Lavine JE, Schwimmer JB, Van Natta ML, Molleston JP, Murray KF, Rosenthal P, Abrams SH, et al. Effect of vitamin E or metformin for treatment of nonalcoholic fatty liver disease in children and adolescents: the TONIC randomized controlled trial. JAMA 2011;305:1659-1668. (Reprinted with permission).
Context: Nonalcoholic fatty liver disease (NAFLD) is the most common chronic liver disease in US children and adolescents and can present with advanced fibrosis or nonalcoholic steatohepatitis (NASH). No treatment has been established. Objective: To determine whether children with NAFLD would improve from therapeutic intervention with vitamin E or metformin. Design, Setting, and Patients: Randomized, double-blind, double-dummy, placebo-controlled clinical trial conducted at 10 university clinical research centers in 173 patients (aged 8-17 years) with biopsy-confirmed NAFLD conducted between September 2005 and March 2010. Interventions Daily dosing of 800 IU of vitamin E (58 patients), 1000 mg of metformin (57 patients), or placebo (58 patients) for 96 weeks. Main Outcome Measures: The primary outcome was sustained reduction in alanine aminotransferase (ALT) defined as 50% or less of the baseline level or 40 U/L or less at visits every 12 weeks from 48 to 96 weeks of treatment. Improvements in histological features of NAFLD and resolution of NASH were secondary outcome measures. Results: Sustained reduction in ALT level was similar to placebo (10/58; 17%; 95% CI, 9% to 29%) in both the vitamin E (15/58; 26%; 95% CI, 15% to 39%; P = .26) and metformin treatment groups (9/57; 16%; 95% CI, 7% to 28%; P = .83). The mean change in ALT level from baseline to 96 weeks was -35.2 U/L (95% CI, -56.9 to -13.5) with placebo vs -48.3 U/L (95% CI, -66.8 to -29.8) with vitamin E (P = .07) and -41.7 U/L (95% CI, -62.9 to -20.5) with metformin (P = .40). The mean change at 96 weeks in hepatocellular ballooning scores was 0.1 with placebo (95% CI, -0.2 to 0.3) vs -0.5 with vitamin E (95% CI, -0.8 to -0.3; P = .006) and -0.3 with metformin (95% CI, -0.6 to -0.0; P = .04); and in NAFLD activity score, -0.7 with placebo (95% CI, -1.3 to -0.2) vs -1.8 with vitamin E (95% CI, -2.4 to -1.2; P = .02) and -1.1 with metformin (95% CI, -1.7 to -0.5; P = .25). Among children with NASH, the proportion who resolved at 96 weeks was 28% with placebo (95% CI, 15% to 45%; 11/39) vs 58% with vitamin E (95% CI, 42% to 73%; 25/43; P = .006) and 41% with metformin (95% CI, 26% to 58%; 16/39; P = .23). Compared with placebo, neither therapy demonstrated significant improvements in other histological features. Conclusion: Neither vitamin E nor metformin was superior to placebo in attaining the primary outcome of sustained reduction in ALT level in patients with pediatric NAFLD. TRIAL REGISTRATION: clinicaltrials.gov Identifier: NCT00063635.
Nonalcoholic fatty liver disease (NAFLD) has rapidly become the most common form of chronic liver disease in children1 and is a major reason for referral to pediatric gastroenterologists and hepatologists. This dramatic increase in the prevalence of NAFLD is a direct consequence of the childhood obesity epidemic, with an estimated 10% of U.S. children having fatty liver.2 The spectrum of NAFLD ranges from the benign form of hepatic steatosis, with the accumulation of lipids in the liver, to the progressive form of nonalcoholic steatohepatitis (NASH), which is characterized by steatosis along with inflammation, hepatocyte injury, and variable degrees of fibrosis.3 It is important to recognize that NASH may present with a distinct histological pattern in children, mainly characterized by increased portal inflammation and fibrosis, as opposed to the predominantly lobular inflammation and perisinusoidal fibrosis observed in adults.4, 5
The progression of NASH to cirrhosis that requires liver transplantation during childhood is well documented,6 making the identification of effective therapy for this condition a pressing issue. Weight loss may improve the liver disease or even lead to the resolution of NAFLD in some children,7 but other than lifestyle advice on diet and exercise, there are no approved therapies for pediatric NAFLD. Insulin resistance (IR) and oxidative stress (OS) play a significant role in disease development and progression (Fig. 1), and therefore pilot studies have assessed the use of metformin (an insulin sensitizer) or vitamin E (an antioxidant) in the treatment of NAFLD in children with mixed results.8-11
Thus, the NASH Clinical Research Network performed the Treatment of NAFLD in Children (TONIC) trial,12 a multicenter, double-blind, double-placebo, randomized, clinical trial in which 173 pediatric patients received metformin (500 mg twice-daily), vitamin E (400 IU twice-daily), or placebo twice-daily for 96 weeks. All three groups received standardized recommendations regarding lifestyle modifications, use of other medications, alcohol avoidance, and management of comorbid illnesses. The primary outcome was sustained reduction in alanine aminotransferase (ALT) level, defined as reduction in serum ALT levels to below 50% of the baseline values or into the normal range (40 U/L or less) during the last 48 weeks of treatment. Secondary histologic outcomes included changes in total NAFLD activity score and individual histological features, and the resolution of NASH. Disappointingly, neither vitamin E nor metformin was superior to placebo in achieving sustained ALT reduction or in improving steatosis, lobular inflammation, or fibrosis scores. The only histologic feature of NASH that improved with both medications was ballooning. Compared to placebo, only vitamin E significantly improved NAFLD activity score and was associated with improved resolution of NASH on the repeat liver biopsy (58% versus 28%; P = 0.006). The investigators suggested that vitamin E should be considered in a subset of children with biopsy-proven NASH and evidence of hepatocellular ballooning degeneration, keeping in mind that the risk of biopsy may outweigh the benefits of therapy.
