancestry.
| ≤28% Native American Ancestry | >28% Native American Ancestry | Interaction P-value (raw, adjusted)2 | |||||
|---|---|---|---|---|---|---|---|
| Deaths/Person Years | HR1 (95% CI) | Deaths/Person Years | HR (95% CI) | Deaths/Person Years | HR (95% CI) | ||
| TLR1 (rs7696175)3 | 0.880, 0.880 | ||||||
| CC/CT | 198 / 20050 | 1.00 | 114 / 11827 | 1.00 | 84 / 8223 | 1.00 | |
| TT | 51 / 3708 | 1.40 (1.03, 1.91) | 31 / 2374 | 1.40 (0.94, 2.09) | 20 / 1334 | 1.48 (0.91, 2.42) | |
| P-value (raw; adjusted)2 | 0.032, 0.032 | 0.100, 0.100 | 0.116, 0.116 | ||||
| TLR4 (rs5030728) | 0.252, 1.000 | ||||||
| GG | 114 / 12578 | 1.00 | 65 / 6971 | 1.00 | 49 / 5608 | 1.00 | |
| GA | 106 / 9413 | 1.24 (0.95, 1.61) | 64 / 6015 | 1.18 (0.83, 1.66) | 42 / 3398 | 1.32 (0.87, 2.01) | |
| AA | 29 / 1767 | 1.96 (1.30, 2.95) | 16 / 1216 | 1.59 (0.92, 2.76) | 13 / 552 | 2.70 (1.46, 5.01) | |
| P-value (raw; adjusted) | 0.001, 0.008 | 0.096, 0.506 | 0.002, 0.010 | ||||
| TNFRSF1A (rs4149578) | 0.356, 1.000 | ||||||
| GG/GA | 240 / 23463 | 1.00 | 140 / 14012 | 1.00 | 100 / 9451 | 1.00 | |
| AA | 7 / 253 | 2.71 (1.27, 5.76) | 4 / 174 | 1.97 (0.72, 5.37) | 3 / 80 | 4.34 (1.34, 14.06) | |
| P-value (raw; adjusted) | 0.010, 0.029 | 0.184, 0.553 | 0.014, 0.043 | ||||
| TNFSF10 (rs231985) | 0.348, 1.000 | ||||||
| AA/AT | 243 / 23464 | 1.00 | 143 / 13979 | 1.00 | 100 / 9486 | 1.00 | |
| TT | 5 / 219 | 2.99 (1.23, 7.28) | 2 / 169 | 2.15 (0.52, 8.78) | 3 / 50 | 5.26 (1.63, 16.93) | |
| P-value (raw; adjusted) | 0.016, 0.096 | 0.288, 0.622 | 0.005, 0.038 | ||||
| TNFSF10 (rs3136597) | 0.354, 1.000 | ||||||
| CC | 153 / 15330 | 1.00 | 81 / 8852 | 1.00 | 72 / 6478 | 1.00 | |
| CA | 78 / 7407 | 1.14 (0.86, 1.50) | 53 / 4694 | 1.34 (0.94, 1.89) | 25 / 2713 | 0.88 (0.55, 1.40) | |
| AA | 18 / 1021 | 2.01 (1.23, 3.29) | 11 / 655 | 2.14 (1.13, 4.06) | 7 / 366 | 1.99 (0.91, 4.36) | |
| P-value (raw; adjusted) | 0.006, 0.039 | 0.020, 0.140 | 0.084, 0.508 | ||||
| TNFSF10 (rs231983) | 0.326, 1.000 | ||||||
| CC | 48 / 5967 | 1.00 | 17 / 2607 | 1.00 | 31 / 3360 | 1.00 | |
| CA/AA | 200 / 17713 | 1.44 (1.04, 1.98) | 128 / 11541 | 1.76 (1.06, 2.92) | 72 / 6172 | 1.28 (0.83, 1.97) | |
| P-value (raw; adjusted) | 0.027, 0.138 | 0.030, 0.179 | 0.269, 1.000 | ||||
Breast cancer survival among primary invasive cases; Hazard Ratios (HR) and 95% Confidence Intervals (CI) adjusted for age, study, BMI during referent year, genetic ancestry, and SEER summary stage. Deaths are for breast cancer deaths only; other causes of death are censored.
