The Comparative Effectiveness of A+AVD vs PET-guided ABVD for the Management of Patients with Advanced Hodgkin Lymphoma: A Systematic Review and Matching-Adjusted Indirect Treatment Comparison

Introduction: In advanced Hodgkin lymphoma (aHL), adriamycin (doxorubicin), bleomycin, vinblastine, and dacarbazine (ABVD) is a recommended chemotherapy regimen despite a substantial proportion of patients experiencing relapse or incomplete disease control. Per the National Comprehensive Care Network (NCCN) guidelines, positron-emission tomography (PET)-guided ABVD and brentuximab vedotin + adriamycin (doxorubicin), vinblastine and dacarbazine (A+AVD) are preferred options for the front-line treatment of aHL (NCCN Guidelines. Hodgkin lymphoma. Version 2. 2023). In the ECHELON-1 (E1) trial, superiority of A+AVD vs ABVD x6 cycles (without PET adaptation at cycle 2) was demonstrated statistically in terms of overall survival (OS) and progression-free survival (PFS). Six-year follow-up of E1 patients showed improved OS with A+AVD vs ABVD. The aim of this study was to indirectly compare the effectiveness (OS and PFS) of A+AVD vs PET-guided ABVD as front-line treatments in the management of patients aged ≤60 years with aHL (Stage III or IV).

Methods: A systematic literature review was conducted using electronic databases such as MEDLINE, Embase, and the Cochrane Library (database inception to July 2022), supplementary sources and data on file for relevant materials reporting on the clinical efficacy of front-line treatments for aHL. Following the identification of relevant randomized controlled trials (RCTs), unanchored matching-adjusted indirect comparison was performed to compare the efficacy of A+AVD (the E1 trial) with PET-guided ABVD (the Risk-Adapted Therapy for aHL [RATHL] trial). As the outcomes used in the analyses were reported among the overall population in RATHL, rather than stratified by specific treatment, there was no potential anchor for the analysis. PFS and OS data from RATHL were only available for adult patients ≤60 years old with stage III/IV aHL, therefore the treatment comparison was limited to patients within this age group. Matching variables were selected based on literature review and the results of Cox regression analyses of the E1 trial data. The International Prognostic Score and Eastern Cooperative Oncology Group scores were used as matching factors as available. The weights generated from the matching procedure were incorporated in weighted Cox regression models to generate hazard ratios (HR) and corresponding 95% confidence intervals (CI). To account for the potential of violating the proportional hazards assumption, additional metrics were calculated including piecewise HRs (95% CI) using Cox regression models for specific time intervals (i.e., 0 to 24, 24 months +) and restricted mean survival times (RMST).

Results: In adult patients aged ≤60 years with Stage III/IV aHL, treatment with A+AVD generally led to increased PFS and OS vs PET-guided ABVD (Table 1). As the proportional hazard assumption was violated, additional metrics were calculated (RMST and piecewise HRs). HRs and RMST for survival outcomes are reported in Table 1.

Conclusions: The findings of this indirect treatment comparison support the improved efficacy of A+AVD relative to PET-guided ABVD with respect to OS and PFS in patients with stage III and IV aHL consistent with the results of E1 study.