Anti-CD70 Monoclonal Antibodies Show Efficacy in Preclinical Models for Chronic Lymphoblastic Leukemia

Abstract

Although therapy for chronic lymphocytic leukemia (CLL) has improved significantly with combination therapies that include ibrutinib and/or anti-B cell mAbs, resistance can develop, and patients can discontinue therapy for various reasons. Therefore, effective, alternative therapies are needed. CD70, a TNF family member, and its receptor, CD27, are mainly expressed on B and T lymphocytes. Different from normal B cells that only transiently express CD70 upon activation, CLL B cells stably express the molecule at relatively high levels, suggesting a fundamental biologic abnormality and adding another example of the dysregulation of the immune system in CLL. Here we investigated the role of CD70 in CLL using blocking anti-CD70 mAbs in preclinical mouse models of the disease. We first confirmed stable expression of CD27 and CD70 in patient samples as well as on murine TCL1 CLL B cells, including TCL1-192 cells, a set of primary murine leukemic cells within which > 95% cells react to autoantigen stimulation via the B-cell receptor (BCR). Comparing with healthy BM, we found that primary CLL patient B cells express significantly higher levels of the CD27 - CD70 receptor-ligand pair at both mRNA and protein levels, as do TCL-1 mice. We next investigated the effects of an anti-mouse CD70 effector function silenced/blocking mAb, FcD-FR70, on TCL1-192 CLL cells in vitro and in vivo in preclinical models. In vitro, treating TCL1 CLL cells with FcD-FR70 significantly inhibited CD40L-induced CLL cell growth. In vivo, twice weekly 10 mg/kg treatments with FcD-FR70 antibody significantly reduced the number of murine CLL B cells in spleens, lymph nodes and blood. The levels of reduction were comparable to those observed in anti-CD19 mAb or ibrutinib treated mice. There was a significantly better treatment outcome when therapy was initiated earlier. Moreover, the level of soluble mouse CD27, the product of CD27 cleavage after interaction with CD70, was significantly reduced in sera of animals treated with FcD-FR70. Notably, FcD-FR70, but not anti-CD19 mAb or ibrutinib, significantly reduced CD70 membrane levels. CD27 expression levels were not altered by all treatments. As CD27 and CD70 are inversely expressed in B cells, due to tight control at the transcriptional and posttranslational levels, this result strongly suggests preferential depletion of malignant cells with the high CD70 expression, which also affects levels of soluble CD27. We then performed bulk RNAseq analysis in CD19+CD5+ splenic cells from mice treated with control or FcD-FR70 antibody. Genes differing by >1.5 fold (P < 0.01) by FcD-FR70 treatment between the two groups included upregulated cell cycle regulator cyren and interferon gamma inducible protein IfI211, as well as downregulated IL10, CCR7, PIM3, PIM1, TGFb3, TNFaip3, JunB and CD69. Importantly, IPA analysis of the altered genes suggested inhibition of TLR7, TLR9, TNF, NFKB, IL4, BCR, BTK, and CD40L signaling pathways. Thus, FcD-FR70 treatment blocked key survival pathways, inducing death of CLL B cells. Finally, we studied the effect of anti-CD70 antibodies on CLL B-T cell interactions. Adding the FcD-FR70 mAb to an allo-MLR significantly blocked T cell proliferation and survival induced by TCL1-192 leukemic B cells (P < 0.01). We then tested the effects of 41D12-FcD, a blocking mAb in human IgG1 format that binds with picomolar affinity to human CD70, in a primary CLL patient derived xenograft model. This revealed a significant reduction of malignant cells in the spleen, BM, and peritoneum (P < 0.001). Interestingly, the efficacy of the blocking 41D12-FcD antibody was associated with a significant decrease of human T cells, suggesting that the antibody directly blocked the proliferation of malignant B cells and also inhibited T cell-CLL cell interactions. Notably, no significant changes were seen in the differentiation of T cells, expression of CTLA4, PD1, PD-L1, and levels of Th1/Th2 cytokines after anti-CD70 treatment. Overall, data from our preclinical models suggest targeting CD70 in CLL alters BCR, CD40L, IL4, and TNF receptor signaling and delays CLL disease progression. Although the involvement of CD70 and the relationships between CLL B cells and non-malignant cells needs further study, the availability of clinical stage therapeutic anti-CD70 antibodies supports the use anti-CD70 immunotherapy in combination therapies in CLL.