- Email: [email protected]
Estrogen metabolism as a risk factor for head and neck cancer
Otolaryngology– Head and Neck Surgery MARCH 2001
VOLUME 124
NUMBER 3
ORIGINAL ARTICLES Estrogen metabolism as a risk factor for head and neck cancer HELEN J. YOO, MD, DANIEL W. SEPKOVIC, PhD, H. LEON BRADLOW, PhD, GUO PEI YU, MD, MPH, HENEDIA V. SIRILIAN, RN, and STIMSON P. SCHANTZ, MD, New York, New York
OBJECTIVE: Estrogen metabolites have been associated in the pathogenesis of breast and cervical cancer; 16α-hydroxyestrone(16α-OHE1) demonstrated proliferative effects whereas 2-hydroxyestrone(2-OHE1) had antiproliferative effects. Our study’s objective is to demonstrate that head and neck (H&N) cancer patients metabolize estrogen differently than healthy controls, which may constitute a risk factor for H&N cancer development. STUDY DESIGN: Urinary metabolite levels of 2-OHE1 and 16α-OHE1 from 50 H&N cancer patients and 50 age- and sex-matched controls were measured using enzyme-linked immunosorbent assay (ELISA). Absolute values and 2-/16α-OHE1 ratios were calculated. Conditional logistic regression for univariate and multivariate analysis with odds ratio (OR) and 95% confidence interval (CI) were used. RESULTS: Thirty percent (15 of 50) from the case group had a low 2-/16α-OHE1 ratio compared with From the New York Eye and Ear Infirmary, Department of Otolaryngology–Head and Neck Surgery (Drs Yoo, Yu, and Schantz and Ms Sirilian), and the Strang Cancer Research Laboratory (Drs Sepkovic, Bradlow, and Schantz) Support provided by the Murray and Isabella Rayburn Fund, New York. Presented at the Annual Meeting of the American Academy of Otolaryngology–Head and Neck Surgery, New Orleans, LA, September 26-29, 1999. Reprint requests: Stimson P. Schantz, MD, New York and Eye and Ear Infirmary, Department of Otolaryngology–Head and Neck Surgery, 310 East 14th Street, New York, NY 10003; e-mail, [email protected]. Copyright © 2001 by The American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. 0194-5998/2001/$35.00 + 0 23/1/113507 doi:10.1067/mhn.2001.113507
only 4% (2 of 50) in the control group (OR = 11.1; 1.4-91.5, 95% CI) (P < 0.05). When adjusted for tobacco, OR remained significant at 15.6 (1.1212.5, 95% CI) (P < 0.05). CONCLUSION: H&N cancer patients are more likely to express abnormal estrogen metabolism than healthy controls; 2-/16α-OHE1 may serve as a potential biological marker of individuals at increased risk of H&N cancer. (Otolaryngol Head Neck Surg 2001;124:241-7.)
A
t a given level of tobacco exposure, women are at greater risk than men in developing cancer of the oral cavity.1 Begg et al2 reported a higher incidence of second primary cancers of the head and neck among women, implicating a possible role of estrogens in promoting cancer. Furthermore, the presence of steroid hormone receptors and hormone-related proteins from normal human larynges and laryngeal carcinomas have been demonstrated to support hormonal influence in carcinogenesis.3-5 It is well established that estrogen promotes cancer in estrogen-responsive tissues such as the breast, endometrium, and cervix. Estrogens are believed to act as initiators and promoters in malignancies causing increased growth rates in transformed cells.6 Estradiol is the active form of estrogen that is primarily metabolized into 2 end products, 2-hydroxyestrone (2-OHE1), and 16α-hydroxyestrone (16α-OHE1) (Fig 1).7 Estrone, the oxidized form of estradiol, is hydroxylated at either the C-2 or C-16 by different enzymes in the final metabolic pathway.7 Thus, induction of hydroxylation at C-2 reduces the production of C-16 and vice versa. 241
242
YOO et al
Fig 1. Two primary metabolic pathways of estradiol: C-2 and C16α hydroxylation.
2-OHE1 is inactive and weakly antiestrogenic.8,9 Conversely, 16α-OHE1 exhibits full estrogenic activityforming covalent bonds with estrogen receptors.10 Increased 16α-OHE1 levels have been associated with a greater risk of cancer of the cervix,6 breast,11 endometrium12 as well as recurrent respiratory papillomatosis (RRP) of the larynx caused by human papilloma virus (HPV).13 Several case-control studies comparing urinary 2-OHE1/16α-OHE1 ratios in patients with breast cancer to matched controls showed a significant decrease in these estrogen metabolite ratios in the case group.11,12,14,15 In addition, examination of absolute levels of estrogen metabolites revealed a much lower production of 2-OHE1 levels among patients with cancer.6,14,15 Similarly, patients with cervical intraepithelial neoplasia (CIN) displayed significantly low 2-OHE1/16α-OHE1 ratios compared with matched controls, and the decrease in ratios correlated with increasing CIN severity.6 In the head and neck, increasing severity of laryngeal papillomatosis showed an inverse relationship to 2OHE1/16α-OHE1 ratios.13 Furthermore, in an in vitro study with HPV immortalized keratinocytes (a correlate of the premalignant keratinocytes) estradiol and 16αOHE1 caused abnormal proliferation and anchorageindependent growth in these cells.16 These investigations led us to believe that H&N cancer patients would metabolize estrogen differently than healthy matched controls, and that these differences may constitute a risk factor in H&N cancer development. METHODS AND MATERIALS Subjects Spot urine samples were collected from 50 patients and 50 controls matched for age and sex. Constancy of urine estrogen
Otolaryngology– Head and Neck Surgery March 2001
metabolite ratios over the course of the day and from 24-hour urine collections have been previously demonstrated. Patients were recruited from the head and neck clinic at New York Eye and Ear Infirmary (NYEEI), who had diagnosed cases of squamous cell cancer of the upper aerodigestive tract: oral cavity, pharynx, and larynx. Each patient’s cancer stage was assessed according to the American Joint Committee on Cancer Staging (AJCC). Our study included patients who had or were about to undergo treatment for their disease. Information was gathered from the medical records of treated patients as to the modality of treatment (surgery, radiation therapy, and chemotherapy) and survival time. Control subjects were patients who were matched to cases for sex and age within 4 years. Controls were patients from NYEEI ENT clinic with general ear, nose, and throat problems, such as hearing loss, allergic rhinitis, etc, with no history of neoplasm at any site. Both study groups completed a questionnaire pertaining to each patient’s medical history; steroid use, tobacco use, and alcohol consumption histories were obtained at the time of urine sample collection. An informed consent was obtained from each participant of this study, and our protocol was approved by the NYEEI Institutional Review Board. Determination of Estrogen Metabolites Collected urine samples were preserved with sodium ascorbate (2 mg/mL) and frozen in –70°. Samples were then transferred to Strang Cancer Research Laboratory in New York City where levels of estrogen metabolites, 2-OHE1, and 16α-OHE1, were determined utilizing an enzyme-linked immunosorbent assay (ELISA). In our study, the duration of storage time to analysis of the samples ranged from 1 to 21 days. ELISA kits purchased from Immunacare, Inc (Bethlehem, PA), include monoclonal antibodies, ELISA plates coated with a proprietary reagent that specifically binds antibodies, deconjugating enzymes, buffers, and standards. The high degree of specificity of these antibodies for urinary 2-OHE1 and 16α-OHE1 have been previously described.17,18 The urinary forms of these estrogen metabolites are found as glucuronides and sulfates and are initially hydrolyzed by adding 190 µL of deconjugating enzymes (containing 500 units of Glusulase, a Helix pomatia liver preparation containing both glucuronidase and sulfatase activity, in a pH 5.0 buffer) to 10 µL of urine and incubated for 2 hours. Neutralization buffer is then added to bring the pH back to 7.0. Aliquots (75 µL) of hydrolyzed urine, steroid coupled to phosphorylase and specific antibody are added to wells coated with a proprietary reagent that binds the antibodies and allowed to interact for 3 hours. The plates are then washed using a Bio-Tek automatic plate washer (Bio-tek, Inc, Winooskie, VT) to remove excess phosphorylase. Color is developed by adding p-nitro-phenyl phosphate in buffer that is then read at 2-minute intervals for 10 times by a Bio-Tek 309 kinetic plate reader equipped with Kineticalc software.
