Abstract
Carcinogen–DNA adducts, a marker of DNA damage, are capable of inducing mutations in oncogenes and tumor suppressor genes, resulting in carcinogenesis. We have shown previously that polycyclic aromatic hydrocarbon (PAH)–DNA adduct levels in prostate cancer cases vary by cellular histology and that higher adduct levels are associated with biochemical recurrence. A nested case–control study was conducted in a historical cohort of 6692 men with histopathologically benign prostate specimens. PAH-DNA adduct levels were determined by immunohistochemistry in benign prostate specimens from 536 prostate cancer case-control pairs (59% White and 41% African American). We estimated the overall and race-stratified risk of subsequent prostate cancer associated with higher adduct levels. Prostate cancer risk for men with elevated adduct levels (defined as greater than control group median) was slightly increased [odds ratio (OR) = 1.28, 95% confidence interval (CI) = 0.98–1.67, P = 0.07]. After race stratification, elevated adduct levels were significantly associated with increased risk in African American men (OR = 1.56, CI = 1.00–2.44, *P = 0.05) but not White men (OR = 1.14, CI = 0.82–1.59, P = 0.45). Elevated PAH-DNA adduct levels were significantly associated with 60% increased risk of prostate cancer among cases diagnosed 1–4 years after cohort entry (OR = 1.60, CI = 1.07–2.41) with a greater risk observed in African Americans within the first 4 years of follow-up (OR = 4.71, CI = 1.97–11.26, ***P = 0.0005). Analyses stratified by age or tumor grade revealed no additional significant heterogeneity in risk. Increased prostate cancer risk associated with high PAH-DNA adduct levels in benign prostate was found only in African Americans; risk was greatest within 4 years of follow-up, possibly reflecting a carcinogenic process not yet histologically detectable.
Introduction
Prostate cancer is the most prevalent non-skin cancer and the second leading cause of cancer-related death among men in the USA (1). Among the established risk factors are family history, advancing age, and race. African American men experience significantly higher incidence and are more likely to die of prostate cancer than Whites (2); reasons for this disparity are unclear, but are likely due to a combination of environmental and genetic factors. There are racial differences in genetic, epigenetic, and gene expression patterns in prostate cancer (3–7), as well as differences in metabolizing enzyme activities (8–10) and DNA repair capacities (11)—all of which may influence the carcinogenic potential of chemical exposures between African Americans and Whites.
Polycyclic aromatic hydrocarbons (PAHs)—ubiquitous environmental contaminants that result from incomplete combustion processes—are among the environmental carcinogens that may be involved in prostate carcinogenesis (12). Several retrospective studies report a possible link between occupational exposure to PAHs and the risk of prostate cancer (13,14). The PAH benzo[a]pyrene is listed by the International Agency for Research on Cancer as carcinogenic to humans; recent reports have emerged of a link between high levels of benzo[a]pyrene consumed in grilled meats and increased risk of advanced prostate cancer (15,16). In vitro experiments have detected DNA adducts in human prostate cells after exposure to benzo[a]pyrene (17,18), and shown that exposure levels of benzo[a]pyrene are positively correlated with DNA damage (19). PAHs derive their carcinogenic properties, in part, through their ability to form PAH-DNA adducts (20). We and others have detected PAH-DNA adducts in prostate epithelial cells (21,22). We have also found racial differences in associations of genetic variation in PAH-metabolizing enzymes and cigarette smoke exposure with PAH-DNA adduct levels in the prostate (23), and that African Americans with high levels of PAH-adducts may be at an increased risk for biochemical recurrence compared with Whites (24).
The quantification of PAH exposures from environmental sources is informative but ultimately a poor measure of a biologically effective exposure dose. It does not account for interindividual genetic variation in metabolism or DNA repair enzymes, which can influence adduct formation (25). Cellular and molecular changes are likely to be more relevant to disease outcome than measurement of PAHs in the ambient environment or diet. As such, the detection and quantification of PAH-DNA adducts within the tissue of interest is an important step toward understanding the connection between chemical exposure and cancer development (26). Case-control studies of breast (27), lung (28,29), colon (30), gastrointestinal (31), liver (32), bladder (33,34) and esophageal (35) cancers have found elevated PAH-DNA adduct levels in cases compared with controls; however, most studies have used circulating lymphocytes or other proxy measures of adduct levels. Among studies that measured DNA adducts in the cancer target tissue (27,32,34), adduct levels were measured in these specimens after the development of cancer. Our own previous research on PAH-DNA adduct levels in prostate was performed exclusively using tissue from cases, was cross sectional in nature, and did not include a control group (22–24).
In this study, we advance the molecular epidemiology of DNA adducts and cancer risk by measuring PAH-DNA adduct levels in histopathologically normal tissue samples taken from the target organ before disease onset. We also assess the relationship between adduct levels and subsequent cancer risk using a case–control study nested within a large historical cohort of men with benign prostate tissue specimens. In addition to testing whether adduct levels were associated with incident prostate cancer and tumor aggressiveness, we also explored race-specific cancer associations.
Methods
Study sample and medical record review
After obtaining approval from the Henry Ford Health System Institutional Review Board, we identified a historical cohort of 6692 men with a benign prostate specimen archived in the Henry Ford Department of Pathology collected by needle core biopsy or transurethral resection of the prostate (TURP) between January 1990 and December 2002. Members of the cohort could have had one or more benign prostate specimens collected before January 1990, within the 13-year cohort window, or after December 2002, but the first benign prostate specimen collected between January 1990 and December 2002 was used as the initial date of cohort entry for the purposes of case–control sampling and study analyses. A nested case–control sample was drawn from this cohort based on eligibility criteria that included a recorded prostate-specific antigen (PSA) level within a year of cohort entry and no history of a previous prostate cancer diagnosis. ‘Date of case diagnosis’ was the date of first cancer-positive tissue specimen or the date a clinician first reported a clinical diagnosis of prostate cancer. Patients diagnosed with prostate cancer <1 year from the date of cohort entry were ineligible for the study. We identified 808 potentially eligible cases diagnosed with prostate cancer prior to July 2007.
