Multicentre study to assess the performance of an artificial intelligence instrument to support qualitative diagnosis of colorectal polyps.

OBJECTIVE: Computer-aided diagnosis (CAD) using artificial intelligence (AI) is expected to support the characterisation of colorectal lesions, which is clinically relevant for efficient colorectal cancer prevention. We conducted this study to assess the diagnostic performance of commercially available CAD systems. METHODS: This was a multicentre, prospective performance evaluation study. The endoscopist diagnosed polyps using white light imaging, followed by non-magnified blue light imaging (non-mBLI) and mBLI. AI subsequently assessed the lesions using non-mBLI (non-mAI), followed by mBLI (mAI). Eventually, endoscopists made the final diagnosis by integrating the AI diagnosis (AI+endoscopist). The primary endpoint was the accuracy of the AI diagnosis of neoplastic lesions. The diagnostic performance of each modality (sensitivity, specificity and accuracy) and confidence levels were also assessed. RESULTS: Overall, 380 lesions from 139 patients were included in the analysis. The accuracy of non-mAI was 83%, 95% CI (79% to 87%), which was inferior to that of mBLI (89%, 95% CI (85% to 92%)) and mAI (89%, 95% CI (85% to 92%)). The accuracy (95% CI) of diagnosis by expert endoscopists using mAI (91%, 95% CI (87% to 94%)) was comparable to that of expert endoscopists using mBLI (91%, 95% CI (87% to 94%)) but better than that of non-expert endoscopists using mAI (83%, 95% CI (75% to 90%)). The level of confidence in making a correct diagnosis was increased when using magnification and AI. CONCLUSIONS: The diagnostic performance of mAI for differentiating colonic lesions is comparable to that of endoscopists, regardless of their experience. However, it can be affected by the use of magnification as well as the endoscopists' level of experience.

Microbiota-induced S100A11-RAGE axis underlies immune evasion in right-sided colon adenomas and is a therapeutic target to boost anti-PD1 efficacy.

BACKGROUND: Tumourigenesis in right-sided and left-sided colons demonstrated distinct features. OBJECTIVE: We aimed to characterise the differences between the left-sided and right-sided adenomas (ADs) representing the early stage of colonic tumourigenesis. DESIGN: Single-cell and spatial transcriptomic datasets were analysed to reveal alterations between right-sided and left-sided colon ADs. Cells, animal experiments and clinical specimens were used to verify the results. RESULTS: Single-cell analysis revealed that in right-sided ADs, there was a significant reduction of goblet cells, and these goblet cells were dysfunctional with attenuated mucin biosynthesis and defective antigen presentation. An impairment of the mucus barrier led to biofilm formation in crypts and subsequent bacteria invasion into right-sided ADs. The regions spatially surrounding the crypts with biofilm occupation underwent an inflammatory response by lipopolysaccharide (LPS) and an apoptosis process, as revealed by spatial transcriptomics. A distinct S100A11+ epithelial cell population in the right-sided ADs was identified, and its expression level was induced by bacterial LPS and peptidoglycan. S100A11 expression facilitated tumour growth in syngeneic immunocompetent mice with increased myeloid-derived suppressor cells (MDSC) but reduced cytotoxic CD8+ T cells. Targeting S100A11 with well-tolerated antagonists of its receptor for advanced glycation end product (RAGE) (Azeliragon) significantly impaired tumour growth and MDSC infiltration, thereby boosting the efficacy of anti-programmed cell death protein 1 therapy in colon cancer. CONCLUSION: Our findings unravelled that dysfunctional goblet cells and consequential bacterial translocation activated the S100A11-RAGE axis in right-sided colon ADs, which recruits MDSCs to promote immune evasion. Targeting this axis by Azeliragon improves the efficacy of immunotherapy in colon cancer.

Cold versus hot snare endoscopic mucosal resection for large (≥15 mm) flat non-pedunculated colorectal polyps: a randomised controlled trial.

