Impact of HIV status on clinical outcomes in diffuse large B-cell lymphoma and burkitt lymphoma: A national retrospective analysis

Abstract Background: Patients with Human Immunodeficiency Virus (HIV) are at increased risk for aggressive lymphomas such as diffuse large B-cell lymphoma (DLBCL) and Burkitt lymphoma (BL). Though the risk of developing DLBCL has decreased dramatically, incidence of BL remained stable since combined antiretroviral therapy (cART) era. Real world outcomes in the current cART era are favorable (PMID:Vasseur L et al., Eur J Cancer 2020; Besson C et al., AIDS. 2017). We evaluated the impact of HIV on healthcare utilization and outcomes of patients with DLBCL. Methods: We queried Healthcare Cost and Utilization Project- National Inpatient Sample to identify adult patients hospitalized with DLBCL and BL from 2016-2020. Patients were stratified by HIV status. Cohorts were analysed to assess differences in sociodemographic variables, comorbidity profiles, inpatient mortality, length of stay (LOS), and total hospital charges (THC) as primary outcomes. Secondary outcomes included incidence of sepsis, respiratory failure, intubation, acute kidney injury (AKI), pressor use, pneumonia, urinary tract infection, neutropenia, blod transfusion, pulmonary embolism and Pneumocystis jirovecii pneumonia (PJP). Statistics were performed using t-test, univariate and multinomial logistic regression. Results: A total 14,825 admissions of HIV-DLBCL, 344,840 non-HIV-DLBCL, 5,360 admissions of HIV-BL and 25,610 non-HIV-BL were identified. HIV-positive patients were younger. HIV-DLBCL was more common in ages 18–45 than in non-HIV-DLBCL (38.5% vs. 9.4%), while non-HIV-DLBCL predominated in >65 years (56.5% vs. 6.5%). HIV-BL was more frequent in ages 18–45 and 45–65 than in non-HIV-BL (49.0% vs. 36.7% and 47.5% vs. 34.4%), whereas non-HIV-BL was more common in >65 years (28.9% vs. 3.5%) (all p < 0.001). The HIV cohort was more likely to be Black (DLBCL: 43.4% vs. 7.5%; BL: 33.5% vs. 4.3%), on Medicaid (DLBCL: 41% vs. 9.1%; BL: 36.8% vs. 15.6%), or uninsured (DLBCL: 7% vs. 2.2%; BL: 8.2% vs. 2.7%), treated at a non-teaching hospital (DLBCL: 91.5% vs. 86%; BL: 91.4% vs. 88.6%), and in the lowest income quartile (DLBCL: 41% vs. 21.2%; BL: 36.1% vs. 20.7%) (all p < 0.001). HIV-cohort had a significantly higher comorbidity burden, coinfection with Hepatitis B/C whereas non-HIV cohort had higher rates of diabetes, cardiac diseases and obesity (all p < 0.05). In DLBCL group, the overall adjusted all-cause mortality was not significantly different between cohorts (aOR = 0.9, 95% CI 0.79-1.23, p = 0.90), however on subgroup analysis mortality was significantly increased for Blacks (aOR = 1.30, 95% CI 1.13-1.49, p < 0.001) compared to Caucasians. Mean LOS (9 vs 7 days, p = 0.38) was similar between cohorts. THC was significantly higher in HIV-DLBCL ($113,626 vs $105,538, p < 0.001). Secondary outcomes like PJP infection (1.42% vs 0.25%, p < 0.001), intubation (3.44% vs 3.16%, p = 0.001), pressor support (1.37% vs 0.91%, p = 0.003), acute kidney injury (17.59% vs 16.66%, p = 0.02), pulmonary embolism (2.19% vs 1.75%, p = 0.02) and neutropenia (17.5% vs 14.3%, p = 0.001) were significantly higher in HIV-DLBCL group. In BL group, the overall adjusted all-cause mortality was not significantly different between cohorts (aOR = 1.4, 95% CI 0.85-2.25, p = 0.19). Mean LOS (9 vs 8 days, p = 0.07) were similar between cohorts. THC was significantly higher in HIV-BL ($117,157 vs $102,261, p < 0.001). Secondary outcomes were not significantly different except PJP infection (0.84% vs 0.17%, p = 0.02). Conclusion: HIV patients were younger and more likely in lower income quartile compared to non-HIV patients. THC was significantly higher in HIV-DLBCL and HIV-BL cohorts compared to non-HIV cohort. Although overall mortality and outcomes were similar between the groups, a racial disparity was seen, with higher mortality in African American patients in HIV-DLBCL group, warranting further investigation.