Skip Navigation


American Journal of Epidemiology Advance Access originally published online on July 17, 2006
American Journal of Epidemiology 2006 164(6):556-566; doi:10.1093/aje/kwj233
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
164/6/556    most recent
kwj233v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (7)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Maserejian, N. N.
Right arrow Articles by Joshipura, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Maserejian, N. N.
Right arrow Articles by Joshipura, K.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

American Journal of Epidemiology Copyright © 2006 by the Johns Hopkins Bloomberg School of Public Health All rights reserved; printed in U.S.A.

Original Contribution

Prospective Study of Fruits and Vegetables and Risk of Oral Premalignant Lesions in Men

Nancy Nairi Maserejian1,2,3, Edward Giovannucci1,4,5, Bernard Rosner4, Athanasios Zavras2 and Kaumudi Joshipura1,2,6

1 Department of Epidemiology, Harvard School of Public Health, Boston, MA
2 Department of Oral Health Policy and Epidemiology, Harvard School of Dental Medicine, Boston, MA
3 New England Research Institutes, Watertown, MA
4 Channing Laboratory, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, MA
5 Department of Nutrition, Harvard School of Public Health, Boston, MA
6 Division of Dental Public Health, School of Dentistry, University of Puerto Rico Medical Sciences, San Juan, PR

Correspondence to Dr. Kaumudi Joshipura, Division of Dental Public Health, School of Dentistry, Office A141E, Medical Sciences Campus, GPO Box 3650367, San Juan, PR 00936-5067 (e-mail: kjoshipura{at}rcm.upr.edu; kaumudi_joshipura{at}hms.harvard.edu).

Received for publication December 23, 2005. Accepted for publication March 9, 2006.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
The authors prospectively evaluated fruit and vegetable consumption and the incidence of oral premalignant lesions among 42,311 US men in the Health Professionals Follow-up Study. Diet was assessed every 4 years by food frequency questionnaires. The authors confirmed 207 cases of clinically or histopathologically diagnosed oral premalignant lesions occurring between 1986 and 2002. Multivariate-adjusted relative risks were calculated from proportional hazards models. Significant inverse associations were observed with citrus fruits, citrus fruit juice, and vitamin-C-rich fruits and vegetables, indicating 30–40% lower risks with greater intakes (e.g., citrus fruit juice quintile 5 vs. quintile 1 relative risk = 0.65, 95% confidence interval: 0.42, 0.99). Inverse associations with fruits did not vary by smoking status and were stronger in analyses of baseline consumption, with a 10-year lag time to disease follow-up (quintile 5 vs. quintile 1 relative risk = 0.41, 95% confidence interval: 0.20, 0.82; p = 0.01). No associations were observed with total vegetables or with ß-carotene-rich or lycopene-rich fruits and vegetables. For current smokers, green leafy vegetables (ptrend = 0.05) and ß-carotene-rich fruits and vegetables (ptrend = 0.02) showed significant linear trends of increased risk (one additional serving/day relative risk = 1.7). The risk of oral premalignant lesions was significantly reduced with higher consumption of fruits, particularly citrus fruits and juices, while no consistent associations were apparent for vegetables.

cohort studies; diet; fruit; leukoplakia; mouth neoplasms; precancerous conditions; tobacco

Abbreviations: CI, confidence interval; OPL, oral premalignant lesion; RR, relative risk


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
Virtually all oral squamous cell carcinomas arise from a premalignant precursor (1Go, 2Go). Epidemiologic studies have found that 16–40 percent of oral premalignant lesions (OPLs) transform to cancer (3Go–6Go). The 5-year survival rate among patients with oral cancer has not improved appreciably in the past 30 years, remaining at 36 percent for African Americans and 60 percent for Caucasian Americans (7Go). Prevention of oral cancer at its precursor stage is necessary.

Case-control studies suggest a reduced risk of OPLs and oral cancer with greater consumption of fruits and vegetables (8Go–13Go), but results are inconsistent, particularly for oral premalignancies (14Go–21Go). A recent meta-analysis of nine case-control studies of oral and pharyngeal cancer determined that higher fruit intake was associated with a decreased risk (odds ratio = 0.53, 95 percent confidence interval (CI): 0.37, 0.76), but there was no consistent association with vegetables (22Go). In addition to selection and recall bias, residual influences of tobacco could distort results and contribute to inconsistencies. Smoking is not only the greatest known risk factor for OPLs but also associated with dietary intake, increased oxidative stress, and lower circulating levels of antioxidant nutrients (23Go). Prospective cohort studies of oral cancer are lacking, but meta-analyses of other cancers indicate that case-control studies, which often suffer from selection and recall bias, may overestimate the protective effects of fruits and vegetables (22Go). Still, in chemoprevention studies, supplements of nutrients commonly obtained from fruits and vegetables, such as ß-carotene and lycopene, have been shown to increase the likelihood of regression of leukoplakia (24Go–26Go). Thus, although evidence suggests a potential benefit from fruit and vegetable intake, our knowledge of the exact role of diet in the prevention of OPLs is limited by the lack of prospective data.

Current public health messages, such as the National Cancer Institute's "Eat 5 to 9 a Day for Better Health" program, advocate that increasing fruit and vegetable intake decreases the risk of many diseases (27Go). In this study, we assessed the value of this message with regard to OPLs and oral cancer. We prospectively evaluated fruit and vegetable consumption and the incidence of OPLs in a cohort of men and examined whether associations varied by tobacco use and timing of diet assessment.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Study population: Health Professionals Follow-up Study
The Health Professionals Follow-Up Study is an ongoing prospective cohort study of 51,529 US male health professionals (58 percent dentists, 8 percent pharmacists, 7 percent optometrists, 4 percent osteopaths, 3 percent podiatrists, 20 percent veterinarians) aged 40–75 years when the study began in 1986. At baseline, participants completed detailed questionnaires assessing dietary intake, lifestyle factors, and medical history. Follow-up questionnaires were mailed every 2 years to update exposure information and ascertain newly diagnosed disease. This study received institutional approval from the Human Subjects Committee at the Harvard School of Public Health.

