Gene therapy of cancer

Division Of Basic Sciences - Nci

Investigators

Linked publications & trials

Abstract

My laboratory is involved in studies to genetically modify autologous lymphocytes to improve their anti-tumor activity. In 1990 we reported the first studies of gene transfer in humans which involved the adoptive transfer of TIL transduced with a marker gene encoding neomycin phosphotransferase. These studies suggested the possibility that genes could be inserted into lymphocytes to improve their anti-tumor efficacy using high efficiency retroviral gene insertion. In clinical studies we showed that up to 30% of patients with metastatic melanoma will achieve objective clinical cancer regressions when treated with their autologous lymphocytes transduced with T cell receptors that recognized the MART-1 or gp100 melanoma antigens. These were the first studies showing that TCR transduced lymphocytes could mediate cancer regression. T-cell receptors were identified that recognized NY-ESO-1 and MAGE-A3, the cancer germ-live antigens, epitopes that have also been used to mediate cancer regression inpatients. Using cells transduced with a TCR reactive with the NY-ESO-1 cancer testes antigen 10 of 19 melanoma patients and 10 of 15 synovial cell sarcoma patients have had objective responses. Current clinical trials involve the administration of cells transduced with antigens encoding T-cell receptors that recognize random somatic mutations. T-cells that recognize random somatic mutations have been identified in 151 of 205 (76%) patients with a variety of solid epithelial cancer. 363 neoantigens were identified all of which were unique except for two patients that recognized a KRAS G12D mutation restricted by Cw0802. Adoptive cell transfer (ACT) with neoantigen-reactive T lymphocytes can mediate cancer regression. Here we isolated unique, personalized, neoantigen-reactive T cell receptors (TCRs) from tumor-infiltrating lymphocytes of patients with metastatic gastrointestinal cancers and incorporated the TCR Alpha and Beta chains into gamma retroviral vectors. We transduced autologous peripheral blood lymphocytes and adoptively transferred these cells into patients after lymphodepleting chemotherapy. In a phase 2 single-arm study, we treated seven patients with metastatic, mismatch repair-proficient colorectal cancers who had progressive disease following multiple previous therapies. The primary end point of the study was the objective response rate as measured using RECIST 1.1, and the secondary end points were safety and tolerability. There was no prespecified interim analysis defined in this study. Three patients had objective clinical responses by RECIST criteria including regressions of metastases to the liver, lungs and lymph nodes lasting 4 to 7 months. All patients received T cell populations containing greater than or equal to 50% TCR-transduced cells, and all T cell populations were polyfunctional in that they secreted IFNy, GM-CSF, IL-2 and granzyme B specifically in response to mutant peptides compared with wild-type counterparts. TCR-transduced cells were detected in the peripheral blood of five patients, including the three responders, at levels greater than or equal to 10% of CD3+ cells 1 month post-ACT. In one patient who responded to therapy, ~20% of CD3+ peripheral blood lymphocytes expressed transduced TCRs more than 2 years after treatment. This study provides early results suggesting that ACT with T cells genetically modified to express personalized neoantigen-reactive TCRs can be tolerated and can mediate tumor regression in patients with metastatic colorectal cancers. Chimeric antigen receptors have been developed that recognize CD19 cell surface antigens on B cell malignancies and clinical trials using the transfer of these cells have moderated objective responses in 56% of patients with Diffuse Large B-cell Lymphoma including 50% with durable complete responses. Studies of T-cell transfer targeting the EGFRvIII mutation' expressed on glioblastomas showed no responses in 18 patients. Recent studies have been devoted to generating libraries of genes encoding T cell receptors that recognize shared p53 and KRAS mutations (restricted by multiple MHC antigens) including KRAS G12D, G12V, G12R, G13 D and 17 receptors encoding p53 "hotspot" mutations for use in cell/gene therapy studies. Thus insertion of T-cell receptors recognizing either individual or shared mutations can be used to treat cancer patients. In the past year the extensive studies of freshly resected tumors have revealed gene signatures that can with a high degree of certainty predict lymphocytes and their transcriptome that can rapidly identify T cells with antitumor activity. This 50 gene signature has been shown to prospectively identify T-cell receptors that can recognize mutations that serve as cancer antigen recognized by a patient's T cells. These finding should enable us to more rapidly identify T cell receptors that we can transfer into a patient's own lymphocytes for use in therapy. The ability to insert these T cell receptor genes into more naive stem-like cells is being evaluated for their improved ability to mediate tumor regression in patients with common epithelial cancers.