Otolaryngology– Head and Neck Surgery Volume 124 Number 3
YOO et al
Table 1. Distribution of sex, age, and race between cases and controls Cases No.
Sex Female Male Age (years) <50 50-59 60-69 ≥70 Race White Nonwhite
Controls %
No.
%
Table 2. Clinical characteristics of 50 patients with head and neck cancer Clinical characteristics
P value
10 40
20 80
10 40
20 80
>0.05
8 18 17 7
16 36 34 14
7 18 16 9
14 36 32 18
>0.05
13 37
26 74
17 33
34 66
>0.05
Standards of 0.625 to 20.0 ng/mL as well as medium and low controls are routinely run in every assay. The low-level and medium-level controls were prepared from urine pools of men and premenopausal women, respectively. For aliquots whose values fell off the standard curve, either twice the volume or a dilution of the urine was assayed. The urinary samples are run in triplicate and the results averaged. The within-assay and between-assay coefficient of variation for 2OHE1 and 16α-OHE1 with this kit are reported to be 7.6% and less than 12%, respectively.18,19 Creatinine levels for each urine sample were measured with a Beckman manual creatinine analyzer using reagents purchased from the Beckman Instrument Co. In order to compare and analyze the absolute values of 2-OHE1 and 16αOHE1 between individuals, the estrogen metabolite values are normalized per milligram of creatinine. Analysis The mean and standard deviations (SD) for absolute values of 2-OHE1 and 16α-OHE1 were calculated and compared between cases and controls. Urinary estrogen metabolite ratios of 2-OHE1/16α-OHE1 were then evaluated for each patient, and the ratios were stratified into 3 levels according to a control distribution: ratios 0 to 1.00 (low), 1.01 to 2.00 (medium), and > 2.01 (high). The number of cases compared with control patients for each ratio level was computed. The differences in distribution of sex, age, race, smoking, and alcohol consumption between cases and controls were examined by χ2 test. Odds-ratio (OR) analysis was used to measure the relationship between head and neck cancer risk and urinary estrogen metabolite ratios. OR and 95% confidence intervals (CI) were derived using conditional logistic regression. In the multivariate model, adjustment of ORs used 3 categories of smoking, ie, never, 1 to 39, and ≥40 cigarettes/day. All statistical analysis was performed using the SAS package.
243
No.
Site of tumor Oral cavity Pharynx Larynx Tumor stage I II III IV Treatment No treatment Surgery only Radiation only Surgery + radiation Other Survival time (months) 0-12 13-24 ≥25
%
4 16 30
8 32 60
7 10 10 22
14 20 20 45
4 18 5 20 3
8 36 10 40 6
20 15 15
40 30 30
Table 3. Distribution of smoking and alcohol consumption between cases and controls Cases* No.
Smoking (pack-years) Nonsmokers 1-39 ≥40 Alcohol Nondrinkers Drinkers
Controls
%
No.
%
P value
3 22 24
6 45 49
20 22 8
40 44 16
< 0.001
20 29
41 59
41 9
82 18
<0.001
*One case was excluded due to missing information about tobacco smoking and alcohol consumption.
RESULTS Clinical Characteristics
Forty men and 10 women made up each of the case and control groups for a total of 100 patients. Ages ranged from 42 to 82 years in the case group and 43 to 82 years in the control group, with median ages 59.9 and 59.5, respectively (Table 1). In the case group, there were 13 whites, 15 African Americans, and 22 Hispanics. In the control group, there were 17 whites, 9 African Americans, 21 Hispanics, and 3 Asian Americans. For statistical analysis, ethnicity was simply divided between whites and nonwhites in both groups (Table 1). There was no significant difference in age or ethnicity between the case and control groups. The most common location of cancer was the larynx (60%), followed by oropharynx (32%), and oral cavity
244
Otolaryngology– Head and Neck Surgery March 2001
YOO et al
Fig 2. Urinary estrogen metabolite ratios compare cases with controls.
(8%) (Table 2). At the time of diagnosis, 65% of our head and neck cancer patients exhibited advanced stage III and IV diseases (Table 2). Seven patients presented with stage I, and 10 had stage II disease. Four patients were diagnosed within days of the study, and a form of therapy had not yet been instituted. Table 2 also displays 18 (36%) patients treated with surgery only and 5 (10%) with radiation therapy. Twenty (40%) patients had received combined surgery and radiation, and 3 remaining individuals were postchemotherapy or chemotherapy and radiation. Survival time varied from 0 (those just diagnosed) to 144 months. In our group of 50 patients, 20 (40%) were free of disease for 0 to 12 months, 15 (30%) without disease for 13 to 24 months, and 15 (30%) were survivors for longer than 2 years. There was no apparent clinical correlation with low urinary estrogen metabolite ratios and above disease specific data. Not surprisingly, there was a significant difference in the number of smokers among cases compared with controls (Table 3). Smokers are considered anyone who has a history of 1 to >40 pack-years. Ninety-four percent of the head and neck cancer patients were tobacco users versus 60% in the control population (P < 0.001). Likewise, alcohol consumption varied between the 2 groups (P < 0.001) with 29 (59%) patients in the head and neck cancer group being drinkers compared with only 9 (18%) patients in the control group. One patient’s history regarding tobacco and alcohol consumption could not be obtained and is not included in Table 3. Drinkers are considered anyone who has a history of regularly drinking a minimum of 2 packs of beer/1 bottle of wine/1 bottle of liquor per week.
Estrogen Metabolites
The absolute values of urinary estrogen metabolites, 2-OHE1, and 16α-OHE1 were normalized using urine creatinine concentrations and were compared between cases and controls. The mean values of 2-OHE1 metabolite level between cases and controls were 9.9 ng/mL (SD ± 5.3) and 10.2 ng/mL (SD ± 14.9), respectively. The average values of 16α-OHE1 between cases and controls were 5.5 ng/mL (SD ± 3.6) and 5.8 ng/mL (SD ± 5.2), revealing no statistical difference in the mean values of 2-OHE1 and 16α-OHE1 between the 2 study groups. However, calculated ratios of 2-OHE1/16α-OHE1 for each patient defined a pattern of low ratios among the head and neck cancer group (Fig 2). When the 2OHE1/16α-OHE1 ratios were stratified into 3 levels: low (0 to 1.00), medium (1.01 to 2.00), and high (>2.01), 15 of 50 H&N cancer patients or 30% had low urinary estrogen metabolite ratios compared with only 2 of 50 or 4% from the control group (P < 0.05) (Table 4). The estimated OR of low urinary estrogen metabolite ratio and head and neck cancer is 11.1 (1.4 to 91.5, 95% CI). When adjusted for tobacco, the OR remained significant at 15.6 (1.1 to 212.5, 95% CI) (P < 0.05. However, adjusted OR for both tobacco and alcohol did not achieve statistical significance (OR = 8.0, 0.3 to 232, 95% CI) (P = 0.225) as a result of the small sample numbers. Because the number of women in the study was small, a separate analysis of male subjects between cases and controls was performed. Thirteen (33%) of 40 versus 2 (5%) of 40 still translated into a clinically significant OR and adjusted OR for tobacco of 10 (1.1 to 89.8, 95% CI) and 13.2 (0.9 to 192.2, 95%
Otolaryngology– Head and Neck Surgery Volume 124 Number 3
YOO et al
245
Table 4. Odds ratios for head and neck cancer associated with urinary estrogen metabolite ratios (2-/16α− OHE1), all subjects Ratio
Low: <1.0 Medium: 1.0-2.0 High: >2.0
Cases (%)
Controls (%)
ORc (95% CI)
P value
15 (30) 16 (30) 20 (40)
2 (4) 27 (54) 21 (42)
11.1 (1.4-91.5) 0.7 (0.3-1.7) 1.0 —
0.024 0.445
ORa (95% CI)
15.6 (1.1-212.5) 0.8 (0.3-2.3) 1.0—
P value
0.040 0.698
ORc, Odds ratios were not adjusted. ORa, Odds ratios were adjusted for cigarette smoking.
