Redirecting T lymphocyte antigen specificity by gene transfer can provide large numbers of tumor-reactive T lymphocytes for adoptive immunotherapy. However, safety concerns associated with viral vector production have limited clinical application of T cells expressing chimeric antigen receptors (CAR). T lymphocytes can be gene modified by RNA electroporation without integration-associated safety concerns. To establish a safe platform for adoptive immunotherapy, we first optimized the vector backbone for RNA in vitro transcription to achieve high-level transgene expression. CAR expression and function of RNA-electroporated T cells could be detected up to a week after electroporation. Multiple injections of RNA CAR-electroporated T cells mediated regression of large vascularized flank mesothelioma tumors in NOD/scid/γc(−/−) mice. Dramatic tumor reduction also occurred when the preexisting intraperitoneal human-derived tumors, which had been growing in vivo for >50 days, were treated by multiple injections of autologous human T cells electroporated with anti-mesothelin CAR mRNA. This is the first report using matched patient tumor and lymphocytes showing that autologous T cells from cancer patients can be engineered to provide an effective therapy for a disseminated tumor in a robust preclinical model. Multiple injections of RNA-engineered T cells are a novel approach for adoptive cell transfer, providing flexible platform for the treatment of cancer that may complement the use of retroviral and lentiviral engineered T cells. This approach may increase the therapeutic index of T cells engineered to express powerful activation domains without the associated safety concerns of integrating viral vectors. Cancer Res; 70(22); 9053-61. ©2010 AACR.

Chimeric antigen receptors (CARs) are synthetic molecules that provide new specificities to T cells. Although successful in treatment of hematologic malignancies, CAR T cells are ineffective for solid tumors to date. We found that the cell-surface molecule c-Met was expressed in ~50% of breast tumors, prompting the construction of a CAR T cell specific for c-Met, which halted tumor growth in immune-incompetent mice with tumor xenografts. We then evaluated the safety and feasibility of treating metastatic breast cancer with intratumoral administration of mRNA-transfected c-Met-CAR T cells in a phase 0 clinical trial (NCT01837602). Introducing the CAR construct via mRNA ensured safety by limiting the non-tumor cell effects (on-target/off-tumor) of targeting c-Met. Patients with metastatic breast cancer with accessible cutaneous or lymph node metastases received a single intratumoral injection of 3 × 107 or 3 × 108 cells. CAR T mRNA was detectable in peripheral blood and in the injected tumor tissues after intratumoral injection in two and four patients, respectively. mRNA c-Met-CAR T cells cell injections were well tolerated, as none of the patients had study drug–related adverse effects greater than grade 1. Tumors treated with intratumoral injected mRNA c-Met-CAR T cells were excised and analyzed by immunohistochemistry, revealing extensive tumor necrosis at the injection site, cellular debris, loss of c-Met immunoreactivity, all surrounded by macrophages at the leading edges and within necrotic zones. We conclude that intratumoral injections of mRNA c-Met-CAR T cells are well tolerated and evoke an inflammatory response within tumors.

Purpose Myeloma-directed cellular immune responses after autologous stem cell transplantation (ASCT) may reduce relapse rates. We studied whether coinjecting the TLR-3 agonist and vaccine adjuvant Poly-ICLC with a MAGE-A3 peptide vaccine was safe and would elicit a high frequency of vaccine-directed immune responses when combined with vaccine-primed and costimulated autologous T cells. Experimental Design In a phase II clinical trial (NCT01245673), we evaluated the safety and activity of ex vivo expanded autologous T cells primed in vivo using a MAGE-A3 multipeptide vaccine (compound GL-0817) combined with Poly-ICLC (Hiltonol), granulocyte macrophage colony-stimulating factor (GM-CSF) ± montanide. Twenty-seven patients with active and/or high-risk myeloma received autografts followed by anti-CD3/anti-CD28–costimulated autologous T cells, accompanied by MAGE-A3 peptide immunizations before T-cell collection and five times after ASCT. Immune responses to the vaccine were evaluated by cytokine production (all patients), dextramer binding to CD8+ T cells, and ELISA performed serially after transplant. Results T-cell infusions were well tolerated, whereas vaccine injection site reactions occurred in >90% of patients. Two of nine patients who received montanide developed sterile abscesses; however, this did not occur in the 18 patients who did not receive montanide. Dextramer staining demonstrated MAGE-A3–specific CD8 T cells in 7 of 8 evaluable HLA-A2+ patients (88%), whereas vaccine-specific cytokine-producing T cells were generated in 19 of 25 patients (76%). Antibody responses developed in 7 of 9 patients (78%) who received montanide and only weakly in 2 of 18 patients (11%) who did not. The 2-year overall survival was 74% [95% confidence interval (CI), 54%–100%] and 2-year event-free survival was 56% (95% CI, 37%–85%). Conclusions A high frequency of vaccine-specific T-cell responses were generated after transplant by combining costimulated autologous T cells with a Poly-ICLC/GM-CSF–primed MAGE-A3 vaccine.

Key Points• Adoptive transfer of autologous lentiviral-engineered T cells expressing an antisense is safe in chronic HIV infection.• Conditionally replicating lentiviral vector was associated with antiviral effects in patients as assessed by viral evolution and viral load.