Comparative analysis of consolidation strategies after induction chemotherapy in primary central nervous system lymphoma

Abstract Background: Primary central nervous system lymphoma (PCNSL) is an aggressive diffuse large B-cell lymphoma, predominantly affecting elderly patients. Current treatment involves high-dose methotrexate (HD-MTX)-based induction followed by consolidation therapy. High-dose chemotherapy with autologous stem cell transplantation (HDC-ASCT) has shown superior outcomes in young and fit patients and is strongly recommended. However, for elderly patients, transplantation is often not feasible due to toxicity concerns and significant comorbidities. For these patients, alternative approaches including whole brain radiotherapy (WBRT) and reduced-intensity chemotherapy may be considered, although studies comparing these strategies are lacking. Accordingly, we conducted this study to evaluate the role of consolidation strategies in patients with PCNSL. Methods: This retrospective analysis included PCNSL patients newly diagnosed at Asan Medical Center between 2000 and 2023. A total of 394 patients received HD-MTX-based induction: HD-MTX alone (n=130), methotrexate/procarbazine/vincristine (MPV, n=68), or rituximab-MPV (R-MPV, n=123). Of these, 317 patients who completed induction therapy (complete response: 148, partial response: 163, stable disease: 6) were included in the analysis. Patients were categorized into five consolidation strategies: HDC-ASCT (n=160), whole brain radiotherapy (WBRT, n=14), high-dose cytarabine (Ara-C, n=83), etoposide/Ara-C (EA, n=34), and observation without consolidation (n=26). Progression-free survival (PFS) was defined as the time from the post-induction response evaluation date to the date of disease progression or death from any cause, whichever occurred first. Overall survival (OS) was defined as the time from the post-induction response evaluation date to the date of death from any cause. Survival was estimated using the Kaplan–Meier method and compared using multivariable Cox regression. Results: Of the 317 patients, 52.1% were male and the median age at diagnosis was 61 years (range, 17–85). The median follow-up duration was 81 months. Baseline characteristics differed across consolidation groups: patients receiving HDC-ASCT were younger (median 57 years) and more likely to have ECOG performance status (PS) ≤1 (72.5%) than those treated with EA (median 70 years; ECOG PS ≤1: 52.9%), Ara-C (median 66 years; ECOG PS ≤1: 65.1%), or observation (median 75 years; ECOG PS ≤1: 38.5%) (p0.001 for age; p=0.005 for ECOG PS). Compared to the observation group (2-year PFS rate: 31.8%, 2-year OS rate : 51.3%), HDC-ASCT showed significantly better outcomes (2-year PFS rate: 76.1%, 2-year OS rate: 83.6%) (both p0.001). In contrast, no other consolidation strategy offered a significant survival benefit over observation: WBRT (2-year PFS rate: 56.2%, 2-year OS rate: 69.6%), Ara-C (2-year PFS rate: 34.9%, 2-year OS rate: 61.4%), and EA (2-year PFS rate: 30.8%, 2-year OS rate: 39.6%) (all p0.050). In the multivariate analysis, HDC-ASCT consolidation was independently associated with favorable survival outcomes (PFS: hazard ratio HR 0.37, p=0.001; OS: HR 0.32, p=0.001), along with age 60 years (PFS: p=0.022, OS: p=0.015) and ECOG PS ≤1 (PFS: p=0.015, OS: p=0.095). In the elderly subgroup (age ≥60 years, n=187), HDC-ASCT (n=64) remained the only effective consolidation strategy with superior PFS (2-year rate: 73.4%; p0.001) and OS (2-year rate: 76.4%; p0.001), compared to observation (n=21; 2-year PFS rate: 35.7%, 2-year OS rate: 50.0%). No other consolidation strategies (EA n=32, Ara-C n=61, and WBRT n=9) were associated with improved survival outcomes compared with observation (all p0.05 for both PFS and OS). Conclusion: In this large, single-center cohort with long-term follow-up, HDC-ASCT was the only consolidation strategy that provided a significant survival benefit in patients with PCNSL. These findings suggest limited benefit from reduced-intensity chemotherapy or WBRT as consolidation therapy. Our results highlight the need for alternative effective strategies for PCNSL patients who are not eligible for HDC-ASCT.