Anti-CD20 monoclonal antibodies (mAbs) are an important component of treatment regimens for B-cell malignancies. Following administration of anti-CD20 mAbs CD20 expression is down-modulated by trogocytosis and internalization. While cells with low or absent CD20 expression can evade further anti-CD20 mAb binding, they are labelled by complement proteins. In particular, complement component C3d is covalently bound to the cell surface. We hypothesized that C3d constitutes a neoantigen that could be targeted by anti-C3d mAbs to enhance anti-CD20 therapy. We derived several hybridomas from mice immunized with human C3d and found to bind three distinct epitopes on C3d. One epitope was shared by all high affinity mAbs, while two other epitopes were bound by low affinity mAbs. From a high affinity murine mAb we derived a human IgG1 chimeric antibody that is highly selective for C3d, does not bind full length C3, and does not cross react with mouse C3d. The chimeric anti-C3d mAb bound immobilized C3d with a Kd of 17nM and cell-bound C3d with a kD of 3.0nM, similar to the kD of 2.4nM of ofatumumab binding to CD20. The anti-C3d antibody mediated complement-dependent cytotoxicity, NK cellular cytotoxicity, and phagocytosis of C3d opsonized cells. We tested activity of the anti-C3d mAb against primary tumor cells from patients with chronic lymphocytic leukemia (CLL) being treated with the anti-CD20 mAb ofatumumab. The first administration of ofatumumab decreased the tumor cell count in the blood of patients by 50% on average. As expected, CLL cells obtained 24 hours after administration of ofatumumab had lost CD20 antigen and carried abundant C3d on their cell surface. Importantly, the anti-C3d chimeric antibody specifically bound CLL cells but no other blood cells, indicating that targeting of C3d preserves the specificity of the initial complement fixing antibody. These studies provide proof of concept that targeting cell deposited C3d can enhance the potency of mAb therapy. Given that our anti-C3d mAb preserves the specificity of the initial mAb, it could augment the potency of many mAbs currently in clinical use by delivering a one-two punch. We conclude that targeting C3d deposited on cancer cells can eliminate antigen escape variants and potentiate existing therapeutic antibodies. The Brutons tyrosine kinase inhibitor ibrutinib induces high rates of clinical response in patients with CLL. However, there remains a need for adjunct treatments to deepen response and to overcome drug resistance. Blinatumomab, a CD19/CD3 bispecific antibody (bsAb) designed in the BiTE format, is FDA approved for the treatment of relapsed or refractory B-cell precursor acute lymphoblastic leukemia. Due to its short half-life of 2.1 hours, blinatumomab requires continuous intravenous dosing for efficacy. We developed a novel CD19/CD3 bsAb in the single-chain Fv-Fc format (CD19/CD3-scFv-Fc) with a half-life of approximately 5 days. In in vitro experiments, both CD19/CD3-scFv-Fc and blinatumomab induced >90% killing of CLL cells from treatment-nave patients. Anti-leukemic activity was associated with increased autologous CD8 and CD4 T cell proliferation, activation, and granzyme B expression. In the NOD/SCID/IL2Rnull patient-derived xenograft mouse model, once-weekly treatment with CD19/CD3-scFv-Fc eliminated >98% of treatment-nave CLL cells in blood and spleen. By contrast, blinatumomab failed to induce a response, even when administered daily. We next explored the activity of CD19/CD3-scFv-Fc in the context of ibrutinib treatment and ibrutinib resistance. CD19/CD3-scFv-Fc induced more rapid killing of CLL cells from ibrutinib-treated patients than those from treatment-nave patients. CD19/CD3-scFv-Fc also demonstrated potent activity against CLL cells from patients with acquired ibrutinib-resistance harboring BTK and/or PLCG2 mutations in vitro and in vivo using patient-derived xenograft models. Taken together, these data support investigation of CD19/CD3 bsAbs and other T cell-recruiting bsAbs as immunotherapies for CLL, especially in combination with ibrutinib or as rescue therapy in ibrutinib-resistant disease. Despite promising early studies, the development of lenalidomide in CLL has been derailed by significant toxicities, also experienced in our clinical trial. However, lenalidomide can induce sustained responses that may be long-lasting even after discontinuation. The latter studies promoted us to complete investigations into the nature of immune responses in patients treated with lenalidomide on our investigator-initiated study. Prior studies have examined the effect of lenalidomide on circulating T cell subsets in CLL patients. However, in vivo analysis of the tumor-microenvironment, where the effects of lenalidomide arguably matter most, are needed. Among 25 patients who completed 4 cycles of therapy, 9 (36%) achieved 10% reduction in lymphadenopathy by imaging and were considered responders for correlative analyses. We performed gene expression profiling of paired LN samples obtained pre-treatment and on day 8. On lenalidomide, T cell numbers in the tumor-microenvironment increased and showed Th1-type polarization with upregulation of IFN and granzyme. Deep sequencing of TCR genes revealed expansion of select clonotypes in responders. In conclusion, lenalidomide promoted a Th1-type immune response that correlated with clinical response, suggesting that an intrinsic anti-tumor immune response may be present in many patients, waiting to be activated. The TCR repertoire in CLL patients is reported to be more clonal than in healthy individuals.97 It is unknown whether expanded TCR clonotypes reflect expansion of T-cell clones stimulated by tumor, microbial, or auto-antigens. To investigate the effect of ibrutinib on distinct T cell clonotypes, we performed deep sequencing of the TCR repertoire in 26 patients with CLL prior to treatment with ibrutinib, at the time of response, and either at disease progression or during sustained remission. TCR clonality was calculated as the inverse of Shannon entropy normalized to the number of productive clones in a sample. The median productive clonality in CLL patients before treatment was more than double the clonality reported for healthy donors and was higher in patients with relapsed/refractory CLL compared to treatment-nave patients. Notably, CD8+ but not CD4+ T-cell counts were significantly correlated with clonality. Clonality increased at the time of response compared to baseline and remained high in patients with sustained anti-tumor response. In 12 patients who subsequently progressed, clonality decreased from the time of response to relapse. Expanding T cell clonotypes were patient-specific and correlated with CD8+ T cell numbers. Based on these preliminary data, we currently are pursuing the hypothesis that distinct CD8+ T-cell clonotypes recognize tumor antigens in a patient-specific manner and may contribute to anti-tumor activity during treatment with ibrutinib.
Showing the most recent 10 out of 19 publications
