P861: SIALOFUCOSYLATED STRUCTURES ENABLE PLATELET BINDING TO MYELOMA CELLS CONFERRING PROTECTION FROM NK-MEDIATED CYTOTOXICITY

Background: Multiple myeloma (MM) is a plasma cell disorder that develops in the bone marrow (BM), characterized by an unchecked proliferation and by the ability to disseminate in different parts of the skeleton. Several studies have highlighted a prominent role of platelets in the metastatic dissemination of cancer cells. Indeed, by binding to tumor cells, platelets promote adhesion to target organs and evasion from natural killer (NK)-mediated cytotoxicity. Importantly, tumor/platelet interactions are mostly mediated by P-selectin. We have previously shown that MM cells enriched for Sialyl Lewis X (SLex), a sialofucosylated structure recognized by selectins, efficiently home into the BM and become resistant to bortezomib in vivo (Natoni et al, 2017). Aims: We herein hypothesized that sialofucosylation participates in platelet recruitment on the surface of MM cells via P-selectin, forming a “cloak” that protects tumor cells from NK-mediated cytotoxicity. Methods: Platelets were purified from peripheral blood samples obtained from healthy individuals. MM cell lines and their derivative that have been enriched for SLex, which strongly bind to selectins, were co-cultured with platelets for 30 min and platelet binding was assessed by flow cytometry using the CD41/CD61 antibody. In some experiments, cells were incubated with platelets in the presence of an anti-P-selectin blocker antibody. To examine the NK-mediated cytotoxicity towards tumor cells, parental and SLex MM cells were co-cultured with platelets for 24 h and then incubated with autologous NK cells expanded in vitro for 7 days. NK-mediated cytotoxicity was assessed after 4h by flow cytometry using Annexin V/7-AAD staining. To examine platelet/MM binding in patients, BM samples from MM patients at different disease stages were analyzed by flow cytometry using a panel of antibodies to identify platelet binding to malignant plasma cells. Results: In MM cell lines, platelets bound exclusively to SLex enriched cells, suggesting that efficient binding requires sialofucosylation. This binding could be blocked by a P-selectin antibody, indicating that the interactions between MM cells and platelets are partially dependent on P-selectin. Platelets significantly decreased NK-mediated cytotoxicity in the SLex enriched (P=0.0025) but not in the parental MM cells, confirming the important role of sialofucosylation in platelet binding. Platelets also bound to malignant cells from primary BM aspirates. Interestingly, the binding to platelets occurred in SLex positive and negative cells, suggesting that in vivo non-sialofucosylated ligands may also be involved in platelet binding. Importantly, the SLex CD41/CD61 double positive population seems to accumulate in MM patients at relapse (median diagnosis vs relapse: 0.86 vs 3.03; P=0.0034). Summary/Conclusion: We herein report that MM cell lines enriched in SLex interact with platelets in vitro, and that this interaction partially protects MM cells from NK-mediated cytotoxicity. These results suggest that in vivo, platelets may promote immune evasion facilitating the dissemination of MM cells. Moreover, SLex positive primary malignant BM plasma cells bind platelets, particularly from patients at relapse, suggesting that in vivo these cells may accumulate as the disease progresses and becomes resistant to treatment. Finally, the interactions between platelets and MM cells could be disrupted by a P-selectin antibody, indicating a possible therapeutic target to restrict metastasis and re-sensitize MM cells to NK cells.