Project Summary Chemotherapy resistance remains a major barrier to successful treatment of patients with acute myeloid leukemia (AML), contributing to high rates of relapse and mortality. Development of more effective treatments for AML is imperative, particularly therapies with alternative mechanisms of action to circumvent chemoresistance. Adoptive cellular immunotherapy using CD19-targeting chimeric antigen receptor (CAR)- expressing T cells has drastically improved the treatment of patients with multiply relapsed/refractory B-cell lymphoblastic leukemia (B-ALL) and lymphoma and was approved by the FDA. However, successful translation of AML immunotherapies has lagged behind and remains a significant unmet medical need. To date, CAR T cells targeting CD33 or CD123 for AML have shown potent preclinical anti-AML activity, but also induced severe myelotoxicity via on target/off tumor damage to hematopoietic stem cells (HSCs). We have developed an innovative system to isolate single-domain nanobodies (Nb) that preferentially bind AML cells and enable cognate CAR T cells to kill the cancer cells. One of these nanobodies, Nb157, specifically binds to the cell surface protein CD13 (aminopeptidase N), which is often upregulated in adult AML specimens and leukemia stem cells (LSCs). In preliminary studies, we demonstrated that Nb157/CD13 CAR T cells (CD13CARTs) potently eradicated AML cells in preclinical animal models. TIM-3, an inhibitory receptor of certain immune cells, is upregulated in AML blast cells and LSCs, but not expressed in human HSCs. Thus, we generated the 1st generation bispecific and split CD13/TIM-3 CARTs (1st G bCARTs) and demonstrated that the bCARTs potently eradicated AML cells in preclinical animal models, with significantly reduced toxicity to HSCs. To further improve the safety profile of the bCARTs, the 2nd generation bCARTs were generated and they did not induce obvious toxicity to HSCs in our ex vivo analysis. We hypothesize that further development of the bispecific or inducible bispecific CARTs can eradicate AML in patient-derived xenograft (PDX) models with little or tolerable off-tumor toxicity. Three specific aims are proposed to test this hypothesis. Aim 1 will evaluate the 2nd generation bispecific CD13/TIM-3CARTs (bCARTs) in maximizing selective AML killing. Aim 2 will investigate efficacy and specificity of inducible CD13/TIM-3CARTs (ibCARTs) in killing AML cells. Aim 3 will develop bispecific CLL-1/TIM-3 bCARTs to selectively killing AML cells. Results obtained from these studies are imminently translatable to the clinic in the near future for patients with relapsed/refractory AML.