In reading the TONIC trial, one cannot help but compare its results to the adult PIVENS trial that used a similar approach to treating nondiabetic NASH patients with vitamin E or an insulin sensitizer (e.g., pioglitazone).13 Both medications in the adult trial were associated with highly significant reductions in steatosis, inflammation, ballooning, and aminotransferases levels, but only vitamin E (and not pioglitazone) significantly improved NASH. Although vitamin E appeared to be beneficial in both trials, the enthusiasm for its use to treat NASH is tempered by the fact that only half the patients had some histologic improvement, which leaves a significant percentage of patients to be classified as nonresponders to vitamin E. Furthermore, the duration of pharmacologic treatment needs to be defined because of concerns about increased overall mortality in adult patients on high dosage of vitamin E supplements.14 In an ancillary study of the TONIC trial, it was found that children with NAFLD consumed a diet that was insufficient in vitamin E, which may contribute to the pathophysiology of NAFLD.15 These findings suggest that a diet rich in vitamin E should be recommended to children with fatty liver, but the role of vitamin E supplements within the spectrum of NAFLD needs further assessment.
Although the TONIC trial is certainly a step forward in tackling the fast-growing problem of pediatric NAFLD, it has several limitations that should be mentioned. First, the primary outcome was sustained improvement in ALT, and this decision was based on the lack of sufficient information on the histology of NAFLD in children at the time of study design for sample-size calculations.16 However, it has been shown that in NAFLD, circulating aminotransferase levels poorly correlate with histology and tend to fluctuate significantly over time.17 The criteria used to define sustained ALT reduction was highly stringent, which may have contributed to the low response rate noted in all groups. Second, the dose for metformin (500 mg twice-daily) was based on a small pilot study that included only 10 children with NASH.10 This dose may have been too low to assess the real efficacy of metformin, as clearly evidenced by the lack of effects of metformin on insulin-resistance measurements in the study, the primary mechanism of action of this drug (Fig. 1). Third, given the fact that NASH has a unique histologic pattern in children, the effect of therapy on portal inflammation was not reported, and the criteria to define “resolution of NASH” was not clearly stated. Finally, an assessment of significant noninvasive markers of NASH in children, such as circulating cytokeratin-18 fragments and enhanced liver fibrosis test, was not done. A major concern with the design of most antioxidant intervention studies in NAFLD, to date, has been the lack of concomitant assessments of systemic measures of OS. A key reason for this omission has been the lack of validated OS measures that are associated with liver histology. Consequently, antioxidant supplementation studies have yet to utilize OS measures as an enrollment criterion to define populations with proven heightened levels of OS. Randomized vitamin E supplementation trials, including TONIC and PIVENS, have, thus far, failed to concomitantly demonstrate the magnitude of systemic antioxidant effect promoted in subjects included in the antioxidant arms of intervention trials and their correlations with liver outcomes. We have recently identified specific fatty acid oxidation products as novel noninvasive markers for OS in patients with NAFLD18 and have shown that they correlated with the major histologic features of NASH.19 Such markers could be used in the future to stratify patients according to their OS status and to monitor response to treatment, avoiding the need for repeated liver biopsies and their potential complications, as acknowledged by the investigators of the TONIC trial.
In conclusion, the TONIC trial generates more questions than answers. What patients are most likely to benefit from pharmacotherapy? Is a liver biopsy a prerequisite for starting treatment? And, if so, what histological criteria should be used to define NASH in children? How should one monitor response while on treatment? Is monotherapy with antioxidants or insulin sensitizers effective or should we focus on combination therapy that targets different processes that lead to disease progression? Despite all these unanswered questions, it is certain that the results of the TONIC trial will be used as an important building block in the greatly needed efforts to identify effective therapeutic options and reliable tools for risk stratification and monitoring disease progression and response to treatment in children with this highly common, potentially serious disease.