Associations Between a Priori-Defined Diet Quality Indices and Survival in a Longitudinal Cohort Study of Head and Neck Squamous Cell Carcinoma Patients

Abstract

Abstract Objectives Use of diet quality indices has become increasingly common. However, there are no studies, to date, that have examined the relationship of a priori indices with prognostic outcomes in the head and neck squamous cell carcinoma (HNSCC) population. The purpose of this analysis was to examine associations between diet quality in the first two years after diagnosis and mortality in a sample of HNSCC patients. Methods This was a secondary analysis of 396 newly diagnosed HNSCC patients recruited from the University of Michigan Head and Neck Specialized Programs of Research Excellence longitudinal cohort study. Participants completed food frequency questionnaires (FFQs) and health surveys at the time of diagnosis, 1-year post-diagnosis (n = 341), and 2-years post-diagnosis (n = 217). Cox Proportional Hazards models assessed associations between diet quality index score (categorized as quintiles and modeled as a time-varying covariate), all-cause-, and cancer-specific mortality. The indices chosen for examination included the Alternative Mediterranean Diet Index (aMED), Alternate Healthy Eating Index-2010 (AHEI-2010), Dietary Approaches to Stop Hypertension (DASH), and a ketogenic diet index. All models adjusted for age, sex, BMI, HPV-status, tumor site, stage, education, smoking, drinking, and total calorie intake (modeled as time-varying). Results There were 105 total deaths and 67 cancer-related deaths, during a median follow-up time of 3 years. Higher aMED and AHEI-2010 scores were significantly inversely associated with all-cause (HRaMED 0.33, 95% CI: 0.18–0.59, ptrend < 0.0001; HRaHEI 0.34, 95% CI: 0.20–0.58, ptrend < 0.001) and cancer-related mortality (HRaMED 0.28, 95% CI: 0.13–0.59, ptrend = 0.001; HRaHEI 0.35, 95% CI: 0.18–0.71, ptrend < 0.0001). There were no significant associations found with the DASH index. The ketogenic diet index showed an attenuated significant inverse association for the fifth quintile, but there was a lack of linear trend. Conclusions Developing and testing interventions that follow food and quantity guidelines emphasized by the aMED and AHEI-2010 diet quality indices among HNSCC survivors, is warranted. Funding Sources NIH/NCI and a USDA/NIFA Hatch Project.