Medical therapy for nonalcoholic fatty liver disease in children and adolescents

Keywords::

• fatty liver disease
• metabolic syndrome
• nonalcoholic steatohepatitis
• obesity
• TONIC

Nonalcoholic fatty liver disease (NAFLD) is the most common liver disease in children living in developed nations. Estimations of pediatric NAFLD prevalence range from 3 to 10%, depending on the means of ascertainment, which usually employs surrogate serum markers or imaging modalities, such as serum alanine aminotransferase (ALT) or hepatic ultrasound. All surrogate means of assessing NAFLD are relatively insensitive, so estimates likely underestimate prevalence. Epidemiological studies from all over the world, including North and South America, Europe and Asia, uniformly demonstrate a pediatric preponderance in obese peripubertal males, which is often more severe in indigenous American and certain Hispanic and Asian populations [1,2]. NAFLD is also the most common liver disease in adults living in developed nations, although there are critical differences to be recognized other than age. A significant subset of pediatric NAFLD cases have histologic changes not found in adults. Although all NAFLD by definition shows macrovesicular steatosis in at least 5% of hepatocytes, pediatric NAFLD can show a pattern with pronounced portal and periportal features of mixed inflammation and fibrosis. They also have less likelihood of ballooning (indicative of cell injury), which, when present with fat and inflammation, is called nonalcoholic steatohepatitis (NASH). Children are less likely to have had significant or any ingestion of alcohol and, by virtue of their early manifestation of disease, most likely have more penetrant genetic traits [3].

Very little is known of the natural history of NAFLD, particularly NASH, in children. A limited study of five children with follow-up biopsies over a mean of 41 months demonstrated fibrosis progression in four children [4]. Information regarding the natural history of NASH is impaired by the need to perform serial biopsies in children, which is restricted to clinical (not research) need, and the lack of attention paid to pediatric NASH until this past decade. Furthermore, once children reach late adolescence, their care shifts to other providers who may not have an interest in this issue. Owing to the fact that children may have longer exposure to factors contributing to NASH progression, development of effective interventions is imperative to prevent recognized NASH sequelae, such as end-stage liver disease and cardiovascular morbidity [5]. The lack of prior trials for treatment of any liver disease in children using a histology end point rendered power analysis for sample size calculation difficult. Prior treatment trials in pediatric NAFLD were usually open label [6–8], or controlled using surrogate end points involving serum ALT or ultrasound [9]. Lack of information about the rate of progression of pediatric NASH also makes it difficult to know how long an intervention is needed to determine if treatment is effective.

The etiopathogenesis of NAFLD, particularly NASH, is thought to involve two critical steps. The first involves excess accumulation of triglyceride and possibly free fatty acids, of which some may be hepatotoxic. The second step probably involves exposure to oxidative stress in excess of available antioxidant defense [10]. Because most patients with NAFLD exhibit insulin resistance, including pediatric patients [11], strategies were targeted to reduce resistance. Thus, a primary strategy to treat NAFLD is to invoke lifestyle interventions that reduce caloric excess, promote a heart-healthy diet and institute more vigorous exercise. Secondary strategies to treat those unable, unwilling or unresponsive to such intervention have included medical treatments aimed at reducing insulin resistance with thiazolidinediones or metformin, or reducing oxidative stress with supplemental antioxidants.

Until recently, no randomized clinical trial using liver histology as a predesignated end point for any liver disease in children had ever been published. However, given that NASH is a clinicopathologic diagnosis, any rigorous trial testing therapeutic efficacy required that a biopsy for end-of-treatment response be done. The NIH-sponsored NASH Clinical Research Network (NASH CRN) recently published a multicenter randomized controlled trial comparing vitamin E or metformin to placebo in children aged 8–17 years of age at enrollment with biopsy-demonstrated NAFLD and serum ALT >60 IU/l. This trial was called Treatment of NAFLD In Children (TONIC). Treatment was continued for 96 weeks. Children with cirrhosis or diabetes were excluded. Owing to considerations regarding sample size calculations and duration of the trial without sufficient pilot information on natural histology of pediatric NAFLD histology, a primary end point was selected for sustained improvement in serum ALT [12]. This end point was felt by the study designers to be inferior to the major secondary end points: resolution in NASH in those who had NASH or borderline NASH at baseline, and reduction in the NAFLD activity score [13]. Fortunately, 149 of the 173 children in the study agreed with their parents to undergo an end-of-treatment percutaneous liver biopsy. Blind readings of biopsy slides by a consensus of NASH CRN pathologists demonstrated statistically significant changes in resolution of NASH, reduction in NAFLD activity score, and decrease in hepatocellular injury scores in those treated with vitamin E (400 IU by mouth twice daily) compared with those treated with placebo (p = 0.006, p = 0.02 and p = 0.006, respectively). There was no significant improvement in fibrosis, nor was there improvement in BMI or insulin resistance [14]. Although this study was considered to have a negative outcome for failing to reach the primary end point in superior ability to achieve sustained ALT reduction over placebo, it is remarkable that the more germane histologic end points were achieved.

The TONIC trial validates the role of vitamin E for efficacy found in an earlier randomized controlled trial performed by the NASH CRN on treatment of adults with NASH [15]. This study, called PIVENS, enrolled adults aged 18 years of age and older with biopsy-demonstrated NASH (not just NAFLD). The trial was larger, with 247 enrollees, and compared pioglitazone or vitamin E to placebo. Vitamin E, at 800 IU/day, achieved the primary end point, which was improvement in the NAFLD activity score of two or more points with no worsening of fibrosis and at least a one-point reduction in cellular injury. Thus, both TONIC and PIVENS demonstrate that vitamin E at 800 IU/day provides significant histologic improvement in children, adolescents and adults in biopsy-proven NASH.

Some important caveats need to be mentioned. Although no adverse events were reported relevant to obtaining the liver biopsy, it is unclear in children who should undergo a biopsy, and when. Only children with biopsy-proven NASH or borderline NASH demonstrated histologic improvement, so more study needs to be performed to define who might benefit without a biopsy. The potential risk and cost of liver biopsy needs to be considered in light of the potential benefit of vitamin E treatment, which may need to be sustained indefinitely to receive long-term benefit. Further development and validation of diagnostic, prognostic and treatment response biomarkers is much needed.

Not everyone who is treated with vitamin E benefits. Further research into who benefits and why is key. Other treatments that may be more efficacious that may also treat concomitant metabolic syndrome comorbidities need to be explored. Treatments in early-phase studies deemed good candidates need to be validated using the same rigor as trials in adults.

The development of noninvasive biomarkers to assess the likelihood of response to treatment and following response to treatment will be useful in avoiding liver biopsy and identifying those who will benefit from starting and continuing vitamin E treatment. Future Phase IIb or III trials in children and adolescents should utilize clinically meaningful primary histology end points and attempt to minimize inflammation, liver cell injury and fibrosis [16].

Financial & competing interests disclosure

JE Lavine is a consultant for Quark Pharmaceuticals and received investigator-initiated grant support from Raptor Pharmaceuticals. Grant support was received from the NIH (U01DK61734 to JE Lavine) for the research described in this review. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

No writing assistance was utilized in the production of this manuscript.