BACKGROUND AND AIMS: Conventional hot snare endoscopic mucosal resection (H-EMR) is effective for the management of large (≥20 mm) non-pedunculated colon polyps (LNPCPs) however, electrocautery-related complications may incur significant morbidity. With a superior safety profile, cold snare EMR (C-EMR) of LNPCPs is an attractive alternative however evidence is lacking. We conducted a randomised trial to compare the efficacy and safety of C-EMR to H-EMR. METHODS: Flat, 15-50 mm adenomatous LNPCPs were prospectively enrolled and randomly assigned to C-EMR or H-EMR with margin thermal ablation at a single tertiary centre. The primary outcome was endoscopically visible and/or histologically confirmed recurrence at 6 months surveillance colonoscopy. Secondary outcomes were clinically significant post-EMR bleeding (CSPEB), delayed perforation and technical success. RESULTS: 177 LNPCPs in 177 patients were randomised to C-EMR arm (n=87) or H-EMR (n=90). Treatment groups were equivalent for technical success 86/87 (98.9%) C-EMR versus H-EMR 90/90 (100%); p=0.31. Recurrence was significantly greater in C-EMR (16/87, 18.4% vs 1/90, 1.1%; relative risk (RR) 16.6, 95% CI 2.24 to 122; p<0.001).Delayed perforation (1/90 (1.1%) vs 0; p=0.32) only occurred in the H-EMR group. CSPEB was significantly greater in the H-EMR arm (7/90 (7.8%) vs 1/87 (1.1%); RR 6.77, 95% CI 0.85 to 53.9; p=0.034). CONCLUSION: Compared with H-EMR, C-EMR for flat, adenomatous LNPCPs, demonstrates superior safety with equivalent technical success. However, endoscopic recurrence is significantly greater for cold snare resection and is currently a limitation of the technique. TRIAL REGISTRATION NUMBER: NCT04138030.

Impact of AI-aided colonoscopy in clinical practice: a prospective randomised controlled trial.

OBJECTIVE: Colorectal cancer (CRC) has a significant role in cancer-related mortality. Colonoscopy, combined with adenoma removal, has proven effective in reducing CRC incidence. However, suboptimal colonoscopy quality often leads to missed polyps. The impact of artificial intelligence (AI) on adenoma and polyp detection rate (ADR, PDR) is yet to be established. DESIGN: We conducted a randomised controlled trial at Sahlgrenska University Hospital in Sweden. Patients underwent colonoscopy with or without the assistance of AI (AI-C or conventional colonoscopy (CC)). Examinations were performed with two different AI systems, that is, Fujifilm CADEye and Medtronic GI Genius. The primary outcome was ADR. RESULTS: Among 286 patients, 240 underwent analysis (average age: 66 years). The ADR was 42% for all patients, and no significant difference emerged between AI-C and CC groups (41% vs 43%). The overall PDR was 61%, with a trend towards higher PDR in the AI-C group. Subgroup analysis revealed higher detection rates for sessile serrated lesions (SSL) with AI assistance (AI-C 22%, CC 11%, p=0.004). No difference was noticed in the detection of polyps or adenomas per colonoscopy. Examinations were most often performed by experienced endoscopists, 78% (n=86 AI-C, 100 CC). CONCLUSION: Amidst the ongoing AI integration, ADR did not improve with AI. Particularly noteworthy is the enhanced detection rates for SSL by AI assistance, especially since they pose a risk for postcolonoscopy CRC. The integration of AI into standard colonoscopy practice warrants further investigation and the development of improved software might be necessary before enforcing its mandatory implementation. TRIAL REGISTRATION NUMBER: NCT05178095.

Comparison of adenoma miss rate and adenoma detection rate between conventional colonoscopy and colonoscopy with second generation distal attachment cuff: a multicenter, randomized, back-to-back trial.

