Racial Disparities in Survival of Patients with High-Risk LA SCCHN in the U.S.

Abstract

Approximately 50-67% of patients with squamous cell carcinomas of the head and neck (SCCHN) present with locally advanced (LA) disease. Human papillomavirus (HPV) associated SCCHN has improved survival compared to non-HPV SCCHN. Studies have analyzed racial disparities in LA SCCHN but few have adjusted for HPV status and socio-economic status (SES). Our objective was to characterize disparities in LA SCCHN. We identified high-risk patients from 2010-2017 with AJCC v8 Stage IVA/IVB SCCHN of the oral cavity, oropharynx (OP), larynx or hypopharynx, or Stage III SCCHN of the OP in the Surveillance, Epidemiology, and End Results (SEER) with HPV Status and Census Tract-level SES/Rurality Combined Database. We excluded OP patients with missing HPV data. SEER-reported treatment was used to classify initial treatment as definitive therapy (DT) categories: surgical DT, non-surgical DT or non-DT. The Kaplan Meier method was used to estimate overall survival (OS) with 95% confidence intervals (CI). Multivariable cox proportional hazard models were used for associations between covariates and hazard ratio (HR) of death, adjusted for age, sex, race, ethnicity, DT category, marital status, rurality, tumor size, cancer stage, and HPV status. We explored the impact of additionally adjusting for Yost SES index quintiles. We identified 17,818 eligible patients: 79.3% White, 14.4% Black, 5.8% Asian/Pacific Islander (A/PI), and 0.6% American Indian/Alaska Native (AI/AN). Primary tumor sites were oral cavity (36.8%), larynx (29.0%), OP (24.4%), and hypopharynx (9.9%). 10.4% were HPV-associated. Race and SES quintiles were related (chi-squared, p<0.001) and the majority (56.5%) of black patients were in the lowest SES quintile. mOS was shorter and risk of death was significantly higher for black vs white patients in both the all-site and OP-only cohorts. When we added SES to multivariable analyses, Black race was no longer associated with increased risk of death in the all-site or OP-only cohorts. (Table 1) CONCLUSION: We found that when adjusting for sociodemographic and clinical factors, Black race was independently associated with a higher risk of death compared to white patients. When we adjusted for SES in multivariable analysis the association between Black race and risk of death was no longer significant, consistent with previously published analyses and indicative of a complex relationship between race and SES. Further research is needed to identify and address the causative factors of disparities in LA HNSCC.