Table 5. Odds ratios for head and neck cancer associated with urinary estrogen metabolie ratios (2-/16α-OHE1), male subjects Ratio
Low: <1.0 Medium: 1.0-2.0 High: >2.0
Cases (%)
Controls (%)
ORc (95% CI)
P value
ORa (95% CI)
P value
13 (33) 13 (33) 14 (35)
2 (5) 23 (58) 15 (38)
10.0 (1.1-89.8) 0.8 (0.3-2.4) 1.0 —
0.040 0.666
13.2 (0.9-192.2) 0.9 (0.2-3.1) 1.0 —
0.059 0.822
ORc, Odds ratios were not adjusted. ORa, Odds ratios were adjusted for cigarette smoking.
CI) (P = 0.059), respectively (Table 5). Although there were more controls (27) than cases (17) with medium 2/16α-OHE1 ratios, this difference was not statistically significant. About equal number of H&N cancer patients (20) to controls (21) had high ratio levels. DISCUSSION
Steroid hormones control cellular growth, proliferation, differentiation, and metabolic processes.20 Once a mutation or transformation of protooncogenes has occurred through actions of a carcinogen, estrogens act as promotors exerting cell proliferation and tumor growth.6,20 Through its mitogenic stimulus, estrogens may also predispose nontransformed cells for initiation of carcinogenesis.21 Estradiol, the primary active form of estrogen, is first reversibly oxidized to estrone, which is then converted irreversibly to either 2-hydroxyestrone or 16α-hydroxyestrone.7 On one hand, 16α-OHE1 has been shown to bind covalently to estrogen receptors promoting DNA synthesis with hyperproliferation of epithelial cells.10 On the other hand, 2-OHE1 is devoid of peripheral estrogenic activity.9 It does not bind to estrogen receptors and is subsequently eliminated in urine.9 There is growing evidence in the literature to support the role of steroid hormonal regulation and squamous cell cancer of the upper aerodigestive tract. Studies by Ferguson et al,3 Resta et al,4 and Maiorano et al5 have demonstrated the presence of estrogen receptors and hormone-related proteins from laryngeal cancer cells. Newfield et al13 investigated estrogen metabolism of nor-
mal laryngeal and laryngeal papilloma cells and determined that 16α-hydroxylation was constitutively high in the normal larynx but even higher in laryngeal papillomas infected with HPV (type 11). They also demonstrated that 16α-OHE1 stimulated proliferation in both cells whereas 2-OHE1 was antiproliferative. In HPV-immortalized keratinocytes, which are correlates of the premalignant keratinocytes, estradiol and 16α-OHE1 caused abnormal proliferation and anchorage independent growth.16 In clinical case-control studies evaluating the role of estrogen metabolites, low 2-/16α-OHE1 ratios have been associated with increased risk in cancer of the breast and cervix.6,11,14,15 Elevated urinary levels of 16α-OHE1 have also been found in women with high risk of breast cancer, in women with breast cancer19 and endometrial cancer.12 Our study currently demonstrates that patients with H&N cancer metabolize estrogens differently as well. When absolute levels of 16α-OHE1 among cancer patients were compared with controls, there was no statistical difference. However, 30% (15 of 50) of H&N cancer patients exhibited low 2/16α-OHE1 ratios compared with only 4% (2 of 50) among the healthy population. When adjusted for tobacco use, the OR for low 2/16α-OHE1 ratio remained clinically significant at 15.6. Although our study is limited by sample size, this statistically significant difference suggests that low 2/16α-OHE1 ratios may constitute a risk factor in development of H&N cancer. A randomized prospective trial involving a larger cohort of patients is needed to confirm these results. Our observations are based on measured metabolites in urine that correlates with patients’ overall estrogen
246
Otolaryngology– Head and Neck Surgery March 2001
YOO et al
metabolism. This raises the question whether increased levels of 16α-OHE1 and low 2-/16α-OHE1 metabolite ratios are manifestations of carcinogenesis and aberrant state of cancer cells or they are inherent characteristics of cancer patients with a genetic predisposition. Telang et al21 demonstrated that murine mammary explant cultures and immortalized nontumorigenic mammary epithelial cell cultures treated with chemical carcinogens and transfected with oncogenes, resulted in altered cellular metabolism of estradiol and aberrant hyperproliferation in vitro. Alteration in estrogen metabolism was detected by a significant increase in C16α-hydroxylation with a concomitant decrease in C2 hydroxylation.21 These results suggest that increased production of 16α-OHE1 is a biochemical manifestation of transformed or immortalized cells. On the other hand, indirect evidence supporting genetic predisposition for aberrant estrogen metabolism is the presence of genetic polymorphism in enzymes that convert estrone to 2-OHE1; 2-hydroxylation is catalyzed by a P450 enzyme that has at least 3 major cytochrome P450 subfamilies including P-450 1A1.22 The Cyp 1A1 gene plays an important role in C-2 hydroxylation of estradiol. Three polymorphisms of the Cyp 1A1 gene are known: MSP-1 restriction fragment length polymorphism, a mutation in exon 7, and a novel Cyp 1A1 polymorphism that is exclusively present in African Americans.22 In a study assessing the role of Cyp 1A1 polymorphisms on breast cancer risk and estradiol metabolism, women with the wild type Cyp 1A1 gene showed a significant increase in the 2/16α-OHE1 metabolite ratio after treatment with indole-3-carbinol (I3C-a phytochemical compound that induces 2-hydroxylation).22 Women with MSP1 polymorphism showed no increase in 2/16α-OHE1 metabolite ratio.22 Evaluation of genetic polymorphisms and estrogen metabolism at cellular levels of squamous cell carcinoma are interesting areas of research that may help define and clarify the role of estrogen metabolism in H&N cancer. Estrogen is metabolized by cytochrome P450 enzymes present in various tissues and its metabolism is influenced by many variables including diet.23 Indole3-carbinol (I3C), a phytochemical compound naturally found in cruciferous vegetables, such as broccoli, cauliflower, and brussels sprouts, is a potent inducer of 2hydroxylation.23 The ability of I3C to alter estrogen metabolic pathway, specifically to increase 2-OHE1 levels and exhibit antiestrogenic effects, has been demonstrated by various investigations.13,16,21 I3C abrogated the proliferative effects of estradiol in cell culture.16,21 In mice, I3C reduced the development of papilloma cysts where infected HPV 11 laryngeal tissue was implanted under renal capsules; implantation was 100% in controls vs only 25% in mice fed I3C.13 In a
