Comparing CHOP, R-CHOP, R-EPOCH, and DA-EPOCH-R in HIV-associated diffuse large B-cell lymphoma: A meta-analysis and systematic review

Abstract INTRODUCTION Diffuse Large B-Cell lymphoma (DLBCL) is the most common subtype of non-hodgkin lymphoma in patients with human immunodeficiency virus (HIV) infection. These patients typically have a worse outcome due to more aggressive histology and advanced stage at presentation. Additional factors, including immunosuppression and increased susceptibility to treatment-related toxicity, create unique clinical challenges. While treatment regimens for HIV-negative DLBCL patients are well established, the optimal approach in HIV positive patients remains debated. Anthracycline-based treatments, such as CHOP and R-CHOP (cyclophosphamide, doxorubicin, vincristine, and prednisone +/- rituximab) are the current standard of care in HIV-negative patients and have been frequently adopted for HIV positive patients. However, emerging data from several phase II and III studies suggest that infusion regimens like EPOCH-R (rituximab, Etoposide, Prednisone, Vincristine, cyclophosphamide, and doxorubicin) and its dose-adjusted variant, DA-EPOCH-R, may yield superior outcomes in HIV-positive patients. Direct comparisons between these regimens in HIV-positive DLBCL remain limited. Here, we conducted a meta-analysis comparing the efficacy of CHOP, R-CHOP, EPOCH-R, and DA-EPOCH-R in treating HIV-associated DLBCL. METHODS: A systematic review search through PubMed and Embase databases was done to identify studies reporting on the outcomes of HIV-positive DLBCL treated with CHOP, R-CHOP, EPOCH-R, and DA-EPOCH-R. Eligible studies included phase II or III trials or retrospective analysis of treatment-naive HIV positive adults (age >18), measuring survival or response rates as primary outcomes. Studies were also selected based on whether they reported patient demographics such as disease stage, CD4 count, Lactate Dehydrogenase (LDH) levels, Eastern Cooperative Oncology Group (ECOG) functional status, and concurrent/prior antiretroviral therapy (ART). Ninteen studies met inclusion criteria. A random-effects meta-analysis was performed to measure associations between overall survival (OS) and complete response (CR) between the four regimens. Multivariate regression was used to evaluate the effect of ECOG status, disease stage, CD4 count and adjust for potential confounders. Pairwise comparisons were also conducted. Heterogeneity was assessed using the I2 statistic. Effect sizes were reported with 95% confidence interval, and p-values of <0.05 were used for statistical significance. RESULTS: The meta-analysis included 19 studies comprising 1230 patients who were treated with either CHOP, R-CHOP, EPOCH-R, or DA-EPOCH-R regimens. Pooled analysis showed significant difference in CR among the four regimens, with DA-EPOCH-R having highest pooled CR (p<0.05). Similarly, patients in the DA-EPOCH-R group also had the highest pooled OS (p<0.05). Multivariate analysis showed no statistical difference in OS between R-CHOP and CHOP (OR 1.09, CI: 0.59-1.98, p>0.05). In contrast, DA-EPOCH-R showed significant improvement over the other regimens, including R-CHOP (OR 1.95, CI: 1.27-3.00, p<0.05) and CHOP (OR 2.12, CI:1.09-4.12, p<0.05). However, it was not significantly superior to EPOCH-R (OR 1.19, CI: 0.72-1.97, p>0.05) DISCUSSION: In the largest meta-analysis to date evaluating HIV-related DLBCL, DA-EPOCH-R demonstrated the most favorable overall survival and response rates, particularly compared to CHOP and R-CHOP. These findings were consistent when adjusting for prognostic factors such as disease stage, CD4 count, and functional status. Notably, EPOCH-R also showed superior outcomes versus R-CHOP, suggesting that infusion regimens may offer therapeutic advantages in HIV positive immunocompromised patients. A fixed-dose EPOCH-R regimen might be enough to balance potential efficacy benefits with the risk of increased toxicity with DA-EPOCH-R. Interestingly, CHOP and R-CHOP both demonstrated relatively decreased therapeutic advantage, with no statistical OS difference on multivariate analysis - contradicting established literature in HIV-negative DLBCL where the addition of rituximab improves CR, PFS, and OS. This shows a need to re-evaluate their role as standard regimens in HIV patients, particularly in the ART era. Our findings support the use of DA-EPOCH-R or EPOCH-R as the preferred therapy in HIV-positive DLBCL patients, and highlight the need for more prospective trials in this population to better define treatment intensity and individualized care.