Hepatopulmonary syndrome is associated with low sphingosine-1-phosphate levels and can be ameliorated by the functional agonist fingolimod

•Hepatopulmonary syndrome is associated with low levels of sphingosine-1-phosphate (S1P). •Fingolimod (a functional agonist of S1P) improves pulmonary vascular tone, oxygenation, and survival in an experimental model. •Fingolimod improves hepatocyte proliferation and portal pressure and decelerates hepatic fibrosis in a CBDL mouse model. Background & Aims Hepatopulmonary syndrome (HPS) is characterised by a defect in arterial oxygenation induced by pulmonary vascular dilatation in patients with liver disease. Fingolimod, a sphingosine-1-phosphate (S1P) receptor modulator, suppresses vasodilation by reducing nitric oxide (NO) production. We investigated the role of S1P in patients with HPS and the role of fingolimod as a therapeutic option in an experimental model of HPS. Methods Patients with cirrhosis with HPS (n = 44) and without HPS (n = 89) and 25 healthy controls were studied. Plasma levels of S1P, NO, and markers of systemic inflammation were studied. In a murine model of common bile duct ligation (CBDL), variations in pulmonary vasculature, arterial oxygenation, liver fibrosis, and inflammation were estimated before and after administration of S1P and fingolimod. Results Log of plasma S1P levels was significantly lower in patients with HPS than in those without HPS (3.1 ± 1.4 vs. 4.6 ± 0.2; p <0.001) and more so in severe intrapulmonary shunting than in mild and moderate intrapulmonary shunting (p <0.001). Plasma tumour necrosis factor-α (76.5 [30.3–91.6] vs. 52.9 [25.2–82.8]; p = 0.02) and NO (152.9 ± 41.2 vs. 79.2 ± 29.2; p = 0.001) levels were higher in patients with HPS than in those without HPS. An increase in Th17 (p <0.001) and T regulatory cells (p <0.001) was observed; the latter inversely correlated with plasma S1P levels. In the CBDL HPS model, fingolimod restored pulmonary vascular injury by increasing the arterial blood gas exchange and reducing systemic and pulmonary inflammation, resulting in improved survival (p = 0.02). Compared with vehicle treatment, fingolimod reduced portal pressure (p <0.05) and hepatic fibrosis and improved hepatocyte proliferation. It also induced apoptotic death in hepatic stellate cells and reduced collagen formation. Conclusions Plasma S1P levels are low in patients with HPS and even more so in severe cases. Fingolimod, by improving pulmonary vascular tone and oxygenation, improves survival in a murine CBDL HPS model. Impact and implications A low level of plasma sphingosine-1-phosphate (S1P) is associated with severe pulmonary vascular shunting, and hence, it can serve as a marker of disease severity in patients with hepatopulmonary syndrome (HPS). Fingolimod, a functional agonist of S1P, reduces hepatic inflammation, improves vascular tone, and thus retards the progression of fibrosis in a preclinical animal model of HPS. Fingolimod is being proposed as a potential novel therapy for management of patients with HPS. Hepatopulmonary syndrome (HPS) is characterised by a defect in arterial oxygenation induced by pulmonary vascular dilatation in patients with liver disease. Fingolimod, a sphingosine-1-phosphate (S1P) receptor modulator, suppresses vasodilation by reducing nitric oxide (NO) production. We investigated the role of S1P in patients with HPS and the role of fingolimod as a therapeutic option in an experimental model of HPS. Patients with cirrhosis with HPS (n = 44) and without HPS (n = 89) and 25 healthy controls were studied. Plasma levels of S1P, NO, and markers of systemic inflammation were studied. In a murine model of common bile duct ligation (CBDL), variations in pulmonary vasculature, arterial oxygenation, liver fibrosis, and inflammation were estimated before and after administration of S1P and fingolimod. Log of plasma S1P levels was significantly lower in patients with HPS than in those without HPS (3.1 ± 1.4 vs. 4.6 ± 0.2; p <0.001) and more so in severe intrapulmonary shunting than in mild and moderate intrapulmonary shunting (p <0.001). Plasma tumour necrosis factor-α (76.5 [30.3–91.6] vs. 52.9 [25.2–82.8]; p = 0.02) and NO (152.9 ± 41.2 vs. 79.2 ± 29.2; p = 0.001) levels were higher in patients with HPS than in those without HPS. An increase in Th17 (p <0.001) and T regulatory cells (p <0.001) was observed; the latter inversely correlated with plasma S1P levels. In the CBDL HPS model, fingolimod restored pulmonary vascular injury by increasing the arterial blood gas exchange and reducing systemic and pulmonary inflammation, resulting in improved survival (p = 0.02). Compared with vehicle treatment, fingolimod reduced portal pressure (p <0.05) and hepatic fibrosis and improved hepatocyte proliferation. It also induced apoptotic death in hepatic stellate cells and reduced collagen formation. Plasma S1P levels are low in patients with HPS and even more so in severe cases. Fingolimod, by improving pulmonary vascular tone and oxygenation, improves survival in a murine CBDL HPS model.