Incidence density sampling was used to select controls with replacement from all cohort members at risk at the time of case occurrence. Controls were randomly selected from among those cohort members who were free of prostate cancer at a follow-up duration greater than or equal to the time between cohort entry and diagnosis dates of the matched case, with the end of follow-up denoted as the ‘reference date’ for controls. Matching criteria included age at entry into cohort (±2 years), date of entry into cohort (±2 years), race (African American or White), and type of specimen (biopsy or TURP). We were able to match 802 of 808 potentially eligible cases. Further review reduced the final sample to 574 case-control pairs (36); exclusions were primarily due to problems with tissue blocks (n = 126; 55.3%), including lack of analyzable prostate tissue, wrong specimen type or missing specimens. Additional pairs were excluded for lack of a PSA test within 1 year of cohort entry, evidence of malignancy after a second pathologic review of prostate specimens, earlier benign specimens (outside the cohort window) or incomplete records at the time of cohort entry or diagnosis. We were able to analyze the PAH-DNA adduct levels in 536 pairs; absence of sufficient numbers of epithelial cells in the benign tissue specimen was the primary reason for failure to generate adduct data.
Smoking status and clinical, demographic, and comorbidity data were abstracted from patients’ medical records from 5 years before the date of cohort entry through the diagnosis (for prostate cancer cases) or reference date (for controls). All medical data used in study analyses are based on the date of cohort entry unless otherwise noted. A 10% re-abstraction of patient charts was performed to optimize abstraction reliability.
Immunohistochemistry
All benign tissue specimens were evaluated for the presence of cancer, high-grade prostatic intraepithelial neoplasia (HGPIN), atrophy, and inflammation by a single genitourinary pathologist (O.N.K.) blinded to disease progression (36). The immunohistochemical assay for PAH-DNA adducts was carried out as described previously (27,37). This assay uses the monoclonal 5D11 antibody, which in cell culture studies has been shown to produce strongly correlated staining levels (r = 0.99) with the treatment dose of benzo(a)pyrene diol epoxide (38). The detection limit of the immunohistochemical assay we used is about 1–3 adducts per 105 nucleotides (37). The assay is not directly calibrated to a nucleotide count, however, and as such there is no lower bound limit that is considered as a zero adduct count or a non-detect level of staining intensity. Consistent with previous studies, we report our results in optical density (OD) units, a measure of the relative intensity of staining (Figure 1) (27). For each prostate specimen, a laboratory technician scored 50 epithelial cells (5 fields with 10 scored cells per field). Scored cells were selected to be representative, in terms of intensity, of the cells in the field. All case-control pairs were processed, stained and scored together to eliminate possible batch effects. PAH-DNA adduct data were further standardized across experiments using two ‘control’ slides (from two unique prostate specimens) across batches (22).
Fig. 1.
(a and b) Example of PAH-DNA adduct staining of a benign prostate biopsy specimen (a, 10× magnification) and corresponding H&E-stained specimen (b, 10× magnification). Inset in adduct-stained specimen depicts an example of one of five high-power fields (40×) used to count 10 cells for scoring (50 cells counted). Note that adduct staining is concentrated in the nuclear compartment.
Statistical analysis
Conditional logistic regression analyses were used to estimate both unadjusted and adjusted odds ratios (ORs) for prostate cancer incidence during follow-up. Individual matching controlled for age, race, and specimen type (biopsy or TURP). Analyses were performed using PAH-adduct level expressed as both continuous and categorical variables, categorized by their distribution among the control subjects. Both unadjusted and adjusted ORs were estimated. Potential confounders were identified by first testing whether the variable was associated with either PAH-DNA adduct levels or case status; associated variables were next tested in multivariable models to determine whether inclusion of the variable in the model substantively changed the effect estimate for elevated PAH-DNA adducts. Only variables that changed the exposure effect estimate by ±10% were retained in adjusted models. Comparisons between the stratified models were assessed using a conditional logistic regression model that included interaction terms with the stratified variable.
Results
Characteristics of analytic sample
In the analytic sample of 536 case–control pairs, cases were an average of 65.4 years old at cohort entry and African Americans made up 41% of the case–control sample (Table I). Approximately 50% of cases were diagnosed between 1 and 4 years after cohort entry. The remaining cases were diagnosed 4–15 years after cohort entry. Cases had a significantly higher PSA level at the time of cohort entry, averaged two more PSA tests within this same time period and had significantly more additional prostate biopsies or TURPs between the date of cohort entry and diagnosis. The majority of cases presented with stage 1 (39.5%) and stage 2 (52.1%) tumors. Twenty-seven percent of cases had advanced tumor grade defined as either Gleason score 8 and above or Gleason score 7 with a primary grade 4. Mean PAH-DNA adduct levels were slightly elevated in cases compared with controls, but the difference was not statistically significant (0.299±0.046 OD versus 0.298±0.045 OD, P = 0.68). In general, the analytic sample had similar baseline characteristics compared with the 38 pairs that were dropped due to lack of analyzable adduct data. The exception to this was that the date of cohort entry and mean age at cohort entry were significantly older in the 38 unanalyzable pairs compared with the analytic sample.
Table I.
Demographic and clinical characteristics of study sample (n = 536 case–control pairs)
| Variable | Controls | Cases | P value | |
|---|---|---|---|---|
| Race | White | 320 (59.7%) | 320 (59.7%) | a |
| African American | 216 (40.3%) | 216 (40.3%) | ||
| Age at cohort entry | 65.4±7.5 | 65.5±7.5 | a | |
| Median date of cohort entry | 03/15/1995 | 3/15/1995 | a | |
| Type of benign specimen | Biopsy | 504 (94.0%) | 504 (94.0%) | a |
| TURP | 32 (6.0%) | 32 (6.0%) | ||
| Median fu Time (Yrs) | 10.48 | 3.96 | <0.0001*** | |
| PSA level at cohort entry (ng/ml) | 5.7±5.4 | 7.7±7.3 | <0.0001*** | |
| Mean number of PSA tests between cohort entry and diagnosis | 6.8±5.1 | 8.7±5.3 | <0.001*** | |
| Mean number of benign prostate specimens procured between cohort entry and diagnosis | 0.5±1.0 | 1.9±1.4 | <0.0001*** | |
| Mean PAH-DNA adduct levels (OD)b | 0.298±0.045 | 0.299±0.046 | 0.68 | |
| Tumor stagec | 1 | 212 (39.5%) | — | |
| 2 | 279 (52.1%) | |||
| 3 | 39 (7.3%) | |||
| 4 | 5 (0.9%) | |||
| Gleason graded | <=6 | 239 (44.5%) | — | |
| 7 (3+4) | 111 (20.7%) | |||
| 7 (4+3) | 48 (9.0%) | |||
| 8–10 | 96 (17.9%) | |||
a Matching factors.
b OD units.
c One sample was missing stage data.
d Six samples were missing Gleason grade data; 35 had tumor foci too small to grade.
Asterisks represent statistical significance of P values.
