Search

Pro-inflammatory signaling is a hallmark feature of human cancer, including in myeloproliferative neoplasms (MPNs), most notably myelofibrosis (MF). Dysregulated inflammatory signaling contributes to fibrotic progression in MF; however, the individual cytokine mediators elicited by malignant MPN cells to promote collagen-producing fibrosis and disease evolution remain yet to be fully elucidated. Previously we identified a critical role for combined constitutive JAK/STAT and aberrant NF-kB pro-inflammatory signaling in myelofibrosis development. Using single-cell transcriptional and cytokine-secretion studies of primary MF patient cells and the hMPLW515L murine model of myelofibrosis, we extend this previous work and delineate the role of CXCL8/CXCR2 signaling in MF pathogenesis and bone marrow fibrosis progression. MF patient hematopoietic stem/progenitor cells are enriched for a CXCL8/CXCR2 gene signature and display enhanced proliferation and fitness in response to exogenous CXCL8 ligand in vitro. Genetic deletion of Cxcr2 in the hMPLW515L adoptive transfer model abrogates fibrosis and extends overall survival, and pharmacologic inhibition of the CXCR1/2 pathway improves hematologic parameters, attenuates bone marrow fibrosis, and synergizes with JAK inhibitor therapy. Our mechanistic insights provide a rationale for therapeutic targeting of the CXCL8/CXCR2 pathway in MF patients.

A mayor pathobiological role for interleukin 8 in the etiology of myelofibrosis has been suggested by observations indicating that megakaryocytes expanded in culture from these patients express great levels of interleukin 8 1 and that the plasma levels of this cytokine are predictive of poor prognosis 2. In preliminary experiments we demonstrated that the megakaryocytes from the bone marrow of the Gata1 low model of myelofibrosis express not only high levels of TGF-β, but also levels greater than normal of lipokalin-2, a known inducer of IL-8 production, and of CXCL1, the murine equivalent of IL-8. In addition, these megakaryocytes express also high levels of the CXCL1 receptors CXCLR1 and CXCR2 and the bone marrow from these mice express an CXCR1/CXCR2 activated signature. Using these data as a foundation, we tested here the effects of treatment of Gata1 low mice with the CXCR1/R2 inhibitor reparixin on the myelofibrosis phenotype expressed by this models. To these aim, Gata1 low mice (8-month old) were treated either with vehicle (3 males and 3 females) or with reparixin (formerly referred to as repertaxin) 3 (5 males and 5 females) for either 20 or 37 days. The drug was administered by minipumps implanted subcutaneously in the dorsal region set to deliver 7.5mg of drug/hr/Kg of body weight. The mice receiving the drug for 37 days had the minipumps replaced by day 17. The efficiency of drug delivery decreased over time since the plasma levels of reparixin were 13.90±4.18 and 6.71±4.18ug/mL at day 20 and 37, respectively (p References: 1) Emadi S et al. Blood. 2005;105:464; 2) Tefferi et al, J Clin Oncol. 2011;29:1356; 3) Bertini R et al, PNAS 2004; 101:11791 Figure 1 Figure 1. Disclosures Crispino: Forma Therapeutics: Research Funding; Scholar Rock: Research Funding; MPN Research Foundation: Membership on an entity's Board of Directors or advisory committees; Sierra Oncology: Consultancy. Massucci: Dompe Farmaceutici Spa R&D: Current Employment. Brandolini: Dompe farmaceutici Spa R&D: Current Employment. Giorgio: Dompe farmaceutici Spa R&D: Current Employment. Allegretti: Dompe farmaceutici Spa R&D: Current Employment. Migliaccio: Dompe farmaceutici Spa R&D: Other: received funding for reserach .

