Is language barrier a mere proxy?: Psychological distress among non–English-speaking prostate cancer patients receiving radiation therapy—A prospective cohort study.

5113 Background: Cancer patients experience depression and anxiety at elevated rates, with minority and non-English-speaking populations disproportionately affected. Prior research at our institution demonstrated elevated distress among Spanish-speaking breast cancer patients undergoing radiation therapy. However, limited data exist on language-related disparities in prostate cancer patients undergoing radiation therapy. We investigated this while exploring the mediating role of social determinants of health (SDOH). Methods: We enrolled prostate cancer patients initiating radiation therapy at an urban safety-net hospital. We assessed anxiety (GAD-2), depression (PHQ-2), and SDOH including food and housing insecurity, and transportation barriers at baseline and treatment completion. Comparisons used t-tests and Mann-Whitney U tests; correlations used Spearman coefficients with statistical significance p < 0.05. Results: Of 84 patients enrolled (mean age 70.3±8.0 years), 49 (58.3%) were English-speaking and 35 (41.7%) were non-English-speaking (Spanish n = 20, Haitian Creole n = 13, Portuguese n = 2). The sample was predominantly Black/Caribbean (64.3%). Non-English speakers had fewer years in the US (41.1 vs. 50.3 years, p = 0.016) and lower educational attainment (25.7% vs. 0% elementary school only, p = 0.006). At baseline, non-English speakers showed numerically higher anxiety (GAD-2: 1.31±2.13 vs. 1.08±1.72) and depression (PHQ-2: 1.09±1.87 vs. 0.84±1.57), though this was not statistically significant (p > 0.50). Notably, Spanish-speaking patients demonstrated the highest distress (GAD-2: 2.00±2.45; PHQ-2: 1.40±1.98), with 30.0% screening positive for clinical anxiety vs. 16.3% of English speakers. SDOH factors were significantly correlated with distress: housing insecurity (ρ = 0.42, p < 0.001), food insecurity (ρ = 0.37, p = 0.001), and transportation worry (ρ = 0.38, p < 0.001). In multivariable regression adjusting for age, stage, race, ethnicity, and education, language was not independently associated with anxiety (B = -0.64, 95% CI: -1.76 to 0.48, p = 0.26). Among 73 patients with paired data, distress remained stable from baseline to treatment completion. Conclusions: While non-English-speaking prostate cancer patients reported higher psychological distress during radiation therapy, SDOH factors emerged as the strongest predictors of distress regardless of language. Interventions targeting social vulnerability may be more impactful than language services alone. SDOH screening should be integrated into oncology care at safety-net institutions serving diverse populations.