B-cell maturation antigen (BCMA, also termed TNFRSF17) is an attractive therapeutic target due to its restricted expression on normal and malignant plasma cells (PC). GSK2857916 (or J6M0-MMAF) is a BCMA-specific antibody conjugated to the microtubule-disrupting agent monomethyl auristatin F (MMAF) via a protease-resistant linker. To evaluate the clinical potential of this agent, tumour cells from seventy multiple myeloma (MM) patients were assessed for BCMA expression by immunohistochemistry and flow cytometry. All patients tested expressed BCMA, at varying levels, and both surface and intracellular expression were observed. BCMA expression is maintained through relapse, extramedullary spread and in residual disease post therapy. BCMA levels may also be prognostically useful as higher levels of BCMA were associated with poorer outcomes, even taking into account genetic risk. We observed rapid internalization of surface BCMA and newly expressed protein by 1 h, suggesting a mechanism for J6M0-MMAF activity even with low surface antigen. J6M0-MMAF mediated cytotoxicity of MM cells varied with dose and antigen levels, with clonogenic progenitors killed at lower doses than mature cells. In comparison, J6M0-MMAF killing of primary CD138(+) myeloma cells occurred with slower kinetics. Our observations support BCMA to be a promising therapeutic target in MM for novel therapies such as J6M0-MMAF.

SummaryDefects in bone repair contribute to multiple myeloma (MM) bone disease. It is unknown whether this reflects failure of osteogenic differentiation from mesenchymal stromal cells (MSC), inherent stromal defects or mature cell dysfunction. We quantified the number of fibroblast colony-forming units (CFU-f) and osteoblast colony-forming units (CFU-ob) in freshly isolated bone marrow (BM) from healthy individuals (N = 10) and MM patients (N = 54). CFU-f and CFU-ob were present in MM BM, at comparable frequency to normal subjects, irrespective of disease stage, and the presence of bone disease. Adherent cultures from MM BM are able to differentiate into osteoblasts, as indicated by the early upregulation of RUNX2, SP7, AXIN2 and DLX5, and the production of alkaline phosphatase and calcium. Coculture with MM cells failed to prevent osteogenic differentiation of adult human MSC. On the other hand, MM cells induced cell cycle progression in resting MSC in a cell contact dependent manner. This effect was confirmed using both primary CD138+ cells and MM cell lines, and was not seen with B or T cell lines. Our data confirm the presence of osteoblast progenitors and the preservation of osteogenic function in MM, however dysregulation of cell cycle control may contribute to the loss of normal bone homeostasis that ultimately results in osteolytic bone loss.

B-cell maturation antigen (BCMA) is upregulated at the terminal stages of plasma cell (PC) differentiation, and is expressed on normal and malignant PC. Apart from low levels of mRNA detected on dendritic cells, expression appears absent on other tissues, indicating the potential as a target for novel antibody therapeutics for multiple myeloma (MM). We generated a humanised anti-BCMA antibody, modified for Fc-enhanced function, and conjugated to mcMMAF (anti-BCMA antibody drug conjugate, ADC). Flow cytometric studies on human myeloma cell lines (HMCL) showed rapid internalisation of anti-BCMA antibody by flow cytometry and confocal microscopy. The internalised antibody was transported to the lysosome and was nearly undetectable by confocal microscopy after 6 hours, indicating the suitability of BCMA as a target for an antibody drug conjugate (ADC). BCMA expression reached original surface levels by 6 hours post antibody treatment, thus maintaining the cell as a target for effector mediated killing. Evaluation of BCMA expression on HMCL revealed variable surface expression (1/11 high, 5/11 moderate, 5/11 low). Tumour cell killing by the ADC was expression level, dose and time dependent. The highest expressing HMCL, H929, showed significant killing (60% at 100ng/mL) at 24 hours, and up to 90% after 2 days. Cells expressing moderate levels of BCMA required incubation for up to 4 days to show maximal levels of cell death, suggesting the importance of continued internalisation of the antibody/antigen complexes over this period. ARH77 cells were transduced to express varying levels of BCMA, and killing at 3 days (200ng/ml) was directly proportional to level of surface expression (NT 0% killing, Low BCMA 75% killing, High BCMA 90%). We studied surface antigen levels in a cohort of patients to ascertain the need for patient selection. Like the HMCL, patient CD138+ plasma cells (PC) displayed a range of expression. Of 67 patients tested, CD138+ PC from 12 expressed high levels, 52 expressed intermediate, and 3 had low/negative surface BCMA as determined by MFI ratio of specific antibody to isotype control. Non-CD138+ cells from the bone marrow (BM) were negative for BCMA. Immuno-histochemistry (IHC) on paraffin-embedded BM sections, using a murine antibody and dual staining with anti-Blimp1 to identify tumour cells, revealed both membrane and diffuse, or punctate, cytoplasmic staining. Expression levels varied, from high uniform, to heterogeneous and patchy, to uniform low level. In no patient were the tumour cells entirely negative for BCMA. There was broad correlation between FACS analysis and IHC, thus patients were divided into high, moderate and low expressing groups. Examination of patient and disease characteristics revealed no correlation between BCMA expression and disease stage, response to last treatment, time from diagnosis, isotype, CD56 expression, or cyclin D-type, but there was a trend towards higher BCMA levels in tumours with adverse genetics (90% of patients with adverse genetics had high/moderate levels cf 64% of patients with standard CGN (p=0.06, Fisher’s exact test, 2-tailed). CD138+ cells in LPL (n=3) were positive for BCMA, but CD20+ lymphocytes were negative. Serum BCMA levels in MM patients (175.6±242.6ng/mL; mean±SD, n=34) were higher than in normal subjects (9.28±1.9ng/mL; n=38) but no correlation with bone marrow plasmacytosis or surface BCMA was noted. Levels appeared similar between new diagnosis (147.6±190.8ng/mL; mean±SD, n=8) and relapsed disease (184.3±259.1ng/mL; n=26). We tested ADC activity on primary tumour cells in whole BM cultures, enumerating viable CD138+ cells by flow cytometry. As with the HMCL, ADC mediated cytotoxicity was expression level, dose and time dependent, with a slower time course than with HMCL, perhaps reflecting the slower turn-over of these cells. In samples expressing moderate levels of BCMA, ADC-mediated cytotoxicity increased from 23.1±2.9% (mean±SEM, n=6) at 1-2 days to 48.3±5.1% at 4 days, and 61.2±6.2% by 6-7 days. Optimal doses of ADC ranged from 500ng-1ug/ml. In summary, these preclinical data in MM support the potential utility of an anti-BCMA ADC across the whole MM population, perhaps with particular efficacy in patients with adverse genetics, for whom an unmet need remains. Disclosures: Yong: GSK: Research Funding. Germaschewski:GSK: Employment. Mayes:GlaxoSmithKline: Employment. Sully:GlaxoSmithKline: Employment. Seestaller-Wehr:GlaxoSmithKline: Employment. Fieles:GlaxoSmithKline: Employment. Tunstead:GlaxoSmithKline: Employment. McCahon:GlaxoSmithKline: Employment. Craigen:GlaxoSmithKline: Employment.