Nivolumab Combined with Brentuximab Vedotin for Relapsed/Refractory Primary Mediastinal Large B-Cell Lymphoma: Preliminary Results from the Phase 2 CheckMate 436 Trial

Introduction: About 20% of patients (pts) with primary mediastinal B-cell lymphoma (PMBL) are not cured after first-line treatment; those with chemosensitive relapse may benefit from autologous hematopoietic cell transplantation (auto-HCT). However, outcomes remain poor for pts with chemorefractory disease and those who have relapsed after auto-HCT. In PMBL, there is increased expression of the programmed death-1 (PD-1) ligands and expression of CD30 on tumor cells. Evasion of host immune responses by tumor cells may lead to resistance to standard chemotherapy. Nivolumab is a fully human IgG4 monoclonal antibody that binds to immune checkpoint PD-1, abrogates tumor inhibitory signals, and augments host antitumor immune response. Brentuximab vedotin (BV) is an antibody-drug conjugate that binds to CD30-expressing cells and induces apoptosis and immunogenic cell death. Combination of nivolumab and BV may therefore have synergistic antitumor activity against relapsed/refractory PMBL (rrPMBL). This study evaluated the safety and efficacy of nivolumab + BV in this pt population.

Methods: This expansion cohort of a phase 1/2 open-label, single-arm, international study (NCT02581631) enrolled pts with PMBL aged ≥15 years, an Eastern Cooperative Oncology Group (ECOG) performance status ≤1, confirmed CD30 expression in ≥1% of tumor cells, measurable disease per Lugano 2014 classification, and relapsed/refractory disease after either high-dose conditioning chemotherapy and auto-HCT, or ≥2 prior multi-agent chemotherapies if auto-HCT ineligible. Nivolumab (240 mg IV flat dose) and BV (1.8 mg/kg IV, prespecified dose modifications allowed) were received every 3 weeks until disease progression or unacceptable toxicity. Pts were evaluated for safety (primary objective) continuously during treatment and at scheduled intervals during follow-up. Descriptive statistics were used to present pt characteristics and adverse events (AEs). The primary efficacy endpoint was investigator-assessed objective response rate (ORR) according to Lugano 2014 classification.

Results: At database lock, 30 pts were treated with nivolumab + BV and included in this interim analysis. At enrollment, median (range) age was 35.5 (19-83) years. Thirteen pts (43%) had stage III or IV disease at initial diagnosis. Pts had received a median of 2 prior systemic therapies before enrollment, and 4 (13%) received auto-HCT (Table). Treatment-related AEs (TRAEs) were reported in 25 pts (83%). The most frequently reported TRAEs were neutropenia (27%), peripheral neuropathy (20%), and hyperthyroidism (13%). Grade 3-4 TRAEs were reported in 10 pts (33%), including 8 pts (27%) with neutropenia, 2 pts (7%) each with thrombocytopenia or decreased neutrophil count, and 1 pt (3%) each with hypersensitivity, diarrhea, or maculopapular rash. Three pts (10%) had treatment-related serious AEs, including 1 pt with grade 3-4 diarrhea and maculopapular rash, and 1 pt with grade 5 acute kidney injury. At the time of the meeting, ORR and complete remission rate will be presented based on a planned interim analysis.

Conclusions: These results demonstrate that in pts with CD30-expressing rrPMBL, the combination of nivolumab and BV has a manageable safety profile and may provide a potential treatment option for this patient population.

Study support: BMS. Writing support: Janice Zhou, Caudex, funded by BMS.

Table.

View largeDownload slide

Table.

View largeDownload slide

Close modal