Journal: Blood
This international retrospective cohort study evaluated 1,139 patients with secondary CNS large B‑cell lymphoma, including de novo presentations (n=537) and relapsed disease (n=602), subdivided as CNS‑isolated relapse and synchronous CNS/systemic relapse.
Key findings:
- Outcomes by presentation
- Two-year progression-free survival (PFS):
- 40.4% for de novo cases
- 43.9% for CNS‑isolated relapses
- 16.2% for synchronous relapses
- CNS‑isolated relapse had a low rate of subsequent systemic recurrence (24‑month cumulative incidence of relapse ~6%), suggesting that systemic dissemination after isolated CNS relapse is uncommon.
- Impact of thiotepa-based ASCT
- In de novo disease, thiotepa-conditioned autologous stem cell transplant (ASCT) was associated with significantly improved PFS (HR 0.57; P = .005) and overall survival (OS) (HR 0.62; P = .023).
- In CNS‑isolated relapse, thiotepa-ASCT also improved outcomes (PFS HR 0.55; P = .002; OS HR 0.39; P < .0001).
- In synchronous relapse, ASCT (with or without thiotepa) was associated with better PFS (HR 0.57; P = .023) and OS (HR 0.48; P = .019).
- Comparison with CAR T-cell therapy
- Using propensity score matching of patients treated at relapse, thiotepa-ASCT was associated with superior outcomes compared with CAR T-cell therapy:
- PFS: HR 0.45; P = .005
- OS: HR 0.41; P = .014
Clinical implications:
- Thiotepa-based ASCT appears to confer a survival advantage in eligible patients across SCNSL subtypes, particularly de novo disease and CNS‑isolated relapse.
- The low systemic relapse rate after CNS‑isolated relapse supports treatment strategies modeled on primary CNS lymphoma protocols, with emphasis on CNS-directed therapy rather than intensive systemic consolidation alone.
- In this dataset, thiotepa-ASCT outperformed CAR T-cell therapy in the relapse setting after adjustment for baseline factors, supporting ASCT as a preferred consolidative option for fit patients when feasible.