PS1443 BLOCKADE OF TNFR1 RATHER THAN TNFR2 IMPROVES HEMATOCRIT AND REDUCES BLOOD PRO‐INFLAMMATORY CYTOKINE LEVELS IN THE VAV1‐CRE X JAK2+/V617F MPN MOUSE MODEL

Background: Myeloproliferative neoplasms (MPN) are characterized by expansion of myeloid cells, splenomegaly and development of thrombosis. Recent studies show that chronic inflammation induced by MPN promotes disease development and increases symptom burden in patients. Aims: In the Vav1‐Cre x JAK2 +/V617F MPN mouse model, we aimed to determine the role of the TNF‐α receptor (TNFR) 1 and 2 pathways as a potential therapeutic target. Pro‐inflammatory TNF‐α is prominently expressed in MPN and is thought to be critically involved in disease progression and symptom burden. Little is known on the contribution of the two distinct TNFRs (TNFR1 and TNFR2) to MPN pathophysiology. Methods: Vav‐Cre1 x JAK2 +/V617F mice were treated i.p. with α‐TNFR1 (20 mg/kg) or α‐TNFR2 (5 mg/kg) antibodies 3 x per week (α‐TNFR1) or 2 x per week (α‐TNFR2) for 3 weeks. The total blood count was measured weekly during treatment (retro‐orbital bleeding). Upon end of treatment, cell composition of blood, bone marrow and spleen was analyzed by flow cytometry. Serum concentrations of cytokines in blood (heart puncture) were measured by a bead‐based multiplex assay. To study the influence of α‐TNFR1 treatment, Vav‐Cre1 x JAK2 +/V617F mice were crossed with a mouse strain expressing chimeric TNFR1 protein (huTNFR1 ki ), consisting of the extracellular human domain and the murine transmembrane and intracellular domain. huTNFR1 ki x Vav‐Cre1 x JAK2 +/V617F mice were treated with the antagonistic human TNFR1 specific mAb H398 and control IgG, respectively. To study the TNFR2 blockade, Vav1‐Cre x JAK2 +/V617F mice were treated with a mouse TNFR2 specific mAb (TR75–54.7) and control IgG, respectively. Results: The mean HCT of α‐TNFR1 treated mice was significantly reduced from 72.5% to 60% (p = 0.0152) as compared to a 2.0% increase in the control group. Strikingly, the decrease in HCT was found in all α‐TNFR1 antibody treated mice. The WBC was significantly increased in α‐TNFR1 treated mice as compared to controls. There was no difference measured in red blood counts between the two groups and platelet number stayed stable in both groups during treatment phase. Examination of serum pro‐inflammatory cytokines levels revealed a strong decrease in TNFα‐, IL‐1β or IL‐6 in H398 treated mice as compared to controls (15–50%). Flow cytometry analysis of immune cells isolated from bone marrow, blood and spleen showed a minor decrease in T‐cell numbers analyzed in peripheral blood. Of note, α‐TNFR1 therapy did not alter immune cell composition in spleen and bone marrow. α‐TNFR2 treatment reduced HCT from 79.0% to 67.8%, the IgG group also showed a slight decrease in HCT (75.5% to 71.0%; p = 0.1000). RBC and PLT stayed stable during antibody treatment, whereas WBC decreased during α‐TNFR2 treatment. There was almost no change in pro‐inflammatory cytokine levels in comparison of α‐TNFR2 treated and control mice. Under the applied treatment conditions no effect of α‐TNFR1 or α‐TNFR2 treatment on splenomegaly could be observed. In addition, TNFR1 or TNFR2 blockade did not reduce adhesion of isolated granulocytes to VCAM or ICAM coated surfaces. Summary/Conclusion: In conclusion, our study using the Vav1‐Cre x JAK2 +/V617F MPN mouse model revealed an involvement of the TNF‐α‐TNFR1 rather than TNF‐α‐TNFR2 pathway in induction of chronic inflammation. α‐TNFR1 treatment downregulated HCT and the levels of pro‐inflammatory cytokines. Therefore TNFR1 appears a promising therapeutic target in a combination therapy with JAK1/2 inhibitors such as Ruxolitinib, which should be exploited in future studies.