This is the authors' abstract. We don't add key points for this paper.
Secondary central nervous system (CNS) involvement in diffuse large B-cell lymphoma (DLBCL) is an infrequent but almost always fatal complication. Its incidence in the rituximab era ranges from 3·9% to 6·7%, depending on the inclusion criteria. Only elderly patients were included in two studies (Feugier et al, 2004; Boehme et al, 2009), whereas Villa et al (2010) included only high-risk patients with advanced stage and testicular involvement;, patients who received intrathecal (IT) prophylaxis were excluded in two studies (Yamamoto et al, 2010; Tomita et al, 2011), and all patients with DLBCL were included in another two studies (Shimazu et al, 2009; Tai et al, 2011). However, it is not clear whether rituximab has really changed the incidence of secondary CNS involvement or if there are any differences in relapsing patients’ characteristics in the pre- and post-rituximab era. This retrospective study included patients in the Nebraska Lymphoma Study Group database diagnosed with DLBCL and secondary CNS involvement between January 1, 1983 and August 1, 2009. Patients with CNS involvement at presentation and those with human immunodeficiency virus-associated lymphoma were excluded. Evaluation of CNS involvement at diagnosis by computed tomography scan, magnetic resonance imaging or cerebrospinal fluid analysis was performed only if clinically indicated. Of 1197 patients receiving treatment with curative intent, 811 were treated with CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) or CHOP-like regimens in the pre-rituximab era (prior to 2001), and 386 patients were treated with rituximab (R) in combination with CHOP or CNOP (mitoxantrone instead of doxorubicin). The overall incidence of secondary CNS involvement in our cohort was 3·2% (38 patients). Eight patients (2·1%) were identified among those who had been treated with R-CHOP (or R-CNOP) and 30 patients (3·7%) among those who had been treated with CHOP-like regimens. The 5-year cumulative incidence (CI) of CNS relapse was 2·2% for the patients treated with rituximab versus 3·6% for those treated without rituximab; P = 0·049, log-rank test (Fig 1). None of the patients with secondary CNS involvement after R-CHOP received IT prophylaxis, except Patient 5 who was given only two doses (prematurely stopped due to coagulation abnormalities) (Table Ia). In the pre-rituximab group, three patients received IT prophylaxis (Table Ib). Pre-treatment characteristics of the patients between the groups were similar. Most patients had stage IV disease and extranodal involvement. However, patients with secondary CNS involvement who had been treated with R-CHOP were significantly younger (P = 0·038, Wilcoxon rank sum test). This may be a reflection of the younger age of all patients who had received R-CHOP compared to those who had been treated with CHOP-like regimens: median age 67 years (range, 18–96 years) vs. 69 years (range, 15–94 years), respectively; P = 0·005, Wilcoxon rank sum test. We reviewed the diagnostic slides of the patients with secondary CNS involvement treated with R-CHOP. All patients had DLBCL, whereas Patient 3 had primary mediastinal B-cell lymphoma. The cumulative incidence of secondary central nervous system involvement (R)-CHOP, rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone; (R)-CNOP, rituximab, cyclophosphamide, mitoxantrone, vincristine, prednisone. All patients treated with R-CHOP had parenchymal-only CNS involvement (brain biopsy was done in seven of eight cases) as opposed to 12 of 27 patients with leptomeningeal involvement in the pre-rituximab era. Furthermore, none of the R-CHOP-treated patients actually progressed during the treatment, although three patients had an early isolated CNS relapse within a month after systemic remission was documented. In contrast, 14 of 30 patients in the pre-rituximab era progressed during treatment. Three late isolated CNS relapses were observed after a complete remission was achieved, one in the R-CHOP group (44 months) and two in the pre-rituximab era (56 and 100 months). Four patients in the R-CHOP group are alive and in remission with follow-up longer than 2 years. All of them were treated with high dose methotrexate-based treatment followed by peripheral blood stem cell transplantation in three patients and whole brain irradiation in one patient. In contrast, all patients in the pre-rituximab era died and only four patients survived for more than 6 months. The overall incidence of secondary CNS involvement in our study was rather low and comparable to the study by the Southwest Oncology Group (Bernstein et al, 2009). As expected, these patients had more advanced disease and more frequent extranodal localization at presentation. In this study, we demonstrated that the addition of rituximab to CHOP was associated with a reduction in the incidence of secondary CNS involvement. Moreover, compared to the pre-rituximab era, progression during systemic treatment is less frequent, CNS relapses tend to be isolated and, most strikingly, leptomeningeal involvement is uncommon. These findings are similar to the Vancouver group’s study (Villa et al, 2010). The reduced incidence of secondary CNS involvement was also demonstrated by Boehme et al, 2009; and Shimazu et al, 2009;. However, a greater proportion of leptomeningeal relapses in R-CHOP-treated patients was observed in the former study. Conversely, the addition of rituximab did not change the incidence of secondary CNS involvement in the other three studies (Feugier et al, 2004; : Yamamoto et al, 2010; Tai et al, 2011). In contrast to previous studies, our R-CHOP-treated patients with secondary CNS involvement were younger and, probably for that reason, half of them were successfully salvaged with second-line treatment.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Mitrović et al. (2012) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: