Zanubrutinib combined with intrathecal chemotherapy for prevention of CNS relapse in high-risk diffuse large B-cell lymphoma

Abstract

Abstract Background: Secondary central nervous system lymphoma (SCNSL) in diffuse large B-cell lymphoma (DLBCL) carries a dismal prognosis, with a 5-year overall survival (OS) rate of only 26.9%. Prophylactic strategies for high-risk patients remain controversial, with limited efficacy and toxicity concerns for high-dose methotrexate (HD-MTX) and intrathecal (IT) chemotherapy. Zanubrutinib, a next-generation BTK inhibitor, has shown potential for CNS penetration and efficacy in CNS lymphoma. We hypothesized that adding zanubrutinib to standard immunochemotherapy and IT chemotherapy could reduce CNS relapse in high-risk DLBCL. Methods: Treatment-naive DLBCL patients (age 18–75 years, ECOG 0–2) with ≥1 high-risk features: CNS-IPI ≥4 or ≥3 extranodal sites, or involvement of testes/kidney/adrenal/breast/uterus/bone marrow/intravascular/leg-type cutaneous DLBCL were enrolled. Patients received six cycles of standard immunochemotherapy (R-CHOP or Pola-R-CHP) every 21-day per cycle, combined with zanubrutinib (160mg twice daily) and IT chemotherapy (methotrexate 12mg plus dexamethasone 5mg on day 2 of each cycle, ≥4 doses). Primary endpoint was 2-year cumulative incidence of CNS relapse. Secondary endpoints included zanubrutinib CNS penetration and prophylactic efficacy, influence of clinicopathological subtypes and molecular biological characteristics on the preventive effect of zanubrutinib, overall response rate (ORR) and complete response rate (CRR), 2-year progression-free survival (PFS) and 2-year OS, safety and adverse effects. This trial was registered with Chinese Clinical Trial Register, number ChiCTR2300072657. Results: Between 1 October 2022 and 1 July 2025, 26 patients were enrolled. The median age was 60 years (range 30-72), with 65.4% (17/26) being male. 76.9%(20/26) had intermediate to high-risk disease per International Prognostic Index (IPI). Pathologically, the majority (61.5%, 16/26) were non-GCB subtype, and over half (57.7%, 15/26) were double-expressor lymphoma. Molecular classification identified MCD in 30.8% (8/26) of patients. Among 20 efficacy evaluable patients, the ORR was 100% (20/20) and CRR was 95% (19/20). Subgroup analysis revealed CR rates of 91.7% (11/12) in non-GCB, 100% in double-expressor (12/12) and MCD subtypes (5/5). With a median follow-up of 17 months (range 6-31), 2 patients experienced relapse, and 1 patient died due to disease progression. The estimated 2-year PFS and OS were 88.1% and 93.8%, respectively. As of data cut-off, no patients developed CNS relapse, the 2-year cumulative incidence of CNS relapse was 0. Notably, zanubrutinib demonstrated favorable CNS penetration, with a corrected cerebrospinal fluid (CSF) / plasma concentration ratio of 64.9% ± 31.5%, and 81.5% of samples exceeding the IC50. The most common adverse events (AEs) were myelosuppression, infections and hypohepatia. ≥ grade 3 AEs included neutropenia (85%), thrombocytopenia (30%), and pneumonia (25%). While all adverse events were manageable with supportive care. No intrathecal bleeding event or AE related death was reported. Conclusion: Zanubrutinib combined with R-CHOP or Pola-R-CHP and IT chemotherapy demonstrated high response rate and low CNS relapse rate in high-risk DLBCL with favorable safety profiles. High CSF penetration of zanubrutinib support its potential role in CNS prophylaxis.