Tobacco, alcohol, and socioeconomic status and adenocarcinomas of the esophagus and gastric cardia.

BACKGROUND: Incidence rates for adenocarcinomas of the esophagus and gastric cardia have risen steeply over the last few decades. To determine risk factors for these tumors, we conducted a multicenter, population-based, case-control study. METHODS: The study included 554 subjects newly diagnosed with esophageal or gastric cardia adenocarcinomas, 589 subjects newly diagnosed with esophageal squamous cell carcinoma or other gastric adenocarcinomas, and 695 control subjects. Estimates of risk (odds ratios [ORs] and corresponding 95% confidence intervals [CIs]) were calculated for the four tumor types separately and for esophageal and gastric cardia adenocarcinomas combined. RESULTS: Risk of esophageal and gastric cardia adenocarcinomas combined was increased among current cigarette smokers (OR = 2.4; 95% = 1.7-3.4), with little reduction observed until 30 years after smoking cessation; this risk rose with increasing intensity and duration of smoking. Risk of these tumors was not related to beer (OR = 0.8; 95% CI = 0.6-1.1) or liquor (OR = 1.1; 95% CI = 0.8-1.4) consumption, but it was reduced for drinking wine (OR = 0.6; 95% CI = 0.5-0.8). Similar ORs were obtained for the development of noncardia gastric adenocarcinomas in relation to tobacco and alcohol use, but higher ORs were obtained for the development of esophageal squamous cell carcinomas. For all four tumor types, risks were higher among those with low income or education. CONCLUSIONS: Smoking is a major risk factor for esophageal and gastric cardia adenocarcinomas, accounting for approximately 40% of cases. IMPLICATIONS: Because of the long lag time before risk of these tumors is reduced among ex-smokers, smoking may affect early stage carcinogenesis. The increase in smoking prevalence during the first two thirds of this century may be reflected in the rising incidence of these tumors in the past few decades among older individuals. The recent decrease in smoking may not yet have had an impact.

Body mass index and risk of adenocarcinomas of the esophagus and gastric cardia.

BACKGROUND: Incidence rates have risen rapidly for esophageal adenocarcinoma and moderately for gastric cardia adenocarcinoma, while rates have remained stable for esophageal squamous cell carcinoma and have declined steadily for noncardia gastric adenocarcinoma. We examined anthropometric risk factors in a population-based case-control study of esophageal and gastric cancers in Connecticut, New Jersey, and western Washington. METHODS: Healthy control subjects (n = 695) and case patients with esophageal squamous cell carcinoma or noncardia gastric adenocarcinoma (n = 589) were frequency-matched to case patients with adenocarcinomas of esophagus or gastric cardia (n = 554) by 5-year age groups, sex, and race (New Jersey only). Classification of cases by tumor site of origin and histology was determined by review of pathology materials and hospital records. Data were collected using in-person structured interviews. Associations with obesity, measured by body mass index (BMI), were estimated by odds ratios (ORs). All ORs were adjusted for geographic location, age, sex, race, cigarette smoking, and proxy response status. RESULTS: The ORs for esophageal adenocarcinoma rose with increasing adult BMI. The magnitude of association with BMI was greater among the younger age groups and among nonsmokers. The ORs for gastric cardia adenocarcinoma rose moderately with increasing BMI. Adult BMI was not associated with risk of esophageal squamous cell carcinoma or noncardia gastric adenocarcinoma. CONCLUSIONS: Increasing prevalence of obesity in the United States population may have contributed to the upward trends in esophageal and gastric cardia adenocarcinomas.

Glutathione S-transferase M1, T1, and P1 polymorphisms as risk factors for renal cell carcinoma: a case-control study.

