Consolidation Improves Survival in Primary Central Nervous System Lymphoma without Preference for Type of High-Dose Methotrexate-Based Induction Treatment Regimen

Introduction

Primary central nervous system lymphoma (PCNSL) is a rare non-Hodgkin lymphoma with a median overall survival (OS) of 3 years. The current treatment strategies of PCNSL are governed by high-dose methotrexate (HD-MTX)-based (immuno)chemotherapies, and in young and fit patients combined with high-dose cytarabine (HD-AraC). These (immuno)polychemotherapies are preferably consolidated with whole brain radiotherapy (WBRT) or autologous stem cell transplantation (autoSCT). The aim of this large, retrospective multicenter study was to evaluate the outcome of PCNSL patients treated with several HD-MTX-based (immuno)polychemotherapeutic induction regimens with and without consolidation.

Methods

A total of 250 consecutive, newly diagnosed PCNSL patients between 2008 and 2021 from 11 Dutch and Belgian hospitals were included. All patients were treated with ≥1 cycle of HD-MTX ≥3g/m² per cycle, with an intended cumulative dose of ≥12g/m². HD-AraC was defined as 2g/m² per cycle with an intended cumulative dose of ≥8g/m². Patients treated with three different standard first-line regimens with intended cumulative doses of ≥12g/m² HD-MTX and ≥8g/m² HD-AraC were allocated to the ‘intense’ treatment group (Figure 1). Patients receiving two other combination regimens of HD-MTX (intended ≥12g/m²) with procarbazine or teniposide and carmustine as standard first-line therapeutic strategy were assigned to the ‘moderate’ treatment group. WBRT with a cumulative dose of 40 Gy and autoSCT (from 2012 onward) were evaluated as sequential consolidation if received within 8 weeks after good response to (immuno)polychemotherapy. Primary end points were best objective response rate (ORR) at any time and 5-year OS.

Results

Of the 250 patients, 142 (57%) were male and the median age at diagnosis was 67 years (range 28-86). In total, 113 (45%) received ‘intense’ treatment and 137 (55%) ‘moderate’ treatment. Considering ORR, 122 patients achieved complete response (CR, 49%), 88 partial response (PR, 35%), 29 stable/progressive disease (SD/PD, 12%) and 11 unknown response (4%).
During follow-up, 117 (47%) patients experienced progression or relapse and 119 (48%) patients died, of which 97 (82%) were lymphoma-related deaths. Median PFS and OS were 2.4 and 3.5 years, respectively. No significant OS difference was found between the ‘intense’ and ‘moderate’ treatment regimens (OS p=0.14) or between any of the HD-MTX-based (immuno)polychemotherapeutic treatment regimens (Figure 1, p=0.54).

In total, 83 (33%) patients received subsequent consolidation therapy of whom 52 (63%) WBRT and 31 (37%) autoSCT. No OS differences between WBRT and autoSCT were observed (Figure 2, p=0.67). Patients receiving consolidation demonstrated a superior OS (p<0.0001) and a significantly better best ORR (p<0.01) compared to unconsolidated patients. All baseline characteristics were comparable between the consolidated and unconsolidated group, except for age (p<0.01) which was higher in the unconsolidated group. To correct for immortal time bias for the consolidation group, landmark (LM) analyses were performed. The significant effect of consolidation was confirmed at LM 5 months (OS p=0.03) and LM 9 months (OS p=0.03) whereas 80% and 98% of the cohort received consolidation, respectively. To evaluate possible confounders, multivariable analysis was performed and confirmed that older age (HR 1.03, CI 1.01-1.05), WHO performance status ≥2 (HR 1.84, CI 1.31-2.57) and elevated LDH (HR 1.78, CI 1.18-2.69) at diagnosis (all p<0.01) have an adverse effect on risk of mortality and consolidation significantly reduced the risk of mortality (HR 0.44, CI 0.30-0.64), in contrast to treatment intensity (HR 0.96, CI 0.67-1.37).

Conclusion

This multicenter study evaluated several first-line HD-MTX-based regimens in a large cohort of PCNSL patients. Sequential consolidation with WBRT or autoSCT was significantly associated with improved OS, regardless of type of HD-MTX-based induction regimens. Disregarding assessment of toxicity, these results together with other studies call for induction therapy with HD-MTX-based (immuno)polychemotherapy irrespective of type of regimen and sequential consolidation therapy with either WBRT or autoSCT, thereby shaping next treatment algorithms.