Metastatic renal cell carcinoma (RCC) is almost universally fatal, with a five-year survival rate of only 12 percent. Immunotherapy offers new treatment options for patients with metastatic RCC, and complete responses to immune checkpoint inhibitors have been seen in five-to-nine percent of patients. As these responses are thought to be T cell mediated, adoptive T cell transfer may offer an effective strategy. The 5T4 antigen has been previously identified as upregulated on primary RCC cells, as well as metastatic tumor, but is expressed at low or undetectable levels in normal tissues. This makes 5T4 a good candidate for T cell-based therapy, and the Tykodi and Warren labs (Clinical Research Division) have identified T cells specific for a 5T4-derived peptide that binds HLA-A*02:01, an allele prevalent in about half the US population. This previous work was the basis for a new paper recently published in Cancer Immunology, Immunotherapy.
In order to develop 5T4-specific T cells for T-cell receptor (TCR) transgenic immunotherapy, the authors sequenced the TCRs of multiple 5T4-specific T cell clones derived from multiple normal donors. They found that 5T4-specific T cells used unique, but highly homologous TCRs, consistent with previous research on HLA-A*02:01 binding. In order to test the efficacy of various TCRs, the authors cloned each receptor into a lentiviral construct and transduced donor T cells. Expression of each construct was similar, and each TCR could recognize 5T4 peptide-expressing target cells, even at low antigen concentration, indicating high functional avidity. Moreover, transgenic TCRs recognized A*02:01+/5T4+ tumor cell lines derived from RCC, breast cancer, and colon cancer. When the authors tested RCC lines lacking either HLA-A*02:01 or 5T4 antigen, they saw low but present background lysis, indicating some off-target recognition. However, when the authors used CRISPR to knockout the endogenous TCR of donor T cells expressing the transgenic TCR, this background was lost, indicating alloreactivity from the endogenous TCR present in the donor T cells and not cross-reactivity of the 5T4-specific TCR.
As an additional safety precaution, the authors tested the ability of the 5T4-specific TCRs to recognize peptides with alanine substitutions, a reasonable measure of cross-reactivity. In order to test binding of each alanine-substituted peptide to HLA-A*02:01, the authors analyzed stabilization of the HLA molecule on the cell surface, an indicator of peptide binding. Only two alanine-substituted peptides could stabilize HLA-A*02:01 surface expression – at positions one and four. Next, the authors tested whether 5T4 TCRs could recognize these alanine-substituted peptides. Of the seven TCRs tested, three could recognize a peptide with alanine substituted at position one to a high degree, and only one TCR recognized the peptide with alanine substituted at position four. Other TCRs displayed no ability to recognize substituted peptides and therefore limited cross-reactivity.
The Tykodi lab showed that transgenic TCRs specific for 5T4 were well expressed in donor T cells and could recognize 5T4-expressing cancer cell lines, including those derived from RCC. Additionally, these TCRs could not recognize cells lacking the correct HLA type, or those not expressing antigen, indicating these TCRs are specific for their cognate antigen. They also showed that at least four TCRs displayed no cross-reactivity to similar peptides with alanine substitutions, indicating a reasonable level of safety. These are promising steps towards the development of transgenic TCR based immunotherapy for the treatment of RCC.
This work was supported by a generous gift from Dave Jackson.
Fred Hutch/UW Cancer Consortium members Scott S. Tykodi and Edus H. Warren contributed to this work.
Xu Y, Morales AJ Cargill MJ, Towlerton AMH, Coffey DG, Warren EH, Tykodi SS. 2019. Preclinical development of T-cell receptor-engineered T-cell therapy targeting the 5T4 tumor antigen on renal cell carcinoma. Cancer Immunol Immunother. 68(12):1979-1993.