Renal cell carcinoma (RCC) has known environmental risk factors, notably smoking, and enzymes that biotransform carcinogens have high levels of activity in the kidney. However, a possible role of polymorphisms in these enzymes in RCC etiology has received little study. We investigated glutathione S-transferase (GST) polymorphisms in a population-based case-control study of RCC. Subjects completed a structured interview, and DNA was isolated from pathological material or buccal cells for 130 cases, and from blood for 505 controls. Genotypes for GSTM1 and GSTT1 were determined by multiplex PCR, and for GSTP1 by oligonucleotide ligation assay. The frequency of GSTM1 null genotype was 50.0% in cases and 50.5% in controls, with an adjusted odds ratio (OR) of 1.0 [95% confidence interval (CI), 0.6-1.6]. For GSTP1, the frequencies of genotypes AA, AG, and GG representing the Ile104Val variant were: cases, 44.6%, 43.1%, and 12.3%; controls, 43.4%, 44.0%, and 12.6%; OR for AG and GG, 1.0 (95% CI, 0.6-1.6). An excess of the GSTT1 null genotype was observed in cases compared with controls, 28.6% versus 18.5% (OR, 1.9; 95% CI, 1.1-3.4). The association with GSTT1 was present among both smokers and nonsmokers, but was modified by body mass index, a recognized risk factor for RCC; among subjects in the lowest tertile of body mass index, the OR for GSTT1 null was 4.8 (95% CI, 1.8-13.0). The association between GSTT1 null and increased RCC risk in this population-based study suggests that activity of the GSTT1 enzyme protects against RCC. This contrasts with a recent report of reduced risk of RCC associated with GSTT1 null in a cohort of trichloroethene-exposed workers and suggests that specific chemical exposures alter the effect of GSTT1 on cancer risk.

Use of aspirin and other nonsteroidal anti-inflammatory drugs and risk of esophageal and gastric cancer.

Regular users of aspirin and other nonsteroidal anti-inflammatory drugs (NSAIDs) are at reduced risk of colon cancer, but the evidence for protective effects of NSAIDs elsewhere in the digestive tract is scant. We investigated the association between the use of NSAIDs and risk of esophageal and gastric cancer, using data from a large population-based, case-control study. Cases were individuals, ages 30-79 years, diagnosed with esophageal adenocarcinoma (n = 293), esophageal squamous cell carcinoma (n = 221), gastric cardia adenocarcinoma (n = 261), or noncardia gastric adenocarcinoma (n = 368) in three areas with population-based tumor registries. Controls (n = 695) were selected by random digit dialing and through the rosters of the Health Care Financing Administration. After controlling for the major risk factors, we found that current users of aspirin were at decreased risk of esophageal adenocarcinoma [odds ratio (OR), 0.37; 95% confidence interval (CI), 0.24-0.58], esophageal squamous cell carcinoma (OR, 0.49; 95% CI, 0.28-0.87), and noncardia gastric adenocarcinoma (OR, 0.46; 95% CI, 0.31-0.68), but not of gastric cardia adenocarcinoma (OR, 0.80; 95% CI, 0.54-1.19), when compared to never users. Risk was similarly reduced among current users of nonaspirin NSAIDs. The associations with current NSAID use persisted when we excluded use within 2 or 5 years of reference date, which might have been affected by preclinical disease in cases, and when we restricted analyses to subjects reporting no history of chronic gastrointestinal symptoms. Our findings add to the growing evidence that the risk of cancers of the esophagus and stomach is reduced in users of NSAIDs, although whether the association is causal in nature is not clear.

Nutrient intake and risk of subtypes of esophageal and gastric cancer.

