Lysosomal Associated Membrane Protein Family Member 5 (LAMP5) As a Novel Mediator of, and Target in High Risk Multiple Myeloma (MM)

Background: Our recently published studies of single cell clonotypic and transcriptional evolution of multiple myeloma precursor disease (Dang et al. Cancer Cell, 2023), as well as related multi-omic studies of more advanced disease, identified a number of genes as potential contributors to progression. Among these is LAMP5, a gene on chromosome 20 that is markedly and specifically overexpressed in the malignant, monoclonal plasma cells in a high proportion of patients. As LAMP5 expression has also been associated with osteolytic bony disease in myeloma, we sought to delineate its potential contributions to the pathobiology of myeloma, which remain incompletely understood. Methods: The impact of LAMP5 knockdown and over-expression was studied using myeloma cell line models both in vitro and in vivo. Moreover, we generated a monoclonal antibody to LAMP5 after finding it was exposed on the myeloma cell plasma membrane, and converted this agent into an antibody drug conjugate (ADC) linked to monomethyl auristatin E (MMAE) or pyrrolobenzodiazepine ( PBD) dimers. Finally, we used the sequence of the single-chain variable fragment to generate LAMP5-targeted chimeric antigen receptor (CAR)-guided T-cells. Results: Analysis of the Multiple Myeloma Research Foundation's CoMMpass SM database indicated that high LAMP5 expression (defined as in the top quartile) in newly diagnosed myeloma patients was significantly associated with markers of increased disease burden, including a higher beta-2-microglobulin (p = 0.047), C-reactive protein (p = 0.0399), serum calcium (p < 0.0001), marrow plasmacytosis (p < 0.001), and a trend towards increased circulating plasma cells (p = 0.075). Moreover, high LAMP5 expression correlated with a shorter progression-free (p = 0.0006) as well as overall survival (p = 0.0006). Similarly, analysis of the University of Arkansas for Medical Sciences gene expression data indicated LAMP5 was increased in the hyperdiploid and proliferative groups that are associated with a poor prognosis. Inducible shRNA-mediated LAMP5 knockdown in myeloma cell lines conferred a survival disadvantage and activated apoptosis. RNA-sequencing analysis identified multiple dysregulated Hallmark pathways, including the G2M checkpoint, the mitotic spindle, apoptosis, mTORC1 signaling, and the unfolded protein response. Proteomic consequences of LAMP5 knockdown included Caspase-7, -8, and -3 cleavage, and increased abundance of Smac, Bim, and PUMA. By contrast, many G2M checkpoint proteins were down-regulated, including Aurora A, Checkpoint-1, Cell division cycle-6, and WEE1 kinases, and decreased levels of phospho-S807/S811 Rb protein were seen. Also, RNA-sequencing revealed mTORC1 pathway dysregulation, and proteomic studies showed decreased activity of AKT and AKT2, mTOR, and p70 S6 kinase. Notably, LAMP5 suppression was associated with reduced levels of expression of an RFP-LC3 fusion protein consistent with a role for LAMP5 in autophagy. In contrast, high LAMP5 expression was associated with improved colony formation in vitro ( A), and promoted disease progression in vivo. Since the LAMP5 promoter had consensus retinoic acid sites, we tested all-trans retinoic acid and found it suppressed LAMP5 protein expression, and reduced in vitro clonogenicity. Finally, since other LAMPs can under some circumstances be found at the plasma membrane, we performed flow and LAMP5 was indeed detected on the surface of unpermeabilized LAMP5-positive myeloma cells. We thus developed a panel of monoclonal antibodies that recognized the extracellular domains of LAMP5, and converted one of these into antibody drug conjugates (ADCs) linked to either MMAE or PBD. Both ADCs reduced the viability of LAMP5-positive myeloma cells in a time- and concentration-dependent manner ( B), activated immunogenic cell death in vitro, and also showed anti-tumor activity in vivo. Importantly, LAMP5-directed CAR T-cells similarly and specifically eliminated only LAMP5-positive myeloma cells with efficacy that was comparable to that of B-cell maturation antigen-targeted CAR T-cells. Conclusions: These pre-clinical in vitro and in vivo data support the possibility that LAMP5 represents a novel mediator of myeloma clonogenicity and high-risk behavior, and that it could serve as a rational therapeutic target leveraging immune therapies that recognize the extracellular domains of cell surface LAMP5.