Merker et al. ERBB2-CAR-Engineered CIK Cells Against Rhabdomyosarcoma progression compared to untreated controls. ERBB2-CAR CIK cell therapy also supported innate immunity as evidenced by selective accumulation of NK and T-NK cell subpopulations in disseminated RMS tumors, which was not observed for WT CIK cells. Our data underscore the power of heterogenous immune cell populations (T, NK, and T-NK cells) to control solid tumors, which can be further enhanced with CARs, suggesting ERBB2-CAR CIK cells as a potential treatment for high-risk RMS.

Internal tandem duplications (ITD) in the receptor tyrosine kinase FLT3 occur in 25 % of acute myeloid leukemia (AML) patients, drive leukemia progression and confer a poor prognosis. Primary resistance to FLT3 kinase inhibitors (FLT3i) quizartinib, crenolanib and gilteritinib is a frequent clinical challenge and occurs in the absence of identifiable genetic causes. This suggests that adaptive cellular mechanisms mediate primary resistance to on-target FLT3i therapy. Here, we systematically investigated acute cellular responses to on-target therapy with multiple FLT3i in FLT3-ITD + AML using recently developed functional translatome proteomics (measuring changes in the nascent proteome) with phosphoproteomics. This pinpointed AKT-mTORC1-ULK1-dependent autophagy as a dominant resistance mechanism to on-target FLT3i therapy. FLT3i induced autophagy in a concentration- and time-dependent manner specifically in FLT3-ITD + cells in vitro and in primary human AML cells ex vivo. Pharmacological or genetic inhibition of autophagy increased the sensitivity to FLT3-targeted therapy in cell lines, patient-derived xenografts and primary AML cells ex vivo. In mice xenografted with FLT3-ITD + AML cells, co-treatment with oral FLT3 and autophagy inhibitors synergistically impaired leukemia progression and extended overall survival. Our findings identify a molecular mechanism responsible for primary FLT3i treatment resistance and demonstrate the pre-clinical efficacy of a rational combination treatment strategy targeting both FLT3 and autophagy induction.

The dismal prognosis of pediatric and young adult patients with high-risk rhabdomyosarcoma (RMS) underscores the need for novel treatment options for this patient group. In previous studies, the tumor-associated surface antigen ERBB2 (HER2/neu) was identified as targetable in high-risk RMS. As a proof of concept, in this study, a novel treatment approach against RMS tumors using a genetically modified natural killer (NK)-92 cell line (NK-92/5.28.z) as an off-the-shelf ERBB2-chimeric antigen receptor (CAR)-engineered cell product was preclinically explored. In cytotoxicity assays, NK-92/5.28.z cells specifically recognized and efficiently eliminated RMS cell suspensions, tumor cell monolayers, and 3D tumor spheroids via the ERBB2-CAR even at effector-to-target ratios as low as 1:1. In contrast to unmodified parental NK-92 cells, which failed to lyse RMS cells, NK-92/5.28.z cells proliferated and became further activated through contact with ERBB2-positive tumor cells. Furthermore, high amounts of effector molecules, such as proinflammatory and antitumoral cytokines, were found in cocultures of NK-92/5.28.z cells with tumor cells. Taken together, our data suggest the enormous potential of this approach for improving the immunotherapy of treatment-resistant tumors, revealing the dual role of NK-92/5.28.z cells as CAR-targeted killers and modulators of endogenous adaptive immunity even in the inhibitory tumor microenvironment of high-risk RMS.