IMPROVED LESION DETECTION WITH HIGH-QUALITY VERSUS ADEQUATE CLEANSING SUCCESS: A POST HOC ANALYSIS OF 1749 PATIENTS IN RANDOMISED CLINICAL TRIALS USING THE HAREFIELD CLEANSING SCALE

Aims:

Lesion detection requires colon cleansing. On the Harefield Cleansing Scale (HCS), success equals a minimal segmental score of 2/4 in all five segments; cumulative 10/20. Benefits of high-quality cleansing are debated. This post hoc analysis of randomised clinical trials assessed lesion detection in cumulative segmental score groups permitting real-world segmental cleansing score variability in each patient.

Methods:

Three similar phase 3 trials assessed the colon cleansing efficacy and safety of 1L NER1006 (PLENVU) versus standard bowel preparations. Cleansing quality assessment was standardised with treatment-blinded central readers using the HCS. Our treatment-independent analysis included all patients with fully documented segmental cleansing scores and lesion counts. Three cumulative HCS score groups 0 – 10 (failed to adequate), 11 – 13 (adequate with some high-quality), and 14 – 20 (mostly high-quality) were stratified for maximal samples with a comparable size. Polyp detection rates (PDR), adenoma detection rates (ADR) and the mean number of polyps (MPP) or adenomas (MAP) per patient were analysed. One-sided t-tests were used to identify any differences in lesion detection versus the highest cleansing quality group HCS 14 – 20.

Results:

From 1985 randomised patients, 1749 were included (Table). The highest quality cleansing group HCS 14 – 20 (n = 551) was associated with a significantly higher ADR, MPP and MAP than the medium quality cleansing group HCS 11 – 13 (n = 581). HCS 14 – 20 was also associated with a significantly higher PDR, ADR, MPP and MAP than the failed to adequate group HCS 0 – 10 (n = 617).

Tab. 1:

Cumulative HCS scores (0 – 20) versus lesion detection in the overall colon

Cumulative HCS segmental scores 0 – 20	Mostly high-quality 14 – 20	Adequate with some high-quality 11 – 13	Failed to adequate 0 – 10
PDR, n/N (%); P-value vs. HCS 14 – 20	263/551 (0.48)	250/581 (0.43) P = 0.056	261/617 (0.42) P = 0.031
ADR, n/N (%); P-value vs. HCS 14 – 20	189/551 (0.34)	156/581 (0.27) P = 0.003	164/617 (0.27) P = 0.002
MPP, mean (SD); P-value vs. HCS 14 – 20	1.28 (2.81)	0.99 (1.72) P = 0.017	1.14 (2.35) P = 0.171
MAP, mean (SD); P-value vs. HCS 14 – 20	0.75 (2.11)	0.54 (1.26) P = 0.022	0.52 (1.19) P = 0.010

Conclusions:

In large and well-balanced sample sizes of clinical practice relevance, high-quality colon cleansing improves lesion detection over lower cleansing qualities.