Risk associated with higher levels of PAH-DNA adducts
Although mean PAH-DNA adduct levels did not differ significantly between cases and controls, previous studies have shown that the effect of DNA adducts on cancer-related outcomes tends to be non-linear (39,40). Therefore, to determine whether prostate cancer risk was associated with elevated PAH-DNA adduct levels, we tested two models, one in which adduct levels were categorized into quartiles, and another in which levels were dichotomized above and below the median (Table II). In the quartiles model, risk of prostate cancer was increased 21% for the third quartile and 27% for the fourth quartile. Because elevated risk was primarily observed for quartiles 3 and 4, with no elevation in risk found for quartile 2, adduct levels were dichotomized at the median for subsequent analyses. When subjects were classified as having higher or lower adduct levels, African American men with elevated adduct levels had a marginally significant 56% increased risk (OR = 1.56, CI = 1.00–2.44, P = 0.05). No trend in risk with increasing PAH-DNA adduct levels was observed for White men nor was there any noticeable increased risk associated with DNA adduct levels greater than median levels.
Table II.
Association of PAH-DNA adduct levels with prostate cancer
| Sample | Crude modela | Adjusted modelb | ||||
|---|---|---|---|---|---|---|
| OR | 95% CI | P value | OR | 95% CI | P value | |
| Adduct level | ||||||
| Whole sample (n = 536 pairs) | ||||||
| Second quartile | 0.94 | 0.64–1.37 | 0.74 | 0.93 | 0.62–1.40 | 0.73 |
| Third quartile | 1.21 | 0.84–1.74 | 0.31 | 1.09 | 0.74–1.62 | 0.66 |
| Fourth quartile | 1.27 | 0.86–1.86 | 0.23 | 1.17 | 0.78–1.76 | 0.45 |
| High adduct level | 1.28 | 0.98–1.67 | 0.07 | 1.17 | 0.88–1.56 | 0.27 |
| African Americans (n = 216 pairs) | ||||||
| Second quartile | 1.11 | 0.60–2.06 | 0.74 | 0.91 | 0.45–1.81 | 0.78 |
| Third quartile | 2.31 | 1.18–4.53 | 0.015* | 2.10 | 1.01–4.39 | 0.048 |
| Fourth quartile | 1.32 | 0.71–2.45 | 0.37 | 1.14 | 0.59–2.21 | 0.70 |
| High adduct level | 1.56 | 1.00–2.44 | 0.05* | 1.51 | 0.93–2.44 | 0.10 |
| Whites (n = 320 pairs) | ||||||
| Second quartile | 0.85 | 0.52–1.37 | 0.49 | 0.92 | 0.55–1.53 | 0.75 |
| Third quartile | 0.92 | 0.58–1.44 | 0.70 | 0.85 | 0.53–1.36 | 0.50 |
| Fourth quartile | 1.25 | 0.76–2.07 | 0.37 | 1.21 | 0.72–2.05 | 0.47 |
| High adduct level | 1.14 | 0.82–1.59 | 0.45 | 1.03 | 0.72–1.46 | 0.88 |
aCases and controls were one-to-one matched on age (±2 years), race (African American or White), date of cohort entry (±2 years) and specimen type (biopsy or TURP).
bAdjusted for number of PSA tests and inflammation status.
Asterisks represent statistical significance of P values.
We next examined potential confounders of the PAH-DNA adducts-prostate cancer relationship by analyzing histopathological factors previously found to be associated with prostate cancer in this study population (36). Based on results shown in Table I, PSA level at cohort entry and number of PSA tests during follow-up were also analyzed. Men without inflammation in their prostate specimen had slightly higher PAH-DNA adduct levels compared with those with inflammation (0.302±0.044 OD versus 0.296±0.047 OD, P = 0.04). Number of PSA tests were also positively correlated with PAH-DNA adduct levels; men in the lowest adduct quartile had 7.2±4.8 tests during follow-up, compared with 8.3±6.3 tests for men in the highest adduct quartile (P = 0.08). We therefore reanalyzed the quartile and median adduct models, adjusting for these two factors (Table II), and found that the ORs for elevated PAH-DNA adduct levels decreased from 1.28 in the full sample to 1.17 in the adjusted model. Likewise, the ORs for elevated PAH-DNA adduct levels decreased from 1.56 to 1.51 in African Americans and from 1.14 to 1.03 in Whites in the stratified adjusted models. Because these changes were nominal, subsequent models did not include any additional covariates for confounders.
Although we found that risk estimates for elevated PAH-DNA adduct levels varied by race, we observed no differences in PAH-DNA adduct levels by race for either cases or controls. PSA levels at cohort entry were slightly higher for African American compared with Whites cases (8.4ng/ml versus 7.4ng/ml; P = 0.02) as well as controls (6.4ng/ml versus 5.3ng/ml; P = 0.06). At time of diagnosis, PSA levels remained higher for African American compared with Whites cases (33.7ng/ml versus 18.2ng/ml; P = 0.02). African American cases and controls were also more likely to be cigarette smokers at time of cohort entry compared with their White counterparts (17.8% versus 12.2%; P = 0.07 for cases and 21.8%% versus 13.5%; P = 0.01 for controls). None of the measured histologic factors or the other potential confounders we analyzed varied by race.
Table III reports risk estimates stratified by selected matching factors (including race) and tumor grade (for cases). The most striking risk difference by strata was observed for the case-control pairs stratified by time of follow-up. In the full sample, elevated PAH-DNA adduct levels were significantly associated with a 60% increased risk of prostate cancer for those cases diagnosed 1–4 years after cohort entry (OR = 1.60, CI = 1.07–2.41). In race-stratified analyses, this increased risk was found solely in African Americans—men with elevated adduct levels were at 4–5 times increased risk for prostate cancer within the first 4 years of follow-up (OR = 4.71, CI = 1.97–11.26). African American men under age 65 with elevated adduct levels were at a slightly greater risk for prostate cancer than their older counterparts (OR = 1.71 versus 1.21), but neither risk estimate was statistically significant. The only other statistically significant association was in African Americans who entered the cohort after March 1995 (OR = 1.80, CI = 1.00–3.24). ORs for White men did not vary significantly across strata.
Table III.
