Central Nervous System Events In Patients with Diffuse Large B-Cell Lymphoma In the Rituximab Era

The central nervous system (CNS) is considered a sanctuary because drugs at standard doses rarely reach it. CNS event is defined as disease recurrence in the CNS during complete systemic remission or CNS progression as concurrent disease in the CNS during active systemic disease. These CNS events are associated with extremely poor patient prognosis. To evaluate the risk of CNS events in patients with diffuse large B-cell lymphoma (DLBCL) in the rituximab (R) era, we performed a retrospective study with a large cohort in Japan.

All patients were diagnosed as having DLBCL and underwent primary therapy from September 2003 to December 2006. Patients with distinct forms of DLBCL, such as intravascular lymphoma, primary effusion lymphoma, and pyothorax-associated lymphoma were excluded. Those who had received any prophylactic CNS treatment; and those with CNS involvement at the beginning of therapy were also excluded. Primary therapy comprised standard R-CHOP therapy or R-THP-COP therapy (pirarubicin instead of doxorubicin hydrochloride). CNS involvement was considered when malignant cells were detected in cytocentrifuged preparations of cerebrospinal fluid (leptomeningeal type) and/or when an intracranial or spinal mass was detected by radiologic imaging such as computed tomography or magnetic resonance imaging (parenchymal type).

(1) Baseline characteristics. Clinical data from 1,492 patients were collected from 48 institutions. The median age was 67 years (range, 15–93 years). R-CHOP therapy was administered in 1,227 patients (82.2%) and R-THP-COP therapy, in 265 patients (17.8%). Therapy for 6–8 cycles was administered to 1099 patients (73.7%). Local irradiation was included for 344 patients (23.1%). The 5-year overall survival (OS) rate in the entire cohort was 72.8%, and the median observation period for the surviving patients was 47.6 months. (2) Incidence of CNS events. In total, 94 CNS events (6.3%) were recorded. More than half were of the parenchymal type (57.4%); the rest were mostly of the leptomeningeal type (29.8%), and some were of both types (12.8%). The events occurred during the first complete remission (CR) in 40 patients (42.6%) and in the second or later CR in 11 patients (11.6%). Death from any cause was recorded in 64 of the 94 patients (68.1%) during the observation period, and most deaths were due to lymphoma. The cumulative 5-year probability of CNS events was 8.0%. (3) Risk factors for CNS events. Multivariate Cox regression analysis identified involvement of the breast (relative risk (RR), 7.9), adrenal gland (RR, 3.3), paranasal sinus (RR, 2.9), and bone (RR, 2.3). as risk factors for CNS events. Time to CNS event curves differed significantly between patients with and without CNS risk sites (P < 0.001). (4) Survival after CNS events. Patients with CNS events demonstrated significantly poorer OS (P < 0.001). The 2-year OS rate after a CNS event was 25.7%, and the 50% survival duration was 4.6 months. No significant differences were observed between any 2 of the 3 types of CNS events. Depending on the systemic lymphoma status at the time of the CNS events, patients in first CR showed better survival than the others (P = 0.019). Patients in any CR also showed superior survival than those not in CR (P = 0.015).

We concluded that DLBCL patients in the R era with any of these risk factors (breast, adrenal gland, paranasal sinus, or bone involvement), in addition to patients with testicular involvement in whom CNS prophylaxis has been already justified, are at high risk for CNS events. The efficacy and manner of CNS prophylaxis for each involvement sites should be evaluated further.

No relevant conflicts of interest to declare.