At a median follow-up of 22.6 months, two isolated CNS relapses occurred, although the small number of events and absence of a randomized comparison make the findings hypothesis-generating rather than proof that the regimen prevents CNS recurrence. The study enrolled adults with newly diagnosed, histologically confirmed CD20-positive diffuse large B-cell lymphoma, including transformed disease, and a CNS International Prognostic Index score of 4 to 6. This index combines age, lactate dehydrogenase level, performance status, number of extranodal sites and Ann Arbor stage to estimate CNS recurrence risk.

Participants had not received prior anti-lymphoma therapy, apart from corticosteroids for symptom control, and had adequate organ function. Patients with evidence of CNS involvement on imaging or cytology were excluded. In the evaluable group, 34 of 37 patients underwent contrast-enhanced brain MRI and three underwent contrast-enhanced brain CT because MRI was contraindicated.

Lumbar puncture was performed in 29 of 37, and cerebrospinal-fluid cytology was negative in all patients with evaluated samples. CSF flow cytometry was available for 18 patients and showed no evidence of CNS involvement. The median age of the evaluable cohort was 63 years, with a range of 41 to 89 years, and 13 patients were older than 70.

Most participants were men, and 29 of 37 had an Eastern Cooperative Oncology Group performance status from 0 to 2. Immunohistochemical analysis identified 24 patients, or 64.9%, with double-expressor lymphoma involving MYC and BCL-2, while 26, or 70.3%, were classified as non-germinal-center B-cell-like. CNS-IPI scores were 4 in 15 patients, 5 in 16 and 6 in six.

Nine patients had high-risk extranodal sites, including the testis, breast, adrenal gland, kidney, uterus or ovary, and epidural space. The cohort therefore included a substantial number of patients with recognized high-risk features for CNS relapse. Each patient received six 21-day cycles of orelabrutinib, administered orally at 150 mg once daily, alongside R-CDOP.

The chemotherapy backbone comprised rituximab, cyclophosphamide, liposomal doxorubicin, vinorelbine and prednisone. Liposomal doxorubicin replaced conventional doxorubicin and vinorelbine replaced vincristine as part of a rationale aimed at reducing cardiotoxicity and neurotoxicity. Age-stratified reductions in cyclophosphamide and liposomal doxorubicin were prespecified for older adults, while granulocyte colony-stimulating factor support followed institutional guidelines.

Response was assessed by CT or PET-CT at baseline, after cycles two and four, and at the end of induction using Lugano 2014 criteria. Seven patients in complete response subsequently underwent autologous stem-cell transplantation as consolidation rather than salvage treatment. The primary efficacy analysis included all 44 enrolled patients in the intention-to-treat population, with early withdrawals for progressive or stable disease counted as nonresponses.

Objective response at the end of induction occurred in 31 patients, giving an intention-to-treat rate of 70.5% with a 95% confidence interval of 55.3% to 82.1%; 22 patients, or 50.0%, achieved complete response, with a 95% confidence interval of 35.1% to 64.9%. The per-protocol analysis of 37 evaluable patients produced an objective response rate of 75.7%, with a 95% confidence interval of 59.9% to 86.9%, and a complete-response rate of 59.5%, with a 95% confidence interval of 43.4% to 73.9%. Complete response increased across successive assessments, from 13.5% after cycle two to 48.6% after cycle four and 59.5% after cycle six.

Because five of the seven excluded patients withdrew with early progressive disease and had numerically poorer baseline characteristics, the per-protocol results may overestimate efficacy relative to the intention-to-treat estimate. Median progression-free survival had not been reached at the median follow-up of 22.6 months, with a 95% confidence interval of 22.1 months to not estimable. Estimated one- and two-year progression-free survival was 77.6% and 57.4%, respectively.

Sixteen progression-free survival events occurred: 12 systemic progressions, two isolated CNS relapses and two combined CNS and systemic progressions. The CNS relapses occurred 8.2 and 19.4 months after treatment began; both involved the brain parenchyma, and one also included leptomeningeal involvement. Median overall survival was 38.1 months, with a 95% confidence interval of 30.1 months to not estimable.