Associations between high PAH-DNA adduct levels and prostate cancer stratified by matching factors and case characteristicsa
| Sample | Whole sample | African Americans | Whites | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Stratum | OR | 95% CI | P value | OR | 95% CI | P value | OR | 95% CI | P value |
| Low tumor grade | n = 386 pairs | n = 154 pairs | n = 232 pairs | ||||||
| 1.17 | 0.84–1.63 | 0.35 | 1.65 | 0.93–2.92 | 0.09 | 0.99 | 0.65–1.49 | 0.95 | |
| High tumor grade | n = 144 pairs | n = 62 pairs | n = 82 pairs | ||||||
| 1.31 | 0.73–2.37 | 0.36 | 1.60 | 0.54–4.73 | 0.40 | 1.22 | 0.60–2.49 | 0.58 | |
| 1–4 years of follow-up | n = 270 pairs | n = 106 pairs | n = 164 pairs | ||||||
| 1.60 | 1.07–2.41 | 0.02* | 4.71 | 1.97–11.26 | 0.0005*** | 1.05 | 0.64–1.71 | 0.85 | |
| 4–15 years of follow-up | n = 267 pairs | n = 111 pairs | n = 156 pairs | ||||||
| 0.85 | 0.56–1.27 | 0.42 | 0.63 | 0.32–1.25 | 0.19 | 1.02 | 0.60–1.71 | 0.96 | |
| Early cohort entry | n = 264 pairs | n = 88 pairs | n = 176 pairs | ||||||
| 1.08 | 0.70–1.66 | 0.74 | 0.93 | 0.38–2.30 | 0.88 | 1.12 | 0.68–1.84 | 0.65 | |
| Late cohort entry | n = 273 pairs | n = 129 pairs | n = 144 pairs | ||||||
| 1.22 | 0.83–1.77 | 0.31 | 1.80 | 1.00–3.24 | 0.048 | 0.89 | 0.53–1.47 | 0.64 | |
| Age <65 | n = 273 pairs | n = 111 pairs | n = 162 pairs | ||||||
| 1.29 | 0.88–1.90 | 0.19 | 1.71 | 0.92–3.19 | 0.09 | 1.10 | 0.67–1.79 | 0.72 | |
| Age 65+ | n = 264 pairs | n = 106 pairs | n = 158 pairs | ||||||
| 1.02 | 0.67–1.55 | 0.94 | 1.21 | 0.54–2.71 | 0.65 | 0.97 | 0.59–1.60 | 0.90 | |
aCases and controls were one-to-one matched on age (±2 years), race (African American or White), date of cohort entry (±2 years) and specimen type (biopsy or TURP).
Effect modifiers of prostate cancer risk associated with high PAH-DNA adduct levels
We next tested whether clinical and histopathological factors associated with prostate cancer modified the relationship between high PAH-DNA adduct levels and prostate cancer (Table IV). Because we have previously found current smoking to be associated with PAH-DNA adduct levels in human prostate (23), we also tested current smoking as a potential effect modifier. Overall, with the exception of current smoking in Whites, none of these variables significantly modified the risk associated with elevated adduct levels and prostate cancer, although some interesting race-specific trends emerged. PSA level at cohort entry and HGPIN are both associated with prostate cancer in this study population (36), and both appeared to modestly enhance the association between elevated adduct levels and prostate cancer in African Americans. Although each of these factors in isolation increased risk 50–100%, the combination of either high PSA or presence of HGPIN with elevated adduct levels resulted in ORs for prostate cancer risk over 4. This degree of effect modification was not observed in Whites.
Table IV.
Effect modification of PAH-DNA adduct level associations with prostate cancera
| Variable | Whole sample (n = 536 pairs) | African Americans (n = 216 pairs) | Whites (n = 320 pairs) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| OR | 95% CI | P value | OR | 95% CI | P value | OR | 95% CI | P value | |||
| PSA at cohort entry | |||||||||||
| PSA < 4ng ml/low adducts | 1 | 1 | 1 | ||||||||
| PSA < 4ng ml/high adducts | 1.17 | 0.68–2.02 | 0.57 | 1.18 | 0.46–3.00 | 0.73 | 1.20 | 0.61–2.34 | 0.60 | ||
| PSA ≥ 4ng ml/low adducts | 2.51 | 1.59–3.96 | <0.0001*** | 2.44 | 1.15–5.20 | 0.02* | 2.59 | 1.46–4.58 | 0.001** | ||
| PSA ≥ 4ng ml/high adducts | 3.62 | 2.24–5.86 | <0.0001*** | 4.39 | 1.95–9.88 | 0.0004** | 3.19 | 1.75–5.81 | 0.0001*** | ||
| PSA≥4ng ml × high adducts interaction | 1.23 | 0.68–2.24 | 0.49 | 1.53 | 0.57–4.08 | 0.40 | 1.03 | 0.48–2.20 | 0.94 | ||
| NO inflammation | |||||||||||
| Inflammation/low adducts | 1 | 1 | 1 | ||||||||
| Inflammation/high adducts | 1.27 | 0.89–1.80 | 0.19 | 1.82 | 1.02–3.24 | 0.04* | 1.00 | 0.64–1.56 | 0.99 | ||
| No inflammation/low adducts | 1.53 | 1.05–2.22 | 0.025* | 2.11 | 1.16–3.83 | 0.015 | 1.22 | 0.76–1.98 | 0.41 | ||
| No inflammation/high adducts | 1.94 | 1.33–2.81 | 0.0005** | 2.59 | 1.39–4.82 | 0.003** | 1.59 | 0.99–2.55 | 0.05* | ||
| No Inflammation × high adducts interaction | 1.00 | 0.60–1.67 | 1.00 | 0.67 | 0.29 – 1.57 | 0.36 | 1.31 | 0.68–2.50 | 0.42 | ||
| HGPIN | |||||||||||
| No HGPIN/low adducts | 1 | 1 | 1 | ||||||||
| No HGPIN/high adducts | 1.26 | 0.95–1.66 | 0.11 | 1.50 | 0.94–2.41 | 0.09 | 1.14 | 0.81–1.61 | 0.45 | ||
| HGPIN/low adducts | 1.63 | 0.83–3.22 | 0.16 | 1.43 | 0.53–3.83 | 0.48 | 1.95 | 0.75–5.10 | 0.17 | ||
| HGPIN/high adducts | 3.38 | 0.65–6.92 | 0.0009** | 4.25 | 1.32–13.68 | 0.02* | 2.90 | 1.17–7.20 | 0.02* | ||
| HGPIN × high adducts interaction | 1.65 | 0.64–4.26 | 0.30 | 1.98 | 0.46–8.58 | 0.36 | 1.30 | 0.36–4.71 | 0.69 | ||
| Simple atrophy | |||||||||||
| No atrophy/low adducts | |||||||||||
| No atrophy/high adducts | 1.43 | 0.89–2.31 | 0.14 | 1.20 | 0.51–2.69 | 0.65 | 1.69 | 0.92–3.09 | 0.09 | ||
| Atrophy/low adducts | 1.10 | 0.72–1.68 | 0.65 | 1.04 | 0.50–2.15 | 0.92 | 1.15 | 0.69–1.93 | 0.59 | ||
| Atrophy/high adducts | 1.34 | 0.88–2.05 | 0.17 | 1.83 | 0.88–3.80 | 0.11 | 1.10 | 0.66–1.85 | 0.72 | ||
| Atrophy × high adducts interaction | 0.85 | 0.48–1.49 | 0.57 | 1.47 | 0.58–3.72 | 0.42 | 0.56 | 0.27–1.16 | 0.12 | ||
| Current smoker | |||||||||||
| Non-smoker/low adducts | 1 | 1 | 1 | ||||||||
| Non-smoker/high adducts | 1.27 | 0.94–1.71 | 0.12 | 1.53 | 0.92–2.53 | 0.10 | 1.16 | 0.80–1.68 | 0.45 | ||
| Current smoker/low adducts | 1.11 | 0.65–1.88 | 0.71 | 0.91 | 0.42–1.97 | 0.82 | 1.45 | 0.68–3.11 | 0.34 | ||
| Current smoker/high adducts | 0.87 | 0.52–1.47 | 0.61 | 1.38 | 0.64–2.98 | 0.41 | 0.58 | 0.27–1.23 | 0.15 | ||
| Current smoker × high adducts interaction | 0.62 | 0.30–1.30 | 0.20 | 0.99 | 0.34–2.87 | 0.99 | 0.34 | 0.12–1.00 | 0.05* | ||
aCases and controls were one-to-one matched on age (±2 years), race (African American or White), date of cohort entry (± 2 years) and specimen type (biopsy or TURP).
