Introduction. Despite the success of autologous chimeric antigen receptor (CAR)-T cells, barriers to a more widespread use of this potentially curative therapy include manufacturing failures and the high cost of individualized production. There is a strong desire for an immediately available cell therapy option; however, development of “off-the-shelf” T cells is challenging. Alloreactive T cells from unrelated donors can cause graft versus host disease (GvHD) for which researchers have successfully used nucleases to reduce expression of the endogenous T cell receptor (TCR) in the allogeneic product. The recognition of allogeneic cells by the host is a complex issue that has not been fully solved to date. Some approaches utilize prolonged immune suppression to avoid immune rejection and increase persistence. Although showing responses in the clinic, this approach carries the risk of infections and the durability of the adoptive T cells is uncertain. Other strategies include deletion of the B2M gene to remove HLA class I molecules and avoid recognition by host CD8 T cells. However, loss of HLA class I sends a “missing-self” signal to natural killer (NK) cells, which readily eliminate B2Mnull T cells. To overcome this, researchers are exploring insertion of the non-polymorphic HLA-E gene, which can provide partial but not full protection from NK cell-mediated lysis. Because activated T cells upregulate HLA class II, rejection by alloreactive CD4 T cells should also be addressed.

Abstract The implementation of various mouse models is critical to asses safety, efficacy, and short- and long-term persistence of therapeutic modalities, especially for cell-based therapies. To increase our repertoire of viable humanized murine models, we developed two in vivo models by taking advantage of Taconic's immunodeficient mice, one monitoring graft versus host disease (GvHD) and the other addressing human natural killer (NK) cell cytotoxicity. We utilized NOG mice to develop a model of GvHD, by transplanting human PBMCs at varying doses and monitoring mice for changes in body weight over time. We determined a dose of injected PBMCs that allowed for GvHD to occur yet provided a window of opportunity for potential therapies to slow progression. When human natural T regulatory cells (nTregs) were co-injected with PBMCs in NOG mice, there was prolonged survival and a less rapid loss of body weight as compared to PBMCs alone. To create an NK cytotoxicity model, we transplanted human primary NK cells into NOG-hIL15 mice, which are NOG mice that constitutively produce human IL-15. We showed successful engraftment and proliferation of NK cells, with peak engraftment occurring 4-5 weeks post injection, and that these human primary NK cells were able to persist without signs of xenogeneic GvHD. Utilizing K562 tumor cells that express luciferase, we found these engrafted NK cells have fast and potent cytotoxic activity using IVIS imaging, resulting in elimination of tumor cells as compared to non-engrafted mice. Our results collectively suggest that the two in vivo models developed here will be valuable tools for investigating the clinical benefit of immune cell-based therapeutics. Citation Format: Elizabeth McMichael, Utsav Jetley, Christopher Rudulier, Minasri Borah, Nicole Ganci, Vandhana Ragothaman, Ishina Balwani, Amanda Frain, Priya Pajanirassa, Yuko Miki, Jeffrey Jones, Troy Luster, Marie Keenan, Terina Martinez, Yong Zhang, Birgit Schultes. In vivo model development for immune cell-based therapeutics [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 5574.

Clostridium difficile infection (CDI) is the principal cause of nosocomial diarrhea and pseudomembranous colitis associated with antibiotic therapy. The pathological effects of CDI are primarily attributed to toxins A (TcdA) and B (TcdB). Adequate toxin-specific antibody responses are associated with asymptomatic carriage, whereas insufficient humoral responses are associated with recurrent CDI. While the data supporting the importance of anti-toxin antibodies are substantial, clarity about the toxin domain specificity of these antibodies is more limited. To investigate this matter, combinations of human mAbs targeting multiple domains of TcdB were assessed using toxin neutralization assays. These data revealed that a combination of mAbs specific to all major toxin domains had improved neutralizing potency when compared to equivalent concentrations of a single mAb or a combination of mAbs against one or two domains. The function and toxin domain binding specificity of serum antibodies elicited by immunization of hamsters with a toxoid vaccine candidate was also assessed. Immunization with a toxoid vaccine candidate provoked toxin neutralizing antibodies specific to multiple domains of both TcdA and TcdB. When assessed in a toxin neutralization assay, polyclonal sera displayed greater activity against elevated concentrations of toxins than equivalent concentrations of individual mAbs. These data suggest a potential benefit of any antibody based therapeutic or prophylactic treatment that targets multiple toxin domains.

