Abstract
BACKGROUND: Culturally and/or linguistically diverse (CALD) patients have unique health needs and may face multiple barriers when accessing healthcare. This study explored the association between of CALD status on treatment received and outcomes for patients with pancreatic cancer. METHODS: Data were extracted from the multi-site PURPLE Pancreatic cancer Translational Registry between January 2016 and April 2023. Registry data was supplemented by country of birth and preferred language data from linkage with hospital administrative datasets. CALD status was defined by being born overseas in a non-main English-speaking country and/or having a preferred language other than English. Descriptive statistics were used to analyse demographic data. Survival analysis was conducted using Kaplan-Meier estimates to generate survival curves, with comparisons assessed via the log-rank test. Moreover, univariable and multivariable Cox proportional hazards regression were employed. RESULTS: Of 1796 patients with pancreatic cancer enrolled at seven participating institutions, 1451 (80.8%) had their CALD status determined; with 661 (46%) identified as CALD. The CALD population were older (median age 72 vs. 68 years; P < 0.001), with a worse performance status (Eastern Cooperative Oncology Group score > 1: 20 vs. 13%, P = 0.004) and a greater number of comorbidities (Charlson Comorbidity Index > 3: 55 vs. 43%, P < 0.001). The use of neoadjuvant therapy in resectable/borderline resectable disease was similar. However, fewer CALD patients proceeded to curative-intent surgery following neoadjuvant therapy (30% vs. 51%, P = 0.041). In the metastatic setting, a higher proportion of CALD patients were offered best supportive care (50% vs. 41%; P = 0.021). Overall, there were no significant differences identified in the progression-free, recurrence-free or overall survival for CALD patients with pancreatic cancer across all stages of disease. CONCLUSIONS: CALD status was associated with multiple adverse prognostic factors (age, PS, comorbidities), which likely impacted differences in treatment received and challenges analysis of the independent impact of CALD status on outcomes. Notably, however, there were no striking differences by CALD status in treatment delivered or survival outcomes.