BACKGROUND AND AIMS: Endocuff VisionTM has been designed to enhance mucosal visualization thereby improving detection of (pre-)malignant colorectal lesions. This multicenter, international, back-to-back, randomized colonoscopy trial compared adenoma detection rate (ADR) and adenoma miss rate (AMR) between Endocuff Vision-assisted colonoscopy (EVC) and conventional colonoscopy (CC). METHODS: Patients aged 40-75 years referred for non-immunochemical fecal occult blood test-based screening, surveillance, or diagnostic colonoscopy were included at ten hospitals and randomized into four groups: Group 1; 2xCC, Group 2; CC followed by EVC, Group 3; EVC followed CC and Group 4; 2xEVC. Primary outcomes included ADR and AMR. RESULTS: A total of 717 patients were randomized of which 661 patients (92.2%) had one and 646 (90.1%) patients had two completed back-to-back colonoscopies. EVC did not significantly improve ADR compared to CC (41.1% [95%-CI;36.1-46.3] versus 35.5% [95%-CI;30.7-40.6], respectively, P=0.125), but EVC did reduced AMR by 11.7% (29.6% [95%-CI;23.6-36.5] versus 17.9% [95%-CI;12.5-23.5], respectively, P=0.049). AMR of 2xCC compared to 2xEVC was also not significantly different (25.9% [95%-CI;19.3-33.9] versus 18.8% [95%-CI;13.9-24.8], respectively, P=0.172). Only 3.7% of the polyps missed during the first procedures had advanced pathologic features. Factors affecting risk of missing adenomas were age (P=0.002), Boston Bowel Preparation Scale (P=0.008) and region where colonoscopy was performed (P<0.001). CONCLUSIONS: Our trial shows that EVC reduces the risk of missing adenomas but does not lead to a significant improved ADR. Remarkably, 25% of adenomas are still missed during conventional colonoscopies, which is not different from miss rates reported 25 years ago; reassuringly, advanced features were only found in 3.7% of these missed lesions. TRAIL REGISTRATION NUMBER: NCT03418948.

Treatment of adenoma recurrence after endoscopic mucosal resection.

OBJECTIVE: Residual or recurrent adenoma (RRA) after endoscopic mucosal resection (EMR) of large non-pedunculated colorectal polyps (LNPCPs) of ≥20 mm is a major limitation. Data on outcomes of the endoscopic treatment of recurrence are scarce, and no evidence-based standard exists. We investigated the efficacy of endoscopic retreatment over time in a large prospective cohort. DESIGN: Over 139 months, detailed morphological and histological data on consecutive RRA detected after EMR for single LNPCPs at one tertiary endoscopy centre were prospectively recorded during structured surveillance colonoscopy. Endoscopic retreatment was performed on cases with evidence of RRA and was performed predominantly using hot snare resection, cold avulsion forceps with adjuvant snare tip soft coagulation or a combination of the two. RESULTS: 213 (14.6%) patients had RRA (168 (78.9%) at first surveillance and 45 (21.1%) thereafter). RRA was commonly 2.5-5.0 mm (48.0%) and unifocal (78.7%). Of 202 (94.8%) cases which had macroscopic evidence of RRA, 194 (96.0%) underwent successful endoscopic therapy and 161 (83.4%) had a subsequent follow-up colonoscopy. Of the latter, endoscopic therapy of recurrence was successful in 149 (92.5%) of 161 in the per-protocol analysis, and 149 (73.8%) of 202 in the intention-to-treat analysis, with a mean of 1.15 (SD 0.36) retreatment sessions. No adverse events were directly attributable to endoscopic therapy. Further RRA after endoscopic therapy was endoscopically treatable in most cases. Overall, only 9 (4.2%, 95% CI 2.2% to 7.8%) of 213 patients with RRA required surgery.Thus 159 (98.8%, 95% CI 95.1% to 99.8%) of 161 cases with initially successful endoscopic treatment of RRA and follow-up remained surgery-free for a median of 13 months (IQR 25.0) of follow-up. CONCLUSIONS: RRA after EMR of LNPCPs can be effectively treated using simple endoscopic techniques with long-term adenoma remission of >90%; only 16% required retreatment. Therefore, more technically complex, morbid and resource-intensive endoscopic or surgical techniques are required only in selected cases. TRIAL REGISTRATION NUMBERS: NCT01368289 and NCT02000141.

