Donor T cells play a vital role in allogeneic hematopoietic stem cell transplantation (alloSCT). First, na?ve and pathogen-specific donor memory T cells (TM) protect recipients from infection. Second, donor T cells mediate the graft-versus leukemia (GVL) effect. In MHC-matched alloSCT, GVL-inducing T cells target minor histocompatibility antigens (miHAs). Unfortunately, miHA-reactive T cells also attack nonmalignant host tissues, causing graft-vs-host disease (GVHD). A longstanding and elusive goal has been to develop approaches that preserve GVL and immune reconstitution while minimizing GVHD. A second goal is to overcome GVL- resistance. Indeed, relapsed disease is the greatest cause of post-transplant mortality. Leukemia relapse must relate to properties of alloreactive T cells, leukemia cells and the host. Understanding how these contribute to relapse requires clinically relevant models in which effects of each can be isolated. A major barrier to understanding relapse has been an inability to track among polyclonal T cells with all (or even a few) of the allospecificities that contribute to GVL. A second obstacle has been the lack of clinically relevant and genetically manipulable leukemias. We have new data that shed light on the disconnect between leukemia relapse in the face of potent alloimmunity in a model wherein easily trackable polyclonal GVL-inducing T cells target a relevant and genetically manipulable leukemia. We hypothesized that GVL would be augmented if CD8+ TM reactive against miHAs expressed by leukemia cells were induced in the donor. As we hoped, the CD8+ TM from miHA-vaccinated donors were potent mediators of GVL against a GVL-resistant model of blast crisis CML (mBC-CML) induced by the retroviral transfer of bcr-abl and NUP98-HOXA9 fusion cDNAs. MiHA- specific TM underwent dramatic expansion. However, a substantial fraction of mice succumbed to mBC-CML. This was quite unexpected, especially as mice with progressive mBC-CML retained a large number of miHA- reactive CD8 cells. We will use this model to test two nonexclusive hypotheses regarding GVL resistance: 1) alloimmunity selects for GVL-resistant leukemia subclones;and 2) the T cell response is restricted by T cell- extrinsic or -intrinsic factors. Our system has unique strengths that make it ideal for these studies. Using MHCI tetramers we can track the miHA-specific polyclonal T cells which are the only T cells that mediate GVL. mBC- CML is genocopy and phenocopy of human BC-CML/AML and because it is induced by retrovirus, we can readily create gene-modified leukemias. We have also assembled key gene-modified mice and reagents that block T cell inhibitory pathways. These studies address a real clinical concern. We have an ongoing clinical trial wherein leukemia patients receive CD34-selected grafts supplemented with TM purified by the depletion of CD45RA+ TN using a reagent co-developed by the PI. In a subsequent study we plan to incorporate donor miHA-vaccination. These studies will identify GVL resistance mechanisms with this approach and similar resistance mechanisms are likely in play in alloimmune responses without donor vaccination.

Public Health Relevance

Many people with common forms of cancer of blood cells have poor outcomes with standard chemotherapy and radiation therapy. Some of these people can be cured by chemotherapy and/or radiation therapy followed by infusion of immune and blood stem cells from another person. While the immune cells attack and kill the cancer cells in some people, they fail to do so in others, and the goal of this proposal is to better understand why the immune attack fails so as this form of therapy can be improved in ways that will cure more people.

National Institute of Health (NIH)
National Heart, Lung, and Blood Institute (NHLBI)
Research Project (R01)
Project #
Application #
Study Section
Special Emphasis Panel (ZRG1-OTC-C (02))
Program Officer
Welniak, Lisbeth A
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
Yale University
Internal Medicine/Medicine
Schools of Medicine
New Haven
United States
Zip Code
Matte-Martone, Catherine; Liu, Jinling; Zhou, Meng et al. (2017) Differential requirements for myeloid leukemia IFN-? conditioning determine graft-versus-leukemia resistance and sensitivity. J Clin Invest 127:2765-2776