Effect of Active Hepatitis C Infection in Relapse-Free Survival Among Patients with HIV‐Related Diffuse Large B‐Cell Lymphoma

Abstract

Introduction: Diffuse-large B cell lymphoma (DLBCL) is the most common lymphoma associated with Hepatitis C Virus (HCV) and Human Immunodeficiency Virus (HIV). During the past decades, the incidence of HIV-related lymphomas has declined; however, it still remains a major cause of morbidity and mortality in this population. Before highly effective antiretroviral therapy for HIV was introduced, individuals with HIV-related DLBCL had worse complete response and survival rates compared to their uninfected counterparts. Additionally, recent studies have shown that relapse rates are lower for DLBCL patients that achieve HCV sustained virological response (SVR). Nevertheless, outcomes in patients with active HCV infection among patients with HIV-related DLBCL have not been reported yet. This study is aimed to compare the clinical outcomes in HIV-related DLBCL among patients with and without active HCV infection. Methods: Patients with HIV infection and newly diagnosed DLBCL between 2005 and 2017 at Montefiore Medical Center/Albert Einstein College of Medicine were identified using the institutional software Clinical Looking Glass. Cases were grouped into concomitant HIV and HCV infection (HIV + HCV) or HIV alone (HIV) based on HCV antibody and viral load counts at DLBCL diagnosis. Patients with active hepatitis B infection were excluded. Data regarding demographics, laboratory parameters, HCV treatment, lymphoma treatment and clinical outcomes including complete response (CR), progression of disease (POD) and relapse were collected by manual chart review and compared among groups. SVR was defined as undetectable HCV RNA level 12 weeks after HCV treatment completion. Clinical outcomes were compared between patients with and without active HCV at the time of outcome evaluation. For this purpose, patients that achieved SVR for HCV were analyzed in the HIV group. Kaplan-Meier curves were plotted to compare 12-months relapse-free survival (RFS) among groups. Statistical analysis was performed using Stata 14.1. Results: A total of 63 patients were identified, of whom 45 (71.4%) had HIV alone (HIV) and 18 (28.6%) had concomitant active HCV and HIV infection (HIV + HCV). The median age was 49 years (IQ: 40-54), there were 39 (61.9%) males, 27 (42.9%) were Non-Hispanic Black and 27 (42.9%) were Hispanic. At DLBCL diagnosis, the majority of cases had advanced stages (stage 3: 15; 25% and stage 4: 40; 66.7%), the median CD4 count was 117 (IQ: 65-217) and 45 (72.6%) had AIDS. Baseline characteristics did not differ among groups. Of 63 patients, 53 (84.1%) were treated for DLBCL. Treatment regimens included DA-EPOCH (28; 50.9%), CHOP (22; 40%) and others (3; 9.1%). Rituximab was given to 33 cases (57.9%). Within the HIV + HCV group, only 3 (16.7%) patients received treatment for HCV; all of them were treated with ledipasvir/sofosbuvir and achieved SVR. In the full cohort, CR was achieved in 17 (32.1%) treated patients and 19 (35.9%) had POD. The CR rate was lower for patients with active HCV + HIV compared to HIV alone (7.1% vs. 41%, p=0.02) and there was no difference in POD among groups (28.6% vs. 38.5%, p=0.37). Patients with active HCV + HIV had a significantly lower 18-month RFS compared to those with HIV (62.9% vs. 87.4%, p=0.035). In a multivariate model adjusted for stage, AIDS and rituximab use; active HCV + HIV was a predictor for relapse (OR:6.36, 95% CI: 1.01 - 39.9, p=0.048). Conclusions: Despite available and effective treatment against HCV infection, only a minority of patients with DLBCL-related HIV received HCV treatment. Active HCV seems to play a role in the outcomes of patients treated for DLBCL and was found to be a predictor for relapse. Hence, HCV treatment should be considered for patients with HIV-related DLBCL. Disclosures Shah: Physicians' Education Resource: Honoraria. Friedman:Incyte pharmaceutical: Membership on an entity's Board of Directors or advisory committees, Research Funding.