Synergism between CAR-T Cells and a Personalized Tumor Vaccine in Hematological Malignances

Background: CAR-T cells are a tremendous breakthrough in the treatment of certain type of blood cancer, demonstrating impressive and durable responses. However, mechanisms of resistance and relapse have been reported. Mechanisms contributing to relapse after CAR-T therapy include the downregulation of the CAR-T target antigen and the rapid extinguishing of CAR-T cells. We have developed a personalized cancer vaccine whereby patient derived tumor cells are fused to autologous dendritic cells (DC). DC/tumor vaccine induces a broad anti-tumor immunity capable of preventing relapse but may not be effective in patients with advanced disease.

Aims: We sought to overcome relapse and resistance to CAR-T therapy to improve the current response rate to CAR-T cells. To this end, we combined CAR-T cells treatment with our personalized DC/tumor vaccine. We postulated that the DC/tumor vaccine would demonstrate synergy with CAR-T cells in a setting where CAR-T cells reduce the bulky disease and the fusion vaccine prevents relapse by expanding CAR-T cells and tumor antigen specific lymphocytes.

Methods: To investigate the effects of the CAR-T/fusion vaccine combination, we used the A20 murine B-cell lymphoma model. Syngeneic T cells were obtained from BALB/c mice and retrovirally transduced with a second-generation CAR construct composed of an antigen binding domain that recognizes murine CD19, the CD3ζ domain, 4-1BB as costimulatory molecule and GFP (m19BBz-GFP). GFP expression was used to assess gene-transfer efficiency and monitor the CAR-transduced T cells. The DC/tumor vaccine was obtained by PEG-mediated fusions of A20 cells and BALB/c DC.

In the in vitro experiments, T cells transduced with the m19BBz-GFP CAR or non-transduced T cells were co-cultured in the presence or the absence of the DC/A20 vaccine for 3 days. Tumor killing was measured by quantifying Firefly luciferase activity (WT A20) or Renilla luciferase activity (CD19- A20). Vaccine was removed before starting the killing assay. In the in vivo experiment, B-cell lymphoma was induced in BALB/c mice by tail vein injection of A20 cells. The mice were lymphodepleted and treated with m19BBz-GFP CAR-T. The mice were then subcutaneous injected with the DC/A20 fusion vaccine or with PBS (control group). CD8+ T-cells specific for the A20 idiotype epitope were quantified by MHC Class I tetramer analysis.

Results: In vitro co-culture of CAR-T cells and DC/A20 vaccine induced a memory-like CAR-T phenotype and strongly improved CAR-T persistence (measured as % of GFP+ T cells in culture). These results were confirmed in vivo where increased CAR-T percentage was observed in the bone marrow and spleen of vaccinated mice with respect to the unvaccinated control group. Moreover, in the vaccinated mice we detected CD8+ T-cells specific for the A20 idiotype epitope demonstrating the expansion of tumor-specific lymphocytes in response to the fusion vaccine.

To assess whether the vaccine-induced persistence of CAR-T cells was translated in a higher tumor killing capacity, we performed an in vitro killing assay. To mimic the presence of CD19- clones in the tumor bulk, we used a mixture of WT A20 and CD19-A20 as CAR-T target cells. Enhanced killing capacity against CD19+ tumors was induced by education of the CAR-T with the vaccine. No effects on the CAR-T killing capacity against CD19- A20 were elicited by the fusion vaccine. However, when the same killing assay was performed using as effector cells a mixture of CAR-T and naïve T cells or a mixture of CAR-T and vaccine-educated T cells, we observed that the addition of vaccine-educated T cell to the CAR-T, strongly reduced both A20 WT and A20 CD19- in the cultures. Thus, vaccine-educated T cells are able to kill tumor cells independently from the presence or the absence of the CAR-T target antigen on the tumor.

Conclusions: The combination of CAR-T and DC/tumor vaccine, increasing the persistence of CAR-T cells and evoking a polyclonal T cell response against tumor antigens in the non-CAR-transduced T cells, may represent a novel therapeutic strategy to overcame therapeutic resistance and improve current response rate to CAR-T therapy.