Libertas: a phase II placebo-controlled study of NRL001 in patients with faecal incontinence showed an unexpected and sustained placebo response

Abstract

Purpose: Faecal incontinence (FI) is distressing, significantly reduces quality of life (QoL) and has few pharmacological treatments. The α1-adrenoceptor agonist NRL001 (1R,2S-methoxamine hydrochloride) improves anal sphincter tone. NRL001 efficacy was evaluated by changes in Wexner scores at week 4 vs. baseline in NRL001-treated patients compared with placebo. Impact of NRL001 on QoL and safety were also assessed.

Methods: Four hundred sixty-six patients received NRL001 (5, 7.5 or 10 mg) or placebo as suppository, once daily over 8 weeks. Wexner score, Vaizey score and QoL were analysed at baseline, week 4 and week 8. FI episodes and adverse events were recorded in diaries.

Results: At week 4, mean reductions in Wexner scores were -3.0, -2.6, -2.6 and -2.4 for NRL001 5, 7.5, 10 mg and placebo, respectively. All reduced further by week 8. As placebo responses also improved, there was no significant treatment effect at week 4 (p = 0.6867) or week 8 (p = 0.5005). FI episode frequency improved for all patients, but not significantly compared with placebo (week 4: p = 0.2619, week 8: p = 0.5278). All patients' QoL improved, but not significantly for all parameters (p > 0.05) except depression/self-perception at week 4 (p = 0.0102) and week 8 (p = 0.0069), compared with placebo. Most adverse events were mild and judged probably or possibly related to NRL001.

Conclusions: All groups demonstrated improvement in efficacy and QoL compared with baseline. NRL001 was well-tolerated without serious safety concerns. Despite the improvement in all groups, there was no statistically significant treatment effect, underlining the importance of relating results to a placebo arm.

Keywords: Alpha-1 receptor agonist; Episode frequency; Faecal incontinence; NRL001; Quality of life.

Fig. 1

The Libertas study design

Fig. 2

Consort patient flow diagram

Fig. 3

Patients’ Wexner scores at screening, week 4 and week 8 (mITT population). Symbols are the means, boxes are the interquartile range, lines within the boxes are the median and whiskers are the range. No significant treatment effect of the change from screening was observed at week 4 or week 8 (p values calculated using ANCOVA with screening as covariate)

Fig. 4

Patients’ mean FIQoL scores at screening, week 4 and week 8 (mITT population) for the four parameters: lifestyle (a), coping/behaviour (b), depression/self-perception (c) and embarrassment (d). p values for treatment effects calculated using ANCOVA with screening as covariate. *Pairwise differences between placebo and NRL001 5 mg were statistically significant at both week 4 and week 8 at the 95 % CI, calculated using Dunnett’s method within ANCOVA