P values for SNPs are unadjusted and adjusted for multiple comparisons using the step down Bonferroni correction.
ARTP p values for genes among all cases: TLR1 = 0.037; TLR4 = 0.032; TNFRSF1A= 0.151; TNFSF10= 0.122. Pathway p value is 0.152.
Discussion
Major contributions of this paper are the identification of important diet and lifestyle factors that modify associations between breast cancer risk and TNF and TLR-related genes and SNPs and of the finding that variants in these genes are associated with both breast cancer risk and mortality in a genetically admixed population. However, few genes and SNPs were associated with either breast cancer risk or mortality. We confirmed that TLR1, which was previously identified in a breast cancer GWAS, was associated with breast cancer risk and mortality, especially among women with greater NA ancestry. Additionally, TLR4, TNFRSF1A, and TNFS10 were associated with breast cancer-specific mortality. TLR3 rs5743305 was associated with ER−/PR+ tumors and TNFRSF1A rs4149578 was significantly associated with ER−/PR− tumors. Diet and lifestyle factors associated with oxidative stress, inflammation, and insulin significantly interacted with several SNPs in these genes.
TLRs are a set of innate immunity genes involved in the activation of NFKB and MAPK, thereby mediating immune/inflammatory response [29]. TLRs can promote inflammation, cell survival and tumor progression [30]. Studies have shown associations between TLR4 Asp299Gly (rs4986790) with increased breast cancer risk and lower metastasis-free survival, although TLR4 rs1927911 and rs10759932 were not associated with survival [31]. TLR4 rs4986970 affects the extracellular domain of TLR4 and is associated with reduced endotoxin responses [31]; TLR4 rs4986971 (in perfect LD with rs4986970) in the promoter region also has been shown to affect gene function [8]. Reduced expression of TLR4 has been shown to inhibit breast cancer cell proliferation; knock out of TLR4 gene can actively inhibit breast cancer cell survival [30]. TLR3 has been shown to directly trigger apoptosis in human breast cancer cells [30,32]. TLR1 was the only gene in this study associated with breast cancer risk based on the ARTP results while both TLR1 rs7696175 and TLR4 rs5030728 were associated with breast cancer-specific mortality. We did not detect any significant associations with previously identified functional SNPs in either TLR4 or TNF. TLR1 rs7696175, which we and others have previously reported being associated with breast cancer risk [33,34], was identified as being a major contributor to risk within the pathway and also associated with mortality in this study.
The TNF family is a group of cytokines associated with apoptosis and antitumor activity; however, they also are involved in inflammation, immunity, and tumor progression [35]. A previous study in Mexico women found that TNF -308 G>A polymorphism (rs1800629) was associated with breast cancer risk [12]. We did not confirm this association. In our study the MAF was 0.079 among Hispanic controls, the majority of whom were from Mexico, while the MAF in NHWs was 0.17. The Mexico study was based on 294 controls (1% AA genotype) and 465 cases (14% AA genotype). Our data are in HWE and show 1 case and no controls with this genotype in the highest NA ancestry group (423 cases, 608 controls) that is predominately from Mexico; our numbers are consistent with allele frequencies reported for Hispanic populations in National Center for Biotechnology Information. A meta-analysis of this SNP and breast cancer risk showed a null association as we observed in this study [36].
TNF apoptosis inducing ligand (TNFSF10 or TRAIL) has been shown to activate apoptosis upon binding to its receptor and has been shown to influence survival among those with metastatic colon cancer [37–39]; it was associated with breast cancer-specific mortality in this study. Triple negative breast cancer cell lines have been shown to be sensitive to TNFSF10, whereas other tumor phenotypes are not [38,40]. While we did not see an association between any SNPs in this gene and ER−/PR− tumors, HER2 data were not available and we did observe an association between ER−/PR− tumors and TNFRSF1A (TNFR superfamily receptor 1A or p60). TNFRSF1A rs4149570 and rs12426675 have been associated with hepatocellular carcinoma cancer and these SNPs in the promoter have high transcriptional activity [41]. Functionality has been assigned to TNFRSF1A rs4149570 in the promoter that results in repression of TNFR1 [41]. We observed a significant interaction with DOBS and this SNP. TNFRSF1A rs4149578 was associated with breast cancer-specific mortality in our study.