preliminary study by Rosen et al24 in which 16 recurrent respiratory papillomatosis patients were treated with oral I3C, 6 had complete responses with cessation of papilloma growth and 6 patients had reduced growth rates. Urinary increase in 2-OHE1/16α-OHE1 ratios correlated with improved response. Other environmental factors that can increase 2-hydroxylation are low fat diet25 and exercise.26 In a study looking at urinary 2/16α-OHE1 metabolites of estrogen in premenopausal women, smoking also increased 2-hydroxylation.27 Alcohol consumption had no affect on estrogen metabolism. Altering estrogen metabolic pathway with naturally occurring compounds such as I3C with subsequent effect on H&N cancer has important clinical implications and warrants future study. In conclusion, we have demonstrated for the first time, that head and neck cancer patients metabolize estrogen differently than healthy matched controls. Low urinary 2-/16α-OHE1 estrogen metabolite ratios were detected in a cohort of H&N cancer patients. Our study suggests that low 2-/16α-OHE1 may constitute a risk factor in development of cancers of the upper aerodigestive tract; 2-/16α-OHE1 may also serve as a potential biological marker of individuals at increased risk of cancer. The ability to alter estrogen metabolic pathways through chemoprevention may diminish cancer risk at multiple sites including the upper aerodigestive tract. REFERENCES 1. Muscat JE, Richie JP Jr, Thompson S, et al. Gender differences in smoking and risk for oral cancer. Cancer Res 1996;56:5192-7. 2. Begg CB, Zhang ZF, Sun M, et al. Methodology for evaluating the incidence of second primary cancers with application to smokingrelated cancers from the surveillance, epidemiology, and end results (SEER) program. Am J Epidemiol 1995;142:653-65. 3. Ferguson BJ, Hudson WR, McCarty KS Jr. Sex steroid receptor distribution in the human larynx and laryngeal carcinoma. Arch Otolaryngol Head Neck Surg 1987;113:1311-5. 4. Resta J, Marsigliante S, Leo G, et al. Molecular biopathology of metaplastic, dysplastic, and neoplastic laryngeal epithelium. Acta Otolaryngol (Stockh) 1997;Suppl527:39-42. 5. Maiorano E, Botticella MA, Marzullo A, et al. Expression of ERD5 and EGFr in laryngeal carcinoma and premalignant epithelium. Acta Otolaryngol (Stockh) 1997;Suppl 527:95-9. 6. Sepkovic DW, Bradlow HL, Ho G, et al. Estrogen metabolite ratios and risk assessment of hormone-related cancers: assay, validation, and prediction of cervical cancer risk. Ann N Y Acad Sci 1995;768:312-6. 7. Lustig RH, Bradlow HL, Fishman J. Estrogen metabolism in disorders of nutrition and dietary composition. In: The menstrual cycle and its disorders. Berlin/New York: Springer-Verlag. 8. Martucci C, Fishman J. Direction of estradiol metabolism as a control of its hormonal action: uterotropic activity of estradiol metabolites. Endocrinology 1977;101:1709-15. 9. Bradlow HL, Telang NT, Sepkovic DW, et al. 2-Hydroxyestrone: the “good” estrogen. J Endocrinol 1996;150:5259-63. 10. Swaneck, GE, Fishman J. Covalent binding of the endogenous estrogen 16–hydroxyestrone to estradiol receptors in human breast cancer cells: characterization and intranuclear localization. Proc Natl Acad Sci U S A 1988;85:7831-5. 11. Schneider J, Kinne D, Fracchia A, et al. Abnormal oxidative
Otolaryngology– Head and Neck Surgery Volume 124 Number 3
12. 13.
14. 15. 16. 17. 18.
19.
YOO et al
metabolism of estradiol in women with breast cancer. Proc Soc Acad Sci U S A 1982;79:3047-51. Fishman J, Schneider J, Hershcope RJ, et al. Increased estrogen 16 alpha-hydroxylase activity in women with breast and endometrial cancer. J Steroid Biochem Mol Biol 1984;20:1077-81. Newfield L, Goldsmith A, Bradlow HL, et al. Estrogen metabolism and human papillomavirus-induced tumors of the larynx: chemo-prophylaxis with indole-3-carbinol. Anticancer Res 1993; 13:337-42. Kabat GC, Chang CJ, Sparano JA, et al. Urinary estrogen metabolites and breast cancer: a case-control study. Cancer Epidemiol Biomarkers Prev 1997;6:500-4. Zheng W, Dunning L, Gong Y, et al. Ratios of urinary 20 hydroxyestrone to 16 alpha-hydroxyestrone are lower in breast cancer patients than controls. Cancer Epidmiol Biomarkers Prev 1998;7:85-6. Newfield L, Bradlow HL, Sepkovic DW, et al. Estrogen metabolism and the malignant potential of human papillomavirus immortalized keratinocytes. Soc Exp Biol Med 1998;322-6. Bradlow HL, Sepkovic DW, Klug T, et al. Application of an improved ELISA assay to the analysis of urinary estrogen metabolites. Steroids 1998;63:406-13. Ziegler RG, Rossi SC, Fears TR, et al. Quantifying estrogen metabolism: an evaluation of the reproducibility and validity of enzyme immunoassays for 2-hydroxyestrone and 16α-hydroxyestrone in urine. Environ Health Perspect 1997;105:607-14. Meilahn EN, Stavola BD, Allen DS, et al. Do urinary estrogen
20. 21. 22. 23. 24. 25. 26. 27.
metabolites predict breast cancer? Guernsey III cohort follow-up. Br J Cancer 1998;78:1250-5. Sekeris CE. Hormonal steroids act as tumour promoters by modulating oncogene expression. J Cancer Res Clin Oncol 1991;117:96-101. Telang NT, Katdare M, Bradlow HL, et al. Estradiol metabolism: an endocrine biomarker for modulation of human mammary carcinogenesis. Environ Health Perspect 1997;105:559-64. Taioli E, Bradlow HL, Garbers S, et al. Role of estradiol metabolism and CYP1A1 polymorphisms in breast cancer risk. Cancer Detect Prev 1999;23:232-7. Michnovicz JJ, Bradlow HL. Induction of estradiol metabolism by dietary indole-3-carbinol in humans. J Natl Cancer Inst 1990;82:947-9. Rosen CA, Woodson GE, Thompson JW, et al. Preliminary results of use of indole-3-carbinol for recurrent respiratory papillomatosis. Otolaryngol Head Neck Surg 1998;118:810-5. Fishman J, Boyar RM, Hellman L, Influence of body weight on estradiol metabolism in young women. J Clin Endocrinol Metab 1975;41:989-91. Snow RC, Barberi RL, Frisch, RE, Estrogen 2-hydroxylase oxidation and menstrual function among elite oarswomen. J Clin Endocrinol Metab 1989;69:369-76. Michnovicz JJ, Hershcopf RJ, Naganuma H, et al. Increased 2hydroxylation of estradiol as a possible mechanism for the antiestrogenic effect of cigarette smoking. N Engl J Med 1986; 315:1305-9.
Access to Otolaryngology–Head and Neck Surgery Online is reserved for print subscribers! Full-text access to Otolaryngology–Head and Neck Surgery Online is now available for all print subscribers. To activate your individual online subscription, please visit Otolaryngology–Head and Neck Surgery Online, point your browser to http://www.mosby.com/oto, follow the prompts to activate your online access, and follow the instructions. To activate your account, you will need your subscriber account number, which you can find on your mailing label (note: the number of digits in your subscriber account number varies from 6 to 10). See the example below in which the subscriber account number has been circled: Sample mailing label This is your subscription account number
247
**************************3-DIGIT 001 SJ P1 FEB00 J027 C: 1 1234567-89 U 05/00 Q: 1 J. H. DOE 531 MAIN ST CENTER CITY, NY 10001-001
Personal subscriptions to Otolaryngology–Head and Neck Surgery Online are for individual use only and may not be transferred. Use of Otolaryngology–Head and Neck Surgery Online is subject to agreement to the terms and conditions as indicated online.
