g-NK are adaptive NK cells epigenetically reprogrammed to exhibit higher antibody-dependent cytotoxicity following engagement of CD16 than conventional NK cells. In an ongoing, open-label Phase I/II trial of g-NK cells (IDP-023) to treat relapsed multiple myeloma (MM), we recently reported a mean maximum decrease in serum M-protein or light chain of 73% in eight R/R myeloma patients, with three patients achieving a reduction of 84% or greater. We collected marrow samples before and after g-NK treatment for cytometry time-of-flight (CyTOF) and gene expression analysis to explore the impact of the tumor microenvironment (TME) on outcomes.

Of the five patients analyzed by CyTOF, two had responses by International Myeloma Working Group criteria in M-protein and one with stable disease (SD) had significant improvement in PET scans (#001) and remained progression free for 5 months when they started CAR T-therapy.  The patients who had improvement of their lesions on PET saw decreases in T cell expression of the inflammatory markers CXCR3 and CD38 and decreases in myeloid derived suppressor cells (MDSC). The patient with the lowest CD8+ T cells, the highest PD-1 expression, and the highest MDSC at baseline achieved no biochemical response and had progressive disease on PET.

Nanostring analysis was performed on four responders and three non-responders to compare differentially expressed genes. Responders exhibited increased immune molecules such as KIR2DL3, granulysin, and HLA-F, as well as a significant decrease in the chemokine CCL21 that recruits MDSC to the TME. After g-NK therapy, responders had lower levels of DUSP2 (a negative regulator of T cells) and higher levels of integrin b3 which promotes T-cell migration. These data suggest that g-NK cells can alter the TME and promote MM control by an influx of non-exhausted T cells. Higher MDSC and exhausted T cells may predict worse response to g-NK-therapy.