We examined several diet and lifestyle factors that could influence oxidative stress, the functional role of these genes. Two of the three SNPs analyzed in CD40LG and one of four SNPs in TNFRSF1A interacted significantly with DOBS, while TLR1 (1), TLR4 (1 of 4 SNPs) and TNF (1 of 4 SNPs) interacted with cigarette smoking. CD40LG is an immune response gene and involved in thrombo-inflammatory reactions by up-regulating cell adhesion molecules and increased production of pro-inflammatory cytokines and reactive oxygen species [42,43]. Higher intake of dietary antioxidants modified the risk associated with the variant allele in two of the CD40LG SNPs. Vitamin C, which is a component of our DOBS, has been shown to suppress NFκB activation by inhibiting TNF activation of IKK [44]. TNF also has been shown to be able to induce reactive oxygen species [12]. Cigarette smoking has been shown to reduce innate immune response by suppressing inflammatory mediators [45], and a high oxidant/free radical burden in cigarette smoke has been correlated with increased expression of inflammatory mediator TNF [46]. The interaction we observed between TLRs and TNF and cigarette smoking has biological plausibility, given the influence of cigarette smoking on immune response and free radical burden and the key role of TLR and TNF in mediators of immune response.
Two SNPs of TLR3 interacted with aspirin/NSAIDs with the greatest effect found among regular users; TRAF2 rs4880073 also interacted with aspirin/NSAIDs. Aspirin has been shown to interfere with the NFκB complex [47]. TRAF2 is required for an NFκB independent signal that protects against TNF-induced apoptosis and TLR3 signaling activates the transcription of NFκB and interferon regulatory factor 3 [48]. TLR3 rs3775291 has been associated with aspirin-exacerbated respiratory disease where eosinophils are activated via TLR3 and then recruit leukocytes to sites of inflammation as part of an inflammatory response.
Three of four SNPs in TRAF2 were associated with BMI among pre-menopausal women, but not among post-menopausal women. The immune system has been shown to play a role in obesity and insulin resistance. The CD40 signaling intermediary is TRAF2 and it has been shown that CD40-mice have worsened insulin resistance. Thus the CD40/TRAF2 signaling pathway is thought to protect against adipose tissue inflammation and metabolic complications associated with obesity [49].
This study represents one of the largest studies of breast cancer in Hispanics, a genetically admixed population of European and NA ancestry. The pooling of data from three studies allowed us to evaluate associations with risk, mortality, and lifestyle factors that could mediate genetic risk. We have pooled our populations to test the hypothesis that differences in risk are associated with ancestry, thus using the population only as a replication from one to the other could yield misleading results. While we have tried to target key genes and SNPs in the candidate pathway, there may be other important genes and SNPs that are not included here. We utilized the Illuminia platform that was based on a Caucasian population LD structure, which could result in not capturing the entire variation in populations with more NA ancestry. Likewise, since we used a tagSNP approach to capture variation across the gene, we have detected associations with SNPs that we do not know their functional significance. Other variables such as persistent infections or country of nativity might be important confounders which we were not able to adjust in our data. Although we used several statistical methods to adjust for the associations observed among our candidate genes, associations could still be chance findings that need replication in other similar populations.
Several genes and SNPs were associated with breast cancer risk and mortality, although the pathway was only significant for women with the highest NA ancestry Additionally, DOBS, cigarette smoking, pre-menopausal BMI, and use of aspirin/NSAID significantly interacted with several SNPs within the pathway. This study suggests the importance of incorporating diet and lifestyle factors to obtain a better understanding of the total underlying genetic risk associated with breast cancer.
Supplementary Material
Highlights.
Associations were stronger among women with greater Native American ancestry.
TLR1 rs7696175 had the strongest influence on risk.
TLR1, TLR4, TNFRSF1A were associated with increased breast cancer mortality.
Diet and lifestyle factors mediated breast cancer risk associated with these genes.