Incidence rates for adenocarcinoma of the esophagus and gastric cardia have been rising rapidly. We examined nutrient intake as a risk factor for esophageal and gastric cancers in a population-based case-control study in Connecticut, New Jersey, and western Washington state. Interviews were completed for cases with histologically confirmed esophageal adenocarcinoma (n = 282), adenocarcinoma of the gastric cardia (n = 255), esophageal squamous cell carcinoma (n = 206), and noncardia gastric adenocarcinoma (n = 352), along with population controls (n = 687). Associations between nutrient intake and risk of cancer were estimated by adjusted odds ratios (ORs), comparing the 75th versus the 25th percentile of intake. The following nutrients were significantly inversely associated with risk of all four tumor types: fiber, beta-carotene, folate, and vitamins C and B6. In contrast, dietary cholesterol, animal protein, and vitamin B12 were significantly positively associated with risk of all four tumor types. Dietary fat [OR, 2.18; 95% confidence interval (CI), 1.27-3.76] was significantly associated with risk of esophageal adenocarcinoma only. Dietary nitrite (OR, 1.65; 95% CI, 1.26-2.16) was associated with noncardia gastric cancer only. Vitamin C supplement use was associated with a significantly lower risk for noncardia gastric cancer (OR, 0.60; 95% CI, 0.41-0.88). Higher intake of nutrients found primarily in plant-based foods was associated with a reduced risk of adenocarcinomas of the esophagus and gastric cardia, whereas higher intake of nutrients found primarily in foods of animal origin was associated with an increased risk.

Risk of esophageal and gastric adenocarcinomas in relation to use of calcium channel blockers, asthma drugs, and other medications that promote gastroesophageal reflux.

Incidence of adenocarcinomas of the esophagus and gastric cardia has risen dramatically over the past 2 decades in the U. S., for reasons that are not yet clear. A number of common medications (e.g., calcium channel blockers, tricyclic antidepressants, and certain asthma medications) promote gastroesophageal reflux by relaxing the lower esophageal sphincter (LES). Reflux is thought to increase cancer risk by promoting cellular proliferation, and by exposing the esophageal epithelium to potentially genotoxic gastric and intestinal contents. Recent studies have suggested that calcium channel blockers may also increase cancer risk by inhibiting apoptosis. Using personal interview data from a multicenter, population-based case-control study conducted between 1993 and 1995 in three areas of the U. S., we evaluated whether the use of LES-relaxing drugs was associated with increased risk of adenocarcinomas of the esophagus and gastric cardia. Cases of esophageal adenocarcinoma (n = 293) and gastric cardia adenocarcinoma (n = 261) were compared with general population controls (n = 695). Information on additional case groups of esophageal squamous cell carcinoma (n = 221) and noncardia gastric cancer (n = 368) were also available for comparison. Overall, 27.4% of controls had used one or more of these drugs for at least 6 months, compared with 30.2% of esophageal adenocarcinoma and 23.8% of gastric cardia adenocarcinoma cases. The adjusted odds ratios (ORs) for ever use were 1.0 [95% confidence interval (CI) = 0.7-1.5] and 0.8 (95% CI = 0.5-1.1), respectively. There was little evidence of increasing risk with increasing duration of use of all LES-relaxing drugs together. We found an increased risk of esophageal adenocarcinoma among persons reporting use of asthma drugs containing theophylline (OR = 2.5; 95% CI = 1.1-5.6) or beta agonists (OR = 1.7; 95% CI = 0.8-3.8). Risks were higher among long-term users (>5 years) of these drugs (OR = 3.1; 95% CI = 0.9-10.3 and OR = 2.3; 95% CI = 0.8-7.0, respectively). In contrast, there was no evidence that the use of calcium channel blockers or other specific groups of drugs increased the risk of any of the cancers studied. These results provide reassuring evidence that the increases in incidence of adenocarcinomas of the esophagus and gastric cardia are not likely to be related to the use of LES-relaxing drugs as a group, or calcium channel blockers in particular, but they do suggest that persons treated for long-standing asthma may be at increased risk of esophageal adenocarcinoma.

Regional variation in survival following the diagnosis of cancer.

