High-Dose Methotrexate Based Therapy for the Treatment of Bing Neel Syndrome

BACKGROUND

Bing-Neel Syndrome (BNS), characterised by the direct infiltration of central nervous system (CNS) with clonal lymphoplasmacytic lymphoma cells, is a rare cause of neurological complications in patients with Waldenström Macroglobulinaemia (WM). The pathophysiology/neurotropism of BNS is not well understood. Various treatments that appear to cross the blood-brain barrier are used, including chemoimmunotherapy combinations and Ibrutinib. High-dose methotrexate (HDM)-based regimens, including MATRix (methotrexate [MTX], cytarabine, thiotepa, rituximab)¹and R-IDARAM (rituximab, MTX [intrathecal and intravenous], cytarabine, dexamethasone) are also used, based on the experience in high-grade CNS lymphoma; however, there are concerns regarding the appropriateness of these regimens in the indolent setting, largely due to their toxicity.

This study aimed to determine the efficacy and tolerability of HDM-based, frontline treatment in a large cohort of BNS patients.

METHODS

A single-centre retrospective review of consecutive patients with BNS treated with HDM based regimens was undertaken. Comprehensive laboratory and CNS imaging results were collated (including whole-brain/spine MRI, bone marrow biopsy, and CSF biochemistry, cytology and MYD88 L265P mutation). Best responses achieved, according to IWWM response assessment criteria, were determined.

RESULTS

Between 2011 and 2020, 27 patients were treated with HDM-based chemoimmunotherapy: 14 (51.9%) with MATRix, 5 (18.5%) with MTX/cytarabine/rituximab (MAR), 2 (7.4%) with HDM only, and 6 (22.2%) with R-IDARAM. Median time from WM diagnosis to BNS diagnosis was 60 months (0-572 months), including 10 patients (37.0%) who were diagnosed with WM at the time of BNS presentation/diagnosis. Median age at diagnosis with BNS was 66 years (43-75 years), and 15/27 (55.6%) were male.

CR was achieved in 6 patients (22.2%), PR in 7 (25.9%), and non-response in 8 (29.6%); PD on therapy occurred in 4 cases (14.8%) and 2 patients died before response could be assessed. Of the 14 patients who received MATRix, 5 (31.2%) achieved PR, 6 (42.9%) had stable disease and 3 (21.4%) progressed while on treatment (Figure 1A). Of the 7 who received MAR or HDM only, 4 (57.1%) achieved CR, with 1 PR and 1 SD; 1 patient died before response could be assessed. Response was not affected by prior chemotherapy (for WM), and no biological or radiological feature predicted response in this highly heterogenous population.

Predicted overall survival plateaued at 71.5% 1 year after start of treatment (Figure 1B). At a median follow up of 19.4 months, 7 patients (25.9%) had died: 3 (11.1%) from progressive BNS (1 patient had received each of MATRix, R-IDARAM and HDM only), and 4 (14.8%) from treatment-related toxicity (including infection). Of the latter 4 patients (all male), 2 had received MATRix and 1 had received each of MAR and R-IDARAM. No objective response was seen in 6 of the 7 deceased patients; the remaining 1 patient achieved PR following full-dose MATRix, but subsequently relapsed and died of BNS 7 months later.

Regarding dose intensity, 5/15 MATRix recipients (33.3%) received full-dose regimens (i.e. 3.5mg/m² of MTX per cycle for 3-4 cycles); a further 5 patients received a dose reduction of MTX (to 50-75%) and 9 patients (60%) received <3 cycles. Treatment was discontinued early in 6 of these patients primarily due to inadequate response (SD or PD), in 2 patients due to toxicity, and in 1 patient after desirable response had been achieved.

CONCLUSION

In the setting of indolent CNS lymphoma, the rationale for chemoimmunotherapy is unclear. Although HDM-based therapy is used in BNS, its administration is limited by toxicity, and dose reductions may undermine dose intensity in the CSF. We observed higher rates of treatment-related mortality and only modest response rates with MATRix vs other HDM-based regimens; this primarily related to treatment related toxicity necessitating dose reduction requirements. These data suggest MATRix offers no response benefit over MAR or HDM only; indeed, the omission of thiotepa appears to allow dose optimisation of MTX and cytarabine, and does not negatively impact on outcome.

If patients survive the first year of their disease/treatment, survival outcome is excellent.

1. Ferreri, A.J.M., et al. The Lancet Haematology 3, e217-e227 (2016).