Secondary CNS Relapse in DLBCL in the Rituximab Era - an Analysis of Prospective Studies

INTRODUCTION

CNS relapse in DLBCL carries poor prognosis. Some studies have suggested decreased incidence with rituximab, but there are several others reporting otherwise. We analyzed prospective studies in literature to understand the role of rituximab and CNS prophylaxis in DLBCL, in comparison to CHOP based therapy.

METHOD

Extensive searches using PUBMED, EMBASE, CENTRAL and major hematology conferences were conducted for prospective studies. The keywords “CNS”, “diffuse large B-cell lymphoma”, “relapse”, “prophylaxis”, “rituximab”, “CHOP” were used. Inclusion: (i) prospective or randomized trials (ii) Entire study population or a significant majority of patients were newly diagnosed DLBCL, (iii) no evidence of CNS involvement at baseline, (iv) use of rituximab-chemotherapy or CHOP-based chemotherapy, (v) have data relevant to our study. Exclusion: (i) retrospective studies, review article or case reports, (ii) exclusively testicular, mediastinal or double hit lymphoma, (iii) HIV positive patients. Data is presented as mean ± standard error of mean. Significant differences (at P<0.05) between groups and couplets were determined by one-way ANOVA and by two-tailed unpaired t-test respectively. Significance of the Kaplan-Meier survival curves was determined using Log-rank (Mantel-Cox) and Gehan-Breslow-Wilcoxon tests.

RESULTS

The study population characteristics are shown in Table. The mean incidence of CNS relapse with rituximab-chemotherapy (R-CHOP/R-CHOEP) was found to be 5.52% (3.21%-7.73%), while that with CHOP-based chemotherapy alone was 4.43% (3.53% -5.33%). No significant difference was observed by two-tailed unpaired t test (P = 0.94). The median time from diagnosis to CNS relapse was 6.5-7 months. The mean incidence of leptomeningeal, parenchymal and both relapses in the rituximab-chemotherapy group were 38.62(±1.93)%, 57.32(±3.06)%, 5.36(±1.79)% compared to 16.17(±0.44)%, 66.17(±0.44)%, 22.06(±0.41)% with chemotherapy alone. One-way ANOVA also showed that use of rituximab resulted in statistically significant (P<0.0001) differences between the patterns of CNS relapse. About 74.1% of CNS relapses among patients receiving rituximab-chemotherapy were isolated CNS relapse, compared to 69.2% for those receiving CHOP chemotherapy. The mean incidence of CNS relapse with the use of any prophylactic CNS directed therapy was observed to be a significantly (P = 0.044) lower [2.97% (2.32%-3.62%)] compared to that without [6.12% (5.91%-6.335]. Median survival following CNS relapse of 365 days following rituximab based chemotherapy and 75 days after CHOP were significantly different (HR for CHOP use 4.867±0.77 at P<0.05).

CONCLUSION

Rituximab hasn't significantly decreased overall incidence, but causes less parenchymal CNS relapse. CNS prophylaxis has a definite role in high risk population. There is significantly better survival after CNS relapse in the rituximab era.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal

Figure 2

View largeDownload slide

Figure 2

View largeDownload slide

Close modal