VOLUME 124
NUMBER 3
ORIGINAL ARTICLES Estrogen metabolism as a risk factor for head and neck cancer HELEN J. YOO, MD, DANIEL W. SEPKOVIC, PhD, H. LEON BRADLOW, PhD, GUO PEI YU, MD, MPH, HENEDIA V. SIRILIAN, RN, and STIMSON P. SCHANTZ, MD, New York, New York
OBJECTIVE: Estrogen metabolites have been associated in the pathogenesis of breast and cervical cancer; 16α-hydroxyestrone(16α-OHE1) demonstrated proliferative effects whereas 2-hydroxyestrone(2-OHE1) had antiproliferative effects. Our study’s objective is to demonstrate that head and neck (H&N) cancer patients metabolize estrogen differently than healthy controls, which may constitute a risk factor for H&N cancer development. STUDY DESIGN: Urinary metabolite levels of 2-OHE1 and 16α-OHE1 from 50 H&N cancer patients and 50 age- and sex-matched controls were measured using enzyme-linked immunosorbent assay (ELISA). Absolute values and 2-/16α-OHE1 ratios were calculated. Conditional logistic regression for univariate and multivariate analysis with odds ratio (OR) and 95% confidence interval (CI) were used. RESULTS: Thirty percent (15 of 50) from the case group had a low 2-/16α-OHE1 ratio compared with From the New York Eye and Ear Infirmary, Department of Otolaryngology–Head and Neck Surgery (Drs Yoo, Yu, and Schantz and Ms Sirilian), and the Strang Cancer Research Laboratory (Drs Sepkovic, Bradlow, and Schantz) Support provided by the Murray and Isabella Rayburn Fund, New York. Presented at the Annual Meeting of the American Academy of Otolaryngology–Head and Neck Surgery, New Orleans, LA, September 26-29, 1999. Reprint requests: Stimson P. Schantz, MD, New York and Eye and Ear Infirmary, Department of Otolaryngology–Head and Neck Surgery, 310 East 14th Street, New York, NY 10003; e-mail, [email protected]. Copyright © 2001 by The American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. 0194-5998/2001/$35.00 + 0 23/1/113507 doi:10.1067/mhn.2001.113507
only 4% (2 of 50) in the control group (OR = 11.1; 1.4-91.5, 95% CI) (P < 0.05). When adjusted for tobacco, OR remained significant at 15.6 (1.1212.5, 95% CI) (P < 0.05). CONCLUSION: H&N cancer patients are more likely to express abnormal estrogen metabolism than healthy controls; 2-/16α-OHE1 may serve as a potential biological marker of individuals at increased risk of H&N cancer. (Otolaryngol Head Neck Surg 2001;124:241-7.)
A
t a given level of tobacco exposure, women are at greater risk than men in developing cancer of the oral cavity.1 Begg et al2 reported a higher incidence of second primary cancers of the head and neck among women, implicating a possible role of estrogens in promoting cancer. Furthermore, the presence of steroid hormone receptors and hormone-related proteins from normal human larynges and laryngeal carcinomas have been demonstrated to support hormonal influence in carcinogenesis.3-5 It is well established that estrogen promotes cancer in estrogen-responsive tissues such as the breast, endometrium, and cervix. Estrogens are believed to act as initiators and promoters in malignancies causing increased growth rates in transformed cells.6 Estradiol is the active form of estrogen that is primarily metabolized into 2 end products, 2-hydroxyestrone (2-OHE1), and 16α-hydroxyestrone (16α-OHE1) (Fig 1).7 Estrone, the oxidized form of estradiol, is hydroxylated at either the C-2 or C-16 by different enzymes in the final metabolic pathway.7 Thus, induction of hydroxylation at C-2 reduces the production of C-16 and vice versa. 241
242
YOO et al
Fig 1. Two primary metabolic pathways of estradiol: C-2 and C16α hydroxylation.
2-OHE1 is inactive and weakly antiestrogenic.8,9 Conversely, 16α-OHE1 exhibits full estrogenic activityforming covalent bonds with estrogen receptors.10 Increased 16α-OHE1 levels have been associated with a greater risk of cancer of the cervix,6 breast,11 endometrium12 as well as recurrent respiratory papillomatosis (RRP) of the larynx caused by human papilloma virus (HPV).13 Several case-control studies comparing urinary 2-OHE1/16α-OHE1 ratios in patients with breast cancer to matched controls showed a significant decrease in these estrogen metabolite ratios in the case group.11,12,14,15 In addition, examination of absolute levels of estrogen metabolites revealed a much lower production of 2-OHE1 levels among patients with cancer.6,14,15 Similarly, patients with cervical intraepithelial neoplasia (CIN) displayed significantly low 2-OHE1/16α-OHE1 ratios compared with matched controls, and the decrease in ratios correlated with increasing CIN severity.6 In the head and neck, increasing severity of laryngeal papillomatosis showed an inverse relationship to 2OHE1/16α-OHE1 ratios.13 Furthermore, in an in vitro study with HPV immortalized keratinocytes (a correlate of the premalignant keratinocytes) estradiol and 16αOHE1 caused abnormal proliferation and anchorageindependent growth in these cells.16 These investigations led us to believe that H&N cancer patients would metabolize estrogen differently than healthy matched controls, and that these differences may constitute a risk factor in H&N cancer development. METHODS AND MATERIALS Subjects Spot urine samples were collected from 50 patients and 50 controls matched for age and sex. Constancy of urine estrogen
Otolaryngology– Head and Neck Surgery March 2001
metabolite ratios over the course of the day and from 24-hour urine collections have been previously demonstrated. Patients were recruited from the head and neck clinic at New York Eye and Ear Infirmary (NYEEI), who had diagnosed cases of squamous cell cancer of the upper aerodigestive tract: oral cavity, pharynx, and larynx. Each patient’s cancer stage was assessed according to the American Joint Committee on Cancer Staging (AJCC). Our study included patients who had or were about to undergo treatment for their disease. Information was gathered from the medical records of treated patients as to the modality of treatment (surgery, radiation therapy, and chemotherapy) and survival time. Control subjects were patients who were matched to cases for sex and age within 4 years. Controls were patients from NYEEI ENT clinic with general ear, nose, and throat problems, such as hearing loss, allergic rhinitis, etc, with no history of neoplasm at any site. Both study groups completed a questionnaire pertaining to each patient’s medical history; steroid use, tobacco use, and alcohol consumption histories were obtained at the time of urine sample collection. An informed consent was obtained from each participant of this study, and our protocol was approved by the NYEEI Institutional Review Board. Determination of Estrogen Metabolites Collected urine samples were preserved with sodium ascorbate (2 mg/mL) and frozen in –70°. Samples were then transferred to Strang Cancer Research Laboratory in New York City where levels of estrogen metabolites, 2-OHE1, and 16α-OHE1, were determined utilizing an enzyme-linked immunosorbent assay (ELISA). In our study, the duration of storage time to analysis of the samples ranged from 1 to 21 days. ELISA kits purchased from Immunacare, Inc (Bethlehem, PA), include monoclonal antibodies, ELISA plates coated with a proprietary reagent that specifically binds antibodies, deconjugating enzymes, buffers, and standards. The high degree of specificity of these antibodies for urinary 2-OHE1 and 16α-OHE1 have been previously described.17,18 The urinary forms of these estrogen metabolites are found as glucuronides and sulfates and are initially hydrolyzed by adding 190 µL of deconjugating enzymes (containing 500 units of Glusulase, a Helix pomatia liver preparation containing both glucuronidase and sulfatase activity, in a pH 5.0 buffer) to 10 µL of urine and incubated for 2 hours. Neutralization buffer is then added to bring the pH back to 7.0. Aliquots (75 µL) of hydrolyzed urine, steroid coupled to phosphorylase and specific antibody are added to wells coated with a proprietary reagent that binds the antibodies and allowed to interact for 3 hours. The plates are then washed using a Bio-Tek automatic plate washer (Bio-tek, Inc, Winooskie, VT) to remove excess phosphorylase. Color is developed by adding p-nitro-phenyl phosphate in buffer that is then read at 2-minute intervals for 10 times by a Bio-Tek 309 kinetic plate reader equipped with Kineticalc software.
