Background:

Patients with relapsed or refractory diffuse large B-cell lymphoma (r/r DLBCL) who are ineligible for or progressed from autologous stem-cell transplantation have poor outcomes and few treatment options. Programmed cell death-1 (PD-1) inhibitor monotherapy has limited efficacy in pretreated DLBCL. Previous study has demonstrated that pembrolizumab plus R-CHOP was effective and safe as first-line treatment in patients with DLBCL and in vitro data suggests that rituximab acts partly via antibody-dependent cellular cytotoxicity, which may be enhanced by programmed cell death-1 (PD-1) blockade. Here, we report the preliminary results of a real-world, single-arm study investigating efficacy and safety of PD-1 inhibitor plus rituximab for r/r DLBCL.

Methods:

This is a real-world, single-center, single-arm study enrolling patients who received at least one dose of PD-1 inhibitor combined with rituximab as salvage therapeutic regimen after R-CHOP treatment in Cancer Hospital, Chinese Academy of Medical Science & Peking Union Medical College. Eligible patients received PD-inhibitors in combination with rituximab (375 mg/m2, q3w) for 6 cycles followed by maintenance therapy of PD-1 inhibitors plus rituximab (q8w) for 6 cycles until unacceptable toxicity, disease progression, death, or completion of 2-year treatment. The outcomes included objective response rate (ORR), disease control rate (DCR), progression-free survival (PFS), overall survival (OS) and safety. Responses were assessed by Lugano 2014 criteria, and PD-L1 staining (22C3) was performed at a validated central laboratory. Genomic analysis was performed including next generation sequencing (NGS) based sequence variant detection, copy number analysis and structural variant detection.

Results:

Between Oct 16, 2018, and May 10, 2021, 34 patients were enrolled and received at least one dose of both PD-1 inhibitor including toripalimab (n=24), pembrolizumab (n=3), nivolumab (n=3) and sintilimab (n=4), combined with rituximab. Key baseline characteristics included median age 56.5 (range 20-78) years; stage III/IV disease 58.8%; elevated LDH 41.2%; IPI ≥3 47.1%; median number of prior treatment regimen 2 (range 1-4); primary chemotherapy-refractory disease after R-CHOP 73.5%; median interval from last rituximab administration 5 (range 0.7-26) months. Histology included 30 DLBCL NOS (88.2%; 11 GCB, 19 non-GCB by Han's algorithm), 4 primary mediastinal B-cell lymphoma (PMBCL; 11.8%). Patients received a median of four (range 1-24) doses of PD-1 inhibitor and rituximab. Eight patients (6 DLBCL-NOS, 2 PMBCL) remained on treatment at the time of data cutoff on Jul 5, 2021.

With a median follow-up of 21.3 months (95% CI: 9.6-24.2), 16 (53.3%; 95% CI: 34.3%-71.7%) of 30 patients with DLBCL-NOS had an objective response, with 2 (6.7%) having a complete response and 14 (46.7%) achieving a partial response. The median DOR was 16.8 (95%CI: 2.8-NR) months. The median PFS was 2.9 (95%CI: 1.5-18.5) months, and median OS was 25.3 (95%CI: 9.9-NR) months. The 1-year PFS and OS rates were 34.1% (95%CI: 17.3%-51.8%) and 73.6% (95%CI: 49.8%-87.4%), respectively. All of 4 PMBCL cases achieved CR, and their PFS were 21.2, 21.6, 22.9 and 29.2 months, respectively. Overall, 19 of 34 patients had available baseline biopsy samples for NGS analysis. Patients with TET2 mutations had higher ORR (100% versus 31.3%, p=0.058) and longer PFS (NR versus 1.6 months, p=0.019) while patients with β2-MG mutations had lower ORR (0% versus 53.3%, p=0.1) and shorter PFS (1.0 versus 2.8 months, p=0.0025). In addition, patients with BCL2 alteration/amplification/rearrangement had inferior OS than those BCL2 wild-type patients (5.0 versus 17.4 months, p=0.024). PD-L1 expression at baseline had no impact on response and survival. Most treatment-related adverse events were grade 1-2 including rash and neutropenia. Treatment was discontinued in 3 patients (2 interstitial pneumonia and 1 hypophysitis).

Conclusion:

PD-1 inhibitor combined with rituximab was effective and well tolerated in r/r DLBCL, achieving promising ORR and long-term remission. Further investigation is warranted.

Disclosures

No relevant conflicts of interest to declare.

Sign in via your Institution