We excluded participants from analyses if they reported an implausible daily energy intake (outside the range of 800–4,200 kcal/day) or did not answer 70 or more of the 131 dietary questions asked at baseline (3 percent of men). In addition, we excluded those who had been diagnosed with any cancer other than nonmelanoma skin cancer (4 percent) or who reported OPLs (0.04 percent) prior to 1986. After exclusions, 42,311 men remained in the study.

Assessment of OPLs
A specific question on the lifetime occurrence of leukoplakia or any other OPL was first included in the 1996 survey, and any new diagnosis was reported on subsequent biennial questionnaires. Participants who reported a professional diagnosis of OPL from 1986 to 2002 were mailed an additional questionnaire to confirm the diagnosis and provide consent for records release. We collected dental records and pathology reports from treating dentists and oral surgeons. A study investigator blinded to the exposure status of participants reviewed records to confirm diagnoses.

Primary outcome definition was based on a clinical diagnosis, because dysplasia is not necessary to define OPLs. Lesions meeting one of the diagnosis criteria described in appendix table 1 were included. Oral malignancies that were not first diagnosed as premalignancies were also included because they likely developed from premalignant precursors (1Go, 2Go); omitting these cases would have resulted in selective exclusion of premalignancies, particularly those with greater malignant potential. In addition, participants whose medical records indicated oral squamous cell carcinoma in situ were considered cases because severe oral epithelial dysplasia is consistent with carcinoma in situ. During follow-up, we verified a total of 207 new OPL events (appendix table 1). Nondentist participants who attested to their OPLs on the additional questionnaire but for whom dental/medical records were unobtainable were considered "probable" OPL cases. The diagnosis remained "probable" for 22 percent of men and "confirmed" for 78 percent of men.

In the main analyses, we included all OPL events (n = 207) and then applied various restrictions to verify overall findings. First, we conducted an analysis with data for only confirmed OPL cases (n = 162). To further minimize potential misclassification and detection bias, we restricted the entire sample of participants to the 24,619 men who were dentists by profession (94 events). We also conducted analyses by excluding 19 cases diagnosed with oral lichen planus or 46 cases first diagnosed as having oral cancer upon detection. An analysis aiming to examine a longer induction period also served to ensure the prospective quality of the investigation; here, we included only those OPL events that occurred after 1996, the first year that the diagnosis appeared on a Health Professionals Follow-Up Study questionnaire (102 events). Finally, because the clinical definition of OPL is prone to subjectivity and is waning in popularity, we analyzed the outcome of histologically confirmed oral epithelial dysplasia or cancer (91 events). Because results were similar across these various restricted analyses, the results we report here are from analyses including all probable OPL events, unless otherwise specified.

Assessment of dietary intake
A semiquantitative food frequency questionnaire containing prespecified portion sizes for 131 food and beverage items, including 15 fruits and 30 vegetables, was administered as part of the 1986, 1990, 1994, and 1998 Health Professionals Follow-Up Study questionnaires. Nine mutually exclusive response possibilities were provided for average frequency of intake over the previous year, with choices ranging from "almost never or less than once per month" to "six or more times per day." The reproducibility and validity of the food frequency questionnaire has been demonstrated by comparison with multiple 1-week dietary records (28Go, 29Go). Compared with those for two 1-week diet records completed 6 months apart, the median Pearson correlation coefficients were 0.8 for fruits and 0.5 for vegetables.

For each participant, average daily intake of each fruit and vegetable item was combined to compute total fruit and vegetable intake, as well as intake of composite fruit and vegetable groups (appendix table 2). When aggregating items to compute average intakes, we considered missing values for individual foods listed in food frequency questionnaires that were substantially complete to imply no intake (<1 percent for most foods) (30Go, 31Go).

Statistical analysis
Each eligible participant contributed person-time from the date that he returned the baseline questionnaire to the month of diagnosis of an oral premalignant or malignant lesion, death, or the end of follow-up (January 31, 2002), whichever occurred first. Each participant could contribute only one endpoint, and the cohort at risk included only those free of the outcome.

To best represent long-term diet and to reduce within-person variation, the main analysis used the cumulative average daily intake of fruits and vegetables from all available questionnaires up to the start of each 2-year follow-up interval (32Go). If participants were diagnosed with cancer during follow-up, diet information was not updated after the beginning of the interval in which they developed the diagnosis. Because the induction period for any association of fruits and vegetables with OPLs is uncertain, in additional analyses, we examined the possibility of a long latency for OPLs by using baseline intake, then applying a lag period of at least 10 years (excluding 105 cases occurring before 1996). We also examined associations with recent diet within 0–4 years of each follow-up period. Participants were grouped into quintiles of average servings per day, with the lowest quintile as the reference category. To minimize the influence of outliers, linear association and tests for trend were assessed by using the median values of deciles of intake to represent the exposure of all participants in the same decile (32Go).

We used multivariate Cox proportional hazards models with time-dependent covariates to calculate hazard ratio estimates and 95 percent confidence intervals as estimates of relative risk. All statistical tests were two sided. All nonnutrient covariates were updated in the analysis by using simple updating for each 2-year period. Multivariate models adjusted for age (months), time period (2-year intervals), cigarette use (never, former, current), age at which smoking started (<15, 15–19, 20–29, ≥30 years of age), quantity smoked during years of active smoking (1–4, 5–14, 15–24, 25–34, 35–44, ≥45 cigarettes/day), time since quitting among past smokers (<10 or ≥10 years), pipe or cigar use (never, former, current), ever use of chewing tobacco (yes/no), alcohol consumption (g/day), total energy intake (quintiles), olive oil as the main type of cooking oil used (yes/no), multivitamin use (yes/no), and profession (dentist vs. nondentist). We also included the following factors and assessed change in estimates, but they did not have an impact on the results and so were not included in the final multivariate analyses: race/ethnicity; family history of cancer; recent physical examination (as a proxy for health-care utilization); number of natural teeth; intake of other groups of food (red meat, fish, processed meats, or cereals); use of vitamin supplements A, C, or E; and an interaction term between tobacco and alcohol. We considered differences in the effect of fruit and vegetable intake by categories of tobacco use and multivitamin use in stratified analyses.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Based on cumulative average consumption during 16 years of follow-up, median intake was 2.2 servings of fruit per day and 3.1 servings of vegetables per day. Men with higher total fruit and vegetable intakes were more likely to have had a recent physical examination, take multivitamin supplements, and use olive oil for cooking, while men whose intakes were lower were younger, consumed more alcohol per day, and were more likely to have used tobacco (table 1).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Number of oral premalignant lesion cases from 1986 to 2002 and age-standardized characteristics at baseline, by quintile of total fruit and vegetable intake, among 42,311 US men in the Health Professionals Follow-up Study

 
Main analysis of cumulative average fruit and vegetable intake
Although total fruit and vegetable intake was associated with lower risk in the age-adjusted model (ptrend = 0.01), the association became null in the multivariate model (ptrend = 0.88) (table 2). Likewise, none of the vegetable composite groups had significant associations, and, on the contrary, green leafy vegetables seemed to increase risk (ptrend = 0.09, one serving/day relative risk (RR) = 1.31, 95 percent CI: 0.96, 1.78). However, all intake categories of total fruits, noncitrus fruits, and citrus fruits were associated with a lower risk of OPLs, indicating approximately a 40 percent lower risk with greater intake of these fruits. Vitamin-C-rich fruits and vegetables were also significantly associated with reduced risk, whereas ß-carotene-rich or lycopene-rich foods showed no association. Citrus fruit juice and oranges in particular showed statistically significant inverse associations; for example, men with the highest intake of oranges had a 43 percent lower risk of OPLs compared with men with the lowest intake (95 percent CI: 0.36, 0.90; p = 0.02). Among noncitrus fruits, peaches had the strongest inverse associations with OPL risk (quintile 5 vs. quintile 1 RR = 0.64, 95 percent CI: 0.42, 0.98; p = 0.04). After we controlled for citrus fruit intake, the risk reduction associated with noncitrus fruit slightly diminished, whereas citrus fruit was not as affected by simultaneous control for intake of other fruits (data not shown). While linear trends as measured by the median values of deciles of fruit intake failed to achieve statistical significance, a continuous measure of servings per day showed significant trends, particularly for vitamin-C-rich fruits and vegetables (ptrend = 0.02) and citrus fruit juice (ptrend = 0.05) (data not shown).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Multivariate relative risks and 95% confidence intervals for oral premalignant lesions, by updated cumulative average fruit and vegetable intake, in 42,311 US men and for 207 events in the Health Professionals Follow-up Study, 1986–2002{dagger}

 
In subanalyses restricted to dysplasia or cancer (n = 91 events), the magnitude of most associations remained similar to those shown in table 2, with slightly stronger associations for noncitrus fruits (one additional serving/day RR = 0.89; ptrend = 0.4) and noticeably stronger inverse associations for cruciferous vegetables (quintile 5 vs. quintile 1 RR = 0.47, 95 percent CI: 0.21, 1.02; p = 0.06 and one additional serving/day RR = 0.56; p = 0.14). Similarly, in subanalyses restricted to the 24,465 participants who are dentists by profession (94 events), the risk reductions with fruit groups were greater than in the main analysis, but, because of the smaller sample size, confidence intervals were wider.

Earlier diet versus recent diet, and possible induction periods
Considering that earlier diet may play an important role in the etiology of OPLs, we used various lag times between the diet measures and follow-up in additional analyses. First, we examined the possibility of a longer induction period by using fruit and vegetable intake at baseline (table 3). Results were similar to those in the main analysis using cumulative average intake through follow-up, in that neither total fruit nor total vegetable intake at baseline was significantly associated with OPL risk. However, baseline intake levels of citrus fruits were more significantly protective than were cumulative average intakes (baseline quintile 5 vs. quintile 1 RR = 0.54, 95 percent CI: 0.35, 0.85; p = 0.007), and the linear trend with increasing citrus fruit juice became significant (one serving/day RR = 0.69, 95 percent CI: 0.49, 0.97; ptrend = 0.03). Overall, greater intake of various fruits at baseline was associated with a 30–46 percent lower OPL risk. When we examined the possibility of a shorter induction period by using recent intake levels, results were also similar to those in the main analysis. To determine whether baseline diet or recent diet was more relevant to OPL risk, we modeled both intakes simultaneously. Baseline levels retained significance to a greater extent than did recent intake levels, indicating that earlier diet may be more important in the development of OPLs.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Multivariate relative risks and 95% confidence intervals for oral premalignant lesions, by baseline (1986) fruit and vegetable intake, overall and with a lag time of 10–16 years between diet assessment and follow-up, in US men in the Health Professionals Follow-up Study{dagger}

 
To further examine the possibility of a longer induction period, we conducted an analysis by using baseline intake levels and excluded the first 10 years of follow-up, thereby restricting to the 102 events that occurred after 1996 (table 3). Compared with those from other time-period analyses, the protective associations were greater here, not only with citrus fruit but also with noncitrus fruit. Men with the highest intake of noncitrus fruit had a 73 percent lower risk of OPLs (RR = 0.27, 95 percent CI: 0.12, 0.60; p = 0.02) than those in the lowest intake quintile, and the linear trend was significant, indicating a 30 percent decreased risk with each additional serving (ptrend = 0.01). An additional three servings per day of all fruits was associated with a 44 percent reduction in OPL risk (95 percent CI: 0.33, 0.96; ptrend = 0.03).