Recent studies have documented substantial geographical variation in patterns of treatment of cancer and other diseases. Because cancer treatment is not uniform nationwide, survival following the diagnosis of cancer might also be expected to vary geographically. Survival data from the nine population-based registries in the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program were analyzed for cancers of the stomach, colon, rectum, lung, breast, uterus, ovary, prostate, and bladder (n = 401,861). The patients included all non-Hispanic white patients diagnosed with cancer of one of the selected sites during 1983-1991. Regional variation in crude five-year survival rates across the nine SEER areas was most marked for cancers of the uterus and prostate. For uterine cancer, for example, five-year survival ranged from 73.2% in Connecticut to 84.0% in Hawaii. Less marked variation was observed for cancers of the colon, rectum, and breast. For cancers of the bladder, ovary, stomach, and lung, survival rates five years after diagnosis were relatively invariant across the SEER areas. Observed differences in survival rates, although statistically significant, were relatively modest from the standpoint of the practicing physician. Nonetheless, the general pattern of regional variation was unchanged after adjustment for sex, age, stage, and surgical treatment and when analysis was limited to patients younger than 70 years, who would be anticipated to have the least comorbidity.

Temporal and regional variability in the surgical treatment of cancer among older people.

OBJECTIVES: Numerous studies have documented that older individuals with cancer have been treated less aggressively than younger individuals. We utilized data from the National Cancer Institute's Surveillance, Epidemiology, and End Results (SEER) Program to assess geographic variation in the effects of age on the surgical treatment of cancer and to examine temporal trends in the treatment of older people between 1973 and 1991. DESIGN: Ecological analyses comparing time trends of treatment in nine regions over a 19-year period. SETTING: Population-based data for nine geographic areas of the United States. PARTICIPANTS: Persons with incident cancer of the breast, colon, rectum, lung, ovary, uterus, pancreas, and stomach residing in these regions during the selected years. RESULTS: After age 64, the percentage of patients treated surgically decreased with increasing age for every cancer site studied. Between 1973 and 1991, the likelihood of receiving surgery for cancers of the uterus, colon, rectum, ovary, and breast increased more rapidly among patients ages 65 years and older than among those under age 65. This relative increase occurred consistently across most geographic areas studied. For cancers of the lung, stomach, and pancreas, the effect of age on the likelihood of receiving surgery did not diminish through the period under study.: CONCLUSIONS: Although older people remain less likely to receive surgical therapy than younger persons, for some important cancer sites the gap between the treatment of older and younger individuals narrowed from 1973 through 1991.

The impact of extended maternity services on prenatal care use among Medicaid women.

The goals of this study were to examine the use of maternity support services (MSS) and maternity case management (MCM) by Medicaid-eligible low-income pregnant women in Washington state, and to determine whether receipt of the services was associated with improved prenatal care use. We obtained data from linked birth certificates and Medicaid eligibility and claims files for women delivering between August 1989 and December 1991. Unconditional logistic regression was used to assess the programs' effects independent of other variables associated with prenatal care adequacy. The percentage of women receiving MSS and MCM was highest among women with demographic risks for adverse birth outcomes. Women receiving prenatal care from health departments or community clinics were more likely to receive MSS and MCM than those seen by private physicians or midwives. After adjustment for multiple confounding factors, we found that recipients of MSS, but not MCM, were significantly less likely than other women to receive an inadequate number of prenatal visits. Our findings suggest that public policies that pay for support services to low-income pregnant women can improve the use of prenatal care.

Identification of the high-risk smoker.

The burden of smoking-related disease can be reduced through smoking prevention and cessation. While physicians should counsel all patients who smoke, disease risks are not uniform among smokers. The smokers of the largest numbers of cigarettes are at greatest risk, as are those who started at younger ages. In addition, other factors determined the likelihood of developing a smoking-related disease. Physicians can use the combination of smoking with these additional risk factors to target smokers for intervention, whether education, cessation, or appropriate screening.

Diet and the risk of pancreatic cancer in men.

