No abstract

Acute lymphoblastic leukemia (ALL) is the most common malignant disease affecting children. Although therapy strategies improved, T-cell acute lymphoblastic leukemia (T-ALL) relapse is associated with chemoresistance and a poor prognosis. One strategy to overcome this obstacle is the application of monoclonal antibodies. Here, we show that leukemic cells from T-ALL patients express surface CD38 and CD47, both attractive targets for antibody therapy. We therefore investigated the commercially available CD38 antibody daratumumab (Dara) in combination with a proprietary modified CD47 antibody (Hu5F9-IgG2σ) in vitro and in vivo. Compared to single treatments, this combination significantly increased in vitro antibody-dependent cellular phagocytosis (ADCP) in T-ALL cell lines as well as in random de novo and in relapsed/refractory T-ALL patient derived xenograft (PDX) samples. Similarly, enhanced ADCP was observed when combining Dara with pharmacological inhibition of CD47 interactions using a glutaminyl cyclase inhibitor. Phase II-like preclinical in vivo trials using T-ALL PDX samples in experimental minimal residual disease like (MRD-like) and overt leukemia models revealed a high anti-leukemic efficacy of CD47 blockade alone. However, T-ALL xenograft mice subjected to chemotherapy first (post-chemo MRD) and subsequently co-treated with Dara and Hu5F9-IgG2σ displayed significantly reduced bone marrow infiltration as compared to single treatments. In relapsed and highly refractory T-ALL PDX combined treatment with Dara and Hu5F9-IgG2σ was required to substantially prolong survival as compared to single treatments. These findings suggest that combining CD47 blockade with Dara is a promising therapy for T-ALL, especially for relapsed/refractory disease harbouring a dismal prognosis in patients.

Central nervous system (CNS) involvement remains a challenge in the diagnosis and treatment of acute lymphoblastic leukemia (ALL). In this study, we identify CD79a (also known as Igα), a signaling component of the preB cell receptor (preBCR), to be associated with CNS-infiltration and –relapse in B-cell precursor (BCP)-ALL patients. Furthermore, we show that downregulation of CD79a hampers the engraftment of leukemia cells in different murine xenograft models, particularly in the CNS.

Background: Native cluster of differentiation (CD) 19 targeting antibodies are poorly effective in triggering antibody-dependent cell-mediated cytotoxicity (ADCC) and complement-dependent cytotoxicity (CDC), which are crucial effector functions of therapeutic antibodies in cancer immunotherapy. Both functions can be enhanced by engineering the antibody’s Fc region by altering the amino acid sequence (Fc protein-engineering) or the Fc-linked glycan (Fc glyco-engineering). We hypothesized that combining Fc glyco-engineering with Fc protein-engineering will rescue ADCC and CDC in CD19 antibodies. Results: Four versions of a CD19 antibody based on tafasitamab’s V-regions were generated: a native IgG1, an Fc protein-engineered version with amino acid exchanges S267E/H268F/S324T/G236A/I332E (EFTAE modification) to enhance CDC, and afucosylated, Fc glyco-engineered versions of both to promote ADCC. Irrespective of fucosylation, antibodies carrying the EFTAE modification had enhanced C1q binding and were superior in inducing CDC. In contrast, afucosylated versions exerted an enhanced affinity to Fcγ receptor IIIA and had increased ADCC activity. Of note, the double-engineered antibody harboring the EFTAE modification and lacking fucose triggered both CDC and ADCC more efficiently. Conclusions: Fc glyco-engineering and protein-engineering could be combined to enhance ADCC and CDC in CD19 antibodies and may allow the generation of antibodies with higher therapeutic efficacy by promoting two key functions simultaneously.