Search

Background: COVID-19 has been associated with hypercoagulability, endothelial cell injury and frequent thrombotic complications resulting both from direct effects of the virus on the endothelium and from the 'cytokine storm' resulting from the host's immune response. Since the COVID-19 vaccines have been shown to effectively prevent symptomatic infection including hospital admissions and severe disease, the risk of COVID-19-related thrombosis should be expected to (almost) disappear in vaccinated individuals. However, some rare cases of venous thrombosis have been reported in individuals vaccinated with mRNA vaccines. Thus, there is a sharp contrast between the clinical or experimental data reported in the literature on COVID-19 and on the rare thrombotic events observed after the vaccination with these vaccines. This phenomenon raised some scepticism of even some fear about the safety of these vaccines which could compromise the adhesion of the citizens in the vaccination program. Aims: We conducted a prospective observational study, to explore the impact of vaccination with the BNT162b2 (Pfizer/BioNTech) on blood hypercoagulability and endothelial cell activation and to investigate if this is modified by the presence of active cancer. Methods: In total 229 subjects were prospectively included in the study from April to June 2021. Subjects were stratified in three predefined groups: 127 vaccinated patients with active cancer (VOnco group), 72 vaccinated health care workers (VHcw group) and 30 non vaccinated health individuals (Control group). Blood samples were obtained 2 days after the administration of the first dose of BNT162b2 vaccine and collected in Vacutainer® tubes (0.109 mol/L trisodium citrate). Platelet poor plasma (PPP) was prepared by double centrifugation at 2000 g for 20 minutes at room temperature and plasma aliquots were stored at -80°C until assayed. Samples of PPP were assessed for thrombin generation (TG) with PPP-Reagent® (Thrombogram-Thrombinoscope assay with PPP-Reagent®TF 5pM), E-selectin, D-dimers, (D-Di), Tissue Factor (TFa), procoagulant phospholipid-dependent clotting time (Procag-PPL) and von Willebrand factor (vWF), thrombomodulin (TM), tissue factor pathway inhibitor (TFPI), and platelet factor 4 (PF4). All assays were from Diagnostica Stago (France). The upper and lower normal limits (UNL and LNL) for each biomarker were calculated by the mean±2SD for the control group. Results: All vaccinated subjects showed significantly increased levels of PF4 (71% >UNL, pUNL, pUNL, pUNL, pUNL, pUNL, pUNL, p Conclusion: The ROADMAP-COVID-19-Vaccine study shows that administration of the first dose of the BNT162b2 vaccine induced significant platelet activation documented by shorter Procoag-PPL associated with increased levels of PF4. Plasma hypercoagulability was less frequent in vaccinated individuals whereas there was no evidence of significant endothelial cells activation after vaccination. Interestingly, the presence of active cancer was not associated with an enhancement of platelet activation, hypercoagulability, or endothelial cell activation after the vaccination. Probably, the generated antibodies against the spike protein or lead to platelet activation in a FcyRIIa dependent manner that results in PF4 release. The implication of the mild inflammatory reaction triggered by the vaccination could be another possible pathway leading to platelet activation. Nevertheless, vaccination does not provoke endothelial activation even in patients with cancer. The findings of the ROADMAP-COVID-19-Vaccine study support the concept administration of mRNA based vaccines does not directly cause a systematic hypercoagulability. Disclosures Gligorov: Roche-Genentech: Research Funding; Novartis: Research Funding; Onxeo: Research Funding; Daichi: Research Funding; MSD: Research Funding; Eisai: Research Funding; Genomic Heatlh: Research Funding; Ipsen: Research Funding; Macrogenics: Research Funding; Pfizer: Research Funding. Terpos: Novartis: Honoraria; Janssen: Consultancy, Honoraria, Research Funding; Genesis: Consultancy, Honoraria, Research Funding; Celgene: Consultancy, Honoraria, Research Funding; BMS: Honoraria; Amgen: Consultancy, Honoraria, Research Funding; Takeda: Consultancy, Honoraria, Research Funding; Sanofi: Consultancy, Honoraria, Research Funding; GSK: Honoraria, Research Funding. Dimopoulos: Amgen: Honoraria; BMS: Honoraria; Janssen: Honoraria; Beigene: Honoraria; Takeda: Honoraria.

