Modeling Marginal Zone Lymphomagenesis

Introduction: Indolent B-Cell Non-Hodgkin’s Lymphomas (NHL) represent a heterogeneous group of lymphoproliferative malignancies, that remain largely incurable. Marginal zone lymphomas (MZL) are the second most common subtype of indolent NHL, and lack a unique cytogenetic identifying abnormality. The B-cell receptor signaling pathway is activated in B-cell malignancy and mediates its activity mainly through the Phosphoinositide 3-kinase (PI3K) pathway. Furthermore, novel PI3K inhibitors, such as copanlisib and parsaclisib, have shown impressive clinical activity in several indolent lymphomas including MZL. This further supports the important role of the PI3K pathway in the pathogenesis of this tumor. Therefore, we hypothesized that the PI3K-mTOR pathway is sufficient for driving the pathogenesis of MZL.

Methods: In order to test our hypothesis, we generated a genetically engineered mouse model carrying heterozygous global knockout alleles of both the tumor suppressor genes Phosphatase and Tensin Homolog (PTEN) and Liver Kinase B1 (LKB1). This led to over-activation of the PI3K-mTOR pathway in all mouse tissues. We closely monitored these mice for tumor formation via weekly physical examinations for several months. Upon tumor detection, the mouse was sacrificed, and tumors were sectioned for histological characterization. In order to generate a more specific model of B cells, and more accurately mimic the underlying human disease, we used the Cre-LoxP system to create the CD19-Cre-PTEN^fl/fl-LKB1^fl/fl.

Results: Thirty mice of global KO PTEN^+/- LKB1 ^+/- died or were sacrificed due to disease progression, defined as either lymph node enlargement and/or splenomegaly. All mice showed either abnormal lymphadenopathy or splenomegaly. By Kaplan-Meier analysis, we saw a steady decrease in both tumor-free and overall survival after 3 months of age. Utilizing the product limit method, the median survival time was 6 months (95% CI: 6, 8). A total of 51 lymph nodes were sent for immunohistochemistry and pathological characterization. Of the 51 nodes, 61.5% (N=32) showed indolent Non-Hodgkin’s Lymphoma, 25% (N=13) were atypical, and 11.5% (N=6) were reactive. All lymph nodes with indolent NHL were of MZL subtype. Compared to wild type (n=3), the new CD19-Cre-PTEN^fl/fl -LKB1^fl/fl (n=3) showed an overall increase in spleen mass (120 vs 196 mg, p=0.0564), % B1 cells ( 4% vs 59%, p= 0.0075), % MZ cells (5% vs 30%, p=0.0547), % plasma cells (1% vs 12%, p=0.0729), and decrease in % FO cells (80% vs 12%, p=0.0003) by flowcytometry. Further characterization of the new model is currently underway.

Conclusion: Marginal zone lymphoma remains an incurable lymphoma that lacks reliable preclinical models. Our data provides, for the first time, a proof of concept on the role of the PI3K-mTOR pathway in the pathogenesis of marginal zone lymphoma and paves the way for future studies understanding the biology of this disease, and developing rational therapies for this incurable malignancy.