Introduction

Computer-aided diagnosis (CAD) systems are widely implemented in the medical field for the analysis of medical images. PolyDeep is a CAD system that has demonstrated its efficacy in lesion detection.

Aims & Methods

The aim of this study is to determine whether PolyDeep increases the adenoma detection rate (ADR) in the context of expert endoscopists participating in a regional, population-based colorectal cancer screening program in Spain. PolyDeep Advance 3 (NCT05513261) is a multicentric, randomized controlled clinical trial. For the primary endpoint, we aimed to assess whether the ADR is superior in PolyDeep-assisted colonoscopy compared to conventional colonoscopy. As secondary endpoints, we aimed to determine whether PolyDeep-assisted colonoscopy improves the polyp detection rate (PDR), serrated lesion detection rate (SLDR), advanced serrated lesion detection rate (ASLDR), and advanced adenoma detection rate (AADR). We included participants who underwent colonoscopy for screening (i.e. those with a positive faecal immunochemical occult blood test – FIT+) or for surveillance after resection of advanced adenomas. We compared both groups using the chi-square and t-Student test.

Results

Between November 2023 and May 2024, we recruited 857 patients, of whom 30 were excluded. The baseline characteristics of the 827 patients included in the analysis were similar (male: 59.3%, screening colonoscopy: 66.7%, mean age: 61.7 ± 6.5 years, withdrawal time: 10:49 ± 04:32 minutes, colorectal cancer incidence: 2.8%). However, there were significant differences between both groups in the Boston Bowel Preparation Scale (conventional colonoscopy: 7.7 ± 1.3 vs. PolyDeep-assisted colonoscopy: 7.4 ± 1.3, p = 0.01). Regarding ADR, we did not find statistically significant differences between conventional and PolyDeep-assisted colonoscopy (57.7% vs. 60.8%, p = 0.4). Similarly, for the secondary endpoints, there were no significant differences in PDR (68.7% vs. 69.6%, p = 0.8), SLDR (32.0% vs. 29.4%, p = 0.5), ASLDR (5.3% vs. 5.1%, p = 1.0), or AADR (24.5% vs. 23.6%, p = 0.8). In Table 1, we described lesion detection rates between both groups by indication (i.e., FIT+ or surveillance).

Conclusion

In the context of expert endoscopists PolyDeep did not increase the ADR.

Introduction

The detection and resection of adenomatous polyps during colonoscopy has been shown to be the most effective way to decrease the incidence of colorectal cancer and associated mortality ^{1, 2}, with adenoma detection rate (ADR) being a key quality performance indicator. While higher ADR is associated with lower post-colonoscopy colorectal cancer (PCCRC) rates, the latter varies significantly across the UKand the true ‘baseline’ ADR is unknown ^3,4. Meeting the rising demand for high quality colonoscopy is a significant challenge.
The use of computer assisted detection (CADe) is associated with a significant improvement in ADR across multiple clinical trials^{5, 6}, but real-world performance data is lacking.
We sought to evaluate effect on adenoma detection of implementing CADe in a ‘real-world’ setting, in the largest such study to date.

Aims & Methods

This was a randomised, stepped-wedge, open-label trial, using the GI Genius™ (GIG) module, powered to measure the site effect of CADe. A mix of District General Hospitals (DGH) and Teaching Hospitals (TH) were included, with an emphasis on ‘non-expert’ colonoscopists (lifetime colonoscopy <2000 at study start). ‘Expert’ colonoscopists were eligible (lifetime colonoscopy >2000 and ADR>40%).
Data collection was divided into three, 3-month phases: phase 1, baseline data collection with no CADe; phase 2, the intervention phase, where GIG-CADe was active; phase 3 where GIG-CADe was removed. The endoscopists and procedural information were recorded during the three phases. ADR across all three phases was analysed and compared.
Data will be forthcoming from 29 hospitals (England, Scotland, and Wales), n=124 colonoscopists and n>5000 patients, but we report herein results from the first 8 sites to complete the trial.
All patients undergoing routine (non-BCSP) colonoscopy were eligible, but those having colonoscopy for polyp or IBD surveillance were excluded. The study design was as close to real world practice as possible, with no other changes in procedure or technique.

Results

Across the 8 sites involved (2 TH), there were 34 endoscopists (64.7% non-expert vs 35.3% expert) and 1418 patients (56.43±16.9 years; 49.5% female) eligible for analysis. Comparative results (univariate analysis) through the three study phases are reported in table 1. When adjusted for bowel preparation and the endoscopists’ experience, there was a 9.53% increase in ADR in the CADe group (r² = 0.159; p = 0.0142).
​

Conclusion

NAIAD is the largest trial of CADe to date and, in this interim analysis, establishes a baseline UK ADR of approximately 27.0%. The use of GIG-CADe is associated with a significant increase in ADR, but this returns to baseline when the device is switched off. Further analysis of our result will inform the implementation of CADe in routine practice.

References

Reference:
1) M.F. Kaminski, J. Regula, E. Kraszewska, et al. Quality indicators for colonoscopy and the risk of interval cancer N Engl J Med, 362 (2010), pp. 1795-1803
2) M.F. Kaminski, P. Wieszczy, M. Rupinski, et al. Increased rate of adenoma detection associates with reduced risk of colorectal cancer and death Gastroenterology, 153 (2017), pp. 98-105
3) Burr NE, Derbyshire E, Taylor J, Whalley S, Subramanian V, Finan P J et al. Variation in post-colonoscopy colorectal cancer across colonoscopy providers in English National Health Service: population based cohort study BMJ 2019; 367: l6090 doi:10.1136/bmj.l6090
4) D.A. Corley, C.D. Jensen, A.R. Marks, et al. Adenoma detection rate and risk of colorectal cancer and death N Engl J Med, 370 (2014), pp. 1298-1306
5) Shaukat A, Lichtenstein DR, Somers SC, Chung DC, et al Computer-Aided Detection Improves Adenomas per Colonoscopy for Screening and Surveillance Colonoscopy: A Randomized Trial. Gastroenterology. 2022 Sep;163(3):732-741. doi: 10.1053/j.gastro.2022.05.028. Epub 2022 May 25. PMID: 35643173.
6) Wang, Pu, et al. "Effect of a deep-learning computer-aided detection system on adenoma detection during colonoscopy (CADe-DB trial): a double-blind randomised study." The lancet Gastroenterology & hepatology 5.4 (2020): 343-351.