Acute myeloid leukemia is a prevalent blood cancer affecting both adults and children. Gene fusions result from chromosomal rearrangements where two disparate genes are fused and form a novel gene-gene junction. Such fusions often initiate leukemogenesis and, therefore, can be present in all leukemic cells throughout disease course and at relapse. One such gene fusion arises from joining between CBFB and MYH11 genes formed by an inversion in chromosome 16. A fusion between these genes occurs in 12% of pediatric and 7% of adult AML patients. Depending on the chromosomal breakpoint, the resulting CFBF-MYH11 fusion can come in a few varieties, but 80-90% of patients with these fusions share the Type A breakpoint/fusion event. Dr. Melinda Biernacki of the Bleakley lab (Clinical Research Division) wanted to understand whether this fusion gene could create novel antigens that could be exploited to target leukemic cells with T cells, an important immune cell subset that recognizes foreign antigens. The results of her investigations were recently published in the Journal of Clinical Investigation.
T cells recognize short pieces of proteins, called peptides, brought to the cell surface by human leukocyte antigens (HLA), which are highly polymorphic in the human population. When T cells recognize a foreign peptide, often derived from pathogens or cancerous cells, they are activated and kill the foreign antigen-presenting cell. The high variation in both T cell receptors (TCRs) and HLA molecules allows for broad sampling of peptides and highly specific immune responses. The Bleakley lab uses a screening approach to find novel antigens and responding T cells, using donor human T cells and autologous antigen-presenting cells pulsed with peptides from candidate tumor-associated antigens. Dr. Biernacki and her coauthors first used HLA prediction algorithms to determine candidate peptides derived from the CBFB-MYH11 fusion site before screening T cell responses to candidate peptides. The authors found multiple T cell clones derived from multiple normal donor PBMC that responded to the same peptide derived from the fusion site. These clones could only lyse peptide-pulsed targets with the HLA type B*40:01, indicating the peptide was restricted to presentation by this HLA molecule. Additionally, the authors used alanine scanning to determine the amino acid residues necessary for T cell recognition. A BLAST search was used on the resulting motif to determine whether any similar human peptides could be found in the human genome that might be inappropriately targeted by the T cell clones. However, CBFB-MYH11-specific T cell clones could not recognize target cells pulsed with the two most similar human peptides, indicating the clones were indeed specific for the fusion epitope.
While recognition of synthetic peptide in the screening process allowed the authors to isolate antigen-specific T cells, it does not guarantee that the peptide will be processed and presented in relevant tumor cells. To determine if the candidate peptide is a bonafide cancer-specific antigen, they tested whether the CBFB-MYH11-specific T cell clones could kill AML cell lines containing the CBFB-MYH11 fusion and restricting HLA. Only when the cell lines contained both the fusion gene and the restricting B*40:01 HLA could the T cell clones kill the target cell lines, indicating that the CBFB-MYH11 fusion peptide is a true cancer-specific antigen. To further test their T cell clones, the authors determined whether they could kill primary AML samples. CBFB-MYH11-specific T cell clones were activated by, and able to lyse, HLA B*40:01+ CBFB-MYH11+ primary AML but not primary AML lacking the fusion or restricting HLA, a promising result that these T cells could recognize and kill primary cancer cells. Furthermore, the authors used a novel mouse model developed by the Rongvaux lab (Clinical Research Division) that allows for growth of primary human AML, which does not grow readily in other murine models. After engraftment of primary AML cells containing the fusion and restricting HLA, the authors adoptively transferred CBFB-MYH11-specific T cells, or control T cells against an irrelevant antigen, into the mice and assessed tumor burden over time. By seven days post transfer, mice who received antigen-specific T cells had low or undetectable levels of disease in the blood. In contrast, mice who received irrelevant T cells had increasing disease. This difference continued under terminal endpoint at one-month post T cell transfer. At termination of the study, 3/5 mice who received antigen-specific T cells had very low or undetectable disease in the bone marrow (the primary site of disease), whereas all animals who received irrelevant T cells had detectible disease. These studies indicated that CBFB-MYH11-specific T cells could control tumor growth in a xenograft model, a promising step towards use in humans.
While the potential patient pool that could benefit from this specific therapy remains small (~10% with the fusion and ~4-30% with the HLA type, depending on ethnicity), this study shows that discovery and development of TCRs directed against novel cancer-specific antigens is both feasible and effective. Dr. Biernacki explained: “Although finding a good target like CBFB-MYH11 is a huge step forward, we have more work to do before this work is ready to go into the clinic as an engineered T cell receptor (TCR) T cell immunotherapy. Some of the steps include optimizing the TCR transfer and making sure that the engineered T cells are as safe and as effective as possible. We would like to also have CBFB-MYH11 be one of a large collection of AML targets that will give us a toolbox of T cell immunotherapies so that we have a treatment for most patients with AML. Although of course we do not know yet, we hope that immunotherapies based on this work will offer new hope, and even cures, for patients with AML.”
This study was supported by Hyundai Hope on Wheels, Stand Up To Cancer, the American
Association for Cancer Research, the National Cancer Institute, the Rally Foundation for Childhood
Cancer Research, Alex’s Lemonade Stand–Foundation for Childhood Cancer, the Fred Hutch Immunotherapy Integrated Research Center.
UW/Fred Hutch Cancer Consortium members Melinda Biernacki, Elihu Estey, Soheil Meshinchi, Anthony Rongvaux, and Marie Bleakley contributed to this work.
Biernacki M, Foster K, Woodward K, Coon M, Cummings C, Cunningham T, Dossa R, Brault M, Stokke J, Olsen T, Gardner K, Estey E, Meshinchi S, Rongvaux A, Bleakley M. 2020. CBFB-MYH11 fusion neoantigen enables T cell recognition and killing of acute myeloid leukemia. J Clin Invest. Oct 1;130(10):5127-5141. doi: 10.1172/JCI137723.