A double-blind, randomized, placebo-controlled trial studying the effects of Saccharomyces boulardii on the gastrointestinal tolerability, safety, and pharmacokinetics of miglustat

Background: Gastrointestinal (GI) disturbances such as diarrhea and flatulence are the most frequent adverse effects associated with miglustat therapy in type 1 Gaucher disease (GD1) and Niemann-Pick disease type C (NP-C), and the most common recorded reason for stopping treatment during clinical trials and in clinical practice settings. Miglustat-related GI disturbances are thought to arise from the inhibition of intestinal disaccharidases, mainly sucrase isomaltase. We report the effects of a co-administered dietary probiotic, S. boulardii, on the GI tolerability of miglustat in healthy adult subjects. Methods: In a double-blind, placebo-controlled, two-period, two-treatment cross-over trial, healthy adult male and female subjects were randomly allocated to treatment sequences, A-B and B-A (treatment A - miglustat 100 mg t.i.d. + placebo; treatment B - miglustat 100 mg t.i.d. + S. boulardii [500 mg, b.i.d.]). GI tolerability data were collected in patient diaries. The primary endpoint was the total number of 'diarrhea days' (>= 3 loose stools within a 24-h period meeting Bristol Stool Scores [BSS] 6-7) based on WHO criteria. Secondary endpoints comprised numerous other diarrhea and GI tolerability indices. Results: Twenty-one subjects received randomized therapy in each treatment sequence (total N = 42), and overall, 37 (88 %) subjects completed the study. The total number of diarrhea days was <1.5 for both treatment sequences, and approximately 60 % of subjects did not experience diarrhea during either treatment period. The mean (SD) number of diarrhea days was lower with miglustat + S. boulardii (0.8 [2.4] days) than with miglustat + placebo (1.3 [2.4] days), but the paired treatment difference was not statistically significant (-0.5 [2.4] days; p = 0.159). However, a significant treatment difference (-0.7 [1.9]; p < 0.05) was identified after post hoc exclusion of a clear outlier who had a very high number of diarrhea days (n = 13) and inconsistent GI tolerability reporting. The incidence of the GI AEs was higher with miglustat + placebo (82 %) than with miglustat + S. boulardii (73 %). There were no between-treatment differences in miglustat pharmacokinetics. Conclusions: Although the primary endpoint was not met, the results of the post-hoc analysis suggest that co-administration of miglustat with S. boulardii might improve GI tolerability.