Screening of normal endoscopic large bowel biopsies with interpretable graph learning: a retrospective study.

OBJECTIVE: To develop an interpretable artificial intelligence algorithm to rule out normal large bowel endoscopic biopsies, saving pathologist resources and helping with early diagnosis. DESIGN: A graph neural network was developed incorporating pathologist domain knowledge to classify 6591 whole-slides images (WSIs) of endoscopic large bowel biopsies from 3291 patients (approximately 54% female, 46% male) as normal or abnormal (non-neoplastic and neoplastic) using clinically driven interpretable features. One UK National Health Service (NHS) site was used for model training and internal validation. External validation was conducted on data from two other NHS sites and one Portuguese site. RESULTS: Model training and internal validation were performed on 5054 WSIs of 2080 patients resulting in an area under the curve-receiver operating characteristic (AUC-ROC) of 0.98 (SD=0.004) and AUC-precision-recall (PR) of 0.98 (SD=0.003). The performance of the model, named Interpretable Gland-Graphs using a Neural Aggregator (IGUANA), was consistent in testing over 1537 WSIs of 1211 patients from three independent external datasets with mean AUC-ROC=0.97 (SD=0.007) and AUC-PR=0.97 (SD=0.005). At a high sensitivity threshold of 99%, the proposed model can reduce the number of normal slides to be reviewed by a pathologist by approximately 55%. IGUANA also provides an explainable output highlighting potential abnormalities in a WSI in the form of a heatmap as well as numerical values associating the model prediction with various histological features. CONCLUSION: The model achieved consistently high accuracy showing its potential in optimising increasingly scarce pathologist resources. Explainable predictions can guide pathologists in their diagnostic decision-making and help boost their confidence in the algorithm, paving the way for its future clinical adoption.

Morphology, Histopathology, and Anatomical Distribution of Sporadic Colorectal Polyps in Chinese Patients.

Background and Aims: There are limited data regarding the morphology, histopathology, and anatomical distribution of sporadic colorectal polyps in Chinese patients. We evaluated these characteristics of sporadic polyps to guide the endoscopic detection and excision of colorectal polyps. Methods: This was a retrospective single-center observational study involving 7381 patients with sporadic colorectal polyps. All polyps were removed endoscopically. The morphology and histopathology of polyps were evaluated according to the Paris classification and the World Health Organization classification, respectively. Results: A total of 22,174 polyps removed endoscopically from 7322 patients were included. In the sigmoid colon, 24.70% of colorectal polyps occurred, followed by the transverse colon (18.58%) in frequency. 0-Is type polyps accounted for 60.60% of all sporadic colorectal polyps. Polyps with 0-Ip, 0-Isp, and 0-IIa types were frequently found in the sigmoid colon, but laterally spreading lesions usually occurred in the ascending colon (24.61%) and rectum (20.51%). Irrespective of the Paris classification and anatomical location, as the polyps enlarge, the proportion of adenomatous polyps gradually increases while the proportion of serrated lesions decreases. Polyps with size ≥1 cm located in the left-sided colon were more likely to have villous/tubulovillous or high-grade dysplasia histology than those located in the right-sided colon, and about 1% of them were demonstrated with adenocarcinoma. Conclusion: Sigmoid colon should be detected adequately during colonoscopy, and polyps with size ≥1 cm should be treated carefully, especially in the left-sided colon.

Regular feedback to individual endoscopists is associated with improved adenoma detection rate and other key performance indicators for colonoscopy.

