The objective of this research was to evaluate the association between serum carotenoids and cervical intraepithelial neoplasia (CIN) among Southwestern American Indian women. Cases were American Indian women with biopsy-proven CIN II/III cervical lesions (n = 81) diagnosed between November 1994 and October 1997. Controls were American Indian women from the same clinics with normal cervical epithelium (n = 160). All of the subjects underwent interviews and laboratory evaluations. Interviews evaluated demographic information, sexual history, and cigarette smoking. Serum concentrations of α-carotene, β-carotene, β-cryptoxanthin, lycopene, and lutein/zeaxanthin were measured by high performance liquid chromatography. Cervical human papillomavirus infection was detected using a PCR-based test. Increasing levels of α-carotene, β-cryptoxanthin, and lutein/zeaxanthin were associated with decreasing risk of CIN II/III. In addition, the highest tertiles of β-cryptoxanthin (odds ratio = 0.39, 95% confidence interval = 0.17–0.91) and lutein/zeaxanthin (odds ratio = 0.40, 95% confidence interval = 0.17–0.95) were associated with the lowest risk of CIN. In conclusion, specially targeted intervention efforts to increase consumption of fruits and vegetables may protect Southwestern American Indian women from developing CIN.

Although extensive evidence exists for a role of fruits and vegetables in lowering the risk of human cancers (1), the association of carotenoids and cancer is considerably less clear. Carotenoids are obtained from fruits, vegetables, fortified foods, and nutritional supplements, and many observational and experimental studies support a cancer preventive role for these compounds (1, 2). However, in randomized clinical trials of β-carotene, either no effect (3) or increased cancer incidence among high-risk subjects (4, 5) was reported. To date, the majority of the research has focused on β-carotene and epithelial cell cancer, whereas relatively few studies have investigated the association of other carotenoids with cancer. In addition, a major limitation of many observational studies is reliance on dietary self-report, which has many potential sources of error and bias (6). Conversely, serum carotenoid concentrations provide an objective exposure assessment and improved confidence in observed associations.

Cervical cancer, a common epithelial cell cancer, is the second most common cancer in women (7) and disproportionately affects minority women (8). Although cervical cancer rates have declined in recent years, American Indian women in New Mexico continue to have higher rates of cervical cancer (11.1/100,000) compared with United States black (10.9/100,000) and white women (8.4/100,000) during 1994–1998 (8). Although much CIN3 research has focused on infectious, reproductive, and contraceptive risk factors, less is known about the role of dietary factors in the development of CIN. Studies (9, 10, 11, 12) of CIN and carotenoids indicate that low concentrations of selected serum carotenoids (α-carotene, β-carotene, lycopene, zeaxanthin, and β-cryptoxanthin) are associated with an increased risk of CIN. Although a dietary survey of one Southwestern American Indian tribe reported that fruits and vegetables were consumed less than once per day (13), no studies have been done to evaluate serum carotenoid concentrations associated with the increased risk of CIN in this population. The objective of this study was to investigate the association of serum concentrations of α-carotene, β-carotene, β-cryptoxanthin, lycopene, and lutein/zeaxanthin with the risk of CIN among Southwestern American Indian women.

This study is a subanalysis of a larger case control study evaluating risk factors for CIN (14). We briefly present the methods for this subanalysis. We enrolled American Indian women who were evaluated at three IHS facilities in New Mexico. Cases were women with histologically proven CIN II and CIN III, ages 18–45 years, and not pregnant. Controls were women with normal cervical epithelium, ages 18–45 years, and not pregnant. In addition, controls were required to have had histories of all normal Pap tests, documented by review of their medical records. Controls for this subanalysis were randomly chosen from all of the controls in the larger study in a 2 controls:1 case ratio.

We enrolled subjects during a 3-year period, November 1994 through October 1997. Study participation among eligible study subjects was high, with only 48 refusals (6.2% of the total eligible subjects who were invited into the larger study). All of the study subjects were paid 20 dollars for their time for participation in this investigation. This protocol was approved by the National IHS Human Research Review Board as well as by the appropriate IHS hospital health boards.

