Past, present and future perspectives in nonalcoholic fatty liver disease

April 2019 Arun Sanyal 

Fibrosis is widely recognized as the hallmark of disease progression in NASH, and the mechanisms underlying fibrogenic progression are currently an area of intense research. Although hepatic stellate cells are widely considered to be the primary cellular source of collagenous matrix in NASH, observations that portal inflammation is associated with fibrosis progression suggest a role for portal myofibroblasts as well⁴³.

In those who have developed cirrhosis, the rates of clinical decompensation have been reported to be ~3–4% annually⁹⁶. In 2017, two rigorously performed controlled trials of simtuzumab involving 219 and 258 individuals provided prospective data on the rates of progression from bridging fibrosis to cirrhosis, and from cirrhosis to decompensation in patients with NASH, respectively⁹⁷. Over a duration of 24 months, 21% of patients with NASH with bridging fibrosis developed cirrhosis and 19% of those with compensated cirrhosis due to NASH developed clinical decompensation. Approximately two-thirds of the patients included in the trial in those with cirrhosis had clinically significant portal hypertension defined by a hepatic venous pressure gradient >10 mmHg (ref.⁹⁸). This finding might explain the higher rate of portal hypertensive complications than those reported in prior studies⁹⁶.

Numerous studies have reported a very high prevalence of NAFLD in the general population, and it is estimated that ~25% of adults and 10% of children have NAFLD in the USA^68,69.

One less controversial aspect of NAFLD progression is the linkage between the disease and HCC. HCC can develop in the absence of cirrhosis, although the absolute risk is low (0.44 per 1,000 person-years of exposure)^1,92,93. Overall, NASH is increasing as an aetiology for HCC and is the second most common cause for HCC requiring transplantation evaluation in the USA⁹⁴. Of note, whereas the incidences of several obesity-associated cancers such as breast cancer and colon cancer have remained stable or declined over the past 10 years, the incidence of HCC has increased by 3% annually for the past 10 years⁹⁵. The growing prevalence of NASH is probably related to this phenomenon⁹⁵.

A high prevalence of NAFLD has also been reported in those of Hispanic origin, in elderly individuals and in those with diabetes^71,72.

The early descriptions of NASH led to efforts to identify the biological basis of NAFLD. It was quickly noted that steatosis could be induced by a high-fat diet or by leptin receptor deficiency in the ob/ob mouse¹⁴. Although these models did not produce steatohepatitis, it was observed that injections of bacterial lipopolysaccharide could induce inflammation along with steatosis and that steatotic hepatocytes developed heightened susceptibility to injury from TNF¹⁵. This work led Day and James to postulate the ‘two-hit’ hypothesis of NASH in 1998, in which steatosis was the first hit and exposure to inflammatory cytokines was the second hit, causing cell death and inflammation in NASH¹⁶. They considered fat accumulation to be fairly benign, although studies have since found fibrosis to develop even in those with steatosis alone¹⁷.

Using euglycaemic–hyperinsulinaemic clamps in humans, it was then demonstrated that, even in the absence of T2DM, glucose disposal was progressively impaired from healthy controls to those with histologically confirmed nonalcoholic fatty liver (NAFL) and then NASH with both low-dose and high-dose insulin infusions¹⁹.

Early studies were accompanied by a description of hepatic oxidative stress associated with NAFLD in humans¹⁹, which occurred with the development of NAFL and increased further in NASH. Mitochondrial injury and both morphological and functional changes in mitochondria in the liver in humans with NASH were described in 1999 and confirmed in 2001, raising the possibility that these changes were involved in driving the oxidative stress in the liver^19,22. 

Lipidomic studies in humans with NAFLD further revealed widespread perturbations in multiple lipid classes such as triglycerides, cholesterol and eicosanoids^29,30. Additional studies demonstrated that hepatic cholesterol synthesis is inappropriately increased in NASH despite an accumulation of free cholesterol, suggesting a defect in intracellular lipid sensing³¹.

Another landmark in the evolution of knowledge on NASH pathogenesis was the observation of an activated innate immune system in this condition. With the discovery of Toll-like receptors (TLRs) in the 1990s and the recognition of increased inflammation in response to their activation by bacterial lipopolysaccharide or intracellular products associated with cellular injury, increased TLR signalling was noted in murine models of NAFLD^34,35. It was further shown that palmitic acid could activate TLRs such as TLR2 and activate the inflammasome in NASH³⁶.

NASH has also been associated with increased systemic bacterial lipopolysaccharide levels, which are known to activate TLR4.

It is now recognized that inflammatory and pro-fibrogenic macrophages probably play a key part in disease progression^41,42, but gaps remain in our knowledge of what drives macrophage infiltration, whether liver-resident macrophages serve similar roles as bone-marrow-derived macrophages and what leads to a switch from a pro-inflammatory to a pro-fibrogenic macrophage profile. 

Introduction

A key finding in the past few years is that NASH, atherosclerosis and T2DM share many common pathogenic features such as ectopic fat deposits, inflammation, cell stress and death and fibrosis in affected organs, which are expected to drive common diagnostic tools that will inform the assessment of all of these diseases. Therapies that will beneficially affect all of these end-organ diseases will emerge as the first-line treatments of the condition.

Finally, it is becoming clear that weight loss, specifically reduced adiposity, is an important driver of histological improvement¹³³. The benefits of weight loss will extend beyond those expected from drug treatment of high-risk NASH to include decreased cardiovascular and metabolic outcomes and potentially even cancer. Public health strategies will need to recognize the common elements driving the growing health consequences of T2DM, NAFLD, heart failure and several obesity-related cancers to reduce the burden of disease related to all of these conditions in the future.

In the past 5 years, the potential to halt disease progression by the use of specific anti-inflammatory and anti-fibrotic agents has been tested. The 1-year findings from a 2-year trial of the CC-chemokine receptor 2 (CCR2)–CCR5 antagonist cenicriviroc, which was designed to reduce NASH fibrosis, has demonstrated a significant (P < 0.01) improvement in one-stage or greater fibrosis reduction in patients with NASH without changing upstream aspects of the disease, such as steatosis and hepatocyte ballooning injury¹²⁷. This trial further demonstrated that improvement in inflammation as defined by biochemical and molecular analyses is not recapitulated by traditional histological methods of inflammation assessment, raising questions about the validity of these conventional methods. Unfortunately, trials to inhibit fibrosis with the use of direct anti-fibrotics such as simtuzumab have been disappointing, suggesting that either more potent anti-fibrotic therapies are needed or that anti-fibrotic strategies should be combined with more metabolically targeted therapeutics. The current evidence, to date, supports the need for a metabolic anchor for the treatment of NASH.