Modification of effects by tobacco use
Because tobacco has great potential to both distort results through confounding and also modify effects through biologic interactions, we performed analyses separately for never users, past users, and current users of tobacco in the form of cigarettes, cigars, pipes, or chewing tobacco (table 4). Although tests for interactions using indicator terms had limited power, results were slightly different when data for never, past, and current users were separated. Among never users of tobacco, no consistent associations emerged, but protective associations with fruits were apparent in subanalyses restricted to dentists, oral epithelial dysplasia, or application of 10–16-year lags (data not shown).


View this table:
[in this window]
[in a new window]

  		TABLE 4. Multivariate relative risks and 95% confidence intervals for oral premalignant lesions associated with cumulative average daily servings of fruits and vegetables, by tobacco use (chewing tobacco, cigarettes, pipe, cigars) and multivitamin use status, in US men in the Health Professionals Follow-up Study, 1986–2002{dagger}

 
Among past tobacco users, all fruit and vegetable groups were strongly associated with lower risk of OPLs. For example, men with the lowest fruit intake had a threefold higher risk of OPLs compared with men in the top fruit intake quintile (quintile 5 vs. quintile 1 RR = 0.34, 95 percent CI: 0.15, 0.79; p = 0.01). Citrus fruit (one serving/day RR = 0.64, 95 percent CI: 0.44, 0.93; ptrend = 0.02) and citrus fruit juice (one serving/day RR = 0.52, 95 percent CI: 0.31, 0.87; ptrend = 0.01) offered significantly reduced risks, as did fruits and vegetables rich in vitamin C (one serving/day RR = 0.68, 95 percent CI: 0.51, 0.92; ptrend = 0.01). Protective associations were even stronger and more statistically significant when we used only baseline intake levels and excluded the first 10 years of follow-up to evaluate a longer induction period. Here, an increase of five servings per day of total fruits and vegetables decreased risk by more than 60 percent (RR = 0.37, 95 percent CI: 0.19, 0.75; ptrend = 0.005), and men in the lowest quintile of non-citrus-fruit intake had an eightfold greater risk than men in the highest quintile (quintile 5 vs. quintile 1 RR = 0.13, 95 percent CI: 0.04, 0.46; p = 0.001).

Among current users of tobacco, the associations with fruits remained close to null, and the confidence intervals were wide. However, green leafy vegetables and ß-carotene-rich fruits and vegetables showed significant linear trends, with a 65–67 percent greater risk of OPLs with each additional serving per day (green leafy vegetables RR = 1.67, 95 percent CI: 0.99, 2.80; ptrend = 0.05 and ß-carotene-rich foods RR = 1.65, 95 percent CI: 1.08, 2.52; ptrend = 0.02). Considering the correlation between green leafy vegetable intake and intake of other fruits and vegetables, we assessed potential confounding by green leafy vegetables among current smokers. When we controlled for green leafy vegetables, intake of all other fruits and vegetables was inversely associated with OPLs, but results remained nonsignificant (data not shown).

Because tobacco use is the strongest known risk factor for OPLs, we explored how higher intakes of citrus fruits or green leafy vegetables modified the association between tobacco use and OPLs (table 5). For current smokers, the relative risk associated with tobacco use was almost doubled for men with higher intake of green leafy vegetables (current smokers with low intake RR = 2.3, current smokers with higher intake RR = 4.3, reference = never smokers with low intake). In contrast, higher citrus fruit intake considerably attenuated the association between current smoking and OPLs, from 5.0 (low intake) to 3.1 (higher intake), and lowered the OPL risk for never and past smokers as well.


View this table:
[in this window]
[in a new window]

  		TABLE 5. Multivariate relative risks and 95% confidence intervals for oral premalignant lesions, by interaction between tobacco use (chewing tobacco, cigarettes, pipe, cigars) status and intake of green leafy vegetables or citrus fruits, in US men in the Health Professionals Follow-up Study, 1986–2002{dagger}

 
Modification of effects by multivitamin use
Fruit and vegetable consumption was higher among multivitamin supplement users than nonusers. When we stratified by multivitamin use (table 4), the protective associations with consumption of fruit, citrus fruit, and citrus fruit juice were somewhat stronger among nonsupplement users, and intake of total fruit showed a significant linear trend among nonusers (nonusers: three servings/day RR = 0.43, 95 percent CI: 0.19, 0.93; ptrend = 0.03).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
In this prospective study, high consumption of fruits, but not vegetables, was associated with a reduced risk of OPLs in men. Citrus fruits and citrus fruit juice indicated the greatest potential for lowering risk, even after rigorous control for tobacco use. The strongest and most consistent inverse associations were seen for past users of tobacco. Current users of tobacco, however, seemed to have an increased risk with higher consumption of green leafy vegetables or ß-carotene-rich foods.