Objective: Monitoring of key performance indicators (KPIs) is a vital element of endoscopy quality improvement. Adenoma detection rate (ADR) is considered the best marker for colonoscopic quality as it inversely correlates with subsequent colonic cancer incidence and mortality, while polyp detection rate (PDR) is an easier-to-calculate surrogate for ADR. This study assessed whether regular feedback to individual endoscopists about their KPIs improved departmental performance. Methods: Individual KPIs were calculated for a period of 8 years (January 2012-December 2019) and fed back to all endoscopists at 6 monthly intervals, alongside anonymised indicators for other endoscopists, aggregate departmental performance data and benchmarks. An automated natural language processing software (EndoMineR) was used to identify adenomas in pathology reports and calculate ADR. Linear regressions were calculated for departmental ADR, PDR and other KPIs at 6 monthly intervals. Results: 39 359 colonoscopies (average 2460 in every 6-month period, range 1799-3059) were performed by an average of 42 (range 34-50) endoscopists. A continuous improvement in collective performance including ADR (12.7%-21.0%, R2 0.92, p<0.001) and PDR (19.0%-29.6%, R2 0.77, p<0.001) was observed throughout the study. Other KPIs showed similar improvement. The detection of non-neoplastic polyps did not increase. When analysed separately, ADR and PDR appeared to improve for gastroenterologists and nurse endoscopists but not for surgeons. Conclusion: Regular feedback with individual and departmental KPIs was associated with improved ADR and overall performance throughout the 8-year study period. Concomitant monitoring of ADR and PDR may prevent 'gaming' behaviour and ensure that genuine improvement is achieved.

A colonoscopic overtube system that creates a 'therapeutic zone' and permits retraction facilitates endoscopic submucosal dissection, was associated with fewer deep bowel wall injuries and instrument exchanges, and required less lifting solution vs classic endoscopic submucosal dissection in an ex vivo bovine model.

INTRODUCTION: Endoscopic submucosal dissection (ESD) is the 'gold standard' for large flat polyps; nevertheless, the rate of adoption in the USA is low. In ESD, the polyp is 'surgically' detached with a needle knife after a submucosal lift; gravity and the dissection cap are used for retraction. ESD would be easier if active retraction were possible. In an ex vivo bovine colon model, this study assessed an overtube system (Boston Scientific ORISE Tissue Retraction System, TRS) that permits retraction and creates 'an operative field' for removal of rectal/sigmoid lesions. METHOD: Classic ESD (C-ESD) was compared to TRS-facilitated ESD (TRS-ESD). Cleaned/preserved bovine large bowel was used, and two 2-cm 'lesions'/colon were branded onto the mucosal surface 25 and 35 cm from the anus. Submucosal saline lifts were made using a thin catheter and a standard needle knife. We tracked case length, number of instrument exchanges (to refresh lift), the volume of lift solution, the fullness of resection, and deep muscle injuries. RESULTS: Fifty ESDs were carried out in 25 colons (25 C-ESD, 25 TRS-ESD). Complete resections were noted in all cases. The TRS method required fewer instrument exchanges (median 5) vs C-ESD (median 9, p < 0.0001) and less lift solution (median 39 ml) than the C-ESD cases (median 55 ml, p = 0.0003). TRS-ESD was associated with fewer deep muscle injuries (median 2) than C-ESD (median 3, p = 0.0191). Finally, the TRS group's median case length (34.5 min) was shorter than that of C-ESD (41 min, p = 0.0543). CONCLUSION: The TRS system provides retraction and facilitates ESD regarding the number of lift injections, the volume of lift solution needed, and avoidance of muscle injuries. Of note, there is an apparent TRS learning curve, and the device mandates a distal-to-proximal approach and initial 360 degree mucosal incision. Further study is warranted.

Multicentre, prospective, randomised study comparing the diagnostic yield of colon capsule endoscopy versus CT colonography in a screening population (the TOPAZ study).

