Primary Central Nervous System Lymphoma and the Blood-Brain Barrier

Recent reports propose high-dose (HD) methotrexate (MTX)-based cytoreduction followed by consolidation with HD chemotherapy (HDC) with or without autologous stem-cell transplantation (ASCT) as first-line treatment for primary central nervous system lymphoma (PCNSL). The main rationale is that HDC-ASCT may improve disease control by achieving higher therapeutic concentrations of cytotoxic chemotherapy in the CNS, thus circumventing the blood-brain barrier (BBB). We have employed an alternative approach to treat PCNSL patients effectively. This first-line treatment enhances delivery of standard dose (SD) MTX-based chemotherapy administered intra-arterially (i.a.) with osmotic BBB opening plus the anti-CD20 monoclonal antibody (mAb) rituximab, followed in complete responders by anti-CD20 mAb maintenance immunotherapy. The main features that distinguish our approach from HDC-ASCT consolidation are to utilize BBB properties to maximize delivery of SD chemotherapy to the CNS sanctuary, and maintenance immunotherapy to delay recurrence. A multi-center study of 149 PCNSL patients (2,079 i.a. BBB treatments) resulted in one death within 48 hrs after treatment, from pulmonary embolism. Prior to routine use of granulocyte colony stimulating factor, granulocytopenic fever occurred in 3% of BBB treatments (Angelov, J Clin Oncol 2009). This contrasts with HDC-ASCT consolidation which carries substantial morbidity and treatment-related mortality (6% to 9%) (Ferrari, Lancet Hemat 2016; Omuro, Blood 2015). Further, HDC-ASCT is used primarily in younger PCNSL patients (< 60-65 yrs) with good performance status and is not feasible in many PCNSL patients given the median age at diagnosis is 60-65 yrs.

First-line enhanced BBB delivery of SD MTX-based chemotherapy in 149 patients provided an overall response rate (ORR) of 82% (58% CR rate) and median overall survival (OS) of 37 mo. In low risk patients (age < 60 yrs and KPS ≥ 70) median OS was approximately 14 yrs. Survivors in CR a minimum of 2 yrs after diagnosis (n = 24) were evaluated with standard neuropsychological tests. Median follow-up was 12 yrs (range 2 to 26 yrs) and results indicated long-term preserved or improved cognitive function (Doolittle, J Clin Oncol 2013). Recently we compared efficacy from two first-line MTX-based regimens: HD MTX (intravenous, 4gr/m²) (Thiel, Lancet Oncol 2010), and enhanced BBB delivery of MTX (i.a., 3.5gr/m²). After adjusting for patient characteristics, CR rate and progression free survival were higher in patients < 65 yrs treated with enhanced delivery of MTX (p < 0.001). When first-line rituximab was added to the i.a. enhanced delivery regimen in 24 patients (median age 64 yrs; median KPS 65), the ORR improved to 92% (75% CR) and OS to 61 mo.

To prevent recurrence, Ney (Leuk Lymphoma 2009) treated a small group of patients in CR after standard PCNSL treatment, with maintenance rituximab. The addition of maintenance immunotherapy resulted in average disease control of 22 mo or longer. Consistent with the Ney report, we noted prolonged median CR duration of 38 mo (range: 10 to 85 mo) in a small subset treated with maintenance rituximab (n = 9). Increased survival was also seen in translational laboratory studies of a rodent model of intracerebrally implanted MC116 human B-cell lymphoma cells, using single agent rituximab, whether or not delivered with BBB disruption and whether or not MTX was included (Muldoon, Clin Cancer Res 2011). These results suggest that rituximab slowly leaked into main CNS tumor mass even in absence of BBB opening; and given the long half-life, was trapped by binding to CD20+ on tumor cells, attaining sufficient concentration for antitumor efficacy.

Our goal is to maximize first-line enhanced BBB delivery of SD chemotherapy to enable less treatment-related morbidity and mortality than is associated with HDC-ASCT in PCNSL, and increase CR duration. Neurocognitive safety has been demonstrated in long-term survivors treated with BBB delivery. Based on the encouraging results from Ney et al, our pilot data using maintenance rituximab, and translational lab studies of mAb delivery to brain, a new randomized multi-center trial is underway to study the effect of maintenance obinutuzumab on CR duration, neurocognition and quality of life in PCNSL patients who achieve CR following first-line MTX-based chemotherapy (NCT02498951). A trial update will be presented during the ASH 2016 meeting.