Otolaryngology– Head and Neck Surgery Volume 124 Number 3
YOO et al
Table 1. Distribution of sex, age, and race between cases and controls Cases No.
Sex Female Male Age (years) <50 50-59 60-69 ≥70 Race White Nonwhite
Controls %
No.
%
Table 2. Clinical characteristics of 50 patients with head and neck cancer Clinical characteristics
P value
10 40
20 80
10 40
20 80
>0.05
8 18 17 7
16 36 34 14
7 18 16 9
14 36 32 18
>0.05
13 37
26 74
17 33
34 66
>0.05
Standards of 0.625 to 20.0 ng/mL as well as medium and low controls are routinely run in every assay. The low-level and medium-level controls were prepared from urine pools of men and premenopausal women, respectively. For aliquots whose values fell off the standard curve, either twice the volume or a dilution of the urine was assayed. The urinary samples are run in triplicate and the results averaged. The within-assay and between-assay coefficient of variation for 2OHE1 and 16α-OHE1 with this kit are reported to be 7.6% and less than 12%, respectively.18,19 Creatinine levels for each urine sample were measured with a Beckman manual creatinine analyzer using reagents purchased from the Beckman Instrument Co. In order to compare and analyze the absolute values of 2-OHE1 and 16αOHE1 between individuals, the estrogen metabolite values are normalized per milligram of creatinine. Analysis The mean and standard deviations (SD) for absolute values of 2-OHE1 and 16α-OHE1 were calculated and compared between cases and controls. Urinary estrogen metabolite ratios of 2-OHE1/16α-OHE1 were then evaluated for each patient, and the ratios were stratified into 3 levels according to a control distribution: ratios 0 to 1.00 (low), 1.01 to 2.00 (medium), and > 2.01 (high). The number of cases compared with control patients for each ratio level was computed. The differences in distribution of sex, age, race, smoking, and alcohol consumption between cases and controls were examined by χ2 test. Odds-ratio (OR) analysis was used to measure the relationship between head and neck cancer risk and urinary estrogen metabolite ratios. OR and 95% confidence intervals (CI) were derived using conditional logistic regression. In the multivariate model, adjustment of ORs used 3 categories of smoking, ie, never, 1 to 39, and ≥40 cigarettes/day. All statistical analysis was performed using the SAS package.
243
No.
Site of tumor Oral cavity Pharynx Larynx Tumor stage I II III IV Treatment No treatment Surgery only Radiation only Surgery + radiation Other Survival time (months) 0-12 13-24 ≥25
%
4 16 30
8 32 60
7 10 10 22
14 20 20 45
4 18 5 20 3
8 36 10 40 6
20 15 15
40 30 30
Table 3. Distribution of smoking and alcohol consumption between cases and controls Cases* No.
Smoking (pack-years) Nonsmokers 1-39 ≥40 Alcohol Nondrinkers Drinkers
Controls
%
No.
%
P value
3 22 24
6 45 49
20 22 8
40 44 16
< 0.001
20 29
41 59
41 9
82 18
<0.001
*One case was excluded due to missing information about tobacco smoking and alcohol consumption.
RESULTS Clinical Characteristics
Forty men and 10 women made up each of the case and control groups for a total of 100 patients. Ages ranged from 42 to 82 years in the case group and 43 to 82 years in the control group, with median ages 59.9 and 59.5, respectively (Table 1). In the case group, there were 13 whites, 15 African Americans, and 22 Hispanics. In the control group, there were 17 whites, 9 African Americans, 21 Hispanics, and 3 Asian Americans. For statistical analysis, ethnicity was simply divided between whites and nonwhites in both groups (Table 1). There was no significant difference in age or ethnicity between the case and control groups. The most common location of cancer was the larynx (60%), followed by oropharynx (32%), and oral cavity
244
Otolaryngology– Head and Neck Surgery March 2001
YOO et al
Fig 2. Urinary estrogen metabolite ratios compare cases with controls.
(8%) (Table 2). At the time of diagnosis, 65% of our head and neck cancer patients exhibited advanced stage III and IV diseases (Table 2). Seven patients presented with stage I, and 10 had stage II disease. Four patients were diagnosed within days of the study, and a form of therapy had not yet been instituted. Table 2 also displays 18 (36%) patients treated with surgery only and 5 (10%) with radiation therapy. Twenty (40%) patients had received combined surgery and radiation, and 3 remaining individuals were postchemotherapy or chemotherapy and radiation. Survival time varied from 0 (those just diagnosed) to 144 months. In our group of 50 patients, 20 (40%) were free of disease for 0 to 12 months, 15 (30%) without disease for 13 to 24 months, and 15 (30%) were survivors for longer than 2 years. There was no apparent clinical correlation with low urinary estrogen metabolite ratios and above disease specific data. Not surprisingly, there was a significant difference in the number of smokers among cases compared with controls (Table 3). Smokers are considered anyone who has a history of 1 to >40 pack-years. Ninety-four percent of the head and neck cancer patients were tobacco users versus 60% in the control population (P < 0.001). Likewise, alcohol consumption varied between the 2 groups (P < 0.001) with 29 (59%) patients in the head and neck cancer group being drinkers compared with only 9 (18%) patients in the control group. One patient’s history regarding tobacco and alcohol consumption could not be obtained and is not included in Table 3. Drinkers are considered anyone who has a history of regularly drinking a minimum of 2 packs of beer/1 bottle of wine/1 bottle of liquor per week.
Estrogen Metabolites
The absolute values of urinary estrogen metabolites, 2-OHE1, and 16α-OHE1 were normalized using urine creatinine concentrations and were compared between cases and controls. The mean values of 2-OHE1 metabolite level between cases and controls were 9.9 ng/mL (SD ± 5.3) and 10.2 ng/mL (SD ± 14.9), respectively. The average values of 16α-OHE1 between cases and controls were 5.5 ng/mL (SD ± 3.6) and 5.8 ng/mL (SD ± 5.2), revealing no statistical difference in the mean values of 2-OHE1 and 16α-OHE1 between the 2 study groups. However, calculated ratios of 2-OHE1/16α-OHE1 for each patient defined a pattern of low ratios among the head and neck cancer group (Fig 2). When the 2OHE1/16α-OHE1 ratios were stratified into 3 levels: low (0 to 1.00), medium (1.01 to 2.00), and high (>2.01), 15 of 50 H&N cancer patients or 30% had low urinary estrogen metabolite ratios compared with only 2 of 50 or 4% from the control group (P < 0.05) (Table 4). The estimated OR of low urinary estrogen metabolite ratio and head and neck cancer is 11.1 (1.4 to 91.5, 95% CI). When adjusted for tobacco, the OR remained significant at 15.6 (1.1 to 212.5, 95% CI) (P < 0.05. However, adjusted OR for both tobacco and alcohol did not achieve statistical significance (OR = 8.0, 0.3 to 232, 95% CI) (P = 0.225) as a result of the small sample numbers. Because the number of women in the study was small, a separate analysis of male subjects between cases and controls was performed. Thirteen (33%) of 40 versus 2 (5%) of 40 still translated into a clinically significant OR and adjusted OR for tobacco of 10 (1.1 to 89.8, 95% CI) and 13.2 (0.9 to 192.2, 95%
Otolaryngology– Head and Neck Surgery Volume 124 Number 3
YOO et al
245
Table 4. Odds ratios for head and neck cancer associated with urinary estrogen metabolite ratios (2-/16α− OHE1), all subjects Ratio
Low: <1.0 Medium: 1.0-2.0 High: >2.0
Cases (%)
Controls (%)
ORc (95% CI)
P value
15 (30) 16 (30) 20 (40)
2 (4) 27 (54) 21 (42)
11.1 (1.4-91.5) 0.7 (0.3-1.7) 1.0 —
0.024 0.445
ORa (95% CI)
15.6 (1.1-212.5) 0.8 (0.3-2.3) 1.0—
P value
0.040 0.698
ORc, Odds ratios were not adjusted. ORa, Odds ratios were adjusted for cigarette smoking.
