The most efficient therapy for metastatic melanoma is still Adoptive Cell Transfer (ACT) of autologous tumor infiltrating lymphocytes (TIL), yielding 50% objective response rate. Current global efforts for improving ACT focus on engineering T cell receptors or Chimeric Antigen Receptors (CAR). Recently, significant clinical results with CAR-engineered autologous peripheral blood lymphocytes were reported in hematological malignancies by groups from the US. Here, we suggest an alternative approach for enhancing ACT by improving homing of lymphocytes to melanoma cells. We show that CXCL1 and CXCL8 are secreted from almost all melanoma cases and can therefore be used as homing signals for CXCR1- or CXCR2-positive cells. Indeed, the migration of CXCR1-positive TIL towards melanoma cells in modified Boyden chambers is blocked by neutralizing anti-CXCL1/8 antibodies. Importantly, CXCR1 and CXCR2 are only rarely expressed by TIL, which highlights the potential to significantly improve their properties by genetic engineering. We overexpressed CXCR1 or CXCR2 in T cells using viral vectors that were clinically approved and used in previous US clinical trials. The engineered cells demonstrate superior homing towards recombinant chemokines or melanoma cells. In vivo experiments in SCID-NOD mice bearing human melanoma xenografts are ongoing. The main advantage of this approach over current TCR or CAR engineering is the preservation of the polyclonal tumor recognition properties of TIL. We have treated so far more than 70 patients with TIL ACT, thus success in these experiments may pave the way to accelerated protocol development towards a clinical trial for the benefit of metastatic melanoma patients.
