Genomic Profiling Reveals Differences in Primary Central Nervous System Lymphoma and Large B-Cell Lymphoma, With Subtyping Suggesting Sensitivity to BTK Inhibition

Abstract

Background B-cell primary central nervous system (CNS) lymphoma (PCL) is diffuse large B-cell lymphoma (DLBCL) confined to the CNS. Less than 50% of patients with PCL achieve complete remission with current therapies. We describe the findings from comprehensive genomic profiling (CGP) of a cohort of 69 patients with PCL, 36 cases of secondary CNS lymphoma (SCL), and 969 cases of DLBCL to highlight their differences and characterize the PCL cohort. In addition, we highlight the differences in frequency of germinal center B-cell like (GCB) and non-GCB subtypes and molecular subtypes, particularly MCD and EZH subtypes, between PCL and DLBCL. Materials and Methods Sixty-nine cases of B-cell PCL, 36 cases of secondary CNS lymphoma (SCL), and 969 cases of DLBCL were evaluated by CGP of 405 genes via DNAseq and 265 genes via RNAseq for fusions (FoundationOne Heme). Tumor mutational burden (TMB) was calculated from 1.23 Mb of sequenced DNA. Results Genomic alterations with significant differences between PCL and DLBCL included MYD88, ETV6, PIM1, PRDM1, CXCR4, TP53, and CREBBP, while only MYD88 was significantly different between SCL and DLBCL. PCL cases were significantly enriched for the MCD molecular subtypes, which have an excellent response to BTKi. We report a patient with a durable complete response to BTKi consistent with their genomic profile. EBV status, CD274 amplification, and TMB status suggest that 38% of PCL patients may benefit from ICPI; however further study is warranted. Conclusion CGP of PCLs reveals biomarkers, genomic alterations, and molecular classifications predictive of BTKi efficacy and potential ICPI efficacy. Given the limitations of standard of care for PCL, CGP is critical to identify potential therapeutic approaches for patients in this rare form of lymphoma.

New hopes in relapsed refractory primary central nervous system lymphoma

“…Strictly linked to the concept of relapse prevention is the use of newer techniques, such as comprehensive genomic profiling, radiomics and artificial intelligence based approaches [93 ▪▪ ,94 ▪ ]. Indeed, molecular and radiological features merged with clinical data could potentially reveal new classification subgroups that might predict better risk and direct patients to newer and personalized therapeutic approaches [39 ▪▪ ,95 ▪ ]. Along with the monitoring of soluble biomarkers, the acquisition of knowledge regarding the clonal evolution of the disease might help in driving the way to improve the actual treatment regimens with the aim to achieve a larger number of long-term remissions [96 ▪▪ ].…”