All of the participants in the larger study were asked to provide a blood sample (99% compliance). Whole blood was obtained using a Vacutainer system and an empty vacuum-sealed blood tube without preservative. The blood tube was immediately placed on ice and protected from light in an opaque container. At the end of the work day (range, 1–6 h after collection), serum was separated by centrifugation at 3000 rpm for 15 min, aliquoted into cryovials, and immediately stored at −70°C. Plasma concentrations of α-carotene, β- carotene, lycopene, lutein, and β-cryptoxanthin were measured in one batch after completion of the study by high performance liquid chromatography using visible, UV, and fluorescence detection (15). A hexane extract of plasma containing tocopherol acetate and retinol acetate as internal standards was evaporated, reconstituted, and chromatographed. Quantitation was accomplished by comparing analyte internal standard ratios to those from a standard curve prepared from the data produced by chromatography of pure standards spiked with the internal standards. Standardization was done for each batch of samples. Quality control analysis was performed using aliquots of a plasma pool with known mean and SD values. A high and low quality control sample evaluated with each assay was required to have a mean value within 2.5 SDs of the mean value established previously.

All of the study subjects had a cervical specimen taken with a dacron cervical swab of the endocervix and ectocervix and placed in standard transport medium (Digene Diagnostics, Silver Spring, MD) for identification of HPV. The presence of HPV was determined using a reverse blot method described previously (16).

We used logistic regression to model the relationship between CIN and each of the five serum carotenoids: α-carotene, β-carotene, β-cryptoxanthin, lycopene, and lutein/zeaxanthin. For these analyses, we classified our carotenoid data into tertiles based on the distribution of the serum carotenoid values among our control sample. A crude OR and 95% CI was determined for the tertiles for each of the five serum carotenoids, and the association between each carotenoid and CIN was assessed by χ2 test for trend. For all of the adjusted analysis, we controlled for age and cervical infection with any HPV type, because these confounders were biologically plausible and have been noted to confound the relationship between carotenoids and CIN in previous studies (9, 10). We identified additional potential confounders based on biological plausibility, their association with both carotenoid levels and CIN, and a change from the crude OR of 10% or more. Although most but not all of the confounders met the criteria above for each of the five carotenoids, we chose to build a final model for each of the carotenoids that adjusted for the same confounders. The additional confounders were household income (in thousands of dollars, <$10, $10–$19, ≥$20) and urban versus rural residence. Smoking is well known to reduce serum carotenoid concentrations (17) and is, therefore, a potential confounder. Nonetheless, smoking was not included in our final model because <15% of our study subjects were smokers, they smoked an average of five or less cigarettes per day, and smoking had no effect on the crude OR. In addition, we also evaluated smoking as a possible effect modifier of the relationship between each carotenoid and CIN and found no effect modification. We also investigated seasonality as a potential confounder and found no relationship with serum carotenoid concentrations. STATA software was used for all of the analyses (version 6.0; STATA, College Station, TX).

Most of the case and control women were 22–30 years of age, married or living with a partner, and had completed some education beyond high school (Table 1). The majority of the American Indian women enrolled in this study was from southwestern tribes and were of full Indian blood quantum. Cases were more likely to be single, have low annual family income, and state “other” reasons for having the initial Pap smear that led to their CIN diagnosis. In this subanalysis, HPV was a strong risk factor for CIN (crude OR = 9.0, 95% CI = 4.7–17.5).

Mean values in μg/dl (SDs) for the carotenoids among cases and controls, respectively, were 14.9 (7.2) and 17.2 (9.2) for α-carotene, 31.2 (18.2) and 36.1 (23.3) for β-carotene, 8.8 (6.6) and 9.4 (5.3) for β-cryptoxanthin, 33.6 (11.5) and 36.5 (11.3) for lycopene, and 24.1 (8.1) and 26.3 (8.4) for lutein/zeaxanthin.

Table 2 shows the crude and adjusted association of serum carotenoid concentrations with CIN. In the crude analysis, increasing tertiles of α-carotene and lutein/zeaxanthin were associated with a decreasing risk of CIN. After adjustment for age, HPV status, income, and urban versus rural residence, a trend for decreasing risk of CIN with increasing tertiles of β-cryptoxanthin as well as α-carotene and lutein/zeaxanthin was noted. The highest tertiles of β-cryptoxanthin and lutein/zeaxanthin were associated with a 60% decreased risk of CIN compared with the lowest tertile.

We found that decreasing serum carotenoid concentrations were associated with increasing risk of CIN. After adjustment for confounding, we noted that the highest tertiles of β-cryptoxanthin and lutein/zeaxanthin were associated with a 60% decreased risk of CIN.

Mean values of α-carotene, β-carotene, and lutein/zeaxanthin in our study of American Indian women were higher compared with values from studies in United States (18) and Latin American (9) women. Mean β-cryptoxanthin and lycopene values were similar to those in the United States study (18) but differed from the Latin American study (9).

Although previous reports have noted an increased risk of CIN with low plasma concentrations of carotenoids [α-carotene (18), β-carotene (18, 19, 20), lycopene (11, 12, 19), β-cryptoxanthin (11), and zeaxanthin (11)], these studies have been limited by lack of adjustment for HPV, a well-known independent cause of cervical neoplasia. Results of studies of serum carotenoids and cervical neoplasia that adjusted for HPV are not entirely consistent. Similar to our study findings, Nagata et al.(10) noted a decreased risk of CIN with higher levels of α-carotene in Japanese women. Potischman et al.(9) noted a trend of decreasing risk of invasive cervical cancer with higher levels of β-carotene among Latin American women. Our data suggest a similar association with CIN, but our results did not reach statistical significance. Our finding of a decreased risk of CIN associated with higher levels of β-cryptoxanthin and lutein/zeaxanthin has not been noted in other HPV-adjusted studies that have evaluated these carotenoids (9, 10). Unlike our study, Nagata et al.(10) reported an association between high levels of lycopene and a decreased risk of CIN. In summary, high serum carotenoids appear to be associated with a decreased risk of cervical neoplasia, although the association with individual carotenoids may differ because of adjustment for different confounders and different dietary habits among diverse populations.

Although many previous studies have suggested the role of carotenoids as a risk factor for CIN, only one earlier study evaluated this relationship in American Indian women. This study (21) documented a nonsignificant increased risk of CIN among women with the lowest levels of carotene. Our study suggests the importance of low serum carotenoid levels and increased risk of CIN among American Indian women, a population at particularly high risk of cervical neoplasia.

Low serum carotenoid concentrations may be associated with the risk of developing CIN for several reasons. Carotenoids are found in deeply pigmented fruits and vegetables such as carrots, tomatoes, cruciferous vegetables (broccoli and spinach), peaches, oranges, and cantaloupes (22). Thus, serum carotenoid concentrations may serve as a biomarker of fruit and vegetable intake. For example, lutein/zeaxanthin may be an indicator of cruciferous vegetable intake with other compounds in these vegetables producing the biological effect of reducing the risk of CIN (23). Alternatively, serum carotenoids may play a direct role in the pathogenesis of CIN. Serum carotenoids are well known for their antioxidant properties and may scavenge free radicals that can induce carcinogenesis through DNA damage (17). In addition, low serum carotenoid concentrations may interfere with DNA repair associated with malignant cell transformation (17).

Our study had several potential limitations. Some of these limitations include lack of blinding of interviewers to case or control status of the subjects. We included American Indian women from tribes outside of the Southwest (∼10% of all of the study subjects). The proportion of nonsouthwestern tribal subjects is not large enough to allow broad generalization of our findings to all of the American Indian groups. We found high correlation among the five carotenoids we measured (especially α-carotene and β-carotene; r2 = 0.69), making it difficult to isolate the effect of one individual carotenoid compound. This may limit our conclusions about the effects of individual carotenoids on the risk of CIN. Lastly, analyses of serum carotenoids are often adjusted for serum lipid levels, because carotenoids are transported by cholesterol-rich lipoproteins in circulation. Therefore, higher lipoprotein concentrations result in higher carotenoid concentrations, independent of dietary intake or total body pool. We were unable to perform this adjustment, because serum lipid levels were not available on our subjects. Given that no evidence exists that serum cholesterol levels are associated with CIN, this lack of adjustment should introduce only random error into our exposure measurement and may attenuate our ability to find associations.