Our findings of inverse associations with fruit but not vegetables are consistent with a recent meta-analysis of epidemiologic studies of oral cancer, in which the estimated odds ratio for increased fruit consumption was 0.53 (95 percent CI: 0.37, 0.76), and vegetables had no apparent association with risk (odds ratio = 0.84, 95 percent CI: 0.67, 1.07) (22Go). Fruits were found to be significantly protective in many case-control studies of oral premalignancies (15Go, 19Go, 20Go) and malignancies (8Go–10Go, 12Go, 13Go). Although citrus fruits in particular have not been evaluated often, they reduced risk of oral cancer by 50 percent in a study of 512 Italian male oral cancer patients (additional one serving/day, odds ratio = 0.5, 95 percent CI: 0.3, 0.8), similar to our estimates (9Go).

Vegetable consumption seemed modestly protective in many studies of oral premalignant and malignant lesions (9Go–12Go, 15Go, 17Go, 19Go, 20Go), but some studies, such as ours, did not find a clear association (14Go, 21Go, 33Go). Discrepancies may be related to issues particular to the case-control design of previous studies, such as recall and selection bias. For example, participation rates among controls are often lower than those among cases, and those who participate are more likely to be health conscious and consume more vegetables, thereby creating an apparent inverse association with vegetable intake (34Go). Such problems were avoided in our study because of its prospective cohort design. Another possible explanation, as suggested by statistical models for random errors in reported smoking and lung cancer risk, is inadequate assessment of exposure to smoking (35Go). Our adjustment method for cigarette smoking followed recommendations to consider smoking duration, intensity, and time since cessation (36Go), and it was tested in a previous Health Professionals Follow-Up Study report on fruit and vegetable intake and risk of lung cancer, where it effectively eliminated apparent risk reductions in high quintiles of fruit and vegetable intake (37Go). Effect modification by tobacco use may have played a role in the results of previous studies. Our stratified analysis showed no clear associations for never users of tobacco, inverse associations across all fruit and vegetable groups for past users, but increased risks with higher consumption of green leafy vegetables for current smokers.

Although the number of current users of tobacco in our study was relatively small, the notion that current smokers have unique susceptibilities involving nutrients abundant in green leafy vegetables, such as ß-carotene or vitamin E, has been observed in other reports. Large-scale trials examining effects of supplementation with ß-carotene and vitamin A found increased risks of lung cancer or colorectal adenomas in smokers, with temporal effects suggesting that these nutrients somehow accelerated the progression of latent disease (38Go–40Go). In a recent trial, vitamin E supplementation was associated with increased risks of oral cancer recurrence and second primary oral tumors (41Go). Although the mechanisms by which vitamin E may act as a prooxidant are unclear (42Go), a commonly proposed mechanism for ß-carotene is that components of cigarette smoke in the presence of high oxygen tension combine to induce oxidation of the nutrient, resulting in a prooxidant effect (43Go). A recent study demonstrated that ß-carotene exacerbates DNA oxidative damage and modifies p53-related pathways of cell proliferation and apoptosis in cultured cells exposed to tobacco smoke but has no significant effect in cells unexposed to tobacco (44Go). Another possibility is that high intake of ß-carotene enhances potential carcinogenesis by decreasing tissue concentrations of retinoic acid and retinoic acid receptor-ß while increasing activator-protein-1 and proliferating-cellular-nuclear-antigen (45Go, 46Go). Thus, current tobacco users should be considered a unique subgroup in analyses aiming to clarify the association between vegetable consumption and risk of OPLs.

Long-term consumption patterns may be more relevant to the etiology of OPLs than recent intakes, as suggested by the stronger associations we found with baseline intake of fruits in 1986 or cumulative average intake over follow-up. In addition, the relevant time frame for protective effects of fruits may be earlier than is often assumed; the magnitude of protective associations and statistical significance increased when we examined an induction time of 10–16 years.

A disadvantage of the asymptomatic nature of OPLs is that some cases may have been missed. We addressed this issue first by restricting the main analysis to the 58 percent of participants who were dentists by profession, because dentists are highly likely to be aware of the health of their oral cavity, thereby minimizing disease misclassification. Results from the dentists-only analysis were similar to those from the main analysis and were even stronger for the significant inverse associations with citrus fruits. The analysis using a 10–16-year lag period also had the added benefit of minimizing any possible misclassification due to recall bias of the OPL diagnosis, which our questionnaires started assessing in 1996. Observed associations were further confirmed in the analyses restricted to cases with oral epithelial dysplasia or cancer.

In conclusion, after control for detailed tobacco use and alcohol consumption characteristics, risk of OPLs in men was significantly reduced with higher consumption of fruits, particularly citrus fruits and citrus fruit juice, while no consistent association was apparent with consumption of vegetables. Past and current smokers, who have a substantially increased risk of OPLs, may particularly benefit from increased fruit intake, because citrus fruit considerably attenuated the risk of OPLs associated with tobacco use. Although there was an indication of an increased risk of OPLs with high consumption of green leafy vegetables and ß-carotene-rich foods for current smokers, the small number of current tobacco users necessitates that this finding be viewed with caution. Timing of diet may be critical to the prevention of disease, because early high intake of fruits seemed to offer greater protection than did recent intake. Although the exact nutrients in fruits and the mechanisms of prevention of oral premalignancy are still to be determined, dietary recommendations to increase consumption of fruits are appropriate for preventing oral precancer and cancer.