Background Cancer patients with venous thromboembolism (VTE) or at risk of VTE are treated with antithrombotic agents. Cancer cells express procoagulant properties and induce hypercoagulability in the microenvironment, that could impact the efficiency of the antithrombotic agents. Aims In the present study, we investigated the interaction between antithrombotic agents with pancreatic cancer cells, as well as with their microenvironment. The impact of apixaban, fondaparinux, enoxaparin and tinzaparin on the procoagulant properties of pancreatic cancer cells BXPC3 was examinated. Reciprocally, we also investigated the impact of BXPC3 on the potency of these antithrombotic agents. Methods BXPC3 cells (400 cells/μl) were exposed for 48 hours to apixaban (2 µg/ml), fondaparinux (2 µg/ml), enoxaparin, tinzaparin (2 anti-Xa IU/ml) or NaCL (control). Then, conditioned media (CM) and BXPC3 cells were harvested, separated and put in contact with normal platelet-poor plasma (PPP). Subsequently, thrombin generation (TG) was assessed using Thrombogram-Thrombinoscope® assay (Diagnostica Stago). Cells' viability was also assessed with the MTT assay. Gene expression for Tissue Factor (TF), Vascular Endothelial Growth Factor (VEGF), Thrombospondin 1 (THSB1) was assessed with RT-qPCR at the cells exposed or not to the antithrombotic agents. Expression of TF protein and activity of cancer cells was assessed using ELISA method. Residual anti-Xa activity in CM was measured using specific amidolytic assays for each antithrombotic agent. Results Apixaban, fondaparinux, enoxaparin, and tinzaparin significantly reduced cell viability by 25%, 12%, 14%, and 11% respectively. In the control experiment non treated BXPC3 cells enhanced TG. Pre-treatment of BXPC3 with the antithrombotic agents did not significantly modify their capacity to trigger and enhance TG. Among the studied agents only apixaban resulted in significant decrease of TF mRNA expression. However, protein expression of TF was not significantly modified by any of the antithrombotic agents. VEGF's mRNA expression was significantly decreased by fondaparinux and enoxaparin. THBS1's mRNA expression was significantly increased by apixaban. The concentrations of the anti-Xa activity of fondaparinux, enoxaparin and tinzaparin in the CM obtained at 48h after exposure of cells were reduced by 27%, 48% and 26% respectively as compared to those initially added in the culture medium. In contrast, the concentration of apixaban in the CM did not significantly change. The CM obtained by cells exposed to apixaban, fondaparinux, enoxaparin and tinzaparin inhibited TG by 70%, 30%, 40% and 90% respectively. Conclusion. Antithrombotic agents reduced the viability of BXPC3 cells. Among the studied agents, apixaban had the most pronounced effect on cells' viability. The antithrombotic agents had a potential downregulating effect on the proangiogenetic properties of BXPC3 via the decrease of VEGF gene expression (fondaparinux and enoxaparin) and enhancement of THBS1 gene expression (apixaban). Nevertheless, preincubation of BXPC3 with the antithrombotic agents did not alter the expression of TF protein and their effect on thrombin generation. Moreover, BXPCE exerted a "degradation" effect on LMWH and fondaparinux. Apixaban appeared to escape from this effect of the cancer cells. A significant inhibitory effect on thrombin generation was exerted by the residual concentrations of the antithrombotic agents in the microenvironment of cancer cells. The ensemble of these data highlight for the first time that the presence of antithrombotic agents in cancer cell microenvironment alters the biology of cancer cells and offer a constant antithrombotic effect in the microenvironment. Disclosures No relevant conflicts of interest to declare.