Asterisks represent statistical significance of P values.
In the full sample, neither absence of inflammation nor presence of simple atrophy in the benign specimen modified the association between elevated PAH-DNA adduct levels and prostate cancer risk; however, when analyses were stratified by race, we observed interaction ORs suggesting both negative interactions (absence of inflammation in African Americans and atrophy in Whites) and positive interactions (atrophy in African Americans and absence of inflammation in Whites). The strongest (and only significant) interaction was observed for current cigarette smoking in Whites. Paradoxically, although elevated adduct levels and current smoking each modestly increased prostate cancer risk, when combined the two were associated with a decreased prostate cancer risk (OR = 0.34, CI = 0.12–1.00).
Inflammation and simple atrophy and PAH-DNA adduct levels
Given our previous finding of significant decreased and increased prostate cancer risk associated with inflammation and simple atrophy, respectively (36), and their opposite effect on risk associated with elevated adduct levels by race, we examined the association of these two factors with PAH-DNA adduct levels in the full sample. In a regression model, simple atrophy (β = 0.0062, P = 0.08) and inflammation (β = –0.0075, P = 0.08) were the two prostate cancer risk factors most strongly associated with PAH-DNA adduct levels. These associations, however, were strikingly different by race. In Whites, the association between simple atrophy and PAH-DNA adduct levels was greater (β = 0.0075, P = 0.11 versus β = 0.0033, P = 0.55). Conversely, in African Americans, inflammation had a much stronger association with PAH-DNA adduct levels (β = –0.0140, **P = 0.007 versus β = –0.0035, P = 0.42). When PAH-DNA adduct levels were stratified by these two factors, subjects with inflammation but no simple atrophy had significantly lower PAH-DNA adducts compared with two of the other three strata (Figure 2a): simple atrophy/no inflammation (0.287±0.056 OD versus 0.302±0.044 OD; *P = 0.02) and no simple atrophy/no inflammation (0.287±0.056 OD versus 0.301±0.043 OD; *P = 0.05). These same differences were observed to an even greater degree in the African American subset (Figure 2b): no simple atrophy/inflammation and simple atrophy/no inflammation (0.284±0.055 OD versus 0.311±0.035 OD, **P = 0.005), and no simple atrophy/inflammation and no simple atrophy/no inflammation (0.284±0.055 OD versus 0.305±0.050 OD; *P = 0.04). In Whites (Figure 2c), the mean adduct levels across all four simple atrophy/inflammation strata were comparable.
Fig. 2.
(a–c) Distribution of PAH-DNA adduct levels (in OD units) by simple atrophy and inflammation status in the full sample (a, n = 916), African Americans (b, n = 359) and Whites (c, n = 557).
Discussion
We report for the first time a prospective analysis of PAH-DNA adduct levels—a marker of biologically effective exposure to PAH—measured in histopathologically benign target tissue, and subsequent cancer risk for the same organ. Prior prospective studies of adduct levels and cancer risk have used surrogate tissues, such as white blood cells, but the correlation between adduct levels in surrogate tissues and the target organ is questionable (41–43). In this study, we find that higher levels of PAH-DNA adducts in benign prostate specimens were associated with a modest, non-significant increased risk for prostate cancer. This risk, however, was greater in African Americans and greater still for African American cases diagnosed within 4 years from the time of tissue collection. African Americans with high PAH-DNA adduct levels had a 4- to 5-fold increased risk of prostate cancer within the first 4 years of follow-up. In Whites, we observed only nominally increased risk associated with the presence of high PAH-DNA adduct levels, irrespective of the length of follow-up.
These temporal and race-specific findings are consistent with our earlier report of an association between PAH-DNA adducts levels measured in radical prostatectomy specimens and higher risk of biochemical recurrence (24). In that study, we found elevated PAH-DNA levels to be associated with increased risk of recurrence with within 2 years of surgery; the effect of higher adduct levels on recurrence was stronger in African Americans than in Whites.
The reasons behind the large disparity between African American and White men in prostate cancer incidence and mortality are likely complex, involving a combination of biological and sociocultural factors, and have been reviewed extensively (44,45). Levels of carcinogen- DNA adducts are a marker of biologically effective exposure that may portend increased cancer risk (39). Why this marker associates only with increased risk in African American men is unclear, but the answer may involve inherent racial differences in the biology of prostate carcinogenesis.
Few studies have examined racial differences in PAH-DNA adduct levels, but a study of smokers found that African American subjects had higher adduct levels in lymphocytes than other subjects after adjustment for gender, education, alpha-tocopherol and beta-carotene levels, and GSTM1 status (46). We and others have shown racial differences in genetic polymorphisms that influence PAH-DNA adduct levels (23,47). In addition, cigarette smoke exposure, a primary source of PAHs, is known to vary by race (48); African Americans may also metabolize PAH compounds from cigarette smoke exposure in a different dose-response manner than Whites (49). In benign prostate tissue (as measured in this study) as well as in tumor and tumor-adjacent benign tissue (as we reported previously) (22), PAH-DNA adduct levels do not vary by race. Yet, prostate cancer risk at the highest adduct levels was greater for African American men compared with White men.