APPENDIX TABLE 1. Eligible oral premalignant lesions for case definition*

Lesion
Definition
%
No.
Leukoplakia A white patch or plaque that does not rub off and cannot be characterized clinically or pathologically as any other disease 73{dagger} 74
Erythroplakia A red patch that cannot be clinically or pathologically diagnosed as any other condition 3{dagger} 3
Erythroleukoplakia An area of leukoplakia that has red patches; also known as "speckled leukoplakia" or "leukoerythroplakia" 2{dagger} 2
Lichen planus Chronic dermatologic disease that also affects the mucosa of the mouth, of either reticular (interlacing white lines) or erosive (ulceration in center) type 19{dagger} 19
Oral epithelial dysplasia Histopathologically verified abnormality of development; in pathology, alteration in size, shape, and organization of adult cells 44{ddagger} 91
Oral squamous cell carcinoma
 Histopathologically verified malignant neoplasm characterized by proliferation of anaplastic cells that tend to invade surrounding tissue and metastasize; virtually all arise from premalignant precursor lesions
 22{ddagger}
 46

* Lesions on the external lip vermilion, which are likely to resemble skin cancer in etiology, and lesions of the oropharynx were ineligible. There were no cases of proliferative verrucous leukoplakia or oral submucous fibrosis.

{dagger} Calculated for the 102 cases who had a clinical diagnosis term assigned.

{ddagger} Calculated for the total number of cases (n = 207). A biopsy was performed on 142 cases; thus, 64% of biopsied lesions displayed oral epithelial dysplasia.


APPENDIX TABLE 2. Composite fruit and vegetable groups*,{dagger}

All fruits
Apples/applesauce, 1
Apricots, 1
Avocado, 1/2
Bananas, 1
Blueberries, 1/2 cup (120 g)
Cantaloupe, 1/4
Grapefruit, 1/2
Oranges, 1
Peaches, 1
Pears, 1
Plums, 1
Prunes, 1/2 cup (120 g)
Raisins/grapes, 1-ounce/bunch (28.3 g)
Strawberries, 1/2 cup (120 g)
Watermelon, 1 slice
Apple juice, 6 fluid ounces (180 ml)
Apple cider, 6 fluid ounces (180 ml)
Orange juice, 6 fluid ounces (180 ml)
Grapefruit juice, 6 fluid ounces (180 ml)
Other fruit juices, 6 fluid ounces (180 ml)
All vegetables Alfalfa sprouts, 1/2 cup (120 g)
Beets, 1/2 cup (120 g)
Broccoli, 1/2 cup (120 g)
Brussels sprouts, 1/2 cup (120 g)
Carrots, 1 or 1/2 cup (120 g)
Cauliflower, 1/2 cup (120 g)
Celery, 4-inch stick (10.2 cm)
Cole slaw/uncooked cabbage/sauerkraut, 1/2 cup (120 g)
Cooked cabbage, 1/2 cup (120 g)
Corn, 1 ear or 1/2 cup (120 g)
Eggplant/zucchini/summer squash, 1/2 cup (120 g)
Green pepper, 1/2
Iceberg or head lettuce, 1 cup (240 g)
Kale/mustard or chard greens, 1/2 cup (120 g)
Mixed vegetables, 1/2 cup (120 g)
Mushrooms, 1
Peas/lima beans, 1/2 cup (120 g)
Romaine or leaf lettuce, 1 cup (240 g)
Spinach, cooked, 1/2 cup (120 g)
Spinach, raw, 1 cup (240 g)
String beans, 1/2 cup (120 g)
Sweet potatoes/yams, 1/2 cup (120 g)
Tomatoes, 1
Tomato juice, 4 fluid ounces (120 ml)
Tomato sauce, 1/2 cup (120 g)
Winter squash, 1/2 cup (120 g)
Total noncitrus fruit Apples/applesauce, apple cider, apple juice, apricots, avocado, bananas, blueberries, cantaloupe, peaches, pears, plums, prunes, raisins/grapes, strawberries, and watermelon
Total citrus fruit Oranges, orange juice, grapefruit, and grapefruit juice
Citrus fruit juice Orange juice and grapefruit juice
Cruciferous vegetables Broccoli, cabbage, cole slaw/sauerkraut, cauliflower, Brussels sprouts, and kale/mustard or chard greens
Green leafy vegetables Raw spinach, cooked spinach, kale/mustard or chard greens, and romaine or leaf lettuce
Fruits and vegetables rich in vitamin C (≥30 mg/serving){ddagger} Oranges, orange juice, grapefruit, grapefruit juice, other fruit juice, cantaloupe, strawberries, broccoli, and Brussels sprouts
Fruits and vegetables rich in ß-carotene (≥2,000 µg/serving){ddagger} Cantaloupe, carrots, broccoli, mixed vegetables, yams/sweet potatoes, raw spinach, cooked spinach, kale/mustard, chard greens, and romaine lettuce
Fruits and vegetables rich in lycopene (≥10 mg/serving){ddagger}
 Tomatoes, tomato juice, and tomato sauce

* Definitions of the composite groups are based on the criteria of Smith et al. (47Go) and were modified to conform to the Health Professionals Follow-up Study questionnaire.

{dagger} The fruits and vegetables were categorized according to methods outlined in Feskanich et al. (37Go), Smith et al. (47Go), Michels et al. (48Go), and the US Department of Agriculture (49Go).

{ddagger} The US Department of Agriculture National Nutrient Database for Standard Reference (49Go) was used to determine which fruits and vegetables were included in this category.


   ACKNOWLEDGMENTS
 
This research was supported by grants from the National Institutes of Health (T32DE07151, HL035464, and CA055075).

The authors thank Dr. Walter Willett and Dr. Eric Rimm for their support and advice; Al Wing, Mira Kaufman, and Lydia Liu for statistical programming assistance; and Elizabeth Frost-Hayes, Stacey DeCaro, Katherine Smith, and Melanie Gerard for administrative assistance.

