Nutritional factors and non-Hodgkin lymphoma survival in an ethnically diverse population: the Multiethnic Cohort.

BACKGROUND/OBJECTIVES: To understand the possible effect of modifiable health behaviors on the prognosis of the increasing number of non-Hodgkin lymphoma (NHL) survivors, we examined the pre-diagnostic intake of major food groups with all-cause and NHL-specific survival in the Multiethnic Cohort (MEC). SUBJECTS/METHODS: This analysis included 2339 participants free of NHL at cohort entry and diagnosed with NHL as identified by cancer registries during follow-up. Deaths were ascertained through routine linkages to state and national death registries. Cox proportional hazards regression was applied to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for overall and NHL-specific mortality according to pre-diagnostic intake of vegetables, fruits, red meat, processed meat, fish, legumes, dietary fiber, dairy products and soy foods assessed by food frequency questionnaire. RESULTS: The mean age at diagnosis was 71.8±8.5 years. During 4.5±4.1 years of follow-up, 1348 deaths, including 903 NHL-specific deaths, occurred. In multivariable models, dairy intake was associated with higher all-cause mortality (highest vs lowest tertile: HR=1.14, 95% CI 1.00-1.31, Ptrend=0.03) and NHL-specific (HR=1.16, 95% CI 0.98-1.37) mortality. Legume intake above the lowest tertile was related to significant 13-16% lower all-cause and NHL-specific mortality, whereas red meat and fish intake in the intermediate tertiles was associated with lower NHL-specific mortality. No association with survival was detected for the other food groups. CONCLUSIONS: These data suggest that pre-diagnostic dietary intake may not appreciably contribute to NHL survival, although the higher mortality for dairy products and the better prognosis associated with legumes agree with known biologic effects of these foods.

Ethnic differences in serum adipokine and C-reactive protein levels: the multiethnic cohort.

BACKGROUND: Ethnic disparities in metabolic disease risk may be the result of differences in circulating adipokines and inflammatory markers related to ethnic variations in obesity and body fat distribution. SUBJECTS/METHODS: In a cross-sectional design, we compared serum levels of leptin, adiponectin, C-reactive protein (CRP), interleukin-6 (IL-6) and tumor necrosis factor-α (TNF-α) in control subjects (321 men and 930 women) from two nested case-control studies conducted within the Multiethnic Cohort Study consisting of whites, Japanese Americans (JA), Latinos, African Americans (AA) and Native Hawaiians (NH). General linear models were applied to evaluate ethnic differences in log-transformed serum biomarker levels before and after adjusting for body mass index (BMI) at cohort entry. RESULTS: In comparison to whites, significant ethnic differences were observed for all biomarkers except TNF-α. JA men and women had significantly lower leptin and CRP levels than whites, and JA women also had lower adiponectin levels. Leptin was significantly higher in AA women (P < 0.01), adiponectin was significantly lower in AA men and women (P = 0.02 and P < 0.001), and CRP and IL-6 were significantly higher in AA men and women. Lower adiponectin (P < 0.0001) and CRP (P = 0.03) levels were the only biomarkers in NH women that differed from whites; no statistically significant differences were seen for NH men and for Latino men and women. When adjusted for BMI at cohort entry, the differences between the lowest and the highest values across ethnic groups decreased for all biomarkers except adiponectin in men indicating that ethnic differences were partially due to weight status. CONCLUSIONS: These findings demonstrate the ethnic variations in circulating adipokine and CRP levels before and after adjustment for BMI. Given the limitation of BMI as a general measure of obesity, further investigation with visceral and subcutaneous adiposity measures are warranted to elucidate ethnicity-related differences in adiposity in relation to disparities in obesity-related disease risk.