One possible explanation for our race-specific findings may be variation in DNA repair capacity by race. A study by Trzeciak et al. measured age-, race-, and sex-specific single-strand DNA repair capacities using a novel comet assay protocol; they found that in African American men, DNA repair capacity decreases with age more than in White men (11). This may explain the effect-modification ORs shown in Table IV; in African Americans, in the presence of each factor thought to increase prostate cancer risk—high PSA, absence of inflammation, simple atrophy and HGPIN—the risk associated with high adduct levels increased as well. In addition, inflammation was inversely associated with PAH-DNA adduct levels exclusively in African American men. Whether this finding was coincidental or due to these two factors residing in a common prostate carcinogenesis pathway is unclear. Although inflammation is thought to occur in concert with environmental insults that lead to DNA damage and ultimately carcinogenesis (50), we have previously reported that absence of inflammation in the benign prostate is a risk factor for subsequent prostate cancer, using this same study sample (36). This absence of an association between simple atrophy and PAH-DNA adduct levels in African Americans suggests that the previously proposed (51) carcinogenic inflammation-simple atrophy DNA damage cascade is unlikely. However, given that our study only provides a cross-sectional view of the benign prostate tissue environment, it is impossible to definitively infer what—if any—biological connection PAH-DNA adducts may have with inflammation, in terms of prostate carcinogenesis.
Our study is unique among studies of PAH-DNA adducts and cancer risk. Previous studies have generally used the case-control approach to estimate risk associated with higher PAH-DNA adduct levels, based on pre-disease adduct measures in blood (28,29). Of the few studies measuring adduct levels in tissue from the cancerous organ (27,32,34), adduct levels in tumor cells tends to be more strongly associated with risk (27,32). However, because specimens were acquired at the time of cancer diagnosis, such studies lack the ability to make a temporal link between adduct levels and the onset of carcinogenesis. Ours is the first study to prospectively estimate the risk associated with PAH-DNA adduct levels in the target tissue prior to cancer diagnosis. Furthermore, the size of our study allows risk estimates to be stratified by factors—such as race and age—that may modify the cancer risk associated with high adduct levels. Race-stratified results need to be interpreted with caution, however, given the smaller sample sizes. Although we cannot rule out some degree of misclassification bias due to undiagnosed cancer at the time of cohort entry, the adduct measures in our study should have, in most instances, preceded onset of invasive carcinoma. In addition, because adducts were measured in histopathologically confirmed benign tissue, our results reflect the potential of this biomarker for predicting onset of future cancer in the benign biopsy setting.
Because our study was limited to one DNA adduct measurement at a single time point, our analysis was confined to a snapshot of what is likely a rapidly changing cellular environment. This may explain, in part, why increased risk associated with elevated PAH-DNA adducts was not observed for cases that developed more than 4 years after cohort entry. Our study was observational, and although the duration of at-risk follow-up was equal for cases and controls, cohort members differed in their medical follow-up and screening behavior. Cases had significantly more PSA tests between cohort entry and diagnosis than controls, and the frequency of PSA tests in our study sample was greater than current screening recommendations, even in controls. Although there is no a priori reason why screening behavior should differ by adduct levels and, indeed, adjustment for number of PSA tests during follow-up did not substantively change our results, the increased frequency of PSA tests in our study population reflects the high risk nature of this cohort. During the 15-year period cases accrued in the study cohort, the standardized prostate cancer incidence was 1.9 times higher for African American men and 1.8 times higher for White men compared with the metropolitan Detroit population from which the cohort was drawn.
To the extent that formation of PAH-DNA adducts are a risk factor for prostate cancer and thus potentially aligned with clinical risk indicators leading to a biopsy referral, we expect the design of the study to bias the results to the null as compared with results that might be expected from a general population cohort; however, construct validity of measuring PAH-DNA adducts in non-tumor target tissue collected prior to diagnosis should outweigh such possible biases to the null. In addition, this cohort of men (collected within a single geographic area across broad demographic categories, including a large proportion of African Americans) is of particular public health and clinical importance; the identification of biomarkers that predict risk of future prostate cancer diagnosis is an important tool for clinical decision making regarding the appropriate intensity of screening and biopsy in this and similar populations.
As a retrospective cohort embedded within a single health system, our study design allowed incident case detection and permitted an investigation of the temporal association of adduct levels with subsequent cancer, providing better understanding of the relationship between a tissue-based marker of DNA damage and subsequent prostate cancer risk. Although the immunohistochemistry methods used were not as specific or sensitive as mass spectrometry and provide a semiquantitative measure of DNA adduct concentration that cannot be directly translated into an adducts per nucleotide result (52), the polyclonal antibody we used has been extensively validated for tissue-based quantification of relative DNA adduct levels (37). Although a correlation has not been shown between immunohistochemical quantitation of PAH-DNA adducts and cigarette smoking in breast (27) and liver (32) tissues, such a correlation has been found in work using lung tissue (53,54) and our own work using prostate (23,55) tissue. Finally, it should be noted that our study was not designed to investigate which routes of PAH exposure lead to higher adduct levels, nor how biologic modifiers, such as inherited capacities for high metabolism of PAH (23,56) or poor DNA repair capacity (57), might influence the effects of those levels. Notably, although these risk factors may explain the underlying reasons for interindividual variation in PAH-DNA adduct levels, as antecedent factors in a causal pathway, the absence of data on risk factors for adduct formation does not undermine the ability of the study to assess the role of adducts in prostate cancer risk (58).
In summary, the risk of prostate cancer conferred by elevated levels of PAH-DNA adducts in pre-malignant benign prostate tissue appears to be race specific and only for a limited time period. However, the short-term increased prostate cancer risk associated with high adducts in African Americans is significantly elevated, and may reflect a carcinogenic process that is underway, but not yet histologically detectable. This theory is supported by the suggestive evidence of a synergistic effect between high PAH-DNA adducts and established risk factors for prostate cancer—such as presence of HGPIN and elevated PSA levels—and risk of prostate cancer in African Americans. Our results suggest that further exploration of racial differences in DNA repair capacities may be warranted, as such differences might provide an explanation of why elevated PAH-DNA adducts confer an increased prostate cancer risk only in African Americans. Such explorations may ultimately lead to a better understanding of both prostate carcinogenesis and racial disparities in prostate cancer incidence and progression.
Funding
National Institutes of Health (5R01-ES011126).
Acknowledgements
The authors wish to thank the medical record abstractors and other study personnel that helped with data collection for this study. The authors especially thank Travis Wheeler and Nancy Lemke, who processed all prostate specimens used in this study.
Conflict of Interest Statement: None declared.
Glossary
Abbreviations:
- CI
confidence interval
- HGPIN
high-grade prostatic intraepithelial neoplasia
- OD
optical density
- OR
odds ratio
- PAH
polycyclic aromatic hydrocarbon
- PSA
prostate-specific antigen
- TURP
transurethral resection of the prostate.