To examine the relation between diet and pancreatic cancer, we conducted a population-based case-control study in western Washington. Cases (n = 148) were married men, aged 20-74 years, who were diagnosed with pancreatic cancer from July 1982 to June 1986. Controls (n = 188), identified by random digit dialing, were frequency matched to the cases by 5-year age groups. Wives responded as surrogates for both cases and controls. Wives were interviewed by telephone, and they completed a mailed, self-administered food frequency questionnaire. Results indicated that pancreatic cancer risk increased with increasing protein intake. The increased risk for heavy consumers of protein was largely confined to individuals aged 65 years and above. In that group, the odds ratio for those in the highest quartile of protein intake, relative to the lowest, was 6.0 (95% confidence interval 1.7-20.6). No association was found between pancreatic cancer risk and the intake of total fat, saturated fat, cholesterol, omega-3 fatty acids, or vitamins A and C. There was an unexpected inverse association between calcium intake and pancreatic cancer risk in these data. These findings are discussed in relation to possible etiologic mechanisms that they suggest.

Risk of pancreatic cancer in relation to medical history and the use of tobacco, alcohol and coffee.

A population-based case-control study was conducted to examine the relationship between certain medical conditions, the use of tobacco, alcohol and coffee, and the incidence of pancreatic cancer. Cases (N = 148) were married men ages 20 through 74 years diagnosed with pancreatic cancer from July 1982 through June 1986. Controls (N = 188) were identified by random digit dialing. Wives, responding as surrogates for both cases and controls, were interviewed by telephone and completed, alone, a food frequency questionnaire. The risk of pancreatic cancer was increased in individuals with a history of diabetes or pancreatitis, and decreased in those with a history of tonsillectomy. Individuals who had ever smoked cigarettes were at elevated risk of disease. This excess risk was confined to current smokers, in whom the odds ratio was 3.2 (95% CI 1.8-5.7); the risk among former smokers resembled that in those who had never smoked. There was no excess risk of pancreatic cancer among those who had ever used other forms of tobacco, including pipe tobacco, cigars and chewing tobacco. After adjustment for demographic and dietary characteristics, there was no association between pancreatic cancer risk and the intake of coffee, beer, red wine, hard liquor or all alcohol combined; a slight reduction in risk was seen among those consuming white wine daily.

Differential survival related to smoking among patients with renal cell carcinoma.

We examined the relation between tobacco smoking and survival among renal cell carcinoma cases, using information from a population-based cancer registry. Current smokers were at increased risk of death compared with non-smokers (hazard ratio (HR) 1.7, 95% CI = 1.2-2.5). The association was strongest within 6 months of diagnosis (HR 2.5, 95% CI = 1.5-4.3). Most of the association was explained by stage at diagnosis, with current smokers more likely (OR 2.2, 95% CI = 1.4-3.5) to have distant disease. Case-control studies may be biased if cases who do not participate owing to short survival have different smoking histories than participating cases.

Determinants of survival following the diagnosis of esophageal adenocarcinoma (United States).

The rapidly rising incidence of esophageal adenocarcinomas in the United States and western Europe remains unexplained. Most persons who develop the disease have had long-standing gastroesophageal reflux symptoms with concomitant Barrett's metaplasia. They are, therefore, potentially identifiable for endoscopic screening and cancer surveillance, which should facilitate the early detection of these tumors. We undertook these analyses to determine the extent to which the opportunity for early diagnosis and treatment of esophageal adenocarcinomas has been realized in the US. Specifically, using data from the Surveillance, Epidemiology, and End Results (SEER) program of the US National Cancer Institute, we examined changes in stage of disease at diagnosis and in survival between 1973 and 1991 and investigated patient characteristics as predictors of survival. Improvements in stage at diagnosis and in survival between 1973 and 1991 were minor and clinically insignificant; overall five-year survival never exceeded 10 percent. Stage of disease at diagnosis was the strongest determinant of subsequent survival; five-year survival with patients with in situ tumors was 68.2 percent. This survival advantage persisted up to 15 years after diagnosis and was independent of other prognostic factors. We conclude that the opportunity for reduction in esophageal cancer mortality has been largely unrealized in the US. In light of the increasing incidence of esophageal adenocarcinoma, efforts should be devoted to identifying those at highest risk of developing Barrett's metaplasia and subsequent adenocarcinoma, and to developing cost-effective primary prevention and cancer surveillance methods targetting them.

