Generation of novel human anti-OX-40 mAbs endowed with different biological properties as tools for cancer therapy

The second generation of Antibody-based Immunotherapy includes monoclonal antibodies against Immune Checkpoints (ICs), to modulate specific T cell responses against cancer or viruses. We recently generated a large repertoire of fully human antibodies targeting ten different ICs through a novel selection strategy based on the combination of phage libraries on human lymphocytes and next generation sequencing (NGS). Here we generated and tested four novel fully human IgG1 mAbs specific for OX-40, an immunostimulatory receptor expressed on immune cells, which has been shown to be a promising target for immune-based therapeutic strategies.

By ELISA and Biolayer Interferometry we demonstrated that they all specifically bind with high affinity to OX-40 and they recognize distinct epitopes. Three of them interfere with the binding of OX-40 and its ligand, thus suggesting that they compete with it for the receptor binding. T cell activation assays confirmed the agonistic properties of these 3 antibodies which are able to mimic the ligand by activating the pathway downstream the receptor. This activation results into an effective proliferation of hPBMCs and secretion of proinflammatory cytokines. Co-culture assays of hPBMCs with tumor cells confirm their ability to induce the activation of immune cells against cancer cells. The fourth antibody, even though non-agonistic, was able to induce the activation of lymphocytes by a different mechanism of action, based on NK-mediated Treg killing in co-culture assays.

Combinations of these anti-OX40 mAbs targeting different epitopes lead to stronger activation of immune cells. Moreover, epitope binning analyses show that they recognize distinct epitopes not overlapping with that of the clinically validated Rocatinlimab, thus they could become potential new therapeutic tools. Taking advantage of the different behaviour of the novel mAbs, we also exploited them to clarify the unclear role of OX-40 on NK cells. We show here for the first time that NK cells express higher levels of a medium glycosylated OX-40 form than T cells, which is preferentially recognized by the novel mAbs but not by OX-40L, which instead binds to a highly glycosylated OX-40 variant absent on non-immune cells. Thus, glycosylation pattern could affect the recognition and biological effects of OX-40 binders and should be considered for the design of novel drugs.