A novel lentiviral vector-based approach to generate chimeric antigen receptor T cells targeting Aspergillus fumigatus

Invasive aspergillosis (IA) is a common and deadly mold infection in immunocompromised patients. As morbidity and mortality of IA are primarily driven by poor immune defense, adjunct immunotherapies, such as chimeric antigen receptor (CAR) T cells, are direly needed. Here, we propose a novel approach to generate Aspergillus fumigatus (AF)-CAR T cells using the single-chain variable fragment domain of monoclonal antibody AF-269-5 and a lentiviral vector system. These cells successfully targeted mature hyphal filaments of representative clinical and reference AF isolates and elicited a potent release of cytotoxic effectors and type 1 T cell cytokines. Furthermore, AF-CAR T cells generated from peripheral blood mononuclear cells of four healthy human donors and expanded with either of three cytokine stimulation regimens (IL-2, IL-2 + IL-21, or IL-7 + IL-15) significantly suppressed mycelial growth of AF-293 after 18 hours of co-culture and synergized with the immunomodulatory antifungal agent caspofungin to control hyphal growth for 36 hours. Moreover, cyclophosphamide-immunosuppressed NSG mice with invasive pulmonary aspergillosis that received two doses of 5 million AF-CAR T cells (6 and 48 hours after AF infection) showed significantly reduced morbidity on day 4 post-infection (P < 0.001) and significantly improved 7-day survival (P = 0.049) compared with mice receiving non-targeting control T cells, even without concomitant antifungal chemotherapy. In conclusion, we developed a novel lentiviral strategy to obtain AF-CAR T cells with high targeting efficacy, yielding significant anti-AF activity in vitro and short-term protection in vivo. Our approach could serve as an important steppingstone for future clinical translation of antifungal CAR T-cell therapy after further refinement and thorough preclinical evaluation.IMPORTANCEInvasive aspergillosis (IA) remains a formidable cause of morbidity and mortality in patients with hematologic malignancies and those undergoing hematopoietic stem cell transplantation. Despite the introduction of several new Aspergillus-active antifungals over the last 30 years, the persisting high mortality of IA in the setting of continuous and profound immunosuppression is a painful reminder of the major unmet need of effective antifungal immune enhancement therapies. The success of chimeric antigen receptor (CAR) T-cell therapy in cancer medicine has inspired researchers to translate this approach to opportunistic infections, including IA. Aiming to refine anti-Aspergillus CAR T-cell therapy and improve its feasibility for future clinical translation, we herein developed and validated a novel antibody-based CAR construct and lentiviral transduction method to accelerate the production of CAR T cells with high targeting efficacy against Aspergillus fumigatus. Our unique approach could provide a promising platform for future clinical translation of CAR T-cell-based antifungal immunotherapy.

B Cells Are Required to Generate Optimal Anti-Melanoma Immunity in Response to Checkpoint Blockade

Immunotherapies such as checkpoint blockade therapies are known to enhance anti-melanoma CD8⁺ T cell immunity, but only a fraction of patients treated with these therapies achieve durable immune response and disease control. It may be that CD8⁺ T cells need help from other immune cells to generate effective and long-lasting anti-tumor immunity or that CD8⁺ T cells alone are insufficient for complete tumor regression and cure. Melanoma contains significant numbers of B cells; however, the role of B cells in anti-melanoma immunity is controversial. In this study, B16 melanoma mouse models were used to determine the role of B cells in anti-melanoma immunity. C57BL/6 mice, B cell knockout (KO) C57BL/6 mice, anti-CD19, and anti-CXCL13 antibody-treated C57BL/6 mice were used to determine treatment efficacy and generation of tumor-specific CD8⁺ T cells in response to PD-L1 blockade alone or combination with TLR-7/8 activation. Whole transcriptome analysis was performed on the tumors from B cell depleted and WT mice, untreated or treated with anti-PD-L1. Both CD40-positive and CD40-negative B cells were isolated from tumors of TLR-7/8 agonist-treated wild-type mice and adoptively transferred into tumor-bearing B cell KO mice, which were treated with anti-PD-L1 and TLR-7/8 agonist. Therapeutic efficacy was determined in the presence of activated or inactivated B cells. Microarray analysis was performed on TLR-7/8-treated tumors to look for the B cell signatures. We found B cells were required to enhance the therapeutic efficacy of monotherapy with anti-PD-L1 antibody and combination therapy with anti-PD-L1 antibody plus TLR-7/8 agonist. However, B cells were not essential for anti-CTLA-4 antibody activity. Interestingly, CD40-positive but not CD40-negative B cells contributed to anti-melanoma immunity. In addition, melanoma patients' TCGA data showed that the presence of B cell chemokine CXCL13 and B cells together with CD8⁺ T cells in tumors were strongly associated with improved overall survival. Our transcriptome data suggest that the absence of B cells enhances immune checkpoints expression in the tumors microenvironment. These results revealed the importance of B cells in the generation of effective anti-melanoma immunity in response to PD-1-PD-L1 blockade immunotherapy. Our findings may facilitate the design of more effective anti-melanoma immunotherapy.