Because the role of infectious, reproductive and contraceptive risk factors in the development of CIN has been better defined over the past decade, the impact of dietary risk factors has received more attention. Our study confirmed previous research documenting a significant role of low serum carotenoids in the development of CIN. In addition, our findings add to the growing body of evidence indicating diets high in fruits and vegetables (and carotenoids in food) reduce the risk of several cancers (1, 2). American Indian women in the Southwest may have dietary patterns that put them at unusually high risk for CIN and require specially targeted intervention efforts.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

        
1

This research was supported by National Cancer Institute Grants R01-CA-55348 and Training Grant T32-CA09168. T. M. B. was supported by a Faculty Research Award from the American Cancer Society.

                
3

The abbreviations used are: CIN, cervical intraepithelial neoplasia; OR, odds ratio; CI, confidence interval; HPV, human papillomavirus; IHS, Indian Health Service.

Table 1

Demographic characteristics of study subjects, American Indian dysplasia study, 1994–1997

Controls (n = 160) No. (%)Cases (n = 81) No. (%)Ps
Age (years)    
 18–21 29 (18.1) 17 (21.0) 0.86 
 22–30 99 (61.9) 49 (60.5)  
 31–45 32 (20.0) 15 (18.5)  
Years of education    
 12 years or below 47 (29.3) 31 (38.3) 0.35 
 13–15 years 89 (55.6) 38 (46.9)  
 16+ years 24 (15.1) 12 (14.8)  
Annual family income    
 <$10,000 38 (23.8) 32 (39.5) 0.005 
 $10,000–$19,999 41 (25.6) 25 (30.9)  
 ≥$20,000 81 (50.6) 24 (29.6)  
Marital status    
 Single, never married 46 (28.8) 26 (32.1) 0.03 
 Married or with partner 106 (66.2) 45 (55.6)  
 Divorced 6 (3.8) 3 (3.7)  
 Separated 2 (1.2) 7 (8.6)  
Tribal affiliation    
 Southwestern 146 (91.3) 72 (88.9) 0.56 
 Other 14 (8.7) 9 (11.1)  
Blood quantuma    
 Full 119 (74.4) 63 (77.8) 0.26 
 ¾ 15 (9.4) 3 (3.7)  
 ½ 17 (10.6) 12 (14.8)  
 ¼ 7 (4.4) 1 (1.2)  
 <¼ 2 (1.2) 2 (2.5)  
Reason for screening    
 Family planning 2 (2.5) <0.001 
 Annual exam 140 (87.5) 45 (55.6)  
 Suspected STD 1 (0.6) 1 (1.2)  
 Planned follow-up 1 (0.6) 9 (10.8)  
 Otherb 18 (11.3) 24 (29.9)  
Controls (n = 160) No. (%)Cases (n = 81) No. (%)Ps
Age (years)    
 18–21 29 (18.1) 17 (21.0) 0.86 
 22–30 99 (61.9) 49 (60.5)  
 31–45 32 (20.0) 15 (18.5)  
Years of education    
 12 years or below 47 (29.3) 31 (38.3) 0.35 
 13–15 years 89 (55.6) 38 (46.9)  
 16+ years 24 (15.1) 12 (14.8)  
Annual family income    
 <$10,000 38 (23.8) 32 (39.5) 0.005 
 $10,000–$19,999 41 (25.6) 25 (30.9)  
 ≥$20,000 81 (50.6) 24 (29.6)  
Marital status    
 Single, never married 46 (28.8) 26 (32.1) 0.03 
 Married or with partner 106 (66.2) 45 (55.6)  
 Divorced 6 (3.8) 3 (3.7)  
 Separated 2 (1.2) 7 (8.6)  
Tribal affiliation    
 Southwestern 146 (91.3) 72 (88.9) 0.56 
 Other 14 (8.7) 9 (11.1)  
Blood quantuma    
 Full 119 (74.4) 63 (77.8) 0.26 
 ¾ 15 (9.4) 3 (3.7)  
 ½ 17 (10.6) 12 (14.8)  
 ¼ 7 (4.4) 1 (1.2)  
 <¼ 2 (1.2) 2 (2.5)  
Reason for screening    
 Family planning 2 (2.5) <0.001 
 Annual exam 140 (87.5) 45 (55.6)  
 Suspected STD 1 (0.6) 1 (1.2)  
 Planned follow-up 1 (0.6) 9 (10.8)  
 Otherb 18 (11.3) 24 (29.9)  
a

Blood quantum is amount of Indian blood reported by study subjects.

b

Other reasons for screening included postpartum visits, pregnancy testing, and vaginitis.

