Treatment patterns and outcomes of primary central nervous system lymphoma treated with high-dose methotrexate with or without autologous stem cell transplantation or whole brain radiation in the rea-world setting

Abstract

Abstract Introduction: Primary central nervous system lymphoma (PCNSL) is a rare non-Hodgkin lymphoma with a high relapse risk in the absence of high-dose methotrexate (HDMTX)-based induction and subsequent consolidation therapy. Autologous stem cell transplantation (ASCT) and whole brain radiation therapy (WBRT) are two well-studied consolidation options and are associated with superior overall survival than no consolidation. We aim to describe the treatment patterns and outcomes of PCNSL treated with HDMTX-based regimens in a real-world setting. Methods: This retrospective study was approved by the Wake Forest School of Medicine IRB. Patients with newly diagnosed PCNSL with diffuse large B-cell lymphoma histology, and received HDMTX-based induction chemotherapy for at least 1 cycle at Levine Cancer Institute or Wake Forest Baptist Comprehensive Cancer Center between 3/1/2022 and 11/1/2024, were included. We collected information including patient demographics, baseline characteristics, comorbidities, timing of initiation of induction, HDMTX dosing, response to induction, choice of consolidation, and reasons for not receiving consolidative ASCT. Overall survival (OS) and progression-free survival (PFS) were estimated using Kaplan-Meier methods and stratified by ASCT status (ASCT vs non-ASCT), treatment delay from clinical presentation (≤ 4 vs > 4 weeks from presentation to start of HDMTX), treatment delay from diagnosis (≤ 7 vs > 7 days from diagnosis to start of HDMTX), and induction regimen (MTX+R vs MTX+R+ additional chemotherapy). Log-rank tests were used to compare strata and univariate Cox proportional hazards models estimated hazards of time from presentation to MTX initiation and time from diagnosis to MTX initiation as continuous variables of days. Results: Forty-one patients met inclusion criteria. Median age was 67 years (range, 32-81), 22 (54%) were female, and 7 (18%) were racial minority or Hispanic ethnicity. Chronic renal, cardiac and pulmonary dysfunction were each seen in 12%. The median time from clinical presentation to initiation of HDMTX was 29 days, and from diagnosis to initiation was 7 days. During induction, 59% received HDMTX+rituximab (R), 39% received HDMTX+R+additional (A) chemotherapy such as temozolomide (MTR) (15%) and cytarabine/thiotepa (MATRix) (15%). The median dose of HDMTX was 3.5 g/m2, 29% received a dose >3.5 g/m2. The overall response rate to induction was 80% (75% with HDMTX+R; 94% with HDMTX+R+A). Nine (22%) patients received ASCT. Thiotepa (TT)/BCNU was used for conditioning in 8 patients, and TT/Busulfan/Cyclophosphamide in 1. Compared with non-ASCT patients, more ASCT patients were female (89% v 44%; p=0.02). Between ASCT and non-ASCT patients, overall response rate (100% vs 75%, p=0.16) and complete remission rate (56% vs 38%, p=0.45) to induction were not statistically different. No other statistically significant difference, including age, was observed between ASCT and non-ASCT patients. Reasons for not receiving ASCT (not mutually exclusive) included induction failure (n=8, 31%), patient refusal (n=7, 27%), advanced age (n=4, 15%), challenging comorbidities (n=6, 23%) or performance status (n=6, 23%); two (6%) received WBRT. Median follow up was 20.1 months. Fourteen (33%) deaths and 15 (37%) PFS events were observed. All 9 ASCT recipients remained alive and progression free with a median follow up of 20.1 months (range, 7.7-34.9), one WBRT recipient remained alive and progression free at 7.6 months, whereas the other died from progression at 11 months. Those who did not receive ASCT had a median PFS and OS of 15.8 and 34.5 months, significantly shorter than ASCT recipients (p=0.020 and p=0.021, respectively). No significant associations were observed between dichotomized time from presentation or diagnosis to initiation of MTX in both PFS (p=0.16 and p=0.98, respectively) and OS (p=0.23 and p=0.98, respectively). Moreover, differences in PFS or OS were not observed when the times from diagnosis or presentation to HDMTX were tested as continuous variables. Further, the choice of MTX-based combination therapy did not significantly impact PFS or OS (p=0.47 and p=0.55, respectively). Conclusion: HDMTX-based induction chemotherapy is effective in patients with PCNSL, even in those with delayed diagnosis or initiation of therapy. Although a minority of patients received consolidation with ASCT, it was associated with 100% progression-free survival.