Prophylaxis with Intrathecal Chemotherapy and High Dose Methotrexate in Patients with Diffuse Large B Cell Lymphoma at High Risk for Central Nervous System Relapse: A Single Centre Experience

Abstract Central nervous system involvement (CNS) is a serious and mostly fatal complication of aggressive lymphoma. The incidence of CNS disease in diffuse large B-cell lymphoma (DLBCL) is low (about 5%) and there are not randomised prospective trial which specifically address a decision-making process for CNS prophylaxis. Potentially two methods exist for identifying patients requiring CNS directed treatment. Surveillance lumbar puncture and brain magnetic resonance (MRI) at the time of diagnosis could identify the presence of lymphoma; another method is the identification of patients whose characteristics are indicative of a high risk of CNS disease. Several site-specific risks are described in literature such as testicular, breast, paranasal sinuses, epidural spaces, and intravascular involvement with an incidence of CNS relapse ranging from 15% to 50%. Recently a modified IPI score (CNS IPI) was described to predict risk of CNS relapse. The efficacy of different forms of CNS prophylaxis has never formally been demonstrated. In the RICOVER 60 trial, patients treated with R-CHOP-14 instead of CHOP-14 presented a lower incidence of CNS relapse while intrathecal methotrexate (MTX) has not showed a role in preventing CNS disease for patients treated with combined immunochemotherapy. Recently, combination of intrathecal MTX and high dose MTX infusion after R-CHOP treatment was considered an effective strategy of prevention of CNS relapse. In order to evaluate the efficacy and feasibility of intrathecal MTX administration and high dose MTX after first line chemotherapy, we retrospectively reviewed 27 patients (11 males and 16 females, mean age 61 yrs, range 27-79) with newly diagnosed DLBCL at high risk for CNS relapse, treated from January 2009 to April 2018 at Veneto Institute of Oncology IOV-IRCCS. In our cohort 21 (78%) patients were at advanced stage (III-IV Ann Arbor stage) and 15 (56%) belonged to intermediate-high or high risk IPI categories, two patients presented with orbital localization. Almost all patients received R-CHOP as first line treatments, two patients with paravertebral localization received HyperCVAD as front line approach. For 20 patients CNS-IPI was intermediate or high. In the other cases with low CNS-IPI, disease localization was considered at high risk of CNS relapse (breast, paravertebral, orbit, paranasal sinus). In all patients we performed brain MRI and diagnostic lumbar puncture at diagnosis (all negative at flow cytometry analysis). All 14 patients >65 yrs were evaluated with comprehensive geriatric assessment (CGA) and 8 (57%) were considered fit. Nineteen (19) patients (70%) received at least 3 lumbar punctures with MTX and 24 (89%) two courses of high dose intravenous MTX during first line therapy. At the end of planned first line treatment, 24 (89%) patients obtained a complete remission at PET scan evaluation and 3 patients (11%) presented progressive disease, 2 with CNS involvement and another with peritoneal disease; this last patient had a Double Hit lymphoma with BCL6 and c-MYC rearrangements. Another patient in complete remission after R-CHOP chemotherapy, experienced CNS relapse three year after obtaining complete remission. Among the two patients treated with Hyper CVAD regimen, one is still in complete remission 5 years after the end of treatment; another developed early CNS relapse. All the 3 patient who experienced CNS involvement after R-CHOP, didn't receive prophylaxis because were evaluated frail at CGA. So, at a median follow up of 26.2 months (3.5-100 months) all patients who received the planned treatment at full dose including CNS prophylaxis did not experience central nervous system relapse. In conclusion, CNS prophylaxis including intrathecal MTX administration and high dose MTX infusion after first line chemotherapy is feasible and effective. Larger prospective trials are needed to evaluate the most effective prophylactic therapy and the correct timing of intravenous MTX infusion. Disclosures No relevant conflicts of interest to declare.