Dietary quality in early-onset gastrointestinal malignancies: NHANES 2005-2018.

3549 Background: The incidence of early-onset colorectal, stomach, esophageal, and pancreatic cancer, which occur in individuals <50 years of age, is rising worldwide. However, little is known about why rates have been up trending. Previous studies have linked poor diet and nutrition to an increased risk of typical-onset gastrointestinal (GI) malignancies. In this study, we aimed to describe dietary quality in individuals with early-onset GI malignancies and compare it to typical-onset GI malignancies. Methods: We queried the 2005-2018 National Health and Nutrition Examination Survey (NHANES) database for all survey participants diagnosed with GI malignancies, including colorectal, esophageal, gallbladder, liver, pancreas, rectal, and stomach cancer. We classified early and typical-onset GI cancer based on age of diagnosis of <50 and ≥50 years, respectively. Participants <50 without a history of cancer were classified as controls. We assessed the dietary quality by calculating the healthy eating index score (HEI-2015) from NHANES database for all participants. The primary outcome was mean HEI-2015 score for early onset GI malignancies. We performed propensity score weighted analysis to balance gender, race, income, education level, and BMI between the groups. We used logistic regression analysis to assess the relationship between early-onset GI malignancies and HEI-2015 quintiles. Results: Out of 32,235 participants in NHANES database, 330 individuals (1%) reported a diagnosis of GI malignancy. Sixty-eight (20.5%) individuals had early-onset GI malignancies. We identified 17,420 controls who were <50 years old with no reported history of malignancy. Fifty-three percent of early-onset GI cancers were female vs 47% of typical-onset GI cancers; 59% vs 59% were non-Hispanic White; 66% vs 40% had a college degree; 13% vs 7% had high income and 40% vs 36% were obese. The mean HEI-2015 score in early-onset GI cancer was 52.9 (SD 14.7) compared to 53.6 (SD 13.0) in typical-onset GI cancers, which was not significantly different (p=0.72). There was no significant association between poor HEI-2015 scores in individuals with early-onset vs typical-onset GI malignancies on logistic regression analysis (OR 0.84 [0.35-2.03]). In addition, HEI-2015 quintiles were not significantly associated with those with early-onset GI malignancies compared to controls <50 years old (OR 2.06 [0.72-5.9]). Conclusions: The diet quality of early-onset GI malignancies does not align with the national Dietary Guidelines. However, in this analysis of the NHANES database, we did not find a significant difference in HEI-2015 scores between early-onset versus typical-onset GI malignancies.