CD38 Down-Regulation on Ex Vivo Activated and Expanded NK Cells for Cell Therapy Persists after Infusion

Introduction: Immunotherapies are gaining more and more importance in the treatment of multiple myeloma (MM). Antibodies directed against MM antigens like CD38, SLAMF7 or BCMA are used either in their natural form, conjugated to drugs, or in the form of bispecific T-cell engagers. Cellular therapies make use of cytotoxic lymphocytes, i.e. T cells or NK cells that can also be modified to express chimeric antigen receptors to target MM cells. Combinations of antibody and cellular therapies could further improve the outcome as, for example, NK cells can mediate antibody dependent cellular cytotoxicity (ADCC). However, NK cells also express CD38 and SLAMF7 and would be targeted by the therapeutic antibodies against these antigens. We have recently reported our clinical study infusing multiple doses of ex vivo activated and expanded autologous NK cells in six patients with MM post autologous stem-cell transplantation (EudraCT 2010-022330-83). Here, we report results of a phenotypic analysis of the ex vivo expanded NK cells and peripheral blood NK cells before and after infusion with implications for possible combination therapies.

Methods: Ex vivo activated and expanded NK cells and NK cells in peripheral blood of the patients were analyzed by multiparameter flow cytometry. Peripheral blood cells were taken from the non-NK cell infusion arm before and at three different timepoints after infusion. NK-cell sub-populations within these samples were analyzed using t-SNE clustering.

Results: Upon ex vivo activation and expansion, we observed that the NK cells gained a unique activated phenotype including populations of CD56 ^brightCD16 ⁺Ki67 ⁺HLA-DR ⁺ NK cells. Interestingly, these NK cells showed a reduced expression of CD38 compared to peripheral blood NK cells. Clustering analyses of data from peripheral blood samples revealed the gradual appearance of a new NK cell population with a similar phenotype in a dose-dependent fashion over four hours following infusion of the NK cell product. Infused NK cells could be detected in circulation up to four weeks after the last infusion. Like the NK cell infusion product, these cells expressed little to none CD38, high levels of NKG2D, 2B4, TIM-3, and TIGIT and similar levels of SLAMF7 compared to peripheral blood NK cells.

Conclusions: The persistent high expression of CD16 and the low expression of CD38 in infused NK cells offers the choice to combine ex vivo activated and expanded NK cells with anti-CD38 antibody therapy without concern for antibody-mediated NK-cell death. Based on these findings, we have started a clinical trial testing this combined therapy (NCT04558931).