Polyethylene Glycol 3350 in the Treatment of Chronic Idiopathic Constipation: Post hoc Analysis Using FDA Endpoints

Abstract

Methods: This multicenter, double-blind, placebo-controlled, parallel-group trial included adults with chronic idiopathic constipation randomized to polyethylene glycol 3350 17 g (n = 204) or placebo (n = 100) once daily for 24 weeks. Post hoc analyses were performed using the US Food and Drug Administration endpoint (≥3 complete spontaneous bowel movements/week and an increase of ≥1 complete spontaneous bowel movement/week from baseline for ≥9/12 weeks, including 3 of the last 4 weeks) along with additional efficacy and safety outcomes.

Results: The proportion of patients meeting the new endpoint was significantly higher with polyethylene glycol 3350 vs placebo (42% vs 13%; P < 0.0001). Reductions in the mean number of hard/lumpy stools/week (-2.1 vs -0.9; P = 0.0014) and the weekly mean five-point cramping rating (-0.3 vs -0.1; P = 0.0272) also significantly favored polyethylene glycol 3350. The proportion of subjects with gastrointestinal adverse events decreased markedly after the first week of treatment in the polyethylene glycol 3350 group.

Conclusion: Using the current US Food and Drug Administration-recommended responder definition and other secondary outcomes, once-daily polyethylene glycol 3350 demonstrated substantial and sustained efficacy and safety over 24 weeks in patients with chronic idiopathic constipation. Trial Registration. The original trial was registered with https://ichgcp.net/clinical-trials-registry/NCT00153153" title="See in ClinicalTrials.gov">NCT00153153.

Conflict of interest statement

The authors declare the following relevant interests: SM is a consultant of Takeda Pharmaceuticals. AJL is a consultant of Takeda, Allakos, Arena, Gemelli, Ironwood, RedHill, Bayer, Alkermes, QOL Medical, Vibrant, Mylan, Progenity, and BioAmerica; Stock ownership is of Johnson & Johnson, Bristol Myers Squibb, Rock West, and Allurion. WDC is a consultant of AbbVie, Alfasigma, Allakos, Alnylam, Arena, Biomerica, Ferring, Gemelli, Ironwood, Nestlé, Phathom, Progenity, RedHill, Ritter, Salix/Valeant, QOL Medical, Takeda, Urovant, and Vibrant; Grants include Commonwealth Diagnostics International, Biomerica, Salix, QOL Medical, and Vibrant; Stock options are GI OnDEMAND, ModifyHealth, and Ritter.

Copyright © 2022 Stacy B. Menees et al.

Figures

Figure 1

CSBM and SBM response rates using different time intervals. Abbreviations: CIC, chronic idiopathic constipation; CMH, Cochran–Mantel–Haenszel; CSBM, complete spontaneous bowel movement; FDA, US Food and Drug Administration; PEG, polyethylene glycol; SBM, spontaneous bowel movement. †FDA CIC endpoint: weekly responder for ≥9 weeks out of the 12-week treatment period including 3 of the last 4 weeks of the period. P < 0.0001 for PEG 3350 vs placebo for all CSBM/SBM responder definitions. P value obtained from CMH chi-squared test, adjusting for pooled study sites, to ensure homogeneity/consistency across the treatment arms. A weekly CSBM/SBM responder is a patient who has at least three CSBMs/SBMs and at least one CSBM/SBM more than baseline in a week, and data are available on at least 4 days in a respective week. A continuous CSBM/SBM responder is a responder for ≥9 out of the first 12 weeks and ≥9 out of the last 12 weeks.

Figure 2

Mean weekly CSBMs (a) and SBMs (b) in subjects receiving PEG 3350 or placebo. †The change from baseline in the mean number of SBM and the mean number of CSBM each week was significantly greater with PEG 3350 versus placebo at week 1 (P=0.0040 and P=0.0013, respectively) and highly significantly greater (P < 0.0001 for both outcomes) each week for weeks 2 through 24. Abbreviations: CSBM, complete spontaneous bowel movement; PEG, polyethylene glycol; SBM, spontaneous bowel movement.

Figure 3

Mean change in the number of BMs associated with straining (a), hard/lumpy stools (b), cramping (c), and gas (d) over time. Abbreviations: BM, bowel movement; PEG, polyethylene glycol. †P < 0.05. Baseline is defined as the average of 2 weeks before treatment (i.e., weeks −2 and −1). Analysis based on available non-missing values in each week. P value derived from the analysis of variance with treatment, pooled-site, and their interaction as covariates.