Comparability of information provided by elderly cancer patients and surrogates regarding health and functional status, social network, and life events.

Stressful life events may adversely affect health, but these effects may be mitigated by a strong social support network. The elderly should be a focus for investigation in this area, but epidemiologic studies in the elderly are often complicated by the need to rely upon information collected from surrogates. We assessed the degree to which 622 elderly cancer patients and surrogates agreed in their answers to questions about life events, social network, and health and functional status. We compared spouses, friends, and children as surrogates by evaluating (1) their agreement with index subjects and (2) the proportion of surrogates of each type who could not answer one or more questions. Agreement for most items on health and functional status was at least moderately good, but agreement was more inconsistent for questions about social network and major life events. Agreement was higher when factual information was solicited than when the question regarded more personal information or attitudes. For questions about health status, social network, and life events, spouses agreed more closely with index subjects than did children or friends. For questions about functional status, however, agreement did not vary consistently by respondent type. We found no consistent effects of the index subject's age or cognitive status on levels of agreement, nor did we find evidence of bias in the surrogates' responses.

Geographic variation in the treatment of localized breast cancer.

BACKGROUND: Breast-conserving surgery followed by radiation therapy is as effective as modified radical mastectomy in treating women with localized breast cancer, as demonstrated by clinical trials reported during the 1980s. The extent to which breast-conserving surgery has been adopted in various regions of the United States is not known. METHODS: We assessed variations in the use of breast-conserving surgery from 1983 to 1986 in 18,399 non-Hispanic white, 324 Hispanic, and 1174 black women with breast cancer in nine areas of the United States, using data from the Surveillance, Epidemiology, and End-Results Program of the National Cancer Institute. RESULTS: The proportion of white women with localized breast cancer who underwent breast-conserving surgery ranged from 9.2 percent (Iowa) to 32.1 percent (Seattle) in 1983-1984 and from 19.6 percent (Iowa) to 41.5 percent (Seattle) in 1985-1986. Between 1983 and 1986, the use of breast-conserving surgery increased in each area, but the relative rankings of the areas changed little. The frequency of the use of breast-conserving surgery in black and Hispanic women was comparable to that in white women. The proportion of women who received radiation therapy after breast-conserving surgery also varied geographically and increased from 1983 to 1986 in all areas. Older women in all areas and black women in Atlanta and Detroit were less likely than other women to receive radiotherapy after breast-conserving surgery. CONCLUSIONS: There is marked variation in the United States in the use of breast-conserving surgery for localized breast cancer. The variation is not explained by demographic factors, although race and age affect the use of radiotherapy after breast-conserving surgery.

Determinants of receiving breast-conserving surgery. The Surveillance, Epidemiology, and End Results Program, 1983-1986.

BACKGROUND: Although breast-conserving surgery was used with increasing frequency during the 1980s for management of breast cancer, most women still undergo mastectomy, and a substantial variation has been documented in the proportion of women receiving breast-conserving surgery across regions of the country. Using data from the Surveillance, Epidemiology, and End Results (SEER) Program for 1983-1986, we assessed characteristics of the county of residence as predictors of receipt of breast-conserving surgery and determined whether regional variation persisted after considering these characteristics. METHODS: The data used involved all 19,661 non-Hispanic white women with localized breast cancer diagnosed in 1983 through 1986 in the nine SEER regions. Information on county characteristics was obtained from standard sources and merged with the SEER data. Univariate multivariate statistical methods were used to assess the effects of county characteristics on type of surgery for breast cancer. RESULTS: As anticipated, age was a strong predictor of type of surgery. In analyses that controlled for age, county characteristics that significantly predicted receipt of breast-conserving surgery included physician-to-population ratio, education and income levels, the presence of cancer center, and the presence of a city of at least 100,000. After controlling for these factors using multiple logistic regression, substantial regional variation persisted. CONCLUSIONS: Regional variation in treatment of localized breast cancer across the SEER regions is not explained by patient's age or county characteristics. Research is needed to address the decision making of individual patients and their physicians regarding type of surgery.