References
- 1.Jemal A., et al. Declining death rates reflect progress against cancer. PLoS ONE. (2010);5:e9584. doi: 10.1371/journal.pone.0009584. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Jemal A., et al. (2010). Cancer statistics, 2010. CA. Cancer J. Clin. 60 277–300 [DOI] [PubMed] [Google Scholar]
- 3. Haiman C.A., et al. (2011). Genome-wide association study of prostate cancer in men of African ancestry identifies a susceptibility locus at 17q21. Nat. Genet. 43 570–573 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Kwabi-Addo B., et al. (2010). Identification of differentially methylated genes in normal prostate tissues from African American and Caucasian men. Clin. Cancer Res. 16 3539–3547 [DOI] [PubMed] [Google Scholar]
- 5. Powell I.J., et al. (2010). Evidence supports a faster growth rate and/or earlier transformation to clinically significant prostate cancer in black than in white American men, and influences racial progression and mortality disparity. J. Urol. 183 1792–1796 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Hatcher D., et al. (2009). Molecular mechanisms involving prostate cancer racial disparity. Am. J. Transl. Res. 1 235–248 [PMC free article] [PubMed] [Google Scholar]
- 7. Freedland S.J., et al. (2005). Explaining racial differences in prostate cancer in the United States: sociology or biology? Prostate 62 243–252 [DOI] [PubMed] [Google Scholar]
- 8. Yu M.C., et al. (1995). Glutathione S-transferase M1 genotype affects aminobiphenyl-hemoglobin adduct levels in white, black and Asian smokers and nonsmokers. Cancer Epidemiol. Biomarkers Prev. 4 861–864 [PubMed] [Google Scholar]
- 9. Relling M.V., et al. (1992). Racial and gender differences in N-acetyltransferase, xanthine oxidase, and CYP1A2 activities. Clin. Pharmacol. Ther. 52 643–658 [DOI] [PubMed] [Google Scholar]
- 10. Muscat J.E., et al. (2008). Comparison of CYP1A2 and NAT2 phenotypes between black and white smokers. Biochem. Pharmacol. 76 929–937 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Trzeciak A.R., et al. (2008). Age, sex, and race influence single-strand break repair capacity in a human population. Free Radic. Biol. Med. 45 1631–1641 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Yang SK, Silverman BD. Polycyclic Aromatic Hydrocarbon Carcinogenesis: Structure-activity Relationships. CRC Press; Boca Raton: (1988). [Google Scholar]
- 13. Aronson K.J., et al. (1996). Occupational risk factors for prostate cancer: results from a case-control study in Montréal, Québec, Canada. Am. J. Epidemiol. 143 363–373 [DOI] [PubMed] [Google Scholar]
- 14. Brown D.A., et al. (2000). Motor vehicle manufacturing and prostate cancer. Am. J. Ind. Med. 38 59–70 [DOI] [PubMed] [Google Scholar]
- 15. Sinha R., et al. (2009). Meat and meat-related compounds and risk of prostate cancer in a large prospective cohort study in the United States. Am. J. Epidemiol. 170 1165–1177 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Punnen S., et al. Impact of meat consumption, preparation, and mutagens on aggressive prostate cancer. PLoS ONE. (2011);6:e27711. doi: 10.1371/journal.pone.0027711. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Wang C.Y., et al. (1999). N-Acetyltransferase expression and DNA binding of N-hydroxyheterocyclic amines in human prostate epithelium. Carcinogenesis 20 1591–1595 [DOI] [PubMed] [Google Scholar]
- 18. Williams J.A., et al. (2000). Metabolic activation of carcinogens and expression of various cytochromes P450 in human prostate tissue. Carcinogenesis 21 1683–1689 [DOI] [PubMed] [Google Scholar]
- 19. Kooiman G.G., et al. (2000). The influence of dietary and environmental factors on prostate cancer risk. Prostate Cancer Prostatic Dis. 3 256–258 [DOI] [PubMed] [Google Scholar]
- 20. Miller E.C., et al. (1981). Searches for ultimate chemical carcinogens and their reactions with cellular macromolecules. Cancer 47 2327–2345 [DOI] [PubMed] [Google Scholar]
- 21. Pratt M.M., et al. (2011). Polycyclic aromatic hydrocarbon (PAH) exposure and DNA adduct semi-quantitation in archived human tissues. Int. J. Environ. Res. Public Health 8 2675–2691 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Rybicki B.A., et al. (2004). Polycyclic aromatic hydrocarbon-DNA adducts in prostate cancer. Cancer Res. 64 8854–8859 [DOI] [PubMed] [Google Scholar]
- 23. Nock N.L., et al. (2007). Associations between smoking, polymorphisms in polycyclic aromatic hydrocarbon (PAH) metabolism and conjugation genes and PAH-DNA adducts in prostate tumors differ by race. Cancer Epidemiol. Biomarkers Prev. 16 1236–1245 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Rybicki B.A., et al. (2008). Polycyclic aromatic hydrocarbon–DNA adducts in prostate and biochemical recurrence after prostatectomy. Clin. Cancer Res. 14 750–757 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Ketelslegers H.B., et al. (2006). Interindividual variations in DNA adduct levels assessed by analysis of multiple genetic polymorphisms in smokers. Cancer Epidemiol. Biomarkers Prev. 15 624–629 [DOI] [PubMed] [Google Scholar]
- 26. Pratt M.M., et al. (2007). The importance of carcinogen dose in chemoprevention studies: quantitative interrelationships between, dibenzo[a,l]pyrene dose, chlorophyllin dose, target organ DNA adduct biomarkers and final tumor outcome. Carcinogenesis 28 611–624 [DOI] [PubMed] [Google Scholar]
- 27. Rundle A., et al. (2000). The relationship between genetic damage from polycyclic aromatic hydrocarbons in breast tissue and breast cancer. Carcinogenesis 21 1281–1289 [PubMed] [Google Scholar]
- 28. Peluso M., et al. (2005). DNA adducts and lung cancer risk: a prospective study. Cancer Res. 65 8042–8048 [DOI] [PubMed] [Google Scholar]
- 29. Tang D., et al. (2001). Association between carcinogen-DNA adducts in white blood cells and lung cancer risk in the physicians health study. Cancer Res. 61 6708–6712 [PubMed] [Google Scholar]