Myelofibrosis (MF), the end stage of Philadelphia-negative Myeloproliferative Neoplasms, is still an unmet clinical need. Treatment with the JAK1/2 inhibitor ruxolitinib ameliorates symptoms and improves overall survival but it is unclear whether it is effective in improving the clinical progression of these patients. The bone marrow (BM) and spleen from MF patients contain great numbers of impaired megakaryocytes (MK) which have been hypothesized to be responsible for their clinical manifestation by expressing on their surface high levels of the adhesion receptor P-selectin which triggers a process of pathological neutrophil emperipolesis leading to increased bioavailability of the pro-fibrotic transforming growth factor-β (TGF-β) in the microenvironment1,2. We have peviously reported that ablation of P-selectin cures myelofibrosis in Gata1low mice, a bona fide model for MF3. Using these data as a foundation, we tested whether Crizanlizumab, an antibody targeting human P-selectin that effectively prevents pain crisis in patients with Sickle Cell Disease4, represents an effective option to treat MF. To this aim, we treated Gata1low mice with RB40.34, an antibody against murine P-selectin used in the pre-clinical study that supported the clinical trial with crizanlizumab5, alone or in combination with ruxolitinib. Twenty-four 11-months old Gata1low mice were microchipped, randomly separated in four groups (3 males and 3 females/group) and treated as follow: Group 1: vehicle; Group 2: biotinylated RB40.34 (30 μg/mouse/day per 3 days/week by tail vein injection); Group 3: ruxolitinib (45mg/Kg twice per day by gavage); Group 4: biotinylated RB40.34 and ruxolitinib in combination. Mice were treated for 2 cycles of 5 consecutive days with two days of rest in between and sacrificed on day 5 (I end-point) and 12 (II end-point). End-points: Day 5: all mice were weighed and bled for blood counts determinations and presence of labeled-RB34.40 on platelets. Males were sacrificed for signaling studies and determination of the presence of RB40.34 on MK from BM and spleen. Day 12: females were weighed, bled for blood counts determinations and sacrificed for histopathology analyses. None of the treatments induced changes in body weight that remained similar to those observed before treatment, suggesting that overall all the treatments were well tolerated. None of the treatments rescued the platelet deficiency of Gata1low mice, which remained thrombocytopenic, nor induced anemia by day 12, inspite they had been bleed 1 week earlier. This lack of anemia indicates that the treatments do not impair the ability of the mice to recover from erythroid stress. Atday 5, high levels of RB40.34 were observed on the platelets and on the MK of mice treated with the antibody, alone or with ruxolitinib. However, only the two drugs in combination significantly normalized the TGF-β signaling in BM by reducing the abnormally high content of TFGβRII and SMAD2/3 and the phosphorylation levels of ERK in this organ. These results indicate that when used in combination, the two drugs are effective in rescuing the abnormal TGF-β signaling of BM from these mice. Byday 12, high levels of RB40.34 were observed on BM and spleen of mice treated with the antibody alone or with ruxolitinib but again only treatment with the two drugs in combination greatly improved the cellularity of the BM medulla (Hematoxilin-eosin staining), which is typically reduced in Gata1low mice, and reduced the massive fibrosis (Gomory staining) and high TGF-β content (immunostaining) observed in the diaphysis area of the long bones (Fig. 1). Treatment with the two drugs in combination also reduced fibrosis and TGF-β content in spleen restoring the architecture of this organ which regained a normal appearance with well recognizable red and white pulp zones. When used alone, neither RB40.34 nor ruxolitinib had any effect of the MF phenotype of this mouse model. In conclusion, RB40.34 in combination with ruxolitinib rescues the MF phenotype of Gata1low mice with limited toxicity providing pre-clinical evidence that treatment with crizanlizumab in combination with ruxolitinib may be more effective than that with ruxolitinib alone to cure MF patients. References: 1Schmitt et al Blood 2000;96:1342; 2Centurione et al Blood 2004; 104:3573; 3Spangrude et al Stem Cells 2016; 34:67; 4Ataga et al NEJM 2017; 376:429; 5Embury et al Blood 2004; 104:3378. Disclosures Wilke: Novartis Pharma AG: Current Employment, Other: Stock owner. Campello-Iddison:Novartis: Current Employment. Migliaccio:Novartis: Research Funding.