OBJECTIVE: Colon capsule endoscopy (CCE) has shown promise for colorectal neoplasia detection compared with optical colonoscopy (OC), but has not been compared with other screening tests in average risk screening patients. DESIGN: Patients 50 to 75 years of age (African Americans, 45-75 years) were randomised to CCE or CT colonography (CTC) and subsequent blinded OC. The primary endpoint was diagnostic yield of polyps ≥6 mm with CCE or CTC. Secondary endpoints included accuracy for size and histology, examination completeness, number/proportion of subjects with polyps and adenomas ≥6 mm and ≥10 mm, subject satisfaction and safety. RESULTS: From 320 enrolled subjects, data from 286 (89.4%) were evaluable. The proportion of subjects with any polyp ≥6 mm confirmed by OC was 31.6% for CCE versus 8.6% for CTC (pPr non-inferiority and superiority=0.999). The diagnostic yield of polyps ≥10 mm was 13.5% with CCE versus 6.3% with CTC (pPr non-inferiority=0.9954). The sensitivity and specificity of CCE for polyps ≥6 mm was 79.2% and 96.3% while that of CTC was 26.8% and 98.9%. The sensitivity and specificity of CCE for polyps ≥10 mm was 85.7% and 98.2% compared with 50% and 99.1% for CTC. Both tests were well tolerated/safe. CONCLUSION: CCE was superior to CTC for detection of polyps ≥6 mm and non-inferior for identification of polyps ≥10 mm. CCE should be considered comparable or superior to CTC as a colorectal neoplasia screening test, although neither test is as effective as OC. TRIAL REGISTRATION NUMBER: ClinicalTrials.gov no: NCT02754661.

Efficacy and tolerability of high and low-volume bowel preparation compared: A real-life single-blinded large-population study.

BACKGROUND: Low-volume preparations for colonoscopy have shown similar efficacy compared to high-volume ones in randomized controlled trials (RCT). However, most RCTs do not provide data about clinical outcomes including lesions detection rate. Moreover, real-life comparisons are lacking. AIM: To compare efficacy (both in terms of adequate bowel preparation and detection of colorectal lesions) and tolerability of a high-volume (HV: 4 L polyethylene glycol, PEG) and a low-volume (LV: 2 L PEG plus bisacodyl) bowel preparation in a real-life setting. METHODS: Consecutive outpatients referred for colonoscopy were prospectively enrolled between 1 December 2014 and 31 December 2016. Patients could choose either LV or HV preparation, with a day-before schedule for morning colonoscopies and a split-dose for afternoon procedures. Adequate bowel preparation according to Boston Bowel Preparation Scale (BBPS), clinical outcomes including polyp detection rate (PDR), adenoma detection rate (ADR), advanced adenoma detection rate (AADR), sessile/serrated lesion detection rate (SDR) and cancer detection rate and self-reported tolerability of HV and LV were blindly assessed. RESULTS: Total 2040 patients were enrolled and 1815 (mean age 60.6 years, 50.2% men) finally included. LV was chosen by 52% of patients (50.8% of men, 54.9% of women). Split-dose schedule was more common with HV (44.7% vs 38.2%, P = 0.005). High-definition scopes were used in 33.4% of patients, without difference in the two groups (P = 0.605). HV and LV preparations showed similar adequate bowel preparation rates (89.2% vs 86.6%, P = 0.098), also considering the two different schedules (HV split-dose 93.8% vs LV split-dose 93.6%, P = 1; HV day-before 85.5% vs LV day-before 82.3%, P = 0.182). Mean global BBPS score was higher for HV preparations (7.1 ± 1.7 vs 6.8 ± 1.6, P < 0.001). After adjustment for sex, age and indications for colonoscopy, HV preparation resulted higher in PDR [Odds ratio (OR) 1.32, 95%CI: 1.07-1.63, P = 0.011] and ADR (OR 1.29, 95%CI 1.02-1.63, P = 0.038) and comparable to LV in AADR (OR 1.51, 95%CI 0.97-2.35, P = 0.069), SDR and cancer detection rate. The use of standard-definition colonoscopes was associated to lower PDR (adjusted OR 1.59, 95%CI: 1.22-2.08, P < 0.001), ADR (adjusted OR 1.71, 95%CI: 1.26-2.30, P < 0.001) and AADR (adjusted OR 1.97, 95%CI: 1.09-3.56, P = 0.025) in patients receiving LV preparation. Mean Visual Analogue Scale tolerability scored equally (7, P = 0.627) but a ≥ 75% dose intake was more frequent with LV (94.6% vs 92.1%, P = 0.003). CONCLUSION: In a real-life setting, PEG-based low-volume preparation with bisacodyl showed similar efficacy and tolerability compared to standard HV preparation. However, with higher PDR and ADR, HV should still be considered as the reference standard for clinical trials and the preferred option in screening colonoscopy, especially when colonoscopy is performed with standard resolution imaging.