Introduction: Myeloma is characterized by bone lesions, which are related to both an increased osteoclast activity and a defect in the differentiation of medullary mesenchymal stem cells (MSCs) into osteoblasts. Outside the medullary environment, adipocyte-derived MSCs (ASCs) could represent a source of functional osteoblasts. However, we recently found a defect in the osteoblastic differentiation of ASCs from myeloma patients (MM-ASCs). We therefore examined the effects of plasma from myeloma patients at diagnosis (MM-plasmas) and in complete remission (CR-plasmas) and from healthy donors on the osteoblastic differentiation of healthy donor-derived ASCs (HD-ASCs) and healthy donor-bone marrow derived MSCs (HD-BM-MSCs). Materials and Methods: We studied 11 MM-ASCs, 5 HD-ASCs and 3 HD-BM-MSCs. The plasmas were from myeloma patients with bone lesions at diagnosis (n=12), in complete remission (n=8) and from 5 pools of 100 healthy donors (HD-plasmas). HD-ASCs were differentiated into osteoblasts and adipocytes and HD-BM-MSCs into osteoblasts with the three types of plasmas as well as newly discovered cytokines. Results: Osteoblastogenesis in HD-ASCs was suppressed by MM-plasmas. Alizarin red coloration and alkaline phosphatase activity were strongly decreased along with a decreased RUNX2 and osteocalcin expression. However, adipocyte differentiation was unaltered. The osteoblastic differentiation deficiency was reversible once the plasma-derived factors were removed. Using cytokine array and comparing MM-plasmas with HD-plasmas, we identified seven cytokines (ANG1, ENA-78, EGF, PDGF-AA/AB/BB and TARC), besides DKK1, highly increased in MM-plasmas which was confirmed by ELISA (Figure). They separately inhibited the osteoblastic differentiation of HD-ASCs. In contrast, myeloma patients in remission had a cytokine plasma level almost normal with barely no osteoblastic differentiation inhibition. In addition, the mixture of the 7 cytokines with and without DKK1 inhibited not only the HD-ASCs but also the HD-BM-MSCs. Concomittantly, we observed that MM-plasmas enhanced adipogenesis-related gene expression. Comparison of MM-ASCs and HD-ASCs by RNA sequencing showed that two master genes characterizing adipocyte differentiation, CD36 and PPARγ, were upregulated in MM-ASCs as compared to HD-ASCs. Moreover, we demonstrated a significant increase in CD36 and PPARγ expression in HD-ASCs in the presence of MM-plasmas or the seven cytokines individually, similarly as in MM-ASCs. Finally, we tried to identify the origin of these cytokines. When myeloma patients were in remission, the cytokines levels were strongly decreased suggesting a malignant plasmocyte secretion. This was reinforced by the detection of the 7 cytokines in three different myeloma cell lines with an especially high secretion of PDGF-AA. We conclude that specific cytokines in MM-plasmas, besides the well-known DKK1, inhibit the osteoblastic differentiation of MM- and HD-ASCs with a skewing towards adipocyte differentiation. Of note, this inhibition by the cytokines were also observed on HD-BM-MSCs suggesting that this could also be the case on myeloma-BM-MSCs. Legend to figure: Cytokine expression in MM, CR and HD-plasmas (A) Representative images of cytokine array blots probed with the plasma samples. The red boxes identify the cytokines significantly dysregulated in MM- as compared to HD- or CR-plasmas and further analyzed by ELISA. The blue boxes identified the cytokines similarly expressed in MM-/CR-/HD-plasmas. (B) Cytokine concentrations in the HD-plasmas (n=5), MM-plasmas (n=11) and CR-plasmas (n=8) were measured by ELISA. * p < 0.05, ** p < 0.01, *** p < 0.001, ns (not significant). Figure 1 Figure 1. Disclosures Delhommeau: Novartis: Consultancy; BMS: Consultancy; Celgene: Consultancy. Garderet: Celgene: Consultancy; Janssen: Consultancy; Amgen: Consultancy; Sanofi: Consultancy; Takeda: Consultancy.