Table 2

Serum carotenoid concentrations as associated with cervical intraepithelial neoplasia; American Indian dysplasia study, 1994–1997

CarotenoidsControls n = 160 No. (%)Cases n = 81 No. (%)CrudeAdjusted**
OR95% CIOR95% CI
α-Carotene (μg/dl)       
 6.5–12.7 53 (33.1) 42 (51.8) 1.0  1.0  
 12.8–18.0 53 (33.1) 21 (25.9) 0.50 0.26–0.96 0.52 0.24–1.12 
 18.1–87.3 54 (33.8) 18 (22.2) 0.42 0.23–0.86 0.46 0.21–1.00 
 Test for trend   P = 0.01  P = 0.04  
β-Carotene (μg/dl)       
 11–24 56 (35.0) 32 (39.5) 1.0  1.0  
 25–38 51 (31.9) 33 (40.7) 1.13 0.61–2.1 0.96 0.45–2.03 
 39–174 53 (33.1) 16 (19.8) 0.53 0.26–1.1 0.46 0.20–1.07 
 Test for trend   P = 0.10  P = 0.08  
β-Cryptoxanthin (μg/dl)       
 2.2–6.4 55 (34.4) 34 (42.0) 1.0  1.0  
 6.5–10.3 52 (32.5) 31 (38.3) 0.96 0.52–1.79 0.80 0.38–1.69 
 10.4–30.3 53 (33.1) 16 (19.7) 0.49 0.24–0.99 0.39 0.17–0.91 
 Test for trend   P = 0.06  P = 0.03  
Lycopene (μg/dl)       
 14.8–30.7 53 (33.1) 37 (45.7) 1.0  1.0  
 30.8–40.7 54 (33.8) 25 (30.9) 0.66 0.35–1.25 0.88 0.41 –1.89 
 40.8–69.8 53 (33.1) 19 (23.4) 0.51 0.26–1.00 0.85 0.38–1.91 
 Test for trend   P = 0.05  P = 0.69  
Lutein/Zeaxanthin (μg/dl)       
 12–21 49 (30.6) 34 (42.0) 1.0  1.0  
 22–28 55 (34.4) 31 (38.3) 0.81 0.44–1.51 0.60 0.28–1.28 
 29–76 56 (35.0) 16 (19.7) 0.41 0.20–0.84 0.40 0.17–0.95 
 Test for trend   P = 0.02  P = 0.03  
CarotenoidsControls n = 160 No. (%)Cases n = 81 No. (%)CrudeAdjusted**
OR95% CIOR95% CI
α-Carotene (μg/dl)       
 6.5–12.7 53 (33.1) 42 (51.8) 1.0  1.0  
 12.8–18.0 53 (33.1) 21 (25.9) 0.50 0.26–0.96 0.52 0.24–1.12 
 18.1–87.3 54 (33.8) 18 (22.2) 0.42 0.23–0.86 0.46 0.21–1.00 
 Test for trend   P = 0.01  P = 0.04  
β-Carotene (μg/dl)       
 11–24 56 (35.0) 32 (39.5) 1.0  1.0  
 25–38 51 (31.9) 33 (40.7) 1.13 0.61–2.1 0.96 0.45–2.03 
 39–174 53 (33.1) 16 (19.8) 0.53 0.26–1.1 0.46 0.20–1.07 
 Test for trend   P = 0.10  P = 0.08  
β-Cryptoxanthin (μg/dl)       
 2.2–6.4 55 (34.4) 34 (42.0) 1.0  1.0  
 6.5–10.3 52 (32.5) 31 (38.3) 0.96 0.52–1.79 0.80 0.38–1.69 
 10.4–30.3 53 (33.1) 16 (19.7) 0.49 0.24–0.99 0.39 0.17–0.91 
 Test for trend   P = 0.06  P = 0.03  
Lycopene (μg/dl)       
 14.8–30.7 53 (33.1) 37 (45.7) 1.0  1.0  
 30.8–40.7 54 (33.8) 25 (30.9) 0.66 0.35–1.25 0.88 0.41 –1.89 
 40.8–69.8 53 (33.1) 19 (23.4) 0.51 0.26–1.00 0.85 0.38–1.91 
 Test for trend   P = 0.05  P = 0.69  
Lutein/Zeaxanthin (μg/dl)       
 12–21 49 (30.6) 34 (42.0) 1.0  1.0  
 22–28 55 (34.4) 31 (38.3) 0.81 0.44–1.51 0.60 0.28–1.28 
 29–76 56 (35.0) 16 (19.7) 0.41 0.20–0.84 0.40 0.17–0.95 
 Test for trend   P = 0.02  P = 0.03  

aAdjusted for age, cervical infection with any type of HPV, income, and urban vs rural residence.

