Impact of racial, ethnic, and socioeconomic disparities on presentation and survival of HCC: A multicenter study.

BACKGROUND

Racial and ethnic disparities have been reported for HCC prognosis, although few studies fully account for clinically important factors and social determinants of health, including neighborhood socioeconomic status.

METHODS

We conducted a retrospective multicenter cohort study of patients newly diagnosed with HCC from January 2010 through August 2018 at 4 large health systems in the United States. We used multivariable logistic regression and cause-specific Cox proportional hazard models to identify factors associated with early-stage HCC presentation and overall survival.

RESULTS

Of 2263 patients with HCC (37.6% non-Hispanic White, 23.5% non-Hispanic Black, 32.6% Hispanic, and 6.4% Asian/other), 42.0% of patients presented at an early stage (Barcelona Clinic Liver Cancer stage 0/A). In fully adjusted models, there were persistent Black-White disparities in early-stage presentation (OR: 0.63, 95% CI: 0.45-0.89) but not Hispanic-White disparities (OR: 0.93, 95% CI: 0.70-1.24). Median survival was 16.2 (IQR: 5.8-36.8) months for White patients compared to 15.7 (IQR: 4.6-34.4) months for Hispanic, 10.0 (IQR: 2.9-29.0) months for Black, and 9.5 (IQR: 3.4-31.9) months for Asian/other patients. Black-White disparities in survival persisted after adjusting for individual demographics and clinical factors (HR: 1.30, 95% CI: 1.09-1.53) but were no longer observed after adding HCC stage and treatment (HR: 1.05, 95% CI: 0.88-1.24), or in fully adjusted models (HR: 0.97, 95% CI: 0.79-1.18). In fully adjusted models, Hispanic-White (HR: 0.87, 95% CI: 0.73-1.03) and Asian/other-White (HR: 0.85, 95% CI: 0.63-1.15) differences in survival were not statistically significant, although patients in high-SES neighborhoods had lower mortality (HR: 0.69, 95% CI: 0.48-0.99).

CONCLUSIONS

In a multicenter cohort of patients with HCC, racial and ethnic differences in HCC prognosis were explained in part by differences in tumor stage at diagnosis and neighborhood SES. These data inform targets to intervene and reduce disparities.