Conflict of interest: none declared.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Mehta FS, Gupta PC, Pindborg JJ. Chewing and smoking habits in relation to precancer and oral cancer. J Cancer Res Clin Oncol 1981;99:35–9.[CrossRef][ISI][Medline]
  2. Melrose RJ. Premalignant oral mucosal diseases. J Calif Dent Assoc 2001;29:593–600.[Medline]
  3. Gupta PC, Mehta FS, Daftary DK, et al. Incidence rates of oral cancer and natural history of oral precancerous lesions in a 10-year follow-up study of Indian villagers. Community Dent Oral Epidemiol 1980;8:283–333.[CrossRef][ISI][Medline]
  4. Silverman S Jr, Gorsky M, Lozada F. Oral leukoplakia and malignant transformation. A follow-up study of 257 patients. Cancer 1984;53:563–8.[CrossRef][ISI][Medline]
  5. Bouquot JE, Weiland LH, Kurland LT. Leukoplakia and carcinoma in situ synchronously associated with invasive oral/oropharyngeal carcinoma in Rochester, Minn., 1935–1984. Oral Surg Oral Med Oral Pathol 1988;65:199–207.[CrossRef][ISI][Medline]
  6. Lumerman H, Freedman P, Kerpel S. Oral epithelial dysplasia and the development of invasive squamous cell carcinoma. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 1995;79:321–9.[CrossRef][ISI][Medline]
  7. Surveillance, Epidemiology, and End Results Program. Bethesda, MD: Division of Cancer Control and Population Sciences, National Cancer Institute, 2005. (www.seer.cancer.gov).
  8. Fioretti F, Bosetti C, Tavani A, et al. Risk factors for oral and pharyngeal cancer in never smokers. Oral Oncol 1999;35:375–8.[CrossRef][ISI][Medline]
  9. Franceschi S, Favero A, Conti E, et al. Food groups, oils and butter, and cancer of the oral cavity and pharynx. Br J Cancer 1999;80:614–20.[CrossRef][ISI][Medline]
  10. De Stefani E, Deneo-Pellegrini H, Mendilaharsu M, et al. Diet and risk of cancer of the upper aerodigestive tract—I. Foods. Oral Oncol 1999;35:17–21.[CrossRef][ISI][Medline]
  11. De Stefani E, Oreggia F, Boffetta P, et al. Tomatoes, tomato-rich foods, lycopene and cancer of the upper aerodigestive tract: a case-control in Uruguay. Oral Oncol 2000;36:47–53.[CrossRef][ISI][Medline]
  12. Tavani A, Gallus S, La Vecchia C, et al. Diet and risk of oral and pharyngeal cancer. An Italian case-control study. Eur J Cancer Prev 2001;10:191–5.[CrossRef][ISI][Medline]
  13. Sanchez MJ, Martinez C, Nieto A, et al. Oral and oropharyngeal cancer in Spain: influence of dietary patterns. Eur J Cancer Prev 2003;12:49–56.[CrossRef][ISI][Medline]
  14. Kaugars GE, Brandt RB, Chan W, et al. Evaluation of risk factors in smokeless tobacco-associated oral lesions. Oral Surg Oral Med Oral Pathol 1991;72:326–31.[ISI][Medline]
  15. Carley KW, Puttaiah R, Alvarez JO, et al. Diet and oral premalignancy in female south Indian tobacco and betel chewers: a case-control study. Nutr Cancer 1994;22:73–84.[ISI][Medline]
  16. Kaugars GE, Silverman S Jr, Lovas JG, et al. A clinical trial of antioxidant supplements in the treatment of oral leukoplakia. Oral Surg Oral Med Oral Pathol 1994;78:462–8.[CrossRef][ISI][Medline]
  17. Gupta PC, Hebert JR, Bhonsle RB, et al. Dietary factors in oral leukoplakia and submucous fibrosis in a population-based case control study in Gujarat, India. Oral Dis 1998;4:200–6.[Medline]
  18. Cianfriglia F, Manieri A, Di Gregorio DA, et al. Retinol dietary intake and oral leukoplakia development. J Exp Clin Cancer Res 1998;17:331–6.[ISI][Medline]
  19. Gupta PC, Hebert JR, Bhonsle RB, et al. Influence of dietary factors on oral precancerous lesions in a population-based case-control study in Kerala, India. Cancer 1999;85:1885–93.[ISI][Medline]
  20. Morse DE, Pendrys DG, Katz RV, et al. Food group intake and the risk of oral epithelial dysplasia in a United States population. Cancer Causes Control 2000;11:713–20.[CrossRef][ISI][Medline]
  21. Hebert JR, Gupta PC, Bhonsle RB, et al. Dietary exposures and oral precancerous lesions in Srikakulam District, Andhra Pradesh, India. Public Health Nutr 2002;5:303–12.[CrossRef][ISI][Medline]
  22. Riboli E, Norat T. Epidemiologic evidence of the protective effect of fruit and vegetables on cancer risk. Am J Clin Nutr 2003;78:559S–569S.[Abstract/Free Full Text]
  23. Alberg A. The influence of cigarette smoking on circulating concentrations of antioxidant micronutrients. Toxicology 2002;180:121–37.[CrossRef][ISI][Medline]
  24. Sankaranarayanan R, Mathew B, Varghese C, et al. Chemoprevention of oral leukoplakia with vitamin A and beta carotene: an assessment. Oral Oncol 1997;33:231–6.[ISI][Medline]
  25. Garewal HS, Katz RV, Meyskens F, et al. Beta-carotene produces sustained remissions in patients with oral leukoplakia: results of a multicenter prospective trial. Arch Otolaryngol Head Neck Surg 1999;125:1305–10.[Abstract/Free Full Text]
  26. Singh M, Krishanappa R, Bagewadi A, et al. Efficacy of oral lycopene in the treatment of oral leukoplakia. Oral Oncol 2004;40:591–6.[CrossRef][ISI][Medline]
  27. National Cancer Institute. National Institutes of Health, 2005. (http://www.5aday.gov/research/health_cancer.html).
