Small but mighty: Platelets in NASH and other chronic liver diseases

Recent studies have identified several pleiotropic effects of platelets on liver homeostasis and disease distinct from their long-recognized roles in hemostasis and coagulation. Concerns for bleeding have traditionally limited the use of anti-platelet agents in advanced liver disease. However, recent evidence increasingly suggests that pro-thrombotic tendencies may exceed the bleeding diathesis traditionally associated with advanced liver disease. This implies that anti-platelet therapy (APT) may in fact be safely used in certain patients with chronic liver disease. The therapeutic potential of platelet modulation is of particular interest given recent studies which implicate platelets in the pathogenesis of viral hepatitis, liver fibrosis, and cholestatic liver injury (Figure 1). Studies suggest that platelets orchestrate and propagate inflammatory responses to several different injuries within the liver¹. Platelets derived factors enhance tumor cell proliferation in the setting of HCC and can also impact communication between tumor and the stroma to promote metastases². Once recruited to the liver, platelets can interact with neutrophils to promote formation of neutrophil extracellular traps, or NETs, within liver sinusoids. Sinusoidal NETs promote formation of sinusoidal microvascular thrombosis and portal hypertension³. Although they have been implicated in the pathophysiology of several liver diseases, platelets are also integral to liver regeneration and repair¹. This suggests that platelets play a nuanced, complex role in liver homeostasis and pathophysiology which merits further investigation.

Figure 1.

Platelets play several roles in liver disease and pathophysiology. Degranulation and release of growth factors and serotonin promotes liver regeneration after resection and also drives tumor angiogenesis and metastases. Interactions with immune cells such as T cell, Kupffer cells, and neutrophils compound several forms of liver disease including NASH and viral hepatitis.

Platelets have been implicated in several vascular consequences of the metabolic syndrome, and preliminary studies have found that daily aspirin use is associated with milder histologic features of non-alcoholic steatohepatitis (NASH) and delayed progression to advanced fibrosis⁴. Malehmir et al. use several dietary and genetic murine models to clarify the role of platelets in NASH⁵. The authors employ genetic and pharmacologic methods, including aspirin and clopidogrel (Asa-clo), to study specific platelet functions, including degranulation, adhesion, and aggregation. Asa-clo are clinically-relevant APTs which are commonly used in the management of patients with coronary artery disease. The authors initially demonstrate increased intrahepatic platelet numbers in animals with NASH compared to normal controls or animals with non-alcoholic fatty liver disease (NAFLD) alone. They identify platelets as the first non-resident cell to infiltrate the liver in early NASH and correlate the timing of platelet infiltration with the appearance of morphologic evidence of NASH including steatosis, hepatocyte swelling, and reduced sinusoidal diameter. The authors extend this observation to humans with NAFLD or NASH whose intrahepatic platelet numbers are significantly increased compared with healthy controls. In a pilot case study, the impact of APT on liver steatosis was assessed in patients undergoing cardiac catheterization procedures due to their heightened risk of NAFLD. The authors importantly controlled for statin use and found that treatment with APT decreased liver volume and fat mass. However, this significant finding in human subjects is limited by small sample size. These initial results suggest that platelets infiltrate early in the course of NASH and thus have a potential role as a chemopreventive agent in patients with the metabolic syndrome and other risk factors for NASH.

In mice fed a NASH diet, APT significantly attenuated liver steatosis, AST, ALT, liver/body weight ratio, and platelet number without altering body weight. Treatment with APT also led to a significant decrease in the NAFLD activity score (NAS). Among the most significant findings of this study is the marked protective effect that APT confers against hepatocellular carcinoma (HCC). While 13 of 51 (25%) mice fed a NASH diet developed HCC at 12 months, no APT-treated mice (n=20) developed HCC. Animals treated with a lower dose of APT were less protected against HCC, leading the authors to suggest that APT may be more efficacious if dosed according to actual as opposed to ideal body weights. These results corroborate prior studies which establish a significant association between increased body-mass index (BMI) and impaired platelet inhibition after treatment with clopidogrel⁶. These studies highlight the need for further investigations of the impact of obesity and the metabolic syndrome on the pharmacokinetics and efficacy of APT. It will be particularly important to determine if bleeding risk limits weight-based dosing in obese patients with advanced liver disease secondary to NASH.