We thank Tim Carlson at Pacific Biometrics, Inc., for the timely analyses of our carotenoid specimens. We also thank David Doody at the Fred Hutchinson Cancer and Research Center and Kathy Kimler Altobelli at the New Mexico Tumor Registry for invaluable statistical assistance. In addition, we acknowledge all of the IHS personnel who helped to facilitate this study in the clinic sites. In particular, Drs. Eve Espey, Tony Ogburn, Alan Waxman, Charles Q. North, and Jill Miller, and Rose Rowan were particularly valuable to the success of the study. Drs. Gary Escudera, William Freeman, and Tim Fleming assisted the study team with protocol approvals. George Montoya, Pat Stauber, Sarah Simplicio, and Dr. Phil Garry provided valuable assistance and advice for the clinical nutrition specimens. Darren Schaffer, Helena Frank, and Evelyn Hood were very valuable as laboratory assistants. The investigators also recognize the valuable contributions of Fred Ettcity and Ray Rodgers of the IHS. The guidance provided by the IHS Service Unit Health Boards was also very helpful to the project.

1
Steinmetz K. A., Potter J. D. Vegetables, fruit, and cancer prevention: a review.
J. Am. Diet. Assoc.
,
96
:
1027
-1039,  
1996
.
2
Steinmetz K. A., Potter J. D. Vegetables, fruit, and cancer II: Mechanisms.
Cancer Causes Control
,
2
:
427
-442,  
1991
.
3
Hennekens C. H., Buring J. E., Manson J. E., Stampfer M., Rosner B., Cook N. R., Belanger C., LaMotte F., Gaziano J. M., Ridker P. M., Willett W., Peto R. Lack of effect of long-term supplementation with β carotene on the incidence of malignant neoplasms and cardiovascular disease.
N. Engl. J. Med.
,
334
:
1145
-1149,  
1996
.
4
Albanes D., Heinonen O. P., Huttunen J. K., Taylor P. R., Virtamo J., Edwards B. K., Haapakoski J., Rautalahti M., Hartman A. M., Palmgren J., et al Effects of α-tocopherol and β-carotene supplements on cancer incidence in the α-Tocopherol β-Carotene Cancer Prevention Study.
Am. J. Clin. Nutr.
,
62(Suppl. 6)
:
1427S
-1430S,  
1995
.
5
Omenn G. S., Goodman G. E., Thornquist M. D., Balmes J., Cullen M. R., Glass A., Keogh J. P., Meyskens F. L., Jr, Valanis B., Williams J. H., Jr, Barnhart S., Cherniack M. G., Brodkin C. A., Hammar S. Risk factors for lung cancer and for intervention effects in CARET, the β-Carotene and Retinol Efficacy Trial.
J. Natl. Cancer Inst.
,
88
:
1550
-1559,  
1996
.
6
Patterson, R. E., and Pietinen, P. Assessment of nutritional status in populations and individuals. In: B. Margetts and L. Arab. (eds.), Public Health Nutrition. Blackwell Scientific Nutrition (in press).
7
Parkin D. M., Pisani P., Ferlay J. Estimates of the worldwide incidence of eighteen major cancers in 1985.
Int. J. Cancer
,
54
:
594
-606,  
1993
.
8
Surveillance, Epidemiology, and End Results (SEER) Program Public-Use Data (1973–1998). National Cancer Institute, Division of Cancer Control and Population Sciences, Bethesda, MD, Surveillance Research Program, Cancer Statistics Branch, 2001.
9
Potischman N., Herrero R., Brinton L. A., Reeves W. C., Stacewicz-Sapuntzakis M., Jones C. J., Brenes M. M., Tenorio F., de Britton R. C., Gaitan E. A case control study of nutrient status and invasive cervical cancer. II serologic indicators.
Am. J. Epidemiol.
