The Impact of Body Mass Index and Residence in Food Priority Areas on Patterns-of-Care and Cancer Outcomes in Patients with Stage III Non-Small Cell Lung Cancer

Abstract

<h3>Purpose/Objective(s)</h3> Nutritional deficiencies have been linked to poor prognosis in multiple cancer sites. Patients living in food priority areas (FPAs) where access to healthy meals is scarce may be at higher risk of nutritional deficits, leading to poor cancer outcomes. Currently, there is no published data analyzing how FPA residence impacts treatments rendered or outcomes for patients with locally advanced non-small cell lung cancer (LA-NSCLC). We aim to characterize socioeconomic and cancer demographics of LA- NSCLC patients living in zip code-designated FPAs treated with curative intent at a single institution. We hypothesize that living in a FPA will impact treatments rendered, patterns of failure, and cancer outcomes. <h3>Materials/Methods</h3> This is an IRB-approved retrospective study of 573 LA-NSCLC patients curatively treated January 2000-January 2020. Chi-square tests were done to compare variables stratified by residence in FPAs. Mann-Whitney U-test was used to compare differences between continuous variables. Kaplan-Meier analysis and cox proportional hazard models were used to analyze overall survival (OS) and freedom from recurrence (FFR) for whole cohort and stratified by BMI. Cox regression with forward model selection was used for multivariate analysis (MVA). <h3>Results</h3> Of the 32% of patients living in FPAs (N=183), more were black (80%, p<0.0001), single (p<0.001), <60yo (p=0.001), uninsured (p< 0.0001), and had a lower median income (p< 0.001). Patients in FPAs had lower mean pre-chemoradiation (CRT) albumin (p=.002) and pre-CRT BMI (p= 0.026) and were less likely to receive trimodality therapy (p=<0.001) or consolidative immunotherapy (p=0.012). There was no difference in OS (21.8m vs. 26.2m, p=0.207) or FFR (11.2m vs 11.3m, p=0.739) between FPA vs No FPA groups. In the subset with pre- and post-CRT normal BMI, there was a significant difference in OS for patients living in an FPA (Median OS 18.4m vs 25.0m, p=0.005 and 15.1m vs. 28.1m, p=0.002; respectively) but no difference in FFR (11.0m in pre (p=0.463) and post-CRT (p=0.526)) There was no significant difference in OS or FFR for obese, underweight, or overweight patients. In all patients, nutritional aspects predictive of OS on MVA were pre-CRT albumin (HR 0.753, p=0.010) and post-CRT BMI (Obese as REF; Underweight: HR 1.737, p=0.043). In patients with post-CRT normal BMI (N=152), living in a FPA (HR 1.888, p=0.008) and pre-CRT albumin (HR 0.645, p=0.004) were predictors of OS on MVA. <h3>Conclusion</h3> In this institutional analysis examining impact of FPAs in patients with LA- NSCLC, we demonstrate a socioeconomic divide in this population, where residing in FPAs reflects distinct patient demographics and receipt of less aggressive therapy. Patients with normal BMI before therapy and particularly after CRT had worse OS if living in a FPA but there was no difference in FFR in this group. To overcome limitations of a retrospective analysis, we are currently characterizing nutritional needs of our patients in a prospective study.