Several studies indicate that abnormal megakaryocytes (MK) drive bone marrow fibrosis in primary myelofibrosis (PMF). Since the lung is a site where MK lodge to release platelets (Lefrançais, Nature 2017; 544:105), we investigated whether the MK abnormalities observed in PMF also lead to fibrosis in lung. To this aim, we used Gata1low mice that express MK abnormalities similar to those observed in PMF patients and develop myelofibrosis with age (Zingariello, Blood 2013;121:3345) and compared number/morphology of MK and extent of fibrosis in lungs from Gata1low mice or from their wild-type (WT) littermates. Two age groups were considered: adult, A (5-7 months) and old, O (>12 months). Pulmonary fibrosis was assessed by H&E and Masson's staining and graded from 0 (absent) to 3 (severe). MK were evaluated by immunohistochemistry and immunofluorescent analysis with CD42 and their number expressed per 0.144 mm2. Statistical analysis was performed with Shapiro test for normality while chi-square test and Spearman's correlation test for comparisons. Since the mutation is carried in the CD1 background that is known to develop multisite inflammation with age (Brayton, Vet Pathol 2012; 49:85), preliminary experiments compared the pro-inflammatory signature of the mice groups by ELISA determinations of the plasma levels of CXCL1 (murine equivalent of IL-8) and its inducer lipocalin2 (LCN2). WT mice expressed a pro-inflammatory signature with high levels of LCN2 (90±45ng/mL in A; 50±49 in O) and CXCL1 (44.5±11.4pg/mL in A;31.6±4.6 in O compared to 17.0 ±5.1 in O P-Selnull/Gata1low mice which do not express inflammation, p Mild and localized interstitial pneumonia characterized by infiltration with lymphocytes, plasma cells and macrophages in the alveolar septae and peribronchial-vascular spaces was observed in 66.6% of WT mice. However, mild interstitial pneumonia was detected in 84.6% of Gata1low mice where it was multi-focal to coalescent and extended from alveolar septae to subpleural interstitium. Lungs from 33% of WT mice had mild fibrosis with accumulation of collagen in alveolar septae (grade 1) while those from 70.8% Gata1low mice presented severe fibrosis (37.5% grade 1, 29.4% grade 2 and 17.6% grade 3) (Fig. 1A, 1B). The mild interstitial pneumonia associated to grade 1 fibrosis present both in Gata1low and WT littermates is considered a background lesion due to the pro-inflammatory cytokine profile of the strain investigated. By contrast, despite the severity of pneumonia was similar in Gata1low and WT littermates, grade 2 and 3 fibrosis was only present in Gata1low mice, indicating that it was not due to the baseline inflammatory milieu of their CD1 background. MK were observed intravascularly and within the perivascular interstitium of lungs from both WT and Gata1low mice, the same location where fibrosis was detected. The number of MK was significantly lower in Gata1low than in WT littermates (median=1, range 0.6-2.57 vs 1.8, range 0.4-8.3, respectively, p=0.02) (Fig 1C). The morphology of MK found in the two groups was significantly dissimilar (p In conclusion, the significantly higher number of Gata1low mice that exhibited moderate to severe pulmonary fibrosis compared to WT controls establishes Gata1low mice as a potential new animal model to study human idiopathic pulmonary fibrosis and to identify possible drug targets to treat this disease. Disclosures No relevant conflicts of interest to declare.

Idiopathic pulmonary fibrosis (IPF) is a chronic inflammatory disorder which leads to progressive fibrosis and lung failure within 2-5 years1,2. Approximately 30% of IPF patients develop Chronic Kidney Disease (CKD) with renal dysfunction and fibrosis3. The treatment options for IPF are still limited and the disease is an unmet clinical need. The only animal model for IPF is represented by mice treated with bleomycin, a compound used for chemotherapy which has lung toxicity as its major adverse effect, that develop progressive fibrosis. Recent data indicate that bleomycin triggers lung fibrosis by increasing the release of the pro-inflammatory cytokine TGF-β by the resident megakaryocytes (MK)4. Using these observations as foundation, we investigated whetherGata1lowmice, a model for myelofibrosis induced by a TGF-β/P-selectin (P-sel) circuitry established by the abnormal MK5, includes fibrosis in lungs and kidney and whether these traits are rescued by genetic ablation (P-selnullmutation) and antibody neutralization (by RB40.34) of P-sel. MK were observed in lungs from wild-type (WT, n=9),Gata1low(n=22),P-selnull(n=4) andP-selnullGata1low(n=10) littermates. In frequency, MK were significantly lower inGata1low andP-selnullGata1lowthan in WT andP-selnulllittermates (median=1.3-1.5 vs 3-3.7/0.144mm2, p Great levels of fibrosis were observed in lungs and kidney fromGata1low mice but not in those from WT littermates. There are statistically significant differences in lung (p=0.01) and kidney (p To further clarify the role of P-sel in lung and kidney fibrosis, 14Gata1lowmice (7 males, 7 females) were treated with either vehicle (8 mice) or RB40.34 (30 μg/mouse/day per 3 days/week by tail vein injection as described6). Mice were sacrificed on day 5 (males) and 12 (females) and histopathology, inflammation, fibrosis and TGF-β content in lungs and kidney analyzed. Treatment with RB40.34 restored the morphology and reduced inflammation, fibrosis and TGF-β content in the lungs and in kidney with respect to vehicle (Fig.1). In conclusion, Gata1low mice are a novel animal model for IPF with CKD. The phenotype is associated with highTGF-βcontent in lung and kidney and is rescued by targeting P-sel, suggesting that treatment with antibodies against P-sel may cure IPF in man. REF: 1Richeldi et al.Lancet389: 1941, 2017; 2Hutchinson et al.EurRespirJ46, 795, 2015; 3Ikezoeet al.Respiration94: 346, 2017; 4Zhouet al.CellDeathDis10:648, 2019: 5Spangrudeet al.StemCells34: 67, 2016; 6Emburyet al.Blood104: 3378, 2004. Disclosures Migliaccio: Novartis:Research Funding.