IL-1α Mediates Innate and Acquired Resistance to Immunotherapy in Melanoma

Inflammation has long been associated with cancer initiation and progression; however, how inflammation causes immune suppression in the tumor microenvironment and resistance to immunotherapy is not well understood. In this study, we show that both innate proinflammatory cytokine IL-1α and immunotherapy-induced IL-1α make melanoma resistant to immunotherapy. In a mouse melanoma model, we found that tumor size was inversely correlated with response to immunotherapy. Large tumors had higher levels of IL-1α, Th2 cytokines, polymorphonuclear myeloid-derived suppressor cells (PMN-MDSCs), and regulatory T cells but lower levels of IL-12, Th1 cytokines, and activated T cells. We found that therapy with adenovirus-encoded CD40L (rAd.CD40L) increased tumor levels of IL-1α and PMN-MDSCs. Blocking the IL-1 signaling pathway significantly decreased rAd.CD40L-induced PMN-MDSCs and their associated PD-L1 expression in the tumor microenvironment and enhanced tumor-specific immunity. Similarly, blocking the IL-1 signaling pathway improved the antimelanoma activity of anti-PD-L1 Ab therapy. Our study suggests that blocking the IL-1α signaling pathway may increase the efficacy of immunotherapies against melanoma.

Abstract 941: B cells are required to generate optimal antitumor immunity in response to PD-1 blockade treatment

Background: Melanoma is the most deadly form of skin cancer. Although, the immunogenic nature of melanoma makes this disease susceptible to immunotherapy, metastatic melanoma remains highly resistant to established immunotherapies like cancer vaccines. Anti-PD-1 antibodies alone showed great clinical activity but were still ineffective in many melanoma patients. These reports highlight the need to understand the molecular mechanism of immunotherapy induced anti-melanoma immune responses to develop new approaches for effective therapy against melanoma. B cells augment CD8+ T cell-mediated immunity during autoimmunity and allograft rejection, but the role of B cells in anti-tumor CD8+ T cell immunity and in cancer patient survival is controversial. Recently, we reported that intratumoral stimulation of TLR-7/8 in a mouse model of melanoma induced tumor regression and activation of B cells. B cells were required for tumor suppression in an antibody-independent manner. Also, we found that melanoma patients who responded to anti-PD-1 blockade therapy had significantly more B cells and CD8+ T cells in tumors than non-responders. Based on these evidences, we hypothesized that activated B cells induce and/or enhance anti-melanoma CD8+ T cell immunity in response to immunotherapy.

Methods: B16 melanoma mouse model was used. Wild type and B cell Knockout mice were used to determine treatment efficacy and generation of tumor-specific CD8+T cells in response to PD-L1 blockade and/or TLR-7/8 activation. TLR-7/8 agonist activated B cells were isolated and adoptively transferred into tumor-bearing B cell KO mice, which were treated with anti-PD-L1.Therapeutic efficacy and melanoma-specific CD8+ T cell immunity were determined.

Results: We found that absence of B cells not only decreases the efficacy of TLR-7/8 agonist or anti-PDL-1 abs monotherapy but also TLR-7/8 agonist and anti-PD-L1 abs combination therapy. In addition, mice were unable to develop potent anti- melanoma CD8+T cell immunity in the absence of B cells in response to either mono or combination therapy. Interestingly, we reported prolonged survival of tumor bearing B cell KO mice which received TLR-7/8 agonist activated B cells compared to inactivated B cells recipients in response to anti-PD-L1 antibody therapy.

Conclusions: These results suggest that activated B cells enhance anti-melanoma CD8+ T cell immunity in response to immunotherapy. We expect our findings to lead the field closer to development of biomarkers based on B cell activation that predict response to immunotherapy. Our findings may also facilitate design of more effective anti-melanoma immunotherapy.

Citation Format: Shubhra Singh, Karishma Bavisi, Jason Roszik, Adi Diab, Richard E. Davis, Patrick Hwu, Willem W. Overwijk, Manisha Singh. B cells are required to generate optimal antitumor immunity in response to PD-1 blockade treatment [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 941.