Biased comparisons of lung cancer survival across geographic areas: effects of stage bias.

This report illustrates bias that may affect comparative analyses of cancer survival across geographic areas and describes how it limits the conclusions that can be drawn from such data. Despite the application of a standardized procedure for assigning tumor stage, patients from different areas who are assigned the same stage category may not be homogeneous with respect to extent of disease, if staging is accomplished more aggressively in one group than in another. Stage-specific comparisons of survival may be biased as a result. Cancer sites for which a large majority of patients are treated surgically may be less susceptible to stage bias.

Association of obesity and ovarian cancer in a case-control study.

Data from a population-based case-control study conducted in Washington State and Utah were used to assess whether obesity is associated with an altered risk of epithelial ovarian cancer. Quetelet index, defined as weight (kg) at age 30 years divided by height (m) squared, was calculated for each woman, and the values for all subjects were divided into five categories of approximately equal size. Compared with women in the lowest category, women in the highest category had an odds ratio of 1.7 (95 per cent CI 1.1-2.7). Risks for women in the three intermediate Quetelet index categories also exceeded the risk for women in the lowest group, but to a much smaller degree. Among women with serous tumors, those in the highest Quetelet index category were at a greater than twofold excess risk (OR = 2.2, 95 per cent CI 1.1-4.2), but the risk was not increased in the intermediate categories. For endometrioid tumors, risk increased consistently with increasing Quetelet index, and the odds ratio in the highest category was 4.7 (95 per cent CI 1.0-22.7). For both serous and endometrioid tumors, the excess risk was largely confined to premenopausal women. The results of this analysis suggest that for at least some types of ovarian tumor, obesity may warrant further attention as a possible etiologic factor.

Diet and nasopharyngeal cancer in a low-risk population.

Asian studies have reported that risk of nasopharyngeal cancer (NPC) is increased in individuals who frequently consume salted fish, which contains high levels of N-nitroso compounds. As part of a collaborative, population-based, case-control study in the U.S., where the annual incidence of the disease is low, we investigated whether dietary intake of preformed nitrosamines or nitrosamine precursors, or of antioxidants including vitamin C and carotenoids, was associated with altered risk of NPC overall, or of specific histologic subtypes of disease. Cases (n = 133) identified at 5 population-based cancer registries and controls (n = 212) identified through random digit dialing completed a telephone interview and self-administered food frequency questionnaire. Dietary exposures were expressed as quartiles of intake, and odds ratios (ORs) calculated using the lowest quartile of intake as the reference category. Risk of non-keratinizing and undifferentiated tumors of the nasopharynx was increased in frequent consumers of preserved meats, which contain high levels of added nitrites. ORs in the 2nd, 3rd and highest quartile were 1.99, 4.35 and 4.59, although 95% confidence intervals did not exclude 1.0. Risk of differentiated squamous cell carcinoma, but not other histologic types, was significantly reduced in individuals with vitamin C intake above the lowest quartile (ORs 0.30, 0.33 and 0.30 in the 2nd, 3rd and highest quartiles, respectively). This association was markedly stronger among non-smokers and former smokers than among current smokers. Finally, individuals who reported consuming supplemental vitamins were at an approximately 50% reduced risk of NPC. Our results indicate that future studies should consider the effects of dietary risk factors on the risk of specific histologic subsets of NPC, and not assume that the disease is etiologically homogeneous.