Table 5. Odds ratios for head and neck cancer associated with urinary estrogen metabolie ratios (2-/16α-OHE1), male subjects Ratio
Low: <1.0 Medium: 1.0-2.0 High: >2.0
Cases (%)
Controls (%)
ORc (95% CI)
P value
ORa (95% CI)
P value
13 (33) 13 (33) 14 (35)
2 (5) 23 (58) 15 (38)
10.0 (1.1-89.8) 0.8 (0.3-2.4) 1.0 —
0.040 0.666
13.2 (0.9-192.2) 0.9 (0.2-3.1) 1.0 —
0.059 0.822
ORc, Odds ratios were not adjusted. ORa, Odds ratios were adjusted for cigarette smoking.
CI) (P = 0.059), respectively (Table 5). Although there were more controls (27) than cases (17) with medium 2/16α-OHE1 ratios, this difference was not statistically significant. About equal number of H&N cancer patients (20) to controls (21) had high ratio levels. DISCUSSION
Steroid hormones control cellular growth, proliferation, differentiation, and metabolic processes.20 Once a mutation or transformation of protooncogenes has occurred through actions of a carcinogen, estrogens act as promotors exerting cell proliferation and tumor growth.6,20 Through its mitogenic stimulus, estrogens may also predispose nontransformed cells for initiation of carcinogenesis.21 Estradiol, the primary active form of estrogen, is first reversibly oxidized to estrone, which is then converted irreversibly to either 2-hydroxyestrone or 16α-hydroxyestrone.7 On one hand, 16α-OHE1 has been shown to bind covalently to estrogen receptors promoting DNA synthesis with hyperproliferation of epithelial cells.10 On the other hand, 2-OHE1 is devoid of peripheral estrogenic activity.9 It does not bind to estrogen receptors and is subsequently eliminated in urine.9 There is growing evidence in the literature to support the role of steroid hormonal regulation and squamous cell cancer of the upper aerodigestive tract. Studies by Ferguson et al,3 Resta et al,4 and Maiorano et al5 have demonstrated the presence of estrogen receptors and hormone-related proteins from laryngeal cancer cells. Newfield et al13 investigated estrogen metabolism of nor-
mal laryngeal and laryngeal papilloma cells and determined that 16α-hydroxylation was constitutively high in the normal larynx but even higher in laryngeal papillomas infected with HPV (type 11). They also demonstrated that 16α-OHE1 stimulated proliferation in both cells whereas 2-OHE1 was antiproliferative. In HPV-immortalized keratinocytes, which are correlates of the premalignant keratinocytes, estradiol and 16α-OHE1 caused abnormal proliferation and anchorage independent growth.16 In clinical case-control studies evaluating the role of estrogen metabolites, low 2-/16α-OHE1 ratios have been associated with increased risk in cancer of the breast and cervix.6,11,14,15 Elevated urinary levels of 16α-OHE1 have also been found in women with high risk of breast cancer, in women with breast cancer19 and endometrial cancer.12 Our study currently demonstrates that patients with H&N cancer metabolize estrogens differently as well. When absolute levels of 16α-OHE1 among cancer patients were compared with controls, there was no statistical difference. However, 30% (15 of 50) of H&N cancer patients exhibited low 2/16α-OHE1 ratios compared with only 4% (2 of 50) among the healthy population. When adjusted for tobacco use, the OR for low 2/16α-OHE1 ratio remained clinically significant at 15.6. Although our study is limited by sample size, this statistically significant difference suggests that low 2/16α-OHE1 ratios may constitute a risk factor in development of H&N cancer. A randomized prospective trial involving a larger cohort of patients is needed to confirm these results. Our observations are based on measured metabolites in urine that correlates with patients’ overall estrogen
246
Otolaryngology– Head and Neck Surgery March 2001
YOO et al
metabolism. This raises the question whether increased levels of 16α-OHE1 and low 2-/16α-OHE1 metabolite ratios are manifestations of carcinogenesis and aberrant state of cancer cells or they are inherent characteristics of cancer patients with a genetic predisposition. Telang et al21 demonstrated that murine mammary explant cultures and immortalized nontumorigenic mammary epithelial cell cultures treated with chemical carcinogens and transfected with oncogenes, resulted in altered cellular metabolism of estradiol and aberrant hyperproliferation in vitro. Alteration in estrogen metabolism was detected by a significant increase in C16α-hydroxylation with a concomitant decrease in C2 hydroxylation.21 These results suggest that increased production of 16α-OHE1 is a biochemical manifestation of transformed or immortalized cells. On the other hand, indirect evidence supporting genetic predisposition for aberrant estrogen metabolism is the presence of genetic polymorphism in enzymes that convert estrone to 2-OHE1; 2-hydroxylation is catalyzed by a P450 enzyme that has at least 3 major cytochrome P450 subfamilies including P-450 1A1.22 The Cyp 1A1 gene plays an important role in C-2 hydroxylation of estradiol. Three polymorphisms of the Cyp 1A1 gene are known: MSP-1 restriction fragment length polymorphism, a mutation in exon 7, and a novel Cyp 1A1 polymorphism that is exclusively present in African Americans.22 In a study assessing the role of Cyp 1A1 polymorphisms on breast cancer risk and estradiol metabolism, women with the wild type Cyp 1A1 gene showed a significant increase in the 2/16α-OHE1 metabolite ratio after treatment with indole-3-carbinol (I3C-a phytochemical compound that induces 2-hydroxylation).22 Women with MSP1 polymorphism showed no increase in 2/16α-OHE1 metabolite ratio.22 Evaluation of genetic polymorphisms and estrogen metabolism at cellular levels of squamous cell carcinoma are interesting areas of research that may help define and clarify the role of estrogen metabolism in H&N cancer. Estrogen is metabolized by cytochrome P450 enzymes present in various tissues and its metabolism is influenced by many variables including diet.23 Indole3-carbinol (I3C), a phytochemical compound naturally found in cruciferous vegetables, such as broccoli, cauliflower, and brussels sprouts, is a potent inducer of 2hydroxylation.23 The ability of I3C to alter estrogen metabolic pathway, specifically to increase 2-OHE1 levels and exhibit antiestrogenic effects, has been demonstrated by various investigations.13,16,21 I3C abrogated the proliferative effects of estradiol in cell culture.16,21 In mice, I3C reduced the development of papilloma cysts where infected HPV 11 laryngeal tissue was implanted under renal capsules; implantation was 100% in controls vs only 25% in mice fed I3C.13 In a