  28. Rimm EB, Giovannucci EL, Stampfer MJ, et al. Reproducibility and validity of an expanded self-administered semiquantitative food frequency questionnaire among male health professionals. Am J Epidemiol 1992;135:1114–26; discussion 1127–36.[Abstract/Free Full Text]
  29. Feskanich D, Rimm EB, Giovannucci EL, et al. Reproducibility and validity of food intake measurements from a semiquantitative food frequency questionnaire. J Am Diet Assoc 1993;93:790–6.[CrossRef][ISI][Medline]
  30. Joshipura KJ, Ascherio A, Manson JE, et al. Fruit and vegetable intake in relation to risk of ischemic stroke. JAMA 1999;282:1233–9.[Abstract/Free Full Text]
  31. Caan B, Hiatt RA, Owen AM. Mailed dietary surveys: response rates, error rates, and the effect of omitted food items on nutrient values. Epidemiology 1991;2:430–6.[ISI][Medline]
  32. Willett WC. Issues in analysis and presentation of dietary data. In: Willett WC, ed. Nutritional epidemiology. New York, NY: Oxford University Press, 1998.
  33. Kaugars GE, Riley WT, Brandt RB, et al. The prevalence of oral lesions in smokeless tobacco users and an evaluation of risk factors. Cancer 1992;70:2579–85.[CrossRef][ISI][Medline]
  34. Willett WC. Diet and cancer: an evolving picture. JAMA 2005;293:233–4.[Free Full Text]
  35. Stram DO, Huberman M, Wu AH. Is residual confounding a reasonable explanation for the apparent protective effects of beta-carotene found in epidemiologic studies of lung cancer in smokers? Am J Epidemiol 2002;155:622–8.[Abstract/Free Full Text]
  36. Leffondré K, Abrahamowicz M, Siemiatycki J, et al. Modeling smoking history: a comparison of different approaches. Am J Epidemiol 2002;156:813–23.[Abstract/Free Full Text]
  37. Feskanich D, Ziegler RG, Michaud DS, et al. Prospective study of fruit and vegetable consumption and risk of lung cancer among men and women. J Natl Cancer Inst 2000;92:1812–23.[Abstract/Free Full Text]
  38. Omenn GS, Goodman GE, Thornquist MD, et al. Effects of a combination of beta carotene and vitamin A on lung cancer and cardiovascular disease. N Engl J Med 1996;334:1150–5.[Abstract/Free Full Text]
  39. Baron JA, Cole BF, Mott L, et al. Neoplastic and antineoplastic effects of beta-carotene on colorectal adenoma recurrence: results of a randomized trial. J Natl Cancer Inst 2003;95:717–22.[Abstract/Free Full Text]
  40. Virtamo J, Pietinen P, Huttunen JK, et al. Incidence of cancer and mortality following alpha-tocopherol and beta-carotene supplementation: a postintervention follow-up. JAMA 2003;290:476–85.[Abstract/Free Full Text]
  41. Bairati I, Meyer F, Gelinas M, et al. A randomized trial of antioxidant vitamins to prevent second primary cancers in head and neck cancer patients. J Natl Cancer Inst 2005;97:481–8.[Abstract/Free Full Text]
  42. Schneider C. Chemistry and biology of vitamin E. Mol Nutr Food Res 2005;49:7–30.[CrossRef][ISI][Medline]
  43. Mayne ST, Handelman GJ, Beecher G. Beta-carotene and lung cancer promotion in heavy smokers—a plausible relationship? J Natl Cancer Inst 1996;88:1513–15.[ISI][Medline]
  44. Palozza P, Serini S, Di Nicuolo F, et al. Beta-carotene exacerbates DNA oxidative damage and modifies p53-related pathways of cell proliferation and apoptosis in cultured cells exposed to tobacco smoke condensate. Carcinogenesis 2004;25:1315–25.[Abstract/Free Full Text]
  45. Liu C, Wang XD, Bronson RT, et al. Effects of physiological versus pharmacological beta-carotene supplementation on cell proliferation and histopathological changes in the lungs of cigarette smoke-exposed ferrets. Carcinogenesis 2000;21:2245–53.[Abstract/Free Full Text]
  46. Liu C, Russell RM, Wang XD. Exposing ferrets to cigarette smoke and a pharmacological dose of beta-carotene supplementation enhance in vitro retinoic acid catabolism in lungs via induction of cytochrome P450 enzymes. J Nutr 2003;133:173–9.[Abstract/Free Full Text]
  47. Smith SA, Campbell DR, Elmer PJ, et al. The University of Minnesota Cancer Prevention Research Unit vegetable and fruit classification scheme (United States). Cancer Causes Control 1995;6:292–302.[CrossRef][ISI][Medline]
  48. Michels KB, Giovannucci E, Joshipura KJ, et al. Prospective study of fruit and vegetable consumption and incidence of colon and rectal cancers. J Natl Cancer Inst 2000;92:1740–52.[Abstract/Free Full Text]
  49. US Department of Agriculture, Agricultural Research Service. USDA National Nutrient Database for Standard Reference, release 18, 2005. Nutrient Data Laboratory home page. (http://www.nal.usda.gov/fnic/foodcomp).

Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
164/6/556    most recent
kwj233v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (7)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Maserejian, N. N.
Right arrow Articles by Joshipura, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Maserejian, N. N.
Right arrow Articles by Joshipura, K.
Social Bookmarking
Add to CiteULike