Compliance with dietary guidelines and its relationship with symptoms and clinical outcomes in patients with advanced cancer in early-phase oncology clinical trials.

Abstract

199 Background: We examined compliance with the Dietary Guidelines for Americans and its association with symptom burden and clinical outcomes in patients with advanced cancer in early-phase clinical trials testing novel immunotherapeutic and targeted agents. Methods: Patients starting an early-phase clinical trial (ECOG-PS = 0-1) were recruited into a prospective, longitudinal design with assessments at baseline and at the end of Cycle 1. Diet and symptom burden were assessed using the validated National Cancer Institute Diet History Questionnaire (NCI-DHQ) and the MD Anderson Symptom Inventory, respectively. Compliance with the Dietary Guidelines for Americans recommendations was measured via the Healthy Eating Index (HEI) (a measure of dietary intake per total energy), computed from NCI-DHQ food group and nutrient scores; higher HEI scores indicate greater compliance with dietary guidance recommendations (possible range = 0–100). Statistical tests included Spearman rank correlations (rho), and Cox proportional hazards models. Results: Among early-phase clinical trial patients [N = 40; 50% female; 80% Non-Hispanic White; 80% ECOG = 1; 36% on trials including an immunotherapeutic agent and 64% on targeted therapy trials; mean age = 55y; mean BMI = 28], mean HEI was 69, compared to 59 for the US general population. The proportion of phase I clinical trial patients who met adequacy guidelines were 80% for whole fruit, 73% for total protein foods, 55% for seafood and plant proteins, 55% for total fruit, 50% for greens and beans, 28% for total vegetables, 15% for fatty acids [(PUFAs + MUFAs)/SFAs ≥2.5], 13% for dairy, and 0% for whole grains. The proportion of patients who met moderation guidelines were 28% for refined grains, 28% for added sugar, 13% for saturated fat, and 0% for sodium. Female patients had higher HEI scores than male patients (73 vs. 65, P = 0.004). Patients who were normal weight (BMI < 25) had higher scores for meeting the moderation in sugar intake guideline than overweight patients (BMI≥25) (7.7 vs. 5.5, P = 0.031). Higher intakes of cooked lean meat from beef, pork, veal, lamb, and game were linked to prolonged overall survival (HR = 0.5, 95%CI = 0.26, 0.96, P = 0.039). In immunotherapy patients, greater compliance with seafood and plant protein recommendations was associated with less fatigue at end of Cycle 1 (rho = -0.7, P = 0.008); in targeted therapy patients, higher glycemic load was associated with worse pain (rho = 0.7, P = 0.004). Conclusions: Diets of these early-phase clinical trial patients overall were congruent with recommendations in the Dietary Guidelines for Americans. However, increasing intakes of whole grains and reducing sodium intakes may be useful dietary goals for this population. Also, dietary factors may influence symptoms, such as fatigue and pain, in early-phase clinical trial patients with advanced cancer.