Interval cancer after colonoscopy in the Austrian National Screening Programme: influence of physician and patient factors.

OBJECTIVE: Postscreening colorectal cancer (PSCRC) after screening colonoscopy is associated with endoscopists' performance and characteristics of resected lesions. Prior studies have shown that adenoma detection rate (ADR) is a decisive factor for PSCRC, but correlations with other parameters need further analysis and ADR may change over time. DESIGN: Cohort study including individuals undergoing screening colonoscopy between 1/2008 and 12/2019 performed by physicians participating in a quality assurance programme in Austria. Data were linked with hospitalisation data for the diagnosis of PSCRC (defined as CRC diagnosis >6 months after colonoscopy). ADR was defined dynamically in relation to the time point of subsequent colonoscopies; high-risk groups of patients were those with an adenoma ≥10 mm, or with high-grade dysplasia, or villous or tubulovillous histology, or a serrated lesion ≥10 mm or with dysplasia, or colonoscopies with ≥3 lesions. Main outcome was PSCRC for each risk group (negative colonoscopy, hyperplastic polyps, low-risk and high-risk group of patients) after colonoscopy by endoscopists with an ADR <20% compared with endoscopists with an ADR ≥20%. RESULTS: 352 685 individuals were included in the study (51.0% women, median age 60 years) of which 10.5% were classified as high-risk group. During a median follow-up of 55.4 months, 241 (0.06%) PSCRC were identified; of 387 participating physicians, 19.6% had at least one PSCRC (8.4% two or more). While higher endoscopist ADR decreased PSCRC incidence (HR per 1% increase 0.97, 95% CI 0.95 to 0.98), affiliation to the high-risk group of patients was also associated with higher PSCRC incidence (HR 3.27, 95% CI 2.36 to 4.00). Similar correlations were seen with regards to high-risk, and advanced adenomas. The risk for PSCRC was significantly higher after colonoscopy by an endoscopist with an ADR <20% as compared with an endoscopist with an ADR ≥20% in patients after negative colonoscopy (HR 2.01, 95% CI 1.35 to 3.0, p<0.001) and for the high-risk group of patients (HR 2.51, 95% CI 1.49 to 4.22, p<0.001). CONCLUSION: A dynamic calculation of the ADR takes into account changes over time but confirms the correlation of ADR and interval cancer. Both lesion characteristics and endoscopists ADR may play a similar role for the risk of PSCRC. This should be considered in deciding about appropriate surveillance intervals in the future.

Designs of colonoscopic adenoma detection trials: more positive results with tandem than with parallel studies - an analysis of studies on imaging techniques and mechanical devices.

BACKGROUND AND AIMS: Adenoma detection rate (ADR) has been shown to correlate with interval cancers after screening colonoscopy and is commonly used as surrogate parameter for its outcome quality. ADR improvements by various techniques have been studied in randomised trials using either parallel or tandem methodololgy. METHODS: A systematic literature search was done on randomised trials (full papers, English language) on tandem or parallel studies using either adenoma miss rates (AMR) or ADR as main outcome to test different novel technologies on imaging (new endoscope generation, narrow band imaging, iScan, Fujinon intelligent chromoendoscopy/blue laser imaging and wide angle scopes) and mechanical devices (transparent caps, endocuff, endorings and balloons). Available meta analyses were also screened for randomised studies. RESULTS: Overall, 24 randomised tandem trials with AMR