The Programmed Death-1 (PD-1) receptor delivers inhibitory checkpoint signals to activated T cells upon binding to its ligands PD-L1 and PD-L2 expressed on antigen-presenting cells and cancer cells, resulting in suppression of T-cell effector function and tumor immune evasion. Clinical antibodies blocking the interaction between PD-1 and PD-L1 restore the cytotoxic function of tumor antigen-specific T cells, yielding durable objective responses in multiple cancers. This report describes the preclinical characterization of REGN2810, a fully human hinge-stabilized IgG4 (S228P) high-affinity anti-PD-1 antibody that potently blocks PD-1 interactions with PD-L1 and PD-L2. REGN2810 was characterized in a series of binding, blocking, and functional cell-based assays, and preclinical in vivo studies in mice and monkeys. In cellbased assays, REGN2810 reverses PD-1-dependent attenuation of T-cell receptor signaling in engineered T cells and enhances responses of human primary T cells. To test the in vivo activity of REGN2810, which does not cross-react with murine PD-1, knockin mice were generated to express a hybrid protein containing the extracellular domain of human PD-1, and transmembrane and intracellular domains of mouse PD-1. In these mice, REGN2810 binds the humanized PD-1 receptor and inhibits growth of MC38 murine tumors. As REGN2810 binds to cynomolgus monkey PD-1 with high affinity, pharmacokinetic and toxicologic assessment of REGN2810 was performed in cynomolgus monkeys. High doses of REGN2810 were well tolerated, without adverse immune-related effects. These preclinical studies validate REGN2810 as a potent and promising candidate for cancer immunotherapy.

Oncogenic fusion proteins, such as EWS-FLI1, are excellent therapeutic targets as they are only located within the tumor. However, there are currently no agents targeted toward transcription factors, which are often considered to be ‘undruggable.’ A considerable body of evidence is accruing that refutes this claim based upon the intrinsic disorder of transcription factors. Our previous studies show that RNA Helicase A (RHA) enhances the oncogenesis of EWS-FLI1, a putative intrinsically disordered protein. Interruption of this protein-protein complex by small molecule inhibitors validates this interaction as a unique therapeutic target. Single enantiomer activity from a chiral compound has been recognized as strong evidence for specificity in a small molecule-protein interaction. Our compound, YK-4-279, has a chiral center and can be separated into two enantiomers by chiral HPLC. We show that there is a significant difference in activity between the two enantiomers. (S)-YK-4-279 is able to disrupt binding between EWS-FLI1 and RHA in an immunoprecipitation assay and blocks the transcriptional activity of EWS-FLI1, while (R)-YK-4-279 cannot. Enantiospecific effects are also established in cytotoxicity assays and caspase assays, where up to a log-fold difference is seen between (S)-YK-4-279 and the racemic YK-4-279. Our findings indicate that only one enantiomer of our small molecule is able to specifically target a protein-protein interaction. This work is significant for its identification of a single enantiomer effect upon a protein interaction suggesting that small molecule targeting of intrinsically disordered proteins can be specific. Furthermore, proving YK-4-279 has only one functional enantiomer will be helpful in moving this compound towards clinical trials.

Ewing sarcoma (ES) involves a tumor-specific chromosomal translocation that produces the EWS-FLI1 protein, which is required for the growth of ES cells both in vitro and in vivo. However, an EWS-FLI1-driven transgenic mouse model is not currently available. Here, we present data from six independent laboratories seeking an alternative approach to express EWS-FLI1 in different murine tissues. We used the Runx2, Col1a2.3, Col1a3.6, Prx1, CAG, Nse, NEFL, Dermo1, P0, Sox9 and Osterix promoters to target EWS-FLI1 or Cre expression. Additional approaches included the induction of an endogenous chromosomal translocation, in utero knock-in, and the injection of Cre-expressing adenovirus to induce EWS-FLI1 expression locally in multiple lineages. Most models resulted in embryonic lethality or developmental defects. EWS-FLI1-induced apoptosis, promoter leakiness, the lack of potential cofactors, and the difficulty of expressing EWS-FLI1 in specific sites were considered the primary reasons for the failed attempts to create a transgenic mouse model of ES.

Neuroblastoma (NB) is a pediatric tumor of neural crest origin with heterogeneous phenotypes. While low stage tumors carry a favorable prognosis, over 50% of high risk NB relapses after treatment with a fatal outcome. Thus, developing therapies targeting refractory NB remains an unsolved clinical problem. Brain-derived neurotrophic factor (BDNF) and its TrkB receptor are known to protect NB cells from chemotherapy-induced cell death, while neuropeptide Y (NPY), acting via its Y2 receptor (Y2R), is an autocrine proliferative and angiogenic factor crucial for maintaining NB tumor growth. Here, we show that in NB cells, BDNF stimulates the synthesis of NPY and induces expression of another one of its receptors, Y5R. In human NB tissues, the expression of NPY and Y5R positively correlated with the expression of BDNF and TrkB. Functionally, BDNF triggered Y5R internalization in NB cells, while Y5R antagonist inhibited BDNF-induced p44/42-MAPK activation and its pro-survival activity. These observations suggested TrkB-Y5R transactivation that resulted in cross-talk between their signaling pathways. Additionally, NPY and Y5R were up-regulated in a BDNF-independent manner in NB cells under pro-apoptotic conditions, such as serum deprivation and chemotherapy, as well as in cell lines and tissues derived from post-treatment NB tumors. Blocking Y5R in chemoresistant NB cells rich in this receptor sensitized them to chemotherapy-induced apoptosis and inhibited their growth in vivo by augmenting cell death. In summary, the NPY/Y5R axis is an inducible survival pathway activated in NB by BDNF or cellular stress. Upon such activation, Y5R augments the pro-survival effect of BDNF via its interactions with TrkB receptor and exerts an additional BDNF-independent anti-apoptotic effect, both of which contribute to NB chemoresistance. Therefore, the NPY/Y5R pathway may become a novel therapeutic target for patients with refractory NB, thus far an incurable form of this disease.

In the tumor microenvironment, multiple inhibitory checkpoint receptors can suppress T-cell function, thereby enabling tumor immune evasion. Blockade of one of these checkpoint receptors, PD-1, with therapeutic antibodies has produced positive clinical responses in various cancers; however, the efficacy of this approach can be further improved. Simultaneously targeting multiple inhibitory checkpoint receptors has emerged as a promising therapeutic strategy. Here, we report the development and characterization of REGN3767, a fully human IgG4 antibody targeting LAG-3, another inhibitory receptor on T cells. REGN3767 binds human and monkey LAG-3 with high affinity and specificity and blocks the interaction of LAG-3 with its ligand, MHC class II. In an engineered T-cell/antigenpresenting cell bioassay, REGN3767 alone, or in combina-tion with cemiplimab (REGN2810, human anti-PD-1 antibody), blocked inhibitory signaling to T cells mediated by hLAG-3/MHCII in the presence of PD-1/PD-L1. To test the in vivo activity of REGN3767 alone or in combination with cemiplimab, we generated human PD-1xLAG-3 knockin mice, in which the extracellular domains of mouse Pdcd1 and Lag3 were replaced with their human counterparts. In these humanized mice, treatment with cemiplimab and REGN3767 showed increased efficacy in a mouse tumor model and enhanced the secretion of proinflammatory cytokines by tumor-specific T cells. The favorable pharmacokinetics and toxicology of REGN3767 in nonhuman primates, together with enhancement of antitumor efficacy of anti-PD-1 antibody in preclinical tumor models, support its clinical development.