In Children With Nonalcoholic Fatty Liver Disease, Cysteamine Bitartrate Delayed Release Improves Liver Enzymes but Does Not Reduce Disease Activity Scores

Abstract

Background & aims: No treatment for nonalcoholic fatty liver disease (NAFLD) has been approved by regulatory agencies. We performed a randomized controlled trial to determine whether 52 weeks of cysteamine bitartrate delayed release (CBDR) reduces the severity of liver disease in children with NAFLD.

Methods: We performed a double-masked trial of 169 children with NAFLD activity scores of 4 or higher at 10 centers. From June 2012 to January 2014, the patients were assigned randomly to receive CBDR or placebo twice daily (300 mg for patients weighing ≤65 kg, 375 mg for patients weighing >65 to 80 kg, and 450 mg for patients weighing >80 kg) for 52 weeks. The primary outcome from the intention-to-treat analysis was improvement in liver histology over 52 weeks, defined as a decrease in the NAFLD activity score of 2 points or more without worsening fibrosis; patients without biopsy specimens from week 52 (17 in the CBDR group and 6 in the placebo group) were considered nonresponders. We calculated the relative risks (RR) of improvement using a stratified Cochran-Mantel-Haenszel analysis.

Results: There was no significant difference between groups in the primary outcome (28% of children in the CBDR group vs 22% in the placebo group; RR, 1.3; 95% confidence interval [CI], 0.8-2.1; P = .34). However, children receiving CBDR had significant changes in prespecified secondary outcomes: reduced mean levels of alanine aminotransferase (reduction, 53 ± 88 U/L vs 8 ± 77 U/L in the placebo group; P = .02) and aspartate aminotransferase (reduction, 31 ± 52 vs 4 ± 36 U/L in the placebo group; P = .008), and a larger proportion had reduced lobular inflammation (36% in the CBDR group vs 21% in the placebo group; RR, 1.8; 95% CI, 1.1-2.9; P = .03). In a post hoc analysis of children weighing 65 kg or less, those taking CBDR had a 4-fold better chance of histologic improvement (observed in 50% of children in the CBDR group vs 13% in the placebo group; RR, 4.0; 95% CI, 1.3-12.3; P = .005).

Conclusions: In a randomized trial, we found that 1 year of CBDR did not reduce overall histologic markers of NAFLD compared with placebo in children. Children receiving CBDR, however, had significant reductions in serum aminotransferase levels and lobular inflammation. ClinicalTrials.gov no: NCT01529268.

Keywords: ALT; AST; Obesity; Pediatrics.

Figure 1. CONSORT Flow Diagram of CyNCh Trial Participants

Figure 2. Changes from baseline in liver enzymes and body mass index z-score according to treatment group

Mean values of changes from baseline during treatment with CBDR (88 patients) or Placebo (81 patients) for up to 52 weeks followed by a 24-week off-treatment period are shown. Error bars show 95% confidence intervals. P-values for the overall treatment effect of change over time on treatment (weeks 4–52) were derived from GEE linear regression, modeling change as a function of treatment group, visit code indicators, baseline value of the outcome, and treatment group by visit code interaction terms; p-values for the treatment effect at each visit were derived from linear regression, modeling change as a function of treatment group and the baseline value of the outcome. (A) Alanine aminotransferase (ALT) concentrations decreased in both the CBDR and placebo groups 4 weeks after initiating treatment and were sustained during the remaining 48-week treatment period but with a significantly greater decrease in patients treated with CBDR (p=0.07 for visits at weeks 4 through 52; p=0.02, 0.01, 0.02, 0.02, 0.02 at weeks 4, 12, 24, 36 and 52 respectively). ALT concentrations in the CBDR group were similar to placebo 24 weeks after treatment discontinuation (p=0.49). (B) Aspartate aminotransferase (AST) had a similar pattern (p=0.02 for visits at weeks 4 through 52; p=0.06, 0.006, 0.04, 0.007, 0.008, 0.28 at weeks 4, 12, 24, 36, 52 and 76, respectively). (C) γ-glutamyl transpeptidase (GGT) had a similar pattern with significantly greater decrease in GGT starting at 12 weeks (p=0.01 for visits at weeks 4 through 52; p=0.98, 0.03, 0.03, 0.11, 0.02, 0.24 at weeks 4, 12, 24, 36, 52 and 76, respectively). (D) Body mass index z-score did not have a significantly greater decrease for CBDR compared to placebo at any visit (p=0.54 for visits at weeks 4 through 52; p=0.48, 0.16, 0.11, 0.33, 0.11, 0.32 at weeks 4, 12, 24, 36, 52 and 76, respectively). The number of patients represented in the figures range from 73–88 in the CBDR group and 74–81 in the placebo group.