Acknowledgments
Funding: The Breast Cancer Health Disparities Study was funded by grant CA14002 from the National Cancer Institute to Dr. Slattery. The San Francisco Bay Area Breast Cancer Study was supported by grants CA63446 and CA77305 from the National Cancer Institute, grant DAMD17-96-1-6071 from the U.S. Department of Defense and grant 7PB-0068 from the California Breast Cancer Research Program. The collection of cancer incidence data used in this study was supported by the California Department of Public Health as part of the statewide cancer reporting program mandated by California Health and Safety Code Section 103885; the National Cancer Institute’s Surveillance, Epidemiology and End Results Program under contract HHSN261201000036C awarded to the Cancer Prevention Institute of California; and the Centers for Disease Control and Prevention’s National Program of Cancer Registries, under agreement #1U58 DP000807-01 awarded to the Public Health Institute. The 4-Corner’s Breast Cancer Study was funded by grants CA078682, CA078762, CA078552, and CA078802 from the National Cancer Institute. The research also was supported by the Utah Cancer Registry, which is funded by contract N01-PC-67000 from the National Cancer Institute, with additional support from the State of Utah Department of Health, the New Mexico Tumor Registry, and the Arizona and Colorado cancer registries, funded by the Centers for Disease Control and Prevention National Program of Cancer Registries and additional state support. The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official view of the National Cancer Institute or endorsement by the State of California Department of Public Health, the National Cancer Institute, and the Centers for Disease Control and Prevention or their Contractors and Subcontractors. The Mexico Breast Cancer Study was funded by Consejo Nacional de Ciencia y Tecnología (CONACyT) (SALUD-2002-C01-7462).
We would also like to acknowledge the contributions of the following individuals to the study: Sandra Edwards and Jennifer Herrick for data harmonization oversight; Erica Wolff and Michael Hoffman for laboratory support; Carolina Ortega for her assistance with data management for the Mexico Breast Cancer Study, Jocelyn Koo for data management for the San Francisco Bay Area Breast Cancer Study; Dr. Tim Byers and Dr. Anna Giuliano for their contribution to the 4-Corners Breast Cancer Study; and Dr. Josh Galanter for assistance in selection of AIMs markers.
Footnotes
Conflict of Interest Statement.
The authors have no conflict of interest to report.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Gao Z, Zuberi A, Quon MJ, Dong Z, Ye J. Aspirin Inhibits Serine Phosphorylation of Insulin Receptor Substrate 1 in Tumor Necrosis Factor-treated Cells through Targeting Multiple Serine Kinases. J Biol Chem. 2003;278:24944–24950. doi: 10.1074/jbc.M300423200. [DOI] [PubMed] [Google Scholar]
- 2.Xu H, Hotamisligil GS. Signaling pathways utilized by tumor necrosis factor receptor 1 in adipocytes to suppress differentiation. FEBS Lett. 2001;506:97–102. doi: 10.1016/s0014-5793(01)02889-7. [DOI] [PubMed] [Google Scholar]
- 3.Hotamisligil GS. The role of TNFalpha and TNF receptors in obesity and insulin resistance. J Intern Med. 1999;245:621–625. doi: 10.1046/j.1365-2796.1999.00490.x. [DOI] [PubMed] [Google Scholar]
- 4.Hotamisligil GS, Peraldi P, Budavari A, Ellis R, White MF, et al. IRS-1-mediated inhibition of insulin receptor tyrosine kinase activity in TNF-alpha- and obesity-induced insulin resistance. Science. 1996;271:665–668. doi: 10.1126/science.271.5249.665. [DOI] [PubMed] [Google Scholar]
- 5.Hofmann C, Lorenz K, Braithwaite SS, Colca JR, Palazuk BJ, et al. Altered gene expression for tumor necrosis factor-alpha and its receptors during drug and dietary modulation of insulin resistance. Endocrinology. 1994;134:264–270. doi: 10.1210/endo.134.1.8275942. [DOI] [PubMed] [Google Scholar]