Estimated one- and two-year overall survival was 93.8% and 90.2%, respectively. The divergence between progression-free and overall survival reflects partly on salvage treatment after relapse, including CAR-T therapy, lenalidomide-based regimens, platinum-based chemotherapy and radiotherapy, rather than establishing the independent effect of first-line orelabrutinib. In an analysis with a median follow-up of 19.1 months, two of the 37 evaluable patients experienced CNS relapse, a proportion of 5.4%.

Estimated cumulative CNS-relapse incidence was 0% at one year and 9.3% at two years, with a 95% confidence interval of 0% to 21.5% at two years. The authors emphasized that this estimate was imprecise because only two CNS events were observed. Orelabrutinib was selected partly because earlier pharmacokinetic studies reported a median cerebrospinal-fluid concentration of 28.7 ng/mL and a median unbound CSF-to-plasma ratio of 59.8%, giving it measurable CNS penetration.

However, the trial itself did not measure CSF drug concentrations, so the treatment’s achieved CNS exposure and its contribution to the relapse estimate remain unconfirmed. Treatment toxicity was frequent, particularly hematologic toxicity, but no treatment-related deaths or permanent discontinuations because of adverse events were reported. All 37 patients assessed for safety experienced at least one treatment-related adverse event.

Any-grade neutropenia occurred in 100%, anemia in 91.9% and thrombocytopenia in 75.7%. Grade 3 or 4 treatment-related events occurred in 25 patients, or 67.6%, led by neutropenia at 59.5%, anemia at 24.3% and thrombocytopenia at 16.2%. Common nonhematologic events included nausea in 64.9%, fatigue in 59.5% and elevated aminotransferases in 43.2%.

Investigators reported managing the severe hematologic events with granulocyte colony-stimulating factor and blood-component transfusions. No atrial fibrillation or flutter, clinical tumor-lysis syndrome, fatal toxicity or unexpected toxicity pattern was observed. Several features prevent the results from being interpreted as a direct comparison with standard care.

The study had no concurrent randomized control group, making historical comparisons vulnerable to selection bias, changes in staging and evolving supportive treatment. Lumbar puncture and CSF flow cytometry were not completed in every patient, so occult baseline CNS involvement cannot be completely excluded. Dose reductions in older adults also introduced variation in treatment intensity, while incomplete retrospective collection of post-progression therapies prevents separation of salvage effects from the first-line regimen’s contribution to survival.

Accordingly, the phase II results support further investigation of a methotrexate-free approach but do not establish CNS-prophylactic efficacy or superiority over R-CHOP or R-CDOP. Randomized controlled validation is required.

Subject of Research: Single-arm phase II trial of orelabrutinib plus R-CDOP in untreated diffuse large B-cell lymphoma with high CNS-relapse risk

Article Title: Efficacy and safety of orelabrutinib plus R-CDOP as first-line therapy in diffuse large B-cell lymphoma with high-risk CNS-IPI: a prospective, multicenter, single-arm, phase II study

Article References: Guo, Y., Wang, J., Lei, H., Bao, C., Chen, P., Shou, L., Guo, Q., Hu, B., Xie, Y., Zhong, S., Qian, W., & Xiao, X. (2026). Efficacy and safety of orelabrutinib plus R-CDOP as first-line therapy in diffuse large B-cell lymphoma with high-risk CNS-IPI: a prospective, multicenter, single-arm, phase II study. Frontiers in Oncology, 16, Article 1976858. https://doi.org/10.3389/fonc.2026.1976858

Image Credits: AI Generated

DOI: 10.3389/fonc.2026.1976858

Keywords: orelabrutinib, diffuse large B-cell lymphoma, CNS relapse, CNS-IPI, BTK inhibitor, R-CDOP, phase II trial, blood-brain barrier, chemotherapy, lymphoma prognosis, Efficacy, safety