CENTRAL NERVOUS SYSTEM (CNS) INVOLVEMENT IN DIFFUSE LARGE B-CELL LYMPHOMA. A SINGLE CENTER ANALYSIS OF THE RISK FACTORS AND THE IMPACT OF CNS PROPHYLAXIS IN THE RITUXIMAB ERA

Background: Central nervous system (CNS) relapse is a serious and generally fatal complication of diffuse large B-cell lymphoma (DLBCL) occurring in almost 5-7% of patients in the pre-rituximab era. The median time from diagnosis to detection of CNS disease is less than 1 year, suggesting that seeding in the CNS occurs early in the course of disease, with the exception of primary testis lymphoma in which it can occur up to 10 years after diagnosis. Up to date, there is no general consensus on how to define high-risk patients and what is the optimal CNS prophylaxis. The addition of rituximab to CHOP chemotherapy (RCHOP) has improved the clinical outcome of DLBCL. Nevertheless, the impact of rituximab on the incidence of CNS complications is still unclear. Aims: This retrospective study aimed to analyze the predictive factors of CNS relapse and the impact of CNS prophylaxis in DLBCL patients treated at our Institute over a 12-years period. and 2014. The patients with all stage of disease, regardless of the IPI score, were eligible and all received RCHOP every 14 or 21 days for at least 4/6 cycles. Patients with HIV or CNS involvement at diagnosis were excluded. Prophylaxis with intrathecal methotrexate (IT-MTX) was administered in patients with a high risk of CNS relapse according to the Italian Society of Hematology’s guidelines. The diagnosis of CNS relapse was based on CT/MRI assessment and/or cerebrospinal fluid cytology. Results: The median age was 58 years (IQR 47-68), 54% were males; 212/393 patients (54%) had stage III-IV, a bone marrow (BM) involvement was found in 51/393 patients (13%), while 49 (12%) had an involvement of more than 1 extranodal site, testis involvement was in 20/393 patients (5%). B symptoms were experienced by 100/393 patients (25%) and an elevated LDH was present in 115/393 patients (29%). According to the IPI score, 197/393 patients (50%) had a IPI<2 and 196/393 (50%) a IPI>2. Twelve/393 patients (3%) experienced a CNS relapse: 7 of them had a brain mass, 2 had a leptomeningeal involvement and 3 had both. CNS prophylaxis was carried out in 81/393 patients (21%): 27 with a IPI<2 and 54 (67%) with a IPI≥2. In the low risk group, only 1 patient with a primary testis involvement developed a CNS relapse 26 months from diagnosis, despite prophylaxis. In the high risk group according to IPI, 11 patients experienced a CNS relapse; only 4 of them had received IT-MTX[R1]. At univariate analysis: gender (p=0.029), advanced stage (p=0.004), PS>2 (p=0.009), presence of B symptoms (p=0.000), elevated LDH (p=0.006), involvement of >1 extranodal sites (p=0.000), BM involvement plus elevated LDH (p=0.003), testis localization (p=0.019) and IPI>=2 (p=0.000) were associated with an increased risk of CNS relapse. At multivariate analysis, only PS>2 (p=0.031, 95%CI: 0.009-0.189), testis involvement (p=0.001, 95%CI: 0.048-0.2) and BM involvement plus elevated LDH (p=0.016, 95%CI: 0.021- 0.2) were predictive factors of high risk of CNS relapse. The 2-year overall survival of the CNS relapsed patients was 9%, with a median follow-up of 8 months from diagnosis (IQR 4-28). Summary/Conclusions: CNS relapse remains a fatal event for patients with DLBCL even in the rituximab era. Poor PS, testis and BM involvement combined relapse. In our experience, single IT-MTX prophylaxis did not reduced CNS.