- 6.Kuribayashi K, Krigsfeld G, Wang W, Xu J, Mayes PA, et al. TNFSF10 (TRAIL), a p53 target gene that mediates p53-dependent cell death. Cancer Biol Ther. 2008;7:2034–2038. doi: 10.4161/cbt.7.12.7460. [DOI] [PubMed] [Google Scholar]
- 7.Fukata M, Abreu MT. TLR4 signalling in the intestine in health and disease. Biochem Soc Trans. 2007;35:1473–1478. doi: 10.1042/BST0351473. [DOI] [PubMed] [Google Scholar]
- 8.Resler AJ, Malone KE, Johnson LG, Malkki M, Petersdorf EW, et al. Genetic variation in TLR or NFkappaB pathways and the risk of breast cancer: a case-control study. BMC Cancer. 2013;13:219. doi: 10.1186/1471-2407-13-219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Mai CW, Kang YB, Pichika MR. Should a Toll-like receptor 4 (TLR-4) agonist or antagonist be designed to treat cancer? TLR-4: its expression and effects in the ten most common cancers. Onco Targets Ther. 2013;6:1573–1587. doi: 10.2147/OTT.S50838. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Simiantonaki N, Kurzik-Dumke U, Karyofylli G, Jayasinghe C, Michel-Schmidt R, et al. Reduced expression of TLR4 is associated with the metastatic status of human colorectal cancer. Int J Mol Med. 2007;20:21–29. [PubMed] [Google Scholar]
- 11.Niedzielska I, Niedzielski Z, Tkacz M, Orawczyk T, Ziaja K, et al. Toll-like receptors and the tendency of normal mucous membrane to transform to polyp or colorectal cancer. J Physiol Pharmacol. 2009;60(Suppl 1):65–71. [PubMed] [Google Scholar]
- 12.Gomez Flores-Ramos L, Escoto-De Dios A, Puebla-Perez AM, Figuera-Villanueva LE, Ramos-Silva A, et al. Association of the tumor necrosis factor-alpha -308G>A polymorphism with breast cancer in Mexican women. Genet Mol Res. 2013;12:5680–5693. doi: 10.4238/2013.November.18.17. [DOI] [PubMed] [Google Scholar]
- 13.Slattery ML, John EM, Torres-Mejia G, Lundgreen A, Herrick JS, et al. Genetic variation in genes involved in hormones, inflammation and energetic factors and breast cancer risk in an admixed population. Carcinogenesis. 2012;33:1512–1521. doi: 10.1093/carcin/bgs163. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Slattery ML, John EM, Torres-Mejia G, Lundgreen A, Lewinger JP, et al. Angiogenesis genes, dietary oxidative balance, and breast cancer risk and progression: The breast cancer health disparities study. Int J Cancer. 2013 doi: 10.1002/ijc.28377. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Slattery ML, Sweeney C, Edwards S, Herrick J, Baumgartner K, et al. Body size, weight change, fat distribution and breast cancer risk in Hispanic and non-Hispanic white women. Breast Cancer Res Treat. 2007;102:85–101. doi: 10.1007/s10549-006-9292-y. [DOI] [PubMed] [Google Scholar]
- 16.Angeles-Llerenas A, Ortega-Olvera C, Perez-Rodriguez E, Esparza-Cano JP, Lazcano-Ponce E, et al. Moderate physical activity and breast cancer risk: the effect of menopausal status. Cancer Causes Control. 2010;21:577–586. doi: 10.1007/s10552-009-9487-8. [DOI] [PubMed] [Google Scholar]
- 17.John EM, Horn-Ross PL, Koo J. Lifetime physical activity and breast cancer risk in a multiethnic population: the San Francisco Bay area breast cancer study. Cancer Epidemiol Biomarkers Prev. 2003;12:1143–1152. [PubMed] [Google Scholar]
- 18.John EM, Phipps AI, Davis A, Koo J. Migration history, acculturation, and breast cancer risk in Hispanic women. Cancer Epidemiol Biomarkers Prev. 2005;14:2905–2913. doi: 10.1158/1055-9965.EPI-05-0483. [DOI] [PubMed] [Google Scholar]
- 19.Slattery ML, Caan BJ, Duncan D, Berry TD, Coates A, et al. A computerized diet history questionnaire for epidemiologic studies. J Am Diet Assoc. 1994;94:761–766. doi: 10.1016/0002-8223(94)91944-5. [DOI] [PubMed] [Google Scholar]