The authors found that APT altered the immune cell populations within the liver and hypothesize that APT attenuates the development of NASH and HCC by modulating the local immune environment. NASH is associated with alterations in Kupffer cell (KC) polarization and function⁷. The authors found that the intrahepatic population of KCs in livers of mice on a NASH diet treated with APT more closely resembled the KC population in livers of mice fed a normal diet. The authors then treated mice with clodronate liposomes (CLL) to decrease KCs within the liver. Treatment with CLL led to a significant decrease in both intrahepatic platelet numbers and the NAS. They therefore conclude that KCs recruit platelets to the liver in the setting of NASH.

GPIbα is the major ligand-binding component of the glycoprotein GPIb-IX-V complex on platelets which binds activated platelets, endothelial cells, and Kupffer cells. Inhibition of platelet-derived GPIbα inhibited platelet recruitment and accumulation in the liver and abrogated development of NASH, confirming the importance of KC-mediated platelet recruitment and adhesion in the pathogenesis of NASH. This finding further corroborates the rationale of investigating therapies which target KCs, including peroxisome proliferator-activator receptor (PPAR) agonists, in mitigating NASH.

The authors mechanistically explore the ways that platelets propagate NASH and find an integral role for platelet α-granules in the pathogenesis of NASH. Mice with genetic absence of α-granules (Nbeal^−/−) demonstrate similar weight gain and intrahepatic platelet numbers on a NASH diet but have improved glucose tolerance and lower NAS, serum cholesterol, and serum aminotransferases. These results suggest that, in addition to platelet adhesion to KCs, platelet degranulation is a pivotal step in the pathogenesis of NASH. Platelet α-granules contain approximately 300 bioactive proteins, including cytokines, chemokines, adhesion molecules, and growth factors¹. Further studies to identify the specific proteins which mediate this effect can lead to the development of more specific therapies.

The pathogenesis of NAFLD has long been recognized as a complex process which is influenced by nutritional, genetic, metabolic, and inflammatory factors. In addition to its impact on serum aminotransferases and histologic progression of NASH, the authors prove that APT influences peripheral metabolic signals. Treatment with APT improved glucose tolerance and also led to a reduction in serum total, low-density lipoprotein (LDL), and high-density lipoprotein (HDL) cholesterol in mice fed a NASH diet. The positive impact of APT on cardiovascular risk factors such as LDL is clinically relevant as cardiovascular disease is a leading cause of mortality in patients with NASH. Portal vein thrombosis is a common complication of cirrhosis which increases risk for poorer outcomes after liver transplantation. In addition, recent studies identify sinusoidal microvascular thromboses as integral to the pathophysiology of portal hypertension³. In addition to beneficial effects noted on serum cholesterol and fasting glucose in this study, it is conceivable that treatment with APT may lead to beneficial off-target effects by decreasing the risk of both microvascular and macrovascular thrombosis.

Overall, this study highlights an important role for platelets in the pathogenesis of NASH and NASH-associated HCC. The authors identify platelet infiltration early in the pathogenesis of NASH. This suggests that APT may be particularly impactful as a chemopreventive agent in the early stages of NASH which are less likely to be associated with platelet dysfunction and thrombocytopenia. The significant impact of APT on HCC risk highlights the importance of clarifying the safety of APT in patients with advanced fibrosis who are at increased risk of HCC. The therapeutic potential of inhibiting the platelet GPIbα receptor is of particular interest given low risk of spontaneous bleeding associated with GPIbα-antibody treatment⁸.