- 30. Gunter M.J., et al. (2007). Leukocyte polycyclic aromatic hydrocarbon-DNA adduct formation and colorectal adenoma. Carcinogenesis 28 1426–1429 [DOI] [PubMed] [Google Scholar]
- 31. Agudo A., et al. (2012). Aromatic DNA adducts and risk of gastrointestinal cancers: a case-cohort study within the EPIC-Spain. Cancer Epidemiol. Biomarkers Prev. 21 685–692 [DOI] [PubMed] [Google Scholar]
- 32. Chen S.Y., et al. (2002). Polycyclic aromatic hydrocarbon-DNA adducts in liver tissues of hepatocellular carcinoma patients and controls. Int. J. Cancer 99 14–21 [DOI] [PubMed] [Google Scholar]
- 33. Peluso M., et al. (1998). White blood cell DNA adducts, smoking, and NAT2 and GSTM1 genotypes in bladder cancer: a case-control study. Cancer Epidemiol. Biomarkers Prev. 7 341–346 [PubMed] [Google Scholar]
- 34. Benhamou S., et al. (2003). DNA adducts in normal bladder tissue and bladder cancer risk. Mutagenesis 18 445–448 [DOI] [PubMed] [Google Scholar]
- 35. Marjani H.A., et al. (2010). Investigation of NQO1 genetic polymorphism, NQO1 gene expression and PAH-DNA adducts in ESCC. A case-control study from Iran. Genet. Mol. Res. 9 239–249 [DOI] [PubMed] [Google Scholar]
- 36. Kryvenko O.N., et al. (2012). Inflammation and preneoplastic lesions in benign prostate as risk factors for prostate cancer. Mod. Pathol. 25 1023–1032 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Zhang Y.J., et al. (1995). Immunoperoxidase detection of polycyclic aromatic hydrocarbon-DNA adducts in oral mucosa cells of smokers and nonsmokers. Cancer Epidemiol. Biomarkers Prev. 4 133–138 [PubMed] [Google Scholar]
- 38. Zenzes M.T., et al. (1998). Immunodetection of benzo[a]pyrene adducts in ovarian cells of women exposed to cigarette smoke. Mol. Hum. Reprod. 4 159–165 [DOI] [PubMed] [Google Scholar]
- 39. Kriek E., et al. (1998). Polycyclic aromatic hydrocarbon-DNA adducts in humans: relevance as biomarkers for exposure and cancer risk. Mutat. Res. 400 215–231 [DOI] [PubMed] [Google Scholar]
- 40. Gammon M.D., et al. (2002). Environmental toxins and breast cancer on Long Island. I. Polycyclic aromatic hydrocarbon DNA adducts. Cancer Epidemiol. Biomarkers Prev. 11 677–685 [PubMed] [Google Scholar]
- 41. van Schooten F.J., et al. (1992). Polycyclic aromatic hydrocarbon—DNA adducts in white blood cells from lung cancer patients: no correlation with adduct levels in lung. Carcinogenesis 13 987–993 [DOI] [PubMed] [Google Scholar]
- 42. Wiencke J.K., et al. (1995). Correlation of DNA adducts in blood mononuclear cells with tobacco carcinogen-induced damage in human lung. Cancer Res. 55 4910–4914 [PubMed] [Google Scholar]
- 43. Nesnow S., et al. (1993). Quantitative and temporal relationships between DNA adduct formation in target and surrogate tissues: implications for biomonitoring. Environ. Health Perspect. 101 (suppl. 3) 37–42 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Powell I.J. (2007). Epidemiology and pathophysiology of prostate cancer in African-American men. J. Urol. 177 444–449 [DOI] [PubMed] [Google Scholar]
- 45. Merrill R.M., et al. (2000). Explaining the difference in prostate cancer mortality rates between white and black men in the United States. Urology 55 730–735 [DOI] [PubMed] [Google Scholar]
- 46. Weiserbs K.F., et al. (2003). A cross-sectional study of polycyclic aromatic hydrocarbon-DNA adducts and polymorphism of glutathione S-transferases among heavy smokers by race/ethnicity. Biomarkers 8 142–155 [DOI] [PubMed] [Google Scholar]
- 47. Wang S., et al. (2008). Assessment of interactions between PAH exposure and genetic polymorphisms on PAH-DNA adducts in African American, Dominican, and Caucasian mothers and newborns. Cancer Epidemiol. Biomarkers Prev. 17 405–413 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Caraballo R.S., et al. (1998). Racial and ethnic differences in serum cotinine levels of cigarette smokers: Third National Health and Nutrition Examination Survey, 1988-1991. JAMA 280 135–139 [DOI] [PubMed] [Google Scholar]
- 49. Benowitz N.L., et al. (2011). Racial differences in the relationship between number of cigarettes smoked and nicotine and carcinogen exposure. Nicotine Tob. Res. 13 772–783 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. Sørensen M., et al. (2003). Linking exposure to environmental pollutants with biological effects. Mutat. Res. 544 255–271 [DOI] [PubMed] [Google Scholar]
- 51. De Marzo A.M., et al. (2007). Inflammation in prostate carcinogenesis. Nat. Rev. Cancer 7 256–269 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52. Santella R.M. (1999). Immunological methods for detection of carcinogen-DNA damage in humans. Cancer Epidemiol. Biomarkers Prev. 8 733–739 [PubMed] [Google Scholar]
- 53. Mumford J.L., et al. (1996). A sensitive color ELISA for detecting polycyclic aromatic hydrocarbon-DNA adducts in human tissues. Mutat. Res. 359 171–177 [DOI] [PubMed] [Google Scholar]
- 54. Gyorffy E., et al. (2004). DNA adducts in tumour, normal peripheral lung and bronchus, and peripheral blood lymphocytes from smoking and non-smoking lung cancer patients: correlations between tissues and detection by 32P-postlabelling and immunoassay. Carcinogenesis 25 1201–1209 [DOI] [PubMed] [Google Scholar]
- 55. Rundle A., et al. (2012). Neighborhood socioeconomic status modifies the association between individual smoking status and PAH-DNA adduct levels in prostate tissue. Environ. Mol. Mutagen. 53 384–391 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56. Nock N.L., et al. (2006). Polymorphisms in polycyclic aromatic hydrocarbon metabolism and conjugation genes, interactions with smoking and prostate cancer risk. Cancer Epidemiol. Biomarkers Prev. 15 756–761 [DOI] [PubMed] [Google Scholar]
- 57. Hu J.J., et al. (2004). Deficient nucleotide excision repair capacity enhances human prostate cancer risk. Cancer Res. 64 1197–1201 [DOI] [PubMed] [Google Scholar]
- 58. Rundle A., et al. (2003). Issues in the epidemiological analysis and interpretation of intermediate biomarkers. Cancer Epidemiol. Biomarkers Prev. 12 491–496 [PubMed] [Google Scholar]