- 20.Murtaugh MA, Sweeney C, Giuliano AR, Herrick JS, Hines L, et al. Diet patterns and breast cancer risk in Hispanic and non-Hispanic white women: the Four-Corners Breast Cancer Study. Am J Clin Nutr. 2008;87:978–984. doi: 10.1093/ajcn/87.4.978. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hernandez-Avila M, Romieu I, Parra S, Hernandez-Avila J, Madrigal H, et al. Validity and reproducibility of a food frequency questionnaire to assess dietary intake of women living in Mexico City. Salud publica de Mexico. 1998;40:133–140. doi: 10.1590/s0036-36341998000200005. [DOI] [PubMed] [Google Scholar]
- 22.Horn-Ross PL, John EM, Lee M, Stewart SL, Koo J, et al. Phytoestrogen consumption and breast cancer risk in a multiethnic population: the Bay Area Breast Cancer Study. Am J Epidemiol. 2001;154:434–441. doi: 10.1093/aje/154.5.434. [DOI] [PubMed] [Google Scholar]
- 23.Falush D, Stephens M, Pritchard JK. Inference of population structure using multilocus genotype data: linked loci and correlated allele frequencies. Genetics. 2003;164:1567–1587. doi: 10.1093/genetics/164.4.1567. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000;155:945–959. doi: 10.1093/genetics/155.2.945. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Nyholt DR. A simple correction for multiple testing for single-nucleotide polymorphisms in linkage disequilibrium with each other. American journal of human genetics. 2004;74:765–769. doi: 10.1086/383251. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Li J, Ji L. Adjusting multiple testing in multilocus analyses using the eigenvalues of a correlation matrix. Heredity. 2005;95:221–227. doi: 10.1038/sj.hdy.6800717. [DOI] [PubMed] [Google Scholar]
- 27.Yu K, Li Q, Bergen AW, Pfeiffer RM, Rosenberg PS, et al. Pathway analysis by adaptive combination of P-values. Genetic epidemiology. 2009;33:700–709. doi: 10.1002/gepi.20422. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kai Yu OL, Wheeler William. R package. 2.0.0. 2011. ARTP Gene and Pathway p-values computed using the Adaptive Rank Truncated Product. [Google Scholar]
- 29.Shatz M, Menendez D, Resnick MA. The human TLR innate immune gene family is differentially influenced by DNA stress and p53 status in cancer cells. Cancer Res. 2012;72:3948–3957. doi: 10.1158/0008-5472.CAN-11-4134. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Yang H, Zhou H, Feng P, Zhou X, Wen H, et al. Reduced expression of Toll-like receptor 4 inhibits human breast cancer cells proliferation and inflammatory cytokines secretion. J Exp Clin Cancer Res. 2010;29:92. doi: 10.1186/1756-9966-29-92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Apetoh L, Ghiringhelli F, Tesniere A, Obeid M, Ortiz C, et al. Toll-like receptor 4-dependent contribution of the immune system to anticancer chemotherapy and radiotherapy. Nat Med. 2007;13:1050–1059. doi: 10.1038/nm1622. [DOI] [PubMed] [Google Scholar]
- 32.Salaun B, Coste I, Rissoan MC, Lebecque SJ, Renno T. TLR3 can directly trigger apoptosis in human cancer cells. J Immunol. 2006;176:4894–4901. doi: 10.4049/jimmunol.176.8.4894. [DOI] [PubMed] [Google Scholar]
- 33.Fejerman L, Stern MC, Ziv E, John EM, Torres-Mejia G, et al. Genetic ancestry modifies the association between genetic risk variants and breast cancer risk among Hispanic and non-Hispanic white women. Carcinogenesis. 2013 doi: 10.1093/carcin/bgt110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Barnholtz-Sloan JS, Shetty PB, Guan X, Nyante SJ, Luo J, et al. FGFR2 and other loci identified in genome-wide association studies are associated with breast cancer in African-American and younger women. Carcinogenesis. 2010;31:1417–1423. doi: 10.1093/carcin/bgq128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Wu Y, Zhou BP. TNF-alpha/NF-kappaB/Snail pathway in cancer cell migration and invasion. Br J Cancer. 