preliminary study by Rosen et al24 in which 16 recurrent respiratory papillomatosis patients were treated with oral I3C, 6 had complete responses with cessation of papilloma growth and 6 patients had reduced growth rates. Urinary increase in 2-OHE1/16α-OHE1 ratios correlated with improved response. Other environmental factors that can increase 2-hydroxylation are low fat diet25 and exercise.26 In a study looking at urinary 2/16α-OHE1 metabolites of estrogen in premenopausal women, smoking also increased 2-hydroxylation.27 Alcohol consumption had no affect on estrogen metabolism. Altering estrogen metabolic pathway with naturally occurring compounds such as I3C with subsequent effect on H&N cancer has important clinical implications and warrants future study. In conclusion, we have demonstrated for the first time, that head and neck cancer patients metabolize estrogen differently than healthy matched controls. Low urinary 2-/16α-OHE1 estrogen metabolite ratios were detected in a cohort of H&N cancer patients. Our study suggests that low 2-/16α-OHE1 may constitute a risk factor in development of cancers of the upper aerodigestive tract; 2-/16α-OHE1 may also serve as a potential biological marker of individuals at increased risk of cancer. The ability to alter estrogen metabolic pathways through chemoprevention may diminish cancer risk at multiple sites including the upper aerodigestive tract. REFERENCES 1. Muscat JE, Richie JP Jr, Thompson S, et al. Gender differences in smoking and risk for oral cancer. Cancer Res 1996;56:5192-7. 2. Begg CB, Zhang ZF, Sun M, et al. Methodology for evaluating the incidence of second primary cancers with application to smokingrelated cancers from the surveillance, epidemiology, and end results (SEER) program. Am J Epidemiol 1995;142:653-65. 3. Ferguson BJ, Hudson WR, McCarty KS Jr. Sex steroid receptor distribution in the human larynx and laryngeal carcinoma. Arch Otolaryngol Head Neck Surg 1987;113:1311-5. 4. Resta J, Marsigliante S, Leo G, et al. Molecular biopathology of metaplastic, dysplastic, and neoplastic laryngeal epithelium. Acta Otolaryngol (Stockh) 1997;Suppl527:39-42. 5. Maiorano E, Botticella MA, Marzullo A, et al. Expression of ERD5 and EGFr in laryngeal carcinoma and premalignant epithelium. Acta Otolaryngol (Stockh) 1997;Suppl 527:95-9. 6. Sepkovic DW, Bradlow HL, Ho G, et al. Estrogen metabolite ratios and risk assessment of hormone-related cancers: assay, validation, and prediction of cervical cancer risk. Ann N Y Acad Sci 1995;768:312-6. 7. Lustig RH, Bradlow HL, Fishman J. Estrogen metabolism in disorders of nutrition and dietary composition. In: The menstrual cycle and its disorders. Berlin/New York: Springer-Verlag. 8. Martucci C, Fishman J. Direction of estradiol metabolism as a control of its hormonal action: uterotropic activity of estradiol metabolites. Endocrinology 1977;101:1709-15. 9. Bradlow HL, Telang NT, Sepkovic DW, et al. 2-Hydroxyestrone: the “good” estrogen. J Endocrinol 1996;150:5259-63. 10. Swaneck, GE, Fishman J. Covalent binding of the endogenous estrogen 16–hydroxyestrone to estradiol receptors in human breast cancer cells: characterization and intranuclear localization. Proc Natl Acad Sci U S A 1988;85:7831-5. 11. Schneider J, Kinne D, Fracchia A, et al. Abnormal oxidative
Otolaryngology– Head and Neck Surgery Volume 124 Number 3
12. 13.
14. 15. 16. 17. 18.
19.
YOO et al
metabolism of estradiol in women with breast cancer. Proc Soc Acad Sci U S A 1982;79:3047-51. Fishman J, Schneider J, Hershcope RJ, et al. Increased estrogen 16 alpha-hydroxylase activity in women with breast and endometrial cancer. J Steroid Biochem Mol Biol 1984;20:1077-81. Newfield L, Goldsmith A, Bradlow HL, et al. Estrogen metabolism and human papillomavirus-induced tumors of the larynx: chemo-prophylaxis with indole-3-carbinol. Anticancer Res 1993; 13:337-42. Kabat GC, Chang CJ, Sparano JA, et al. Urinary estrogen metabolites and breast cancer: a case-control study. Cancer Epidemiol Biomarkers Prev 1997;6:500-4. Zheng W, Dunning L, Gong Y, et al. Ratios of urinary 20 hydroxyestrone to 16 alpha-hydroxyestrone are lower in breast cancer patients than controls. Cancer Epidmiol Biomarkers Prev 1998;7:85-6. Newfield L, Bradlow HL, Sepkovic DW, et al. Estrogen metabolism and the malignant potential of human papillomavirus immortalized keratinocytes. Soc Exp Biol Med 1998;322-6. Bradlow HL, Sepkovic DW, Klug T, et al. Application of an improved ELISA assay to the analysis of urinary estrogen metabolites. Steroids 1998;63:406-13. Ziegler RG, Rossi SC, Fears TR, et al. Quantifying estrogen metabolism: an evaluation of the reproducibility and validity of enzyme immunoassays for 2-hydroxyestrone and 16α-hydroxyestrone in urine. Environ Health Perspect 1997;105:607-14. Meilahn EN, Stavola BD, Allen DS, et al. Do urinary estrogen
20. 21. 22. 23. 24. 25. 26. 27.
metabolites predict breast cancer? Guernsey III cohort follow-up. Br J Cancer 1998;78:1250-5. Sekeris CE. Hormonal steroids act as tumour promoters by modulating oncogene expression. J Cancer Res Clin Oncol 1991;117:96-101. Telang NT, Katdare M, Bradlow HL, et al. Estradiol metabolism: an endocrine biomarker for modulation of human mammary carcinogenesis. Environ Health Perspect 1997;105:559-64. Taioli E, Bradlow HL, Garbers S, et al. Role of estradiol metabolism and CYP1A1 polymorphisms in breast cancer risk. Cancer Detect Prev 1999;23:232-7. Michnovicz JJ, Bradlow HL. Induction of estradiol metabolism by dietary indole-3-carbinol in humans. J Natl Cancer Inst 1990;82:947-9. Rosen CA, Woodson GE, Thompson JW, et al. Preliminary results of use of indole-3-carbinol for recurrent respiratory papillomatosis. Otolaryngol Head Neck Surg 1998;118:810-5. Fishman J, Boyar RM, Hellman L, Influence of body weight on estradiol metabolism in young women. J Clin Endocrinol Metab 1975;41:989-91. Snow RC, Barberi RL, Frisch, RE, Estrogen 2-hydroxylase oxidation and menstrual function among elite oarswomen. J Clin Endocrinol Metab 1989;69:369-76. Michnovicz JJ, Hershcopf RJ, Naganuma H, et al. Increased 2hydroxylation of estradiol as a possible mechanism for the antiestrogenic effect of cigarette smoking. N Engl J Med 1986; 315:1305-9.
Access to Otolaryngology–Head and Neck Surgery Online is reserved for print subscribers! Full-text access to Otolaryngology–Head and Neck Surgery Online is now available for all print subscribers. To activate your individual online subscription, please visit Otolaryngology–Head and Neck Surgery Online, point your browser to http://www.mosby.com/oto, follow the prompts to activate your online access, and follow the instructions. To activate your account, you will need your subscriber account number, which you can find on your mailing label (note: the number of digits in your subscriber account number varies from 6 to 10). See the example below in which the subscriber account number has been circled: Sample mailing label This is your subscription account number
247
**************************3-DIGIT 001 SJ P1 FEB00 J027 C: 1 1234567-89 U 05/00 Q: 1 J. H. DOE 531 MAIN ST CENTER CITY, NY 10001-001
Personal subscriptions to Otolaryngology–Head and Neck Surgery Online are for individual use only and may not be transferred. Use of Otolaryngology–Head and Neck Surgery Online is subject to agreement to the terms and conditions as indicated online.