CD38 targeting antibodies are at different phases of clinical development, with daratumumab already approved as monotherapy and in combination with standards of care in multiple myeloma (MM). Anti-CD38 monoclonal antibodies (mAbs) induce tumor cell depletion in part by Fc-dependent effector mechanisms such as antibody dependent cellular cytotoxicity (ADCC), antibody dependent cellular phagocytosis (ADCP), and complement dependent cytotoxicity (CDC). However, not all MM patients achieve minimal residual disease (MRD)-negativity and similar clinical response. In addition, some patients on daratumumab develop resistance due to reduced cell surface CD38 and high levels of complement inhibitors (CD55 and CD59). We have leveraged Fc multimerization technology (Ortiz et al Sci Transl Med. 2016; 8: 365) to generate an optimized platform (SIF; selective immunomodulator of Fc receptors) that utilizes the valency effect of Fc multimerization to enhance binding to the Fcγ receptors and complement. We combined the Fab-region of CD38 targeting mAb to SIF platform to generate an anti-CD38 SIFbody to enhance immune and complement mediated cytotoxicity against tumor cells. In several human tumor cell line-based cytotoxic assays using primary human effector cells (NK cells and macrophages) and complement, the anti-CD38 SIFbody demonstrates up to 10-fold increase in efficacy and ≥16-fold increase in potency compared to daratumumab and the surrogate therapeutic anti-CD38 mAb (TAK-079). In isolated whole human blood incubated with tumor cells, the anti-CD38 SIFbody demonstrated 40-100 fold increase in potency and 2-3 fold increase in efficacy. In bone marrow cells isolated from MM patients with >80% plasma cells anti-CD38 SIFbody showed better potency and a 3-5 fold increased efficacy (with 100% plasma cell elimination) than daratumumab, suggesting the SIFbody may be more suitable molecule for achieving greater MRD-negativity rates in MM patients. Daratumumab fails to induce CDC against tumor cell lines with low CD38 and high CD55 and CD59, however the SIFbody achieves 100% efficacy in such settings, suggesting this molecule may be effective in patients who are developing resistance to treatment. In single dose pharmacodynamic and tolerability studies in cynomolgus monkeys SIFbody demonstrated up to 5-fold increase in B cell depletion from peripheral blood compared to TAK-079 across all dose ranges (0.3, 1, & 3 mg/kg) tested without any adverse events. Therefore, by leveraging our Fc multimerization technology we have generated a differentiated potential best-in-class anti-CD38 therapeutic. Citation Format: Amit Choudhury, Daniel F. Ortiz, Shannon Argueta, Kevin Garofalo, Jonathan C. Lansing, Utsav Jetley, Danice Wilkins, Carlos Bosques, Edward Cochran, Naveen Bhatnagar, Jay Duffner, Abhinav Gupta, Stan Lee, Karunya Srinivasan, Viraj Parge, Radouane Zouaoui, Jason Wang, Anthony M. Manning. Discovery of a potential best-in-class anti-CD38 therapeutic utilizing Fc multimerization [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 561.

Clostridiumdifficile (C.difficile)is a significant human pathogen. C.difficile infection (CDI) causesclinical symptoms ranging from diarrhea to life-threatening fulminant pseudo membranous colitis. The pathogenesis of C. difficile is mediated by two large exotoxins, toxins A and B. These two toxins are highly homologous, single chain proteins consisting of four functional domains: N-terminal glucosyl transferase domain (GTD), cysteine protease domain (CPD), translocation domain (TLD) and a C-terminal receptor-binding domain (RBD). The important role played by anti-toxin sera and antibodies in the prevention of primary and recurrent CDI has been described and demonstrated in both clinical and pre-clinical studies; however, work focused on the impact of the toxin domain specificity of these antibodies is limited. To address this deficit, sera from a C. difficile vaccine immunized hamsters and toxin-specific human monoclonal antibodies (mAbs) were used to assess the impact of toxin domain specificity on antibody mediated inhibition of cytotoxicity in a Vero cell-based functional assay. Results from toxin domain immunoabsorption assays using hamster anti-toxinsera indicated that no single domain fragment from either toxin A or toxin B could inhibit neutralizing activities. Anti-toxin A activity was prevented with a combination of GTD and CTD fragments while anti-toxin B activity required the GTD, CTD and CPD fragments to block activity. Assays with human mAbs demonstrated that combining mAbs that target multiple toxin domains greatly improves neutralizing potency when compared to equivalent concentrations of either a single mAb or a combination of mAbs against a single domain.