Exploiting FOPX3 driven tumor microenvironment adaptation and antigen spreading to enhance adoptive T-cell therapy

Resumen

To favor adaptation of transferred T cells to the hostile TME, we decided to fight fire with fire and exploit the FOXP3-driven adaptation of CD4 Treg to TME. Our data show that FOXP3 overexpressing CD8 T cells proliferated and markedly accumulated in the TME while showing enhanced cytotoxicity and efficacy in ACT (Fig. 26). Interestingly, tumor-infiltrating Foxp3UP cells exhibited a T-cell effector and leukocyte migration genetic signature, as well as positive enrichment in glycolysis, FA metabolism and OXPHOS pathways. Flow-cytometry analyses confirmed that Foxp3UP cells in the TME exhibited a higher expression of GLUT1, enhanced glucose and FA uptake as well as intracellular lipid accumulation. Interestingly, in vitro, Foxp3UP cells compensated for the loss of mitochondrial respiration-driven ATP production by activating aerobic glycolysis. Moreover, in limiting nutrient conditions these cells were able to engage FA oxidation to drive OXPHOS for their energetic demands. Further, their ability to couple glycolysis and OXPHOS gave them a metabolic advantage to sustain their proliferation when glucose was restricted. These cells also showed better survival under low-glucose, high-lactate conditions and while they were more sensitive to IL-2 withdrawal, they were less prone to activation-induced cell death. Importantly, disrupting Foxp3 gene in activated CD8 T cells impaired their tumor recruitment and antitumor efficacy. Therefore, we describe a hitherto unknown role of Foxp3 in the adaptation of CD8 T cells to TME that can be used to enhance their efficacy in ACT. On the other hand, target Ag loss has emerged as a major cause of relapse after CAR T-cell therapy. To counteract this tumor escape mechanism, we combined CAR T cells with an immunostimulatory agent, such as the STING-L 2’3’cGAMP. Using two immune-competent solid tumor models we showed that CAR T-cell treatment led to the emergence of tumor cells that lose the targeted Ag, recreating the cancer immunoediting effect of CAR T-cell therapy. The combination of CAR T cells with the intratumoral delivery of 2’3’cGAMP showed a synergistic effect, being able to restrain the growth of STING-L-treated and non-treated tumors. Interestingly, a secondary immune response against non-CAR-targeted Ags, as determined by MHC-I-tetramer staining, was fostered and the intensity of this Ag spreading effect correlated with the efficacy of the combination. This was consistent with the oligoclonal expansion of host T cells, as revealed by TCR repertoire in-depth analysis. Moreover, only in the combination group did the activation of endogenous T cells translate into a systemic response. Importantly, the Ag spreading and the antitumor effects of the combination therapy were fully dependent on host STING signaling and Batf3-dependent dendritic cells, and partially dependent on perforin release by CAR T cells. This indicated that the CAR T-cell cytotoxic activity together with the immunostimulatory action of STING-L were necessary for the efficacy of the combination (Fig. 27). Interestingly, the effectiveness of CAR T-cell/STING-L therapy also depended on STING signaling in CAR T cells. Our data show that 2'3'cGAMP is a suitable adjuvant to combine with CAR T-cell therapy and evoke an endogenous T-cell response that prevents the outgrowth of tumor variants lacking the target Ag. In summary, our results show that using the same methods as Tregs to adapt to TME and recruiting endogenous T cells to fight cancer cells, the effectiveness of ACT can be improved. I hope these observations may contribute to the global understanding and advance of ACT and immunology.