Editors' ChoiceCancer

Targeting leukemia-specific mutations

See allHide authors and affiliations

Science Translational Medicine  09 Sep 2020:
Vol. 12, Issue 560, eabe1713
DOI: 10.1126/scitranslmed.abe1713


T cells can recognize neoantigens derived from protein products of characteristic genetic fusions and thereby specifically target leukemia cells.

Most cases of acute myeloid leukemia (AML) are incurable with chemotherapy alone. The power of T cell immunotherapy to cure chemotherapy-refractory AML was demonstrated long ago with allogeneic hematopoietic cell transplantation (HCT). However, both allogeneic HCT and engineered T cell therapies generally target antigens expressed on both malignant and normal tissues, with resultant off-target toxicity.

The ideal T cell immunotherapy would target tumor-specific mutations. When tumor-specific mutations are patient-specific, such immunotherapy may be relevant only for each individual patient. Yet, there are some recurrent genetic mutations found in specific subtypes of AML, and these present an opportunity wherein particular tumor-specific neoantigens may be targetable and relevant for a larger subset of patients.

Biernacki and colleagues report on the development of T cell receptors specific for one of these recurrent genetic fusions (core binding factor beta fused with smooth muscle myosin heavy chain 11, CBFB-MYH11). They examined peptide epitopes spanning the fusion but found peptides with high predicted binding affinity only for two (HLA-B*40:01 and HLA-B*44:02) of 20 prevalent class I human leukocyte antigen (HLA) molecules. The peptide with the highest predicted binding affinity (REEMEVHEL) was immunogenic, bound to HLA-B*40:01, and was naturally processed, making it a relevant tumor-specific antigen. The investigative team found six T cell clones that were specific for REEMEVHEL:HLA-B*40:01, four of which had high avidity. In an epitope- and HLA-specific manner, these T cell clones could kill AML cell lines and primary AML blasts and control AML in a xenogeneic model. Lastly, T cells transduced with the identified T cell receptors-maintained epitope- and HLA-specific cytolytic effects, suggesting that these T cell receptors may be directly translatable for patients.

This study provides proof-of-principle for the feasibility of targeting peptides derived from protein products of recurrent tumor-specific genetic fusions in AML. Yet, the scarcity of HLA molecules predicted to bind candidate peptide epitopes makes it possible that this approach may be applicable only to specific ethnic populations. Furthermore, the immunosuppressive microenvironment engendered by AML may be an additional translational barrier. Finally, the CBFB-MYH11 gene fusion is considered a “favorable-risk” leukemia that is often curable with chemotherapy alone, while some other fusions are associated with higher risk. For patients with higher-risk fusion-associated leukemias and for those with “favorable-risk” AML who have nevertheless relapsed, the approach described here has the potential to truly separate immune responses against malignant versus normal tissue, the foremost goal of immunotherapy.

Highlighted Article

Stay Connected to Science Translational Medicine

Navigate This Article