Chemotherapy with high-dose methotrexate is the conventional approach to treat primary CNS lymphomas, but superiority of polychemotherapy compared with high-dose methotrexate alone is unproven. We assessed the effect of adding high-dose cytarabine to methotrexate in patients with newly diagnosed primary CNS lymphoma.

This open, randomised, phase 2 trial was undertaken in 24 centres in six countries. 79 patients with non-Hodgkin lymphoma exclusively localised into the CNS, cranial nerves, or eyes, aged 18-75 years, and with Eastern Cooperative Oncology Group performance status of 3 or lower and measurable disease were centrally randomly assigned by computer to receive four courses of either methotrexate 3.5 g/m(2) on day 1 (n=40) or methotrexate 3.5 g/m(2) on day 1 plus cytarabine 2 g/m(2) twice a day on days 2-3 (n=39). Both regimens were administered every 3 weeks and were followed by whole-brain irradiation. The primary endpoint was complete remission rate after chemotherapy. Analysis was by intention to treat. This study is registered with ClinicalTrials.gov, number NCT00210314.

All randomly assigned participants were analysed. After chemotherapy, seven patients given methotrexate and 18 given methotrexate plus cytarabine achieved a complete remission, with a complete remission rate of 18% (95% CI 6-30) and 46% (31-61), respectively, (p=0.006). Nine patients receiving methotrexate and nine receiving methotrexate plus cytarabine achieved a partial response, with an overall response rate of 40% (25-55) and 69% (55-83), respectively, (p=0.009). Grade 3-4 haematological toxicity was more common in the methotrexate plus cytarabine group than in the methotrexate group (36 [92%] vs six [15%]). Four patients died of toxic effects (three vs one).

In patients aged 75 years and younger with primary CNS lymphoma, the addition of high-dose cytarabine to high-dose methotrexate provides improved outcome with acceptable toxicity compared with high-dose methotrexate alone.

Swiss Cancer League.