Lucy B. Clayton

Background Development of bowel preparation products has been based upon colon cleansing rating by a local endoscopist. It is unclear how bowel preparation scales perform when centrally evaluated. Aims To evaluate the reliability of bowel preparation quality scales when assessed by central readers. Methods Four central readers evaluated 52 videos in triplicate, 2 weeks apart, during the entire endoscopic procedure (insertion/withdrawal of the colonoscope) and exclusively on colonoscope withdrawal using the Boston Bowel Preparation Scale (BBPS), Chicago Bowel Preparation scale, Harefield Cleansing Scale, Ottawa Bowel Preparation Quality Scale (OBPQS), Aronchick score, a visual analogue scale, and additional items proposed in a modified Research and Development/University of California Los Angeles appropriateness process. Reliability was assessed with intraclass correlation coefficients. Results Intraclass correlation coefficients (95% confidence interval) for inter-rater reliability of the quality scales ranged from 0.51 to 0.65 (consistent with moderate to substantial inter-rater reliability) during the entire procedure. Corresponding intraclass correlation coefficients for intra-rater reliability ranged from 0.69 to 0.77 (consistent with substantial intra-rater reliability). Reliability was highest in the right colon and lowest in the left colon. No differences were observed in reliability when assessed for the procedure overall (insertion/withdrawal) relative to assessment on withdrawal alone. Conclusion All five bowel preparation quality scales had moderate to substantial inter-rater reliability. Panelists considered the Aronchick score too simplistic for clinical trials and recognized that assessment of residual fluid in the Ottawa Bowel Preparation Quality Scale was not amenable to central assessment.

Colonoscopy requires colon cleansing. For this, many polyethylene glycol (PEG)-based preparations still require a high preparation-volume intake. Using an increased osmotic load with ascorbate (Asc), five new low-volume PEG-based bowel preparations (LVPEG) were tested for clinical proof of concept. This two-part, open-label study examined preparation-volumes of 1–1.25 L and total required fluid volumes of 2–3 L. Part 1, in healthy volunteers, used mean cumulative 24-h stool weight (target > 2750 g) to identify a lead candidate. Part 2 was endoscopist-blinded: patients undergoing screening colonoscopy were randomized before treatment with the selected lead, one of two variants of it, or the control 2 L PEG + Asc. Two primary endpoints were used for proof of concept demonstration: mean 24-h stool weight and bowel cleansing success (Harefield Cleansing Scale). A total of 120 subjects (30 per group) were enrolled/randomized 1:1:1:1 (max 40:60 gender ratio) per completed Part. In Part 1, LVPEG-3 achieved the largest mean stool weight (3399 g: P < 0.0001 vs target) and was selected for Part 2. In Part 2, stool weights exceeded the target, notably for LVPEG-4 (3215 g: P < 0.001), which achieved 100% cleansing success after a total required fluid intake of 2 L. The control achieved 90% cleansing success. Adverse events were few, gastrointestinal in nature and similar between groups. LVPEG-4 achieved a clinically useful combination of cleansing, safety/tolerability and low consumption volume: 1 L preparation + 1 L required additional fluid. Named NER1006, LVPEG-4 demonstrated clinical proof of concept and warrants further investigation. October 2012. Identifier: NCT01714466 . EudraCT: 2012–003052-37 The trial was prospectively registered.

Background Polyethylene glycol (PEG)-based bowel preparations are effective cleansers but many require high-volume intake. This phase 3, randomized, blinded, multicenter, parallel-group, central reader-assessed study assessed the 1 L PEG NER1006 bowel preparation vs. standard 2 L PEG with ascorbate (2LPEG). Methods Patients undergoing colonoscopy were randomized (1:1:1) to receive NER1006, as an evening/morning (N2D) or morning-only (N1D) regimen, or evening/morning 2LPEG. Cleansing was assessed using the Harefield Cleansing Scale (HCS) and the Boston Bowel Preparation Scale (BBPS). Primary end points were overall bowel cleansing success and high-quality cleansing in the right colon. Modified full analysis set (mFAS) and per protocol (PP) analyses were performed. Mean cleansing scores were analyzed post hoc. Results Of 849 randomized patients, efficacy was analyzed in the following patient numbers (mFAS/PP): total n = 822/670; N2D n = 275/220; N1D n = 275/218; 2LPEG n = 272/232. mFAS established noninferiority. PP showed superiority for N2D on overall success (97.3 % vs. 92.2 %; P = 0.014), and for N2D and N1D on right colon high-quality cleansing (N2D 32.3 % vs. 15.9 %, P < 0.001; N1D 34.4 % vs. 15.9 %, P < 0.001) vs. 2LPEG. Using HCS, N2D and N1D attained superior segmental high-quality cleansing (P ≤ 0.003 per segment). N2D showed superior mean segmental HCS scores (P ≤ 0.007 per segment). Both N2D and N1D achieved superior mean overall (P < 0.001 and P = 0.006) and right colon BBPS scores (P < 0.001 and P = 0.013). N2D demonstrated superior right colon polyp detection (P = 0.024). Adherence, tolerability, and safety were comparable between treatments. Conclusions NER1006 is the first low-volume preparation to demonstrate superior colon cleansing efficacy vs. standard 2LPEG with ascorbate, with comparable safety and tolerability. European Clinical Trials Database (EudraCT)2014-002185-78TRIAL REGISTRATION: Multicenter, randomized, parallel group, phase 3 trial 2014-002185-78 at https://eudract.ema.europa.eu/