Primary diffuse large B-cell lymphoma (DLBCL) of the central nervous (CNS) system is a rare extranodal lymphoma confined to the CNS. Although standard treatment with high-dose methotrexate (HD-MTX) can induce remission, outcomes following relapse are poor. Approximately one-quarter of patients relapse after initial therapy, and salvage options have historically yielded low objective response rates (ORRs), typically less than 30%, with progression-free survival (PFS) often less than 3 months. Against this backdrop, the phase 2 SPECTRUM study evaluated the combination of the PD-1 inhibitor tislelizumab and the antifolate chemotherapy agent pemetrexed.
This multicenter, open-label trial conducted in South Korea enrolled patients with histologically confirmed relapsed/refractory (R/R) DLBCL of the CNS who had progressed after HD-MTX–based therapy. Patients received tislelizumab (200 mg) plus pemetrexed (500 mg/m2) intravenously every 3 weeks until disease progression or unacceptable toxicity. The primary endpoint was ORR, with secondary endpoints including PFS, duration of response (DoR), overall survival (OS), and safety. Exploratory endpoints included LymphGen classification, PDL-1 (22C3) and outcomes, and serial plasma circulating tumor DNA (ctDNA) MYD88L265P and outcomes.
As of December 2025, 29 patients had been enrolled, with a median age of 60 years. Most patients (~86%) had previously received rituximab, reflecting modern treatment patterns. With a median follow-up of 30.4 months, patients received a median of 6 treatment cycles. Efficacy results were notable. Among evaluable patients, the ORR reached 71.4%, with a complete response rate of 50%. These outcomes represent a substantial improvement over historical benchmarks in this setting. The median DoR was 14.5 months, suggesting that responses were not only frequent but also durable. The median PFS was 6.8 months, whereas the median OS reached 27.8 months.
The combination therapy demonstrated a manageable safety profile. Treatment-emergent adverse events (AEs) occurred in approximately 65.5% of patients. Grade 3 or higher AEs were observed in 34.5% of patients, including 1 treatment-related death. Eight patients discontinued therapy due to AEs, all attributed to pemetrexed rather than the immunotherapy component. Immune-related AEs occurred in 17.2% of patients, consistent with the known safety profile of checkpoint inhibitors.
Exploratory molecular analyses provided additional biological insight. The majority of tumors were classified as the MCD subtype by LymphGen classification, and MYD88L265P mutations were detected in over 60% of cases. Notably, detection of this mutation was more sensitive in cerebrospinal fluid (CSF) than in plasma, supporting the value of CSF-based monitoring in CNS disease. In an illustrative case, clearance of the MYD88 mutation in ctDNA correlated with a complete metabolic response on imaging.
Overall, the SPECTRUM trial met its primary endpoint and demonstrated promising activity for tislelizumab plus pemetrexed in R/R primary DLBCL of the CNS. The high response rates, meaningful durability, and manageable toxicity profile suggest this combination could represent a valuable new treatment option in a setting with significant unmet need. Further studies will be needed to confirm these findings and better define the role of immunotherapy-based combinations in CNS lymphoma.
Source: Kim JS, et al. SPECTRUM: phase II study of tislelizumab plus pemetrexed in patients with relapsed or refractory (R/R) primary diffuse large B-cell lymphoma (DLBCL) of the central nervous system (CNS)—The Korean Cancer Study Group LY22-07. Presented at: ASCO Annual Meeting. May 29-June 2, 2026; Chicago, IL. Abstract 2009.