NQO1 modulates CD38 surface expression in multiple myeloma

Abstract Introduction: The efficacy of immunotherapies such as monoclonal antibodies (mAbs), CAR T cells, and bispecific antibodies critically depends on the expression of their target surface antigens. In multiple myeloma (MM), therapies directed against CD38, BCMA, and GPRC5D are clinically approved and widely used. However, modulation or loss of these target antigens can lead to resistance. In RNA sequencing analyses of patient samples from our institution (n = 93), we recently identified significantly higher NAD(P)H:quinone oxidoreductase 1 (NQO1) expression in relapsed/refractory MM (mean log2CPM: –0.682) compared to newly diagnosed MM (mean log2CPM: –1.742; p = 0.0002). NQO1 is an FAD-dependent two-electron reductase with known antioxidant and cytoprotective functions. Given its role in redox-sensitive signaling pathways that influence transcriptional regulation, we hypothesized that NQO1 may modulate CD38 expression and thereby affect the efficacy of CD38-targeted therapies in MM. Methods: A Sleeping Beauty transposon system was utilized to overexpress NQO1 in MM cell models AMO1 and RPMI8226. Direct stochastic optical reconstruction microscopy (dSTORM) was used to quantify surface expression of CD38. Drug sensitivity was assessed via cytotoxicity assays. Bulk RNA-seq data from MM patient samples were obtained to compare NQO1 expression levels, and the DepMap datasets were employed for correlation analysis between NQO1 expression levels and CD38 protein abundance. Results: dSTORM quantification of surface antigens revealed consistent downregulation of CD38 in both NQO1-overexpressing AMO1 (AMO1-NQO1) and RPMI8226 (RPMI8226-NQO1) multiple myeloma (MM) cell lines. CD38 cluster density was significantly reduced in AMO1-NQO1 cells (median: 8.89 vs. 19.55 clusters/µm² in WT, p < 0.0001) and in RPMI8226-NQO1 cells (median: 33.53 vs. 46.02 clusters/µm² in WT, p = 0.0238). To validate these findings in an independent dataset, we analyzed the Cancer Dependency Map (DepMap; n = 299), which confirmed a significant inverse correlation between NQO1 and CD38 levels (Pearson r = –0.318, p = 1.81 × 10⁻⁸). Following the in silico validation, the impact of this inverse association at the functional level was assessed via cytotoxicity assays. At an effector-to-target ratio of 5:1 with 10 µg/mL Daratumumab (Dara) treatment (72 h), the survival rate of cells with induced NQO1 expression was significantly higher than that of WT cells (NQO1: 77.40% ± 3.95; WT: 60.62% ± 4.39; p = 0.0082), indicating that NQO1-mediated suppression of CD38 surface expression reduces the efficacy of Dara. Notably, inhibition of NQO1 activity using ES936 restored CD38 surface expression, as measured by dSTORM. In NQO1-overexpressing cells, treatment with 500 nM ES936 for 72 hours significantly increased CD38 cluster density (DMSO: 16.12 ± 1.74 clusters/µm²; ES936: 39.02 ± 2.72 clusters/µm²; p < 0.0001), demonstrating that CD38 downregulation is reversible and amenable to pharmacological targeting. Conclusion: In this study, we identified NQO1 as a novel regulator of CD38 surface expression in MM. Using in vitro models and publicly available datasets, we demonstrated that NQO1 overexpression suppresses CD38 expression and confers Dara sensitivity. Notably, pharmacological inhibition of NQO1 restored CD38 expression, suggesting NQO1 as a potential therapeutic target to improve the efficacy of CD38-directed therapies in relapsed or treatment-refractory MM.