,
134
:
1347
-1355,  
1991
.
10
Nagata C., Shimizu H., Yoshikawa H., Noda K., Nozawa S., Yajima A., Sekiya S., Sugimori H., Hirai Y., Kanazawa K., Sugase M., Kawana T. Serum carotenoids and vitamins and risk of cervical dysplasia from a case control study in Japan.
Br. J. Cancer
,
81
:
1234
-1237,  
1999
.
11
Peng Y. M., Peng Y. S., Childers J. M., Hatch K. D., Roe D. J., Lin Y., Lin P. Concentrations of carotenoids, tocopherols, and retinol in paired plasma and cervical tissue of patients with cervical cancer, precancer, and noncancerous diseases.
Cancer Epidemiol. Biomark. Prev.
,
7
:
347
-350,  
1998
.
12
VanEenwyk J., Davis F. G., Bowen P. E. Dietary and serum carotenoids and cervical intraepithelial neoplasia.
Int. J. Cancer
,
48
:
34
-38,  
1991
.
13
Ballew C., White L. L., Strauss K. F., Benson L. J., Mendlein J. M., Mokdad A. H. Intake of nutrients and food sources of nutrients among the Navajo: findings from the Navajo Health and Nutrition Survey.
J. Nutr.
,
127(Suppl.)
:
2085S
-2093S,  
1997
.
14
Schiff M., Becker T. M., Masuk M., van Asselt-King L., Wheeler C. M., Altobelli K. K., North C. Q., Nahmias A. J. Risk factors for cervical intraepithelial neoplasia in Southwestern American Indian women.
Am. J. Epideimiol.
,
152
:
716
-726,  
2000
.
15
Talwar D., Ha T. K. K., Cooney J., Brownlee C., O’Reilly D. S. J. A routine method for the simultaneous measurement of retinol, α-tocopherol and five serum carotenoids in human plasma by reverse phase HPLC.
Clin. Chim. Acta.
,
270
:
85
-100,  
1998
.
16
Gravitt P. E., Peyton C. L., Apple R. J., Wheeler C. M. Genotyping of 27 human papillomavirus types by using L1 consensus PCR products by a single-hybridization, reverse line blot detection method.
J. Clin. Microbiol.
,
36
:
3020
-3027,  
1998
.
17
Rock C. L. Carotenoids: biology and treatment.
Pharmacol. Ther.
,
75
:
185
-197,  
1997
.
18
Batieha A. M., Armenian H. K., Norkus E. P., Morris J. S., Spate V. E., Comstock G. W. Serum micronutrients and the subsequent risk of cervical cancer in a population-based nested case-control study.
Cancer Epidemiol. Biomark. Prev.
,
2
:
335
-339,  
1993
.
19
Palan P., Mikhail M. S., Goldberg G., Basu J., Runowicz C. D., Romney S. L. Plasma levels of B-carotene, lycopene, canthaxanthin, retinol, and a- and t-tocopherol in cervical intraepithelial neoplasia and cancer.
Clin. Cancer Res.
,
2
:
181
-185,  
1996
.
20
Brock K. E., Berry G., Mock P. A., MacLennan R., Truswell A. S., Brinton L. A. Nutrients in diet and plasma and risk of in situ cervical cancer.
J. Natl. Cancer Inst.
,
80
:
580
-585,  
1988
.
21
Buckley D. I., McPherson R. S., North C. Q., Becker T. M. Dietary micronutrients and cervical dysplasia in southwestern American Indian women.
Nutr. Cancer
,
17
:
179
-185,  
1992
.
22
Smith S. A., Campbell D. R., Elmer P. J., Martini M. C., Slavin J. L., Potter J. D. The University of Minnesota Cancer Prevention Research Unit vegetable and fruit classification scheme (United States).
Cancer Causes Control
,
6
:
292
-302,  
1995
.
23
Campbell D. R., Gross M. D., Martini M. C., Grandits G. A., Slavin J. L., Potter J. D. Plasma carotenoids as biomarkers of vegetable and fruit intake.
Cancer Epidemiol. Biomark. Prev
,
3
:
493
-500,  
1994
.