Pre-Planned Interim Safety Analysis of the Niveau Trial, a Randomized Phase 3 Study for Patients with Aggressive Non-Hodgkin Lymphoma in First Relapse or Progression Not Eligible for High-Dose Chemotherapy (HDT), Testing Nivolumab in Combination with Gemcitabine, Oxaliplatin (GemOx) Plus Rituximab (R) in Case of B-Cell Lymphoma

Background: Anti-PD1 antibodies have the potential to increase rituximab-mediated effector mechanisms, to target the microenvironment and PD-1 in aggressive Non-Hodgkin Lymphoma (NHL). Addition of Nivolumab (Nivo) might increase efficacy of chemotherapy. The NIVEAU trial is an ongoing international, multicenter, randomized, open label, phase 3 study testing Nivo in combination with Rituximab, Gemcitabine, Oxaliplatin ((R-)GemOx) (NCT03366272). We performed a pre-planned safety analysis comparing the experimental arm (including safety run-in phase) and standard arm.

Methods: Key eligibility criteria include: first relapse or progression of an aggressive lymphoma (B-cell and peripheral T-cell lymphoma (PTCL)), ineligibility for HDT (defined as >65 years of age or older than 18 years if HCT-CI score >2), only 1 prior chemotherapy including an anthracycline and R in case of B-NHL. Patients (Pts) were planned to receive 8 bi-weekly cycles of GemOx (Gem 1000 mg/m², Ox 100 mg/m²) and R 375 mg/m² for B-NHL at d1 of each cycle. Pts randomized into the experimental arm additionally received Nivo 3 mg/kg every two weeks for a total of 26 applications or until progression. Safety analysis was performed after 30 pts had been included into the experimental arm.

Results: The analysis (data cut-off 29-August-2019) included 59 pts (44 with B-NHL, 15 with PTCL). 22 pts with B-NHL were randomized in the standard arm, 22 in the experimental arm. 7 pts with PTCL were randomized in the standard arm, 8 in the experimental arm. Median age was 78 years (range: 59; 87) for B-NHL and 70 years (range: 53; 80) for PTCL. 18 (41%) pts with B-NHL and 7 (47%) with PTCL were male. Twenty-five (57%) pts with B-NHL and 11 (73%) with PTCL had duration of first response ≤ 12 months. Due to the non-randomized safety run-in phase where all pts were treated in the experimental arm, the cumulative observation time of the experimental arm was longer.

The most common adverse events (AE) (≥5%) during (immuno-)chemotherapy for pts treated with (R-)GemOx (grade ≥3) per documented cycles were thrombocytopenia 9/64 (14%), infection 7/64 (11%), leukocytopenia 5/64 (8%), peripheral sensory neuropathy 4/64 (6%) and anemia 4/63 (6%).

Other frequent AEs gr. 3 (n>1) and/or life-threatening AEs (grade ≥4) were 2 gastrointestinal hemorrhages (1x gr. 4, 1x gr. 3), 1 gastric perforation gr. 4 and 2 tumor lysis syndromes gr. 3.

In pts treated with (R-)GemOx 1 therapy-related death occurred after 2 cycles of (immuno-)chemotherapy. One pt with gastric involvement of lymphoma terminated treatment early due to sepsis after surgical intervention after gastrointestinal bleeding.

The most common AEs (≥5%) during immunochemotherapy for pts treated with (R-)GemOx+Nivo (grade ≥3) per documented cycles were thrombocytopenia 37/140 (26%), anemia 22/140 (16%), increased lipase 13/129 (10%), leukocytopenia 11/140 (8%) and infection 7/141 (5%).

Other frequent AEs gr. 3 (n>1) and/or life-threatening AEs (grade ≥4) were 2 gastrointestinal hemorrhages (1x gr. 5, 1x gr. 3), 1 renal failure gr. 5, 1 CNS vasculitis (gr. 4), 9 neutropenia (1x gr. 4, 8x gr. 3), 5 lymphopenias (2x gr. 4, 3x gr. 3), 19 increased yGT (each gr. 3), 2 hypertension (2x gr. 3), 2 syncopes (2x gr. 3) and 2 infusion related reactions (2x gr. 3).

During consolidation with Nivo until 100 days after end of therapy most common AEs (≥5%, grade ≥3) were: lymphopenia 15/114 (13%), neutropenia 9/114 (8%) and leukocytopenia 8/114 (7%).

Other frequent AEs gr. 3 (n>1) and/or life-threatening AEs (grade ≥4) were 6x increased yGT (each gr. 3) and 3 back pains (gr. 3).

In pts treated with (R-)GemOx+Nivo 2 pts terminated treatment early, 1 due to elevated liver enzymes and 1 due to CNS vasculitis.

Conclusion: So far, a moderately higher toxicity of the experimental arm, the combination of (R)-GemOx with Nivo, was observed. However, this does not influence the overall risk profile of the trial. No increased rate of progressions in pts with PTCL treated with Nivo was observed. The randomized phase 3 study continues as planned with continuous evaluation of toxicity. Supported by BMS