Survival outcomes in veterans with hepatocellular carcinoma, with and without HIV infection

BACKGROUND

HIV infection has been associated with survival disparities among persons with hepatocellular carcinoma (HCC). However, most studies examining survival do not control for provider (e.g. type of HCC treatment given) or individual-level factors (e.g. homelessness, substance use) that could impact survival. In this study, we evaluate the effect of HIV status on survival among persons with HCC, in a comprehensive model that accounts for key individual, provider, and systems-level factors.

METHODS

We conducted a retrospective cohort study of persons with HIV (PWH) matched 1 : 1 to HIV-negative controls based on age and year of HCC diagnosis in the national Veterans Administration (VA) health system. The primary outcome was survival. We used Cox regression models to evaluate the effect of HIV status on risk of death.

RESULTS

This cohort included 200 matched pairs diagnosed with HCC between 2009 and 2016. A total of 114 PWH (57.0%) and 115 HIV-negative patients (57.5%) received guideline-concordant therapy ( P  = 0.92). Median survival was 13.4 months [95% confidence interval (CI) 8.7-18.1] among PWH and 19.1 months (95% CI 14.6-24.9) for HIV-negative patients. In adjusted models, older age, homelessness, advanced Barcelona Clinic Liver Cancer (BCLC) stage, and not receiving any HCC treatment predicted risk of death. HIV status was not associated with risk of death [adjusted hazard ratio (aHR) 0.95; 95% CI 0.75-1.20; P  = 0.65].

CONCLUSION

HIV status was not associated with worse survival among HCC patients, in a single-payer, equal access healthcare system. These results suggest that HIV infection alone should not exclude PWH from receiving standard therapy.

Trends in gender-based disparity in incidence, mortality and survival for major digestive disease cancers in the U.S. (2000-2016)

e13621

Background: Males have excess risk of most digestive cancers; whether cancer mainly due to virus (stomach), lifestyle including obesity/alcohol (colon/esophagus), or both (liver). We evaluated for secular trends in gender-specific incidence, mortality and survival for these cancers in the U.S. Methods: We used official U.S. Cancer Statistics [USCS] data (2000-2016) to calculate annual age-adjusted incidence and mortality rates (IR/MR). It covers >98% population/all states (vs. SEER-10 statewide registries, covers <1/3rd African-Americans/Whites). We calculated mortality-to-incidence rate ratios (MIR), an established 5-year survival proxy. We used Joinpoint to assess trends in average annual percentage change (AAPC). Results: Table shows age-adjusted IRs and MRs for colon, esophagus and stomach cancer decreased for men and women (all AAPCs negative, all significant but stomach IR women), with both liver IR/MR significantly increasing in both genders. Gender differences in relative burden were seen (e.g., MRs in colon>esophagus>liver>stomach in men vs. colon>stomach>liver>esophagus in women), with greatest male excess in esophagus (e.g., 4.4-fold IR). MIRs suggested less favorable survival (higher=worse) for men for esophagus/colon. Conclusion: Substantial gender disparity persists in digestive cancers, with male excess gap decreasing for colon and increasing for liver. [Table: see text]