Expression of a Dysfunctional Activation Induced Cytidine Deaminase Is Correlated with Disease Progression in Splenic Marginal Zone Lymphoma.

Abstract 2397

B cells can undergo at least two differentiation pathways, dependent of T cells or not, starting from follicular or marginal zone B cells respectively. The T-independent response, less understood than the germinal center reaction, is triggered by specific antigens and arises from marginal zone B cells. During this development, some B cells undergo somatic hypermutation (SHM) and class switch recombination (CSR), triggered by the same DNA editing enzyme called Activation Induced Cytidine Deaminase (AID).

The splenic marginal zone lymphoma (SMZL) is a rare lymphoproliferative disorder characterized by a clonal expansion of B cells in the marginal zone of the spleen. These B-cells underwent SHM in roughly 60% of the cases but nearly none underwent CSR. These observations suggest that tumor clones originate from a particular activated B cell subset not transiting through the germinal center. In order to confirm this hypothesis, we focused our work on the status and impact of AID in this disease and worked on purified B cells extracted from spleen of well-characterized SMZL cases.

We determined AID status by quantitative RT-PCR analysis on 27 SMZL samples and compared it with 5 controls. In the SMZL group the relative level of expression of AID is heterogeneous but two subgroups could be distinguished: one considered as expressing AID (14 cases out of the 27 analyzed), the remaining considered as not expressing AID. When we compared AID expression rate with occurrence of SHM and CSR, no clear correlation between AID expression and presence of SHM or CSR could be observed suggesting that AID, when expressed, is dysfunctional.

To address this hypothesis, we first analyzed AID protein by immunohistochemistry and a good correlation between IHC signal and AID mRNA expression level has been observed. As AID gene was not mutated, we next focused our work on AID mRNA splicing variants as these variants exhibit different functions according to the domain of the protein they contain in a murine model.

We found that SMZL B cells express various splicing variants of AID mRNA, some of those variants corresponding to the full length isoform (n = 6/17), and other variants corresponding to AID-ΔE4a (n = 2/17) or AID-ΔE4 (n = 7/17) isoforms known to be expressed in normal germinal center B cells as well as in Chronic Lymphocytic and Acute Lymphoblastic Leukemia. These findings indicate that although expressed at the mRNA and protein levels, AID may not be fully functional in SMZL cases.

Finally we addressed the potential clinical significance of AID expression. We identified for that purpose a group of “progressive SMZL” patients that had received immuno-chemotherapy after splenectomy because of a significant risk of progression or transformation into aggressive large B cell lymphoma (n = 8/27) pre-empting outcome differences. We found a higher proportion of AID expressing patients in the defined “progressive SMZL” group (n = 7/8) as compared to the proportion found in the “indolent SMZL” group (n = 5/14, p = 0,03).

Altogether, this data suggest that the B cell clone leading to SMZL originate from the marginal zone and support the hypothesis of a lymphoproliferative disorder affecting the T-independent response. AID expression in SMZL may reflect an advanced stage of the disease and could be correlated with the evolution of the lymphoma into a more clinically or pathologically aggressive form.