2010;102:639–644. doi: 10.1038/sj.bjc.6605530. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Guo J, Meng H, Pei J, Zhu M. Association between the TNF-alpha-238G>A and TGF-beta1 L10P Polymorphisms and Breast Cancer Risk: A Meta-Analysis. Breast Care (Basel) 2011;6:126–129. doi: 10.1159/000327515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Bisgin A, Kargi A, Yalcin AD, Aydin C, Ekinci D, et al. Increased serum sTRAIL levels were correlated with survival in bevacizumab-treated metastatic colon cancer. BMC Cancer. 2012;12:58. doi: 10.1186/1471-2407-12-58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Rahman M, Pumphrey JG, Lipkowitz S. The TRAIL to targeted therapy of breast cancer. Adv Cancer Res. 2009;103:43–73. doi: 10.1016/S0065-230X(09)03003-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Thorburn A, Behbakht K, Ford H. TRAIL receptor-targeted therapeutics: resistance mechanisms and strategies to avoid them. Drug Resist Updat. 2008;11:17–24. doi: 10.1016/j.drup.2008.02.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Rahman M, Davis SR, Pumphrey JG, Bao J, Nau MM, et al. TRAIL induces apoptosis in triple-negative breast cancer cells with a mesenchymal phenotype. Breast Cancer Res Treat. 2009;113:217–230. doi: 10.1007/s10549-008-9924-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Kim S, Moon SM, Kim YS, Kim JJ, Ryu HJ, et al. TNFR1 promoter -329G/T polymorphism results in allele-specific repression of TNFR1 expression. Biochem Biophys Res Commun. 2008;368:395–401. doi: 10.1016/j.bbrc.2008.01.098. [DOI] [PubMed] [Google Scholar]
- 42.Bou Khzam L, Hachem A, Zaid Y, Boulahya R, Mourad W, et al. Soluble CD40 ligand impairs the anti-platelet function of peripheral blood angiogenic outgrowth cells via increased production of reactive oxygen species. Thromb Haemost. 2013;109:940–947. doi: 10.1160/TH12-09-0679. [DOI] [PubMed] [Google Scholar]
- 43.Bou Khzam L, Boulahya R, Abou-Saleh H, Hachem A, Zaid Y, et al. Soluble CD40 ligand stimulates the pro-angiogenic function of peripheral blood angiogenic outgrowth cells via increased release of matrix metalloproteinase-9. PLoS One. 2013;8:e84289. doi: 10.1371/journal.pone.0084289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Carcamo JM, Pedraza A, Borquez-Ojeda O, Golde DW. Vitamin C suppresses TNF alpha-induced NF kappa B activation by inhibiting I kappa B alpha phosphorylation. Biochemistry. 2002;41:12995–13002. doi: 10.1021/bi0263210. [DOI] [PubMed] [Google Scholar]
- 45.Metcalfe HJ, Lea S, Hughes D, Khalaf R, Abbott-Banner K, et al. Effects of cigarette smoke on TLR activation of COPD macrophages. Clin Exp Immunol. 2014 doi: 10.1111/cei.12289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Bozinovski S, Vlahos R, Zhang Y, Lah LC, Seow HJ, et al. Carbonylation caused by cigarette smoke extract is associated with defective macrophage immunity. Am J Respir Cell Mol Biol. 2011;45:229–236. doi: 10.1165/rcmb.2010-0272OC. [DOI] [PubMed] [Google Scholar]
- 47.Brummelkamp TR, Nijman SM, Dirac AM, Bernards R. Loss of the cylindromatosis tumour suppressor inhibits apoptosis by activating NF-kappaB. Nature. 2003;424:797–801. doi: 10.1038/nature01811. [DOI] [PubMed] [Google Scholar]
- 48.Palikhe NS, Kim SH, Kim JH, Losol P, Ye YM, et al. Role of Toll-like Receptor 3 Variants in Aspirin-Exacerbated Respiratory Disease. Allergy Asthma Immunol Res. 2011;3:123–127. doi: 10.4168/aair.2011.3.2.123. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Chatzigeorgiou A, Seijkens T, Zarzycka B, Engel D, Poggi M, et al. Blocking CD40-TRAF6 signaling is a therapeutic target in obesity-associated insulin resistance. Proc Natl Acad Sci U S A. 2014;111:2686–2691. doi: 10.1073/pnas.1400419111. [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.