Search

Natural killer (NK) cell therapy is becoming more and more attractive as a treatment option for various types of cancer due to its high safety profile and proven clinical efficacy. Currently, different sources (e.g. peripheral blood NK (PBNK), umbilical cord blood NK (UCB-NK), hematopoietic stem cells (HSCs), cell lines, induced pluripotent stem cells (iPSCs)) and different culture strategies (e.g. feeder cell-based, cytokines) are being investigated to attain optimal expansion, genetic manipulation and high functionality. It is worth mentioning that the choice of source or culture method can significantly impact the efficacy and characteristics of a cell product. In addition, depending on the starting source, the manufacturing workflow requires different process pipelines. Glycostem has established a platform for a good manufacturing practice (GMP) grade off-the-shelf cryopreserved NK cell product using CD34+ HSCs from UCB. The NK cell product GTA002 is currently tested in a phase I/IIa clinical trial in patients with acute myeloid leukemia (AML). To further improve the product and extend its therapeutical applications, a comprehensive understanding of its characteristics and mechanisms of action is crucial, which is addressed in this thesis from various perspectives. In various solid indications, the activation mechanisms of GTA002 were investigated (Chapter 3-4), characteristics of NK cell batches with high cytotoxic potential were identified (Chapter 4), the synergy of combination therapy was explored (Chapter 5), and the immunomodulatory role of GTA002 was studied (Chapter 6-7).

Allogeneic natural killer (aNK) cell adoptive therapy has the potential to dramatically impact clinical outcomes of glioblastoma multiforme (GBM). However, in order to exert therapeutic activity, NK cells require tumor expression of ligands for activating receptors, such as MHC Class I peptide A/B (MICA/B) and ULBPs. Here, we describe the use of a blood–brain barrier (BBB) permissive supramolecular cationic drug vehicle comprising an inhibitor of the chaperone heat shock protein 90 (Hsp90), which sustains a cytotoxic effect on GBM cells, boosts the expression of MICA/B and ULBPs on the residual population, and augments the activity of clinical-grade aNK cells (GTA002). First, we identify Hsp90 mRNA transcription and gain of function as significantly upregulated in GBM compared to other central nervous system tumors. Through a rational chemical design, we optimize a radicicol supramolecular prodrug containing cationic excipients, SCI-101, which displays >2-fold increase in relative BBB penetration compared to less cationic formulations in organoids, in vitro. Using 2D and 3D biological models, we confirm SCI-101 sustains GBM cytotoxicity 72 h after drug removal and induces cell surface MICA/B protein and ULBP mRNA up to 200% in residual tumor cells compared to the naked drug alone without augmenting the shedding of MICA/B, in vitro. Finally, we generate and test the sequential administration of SCI-101 with a clinical aNK cell therapy, GTA002, differentiated and expanded from healthy umbilical cord blood CD34+ hematopoietic stem cells. Using a longitudinal in vitro model, we demonstrate >350% relative cell killing is achieved in SCI-101–treated cell lines compared to vehicle controls. In summary, these data provide a first-of-its-kind BBB-penetrating, long-acting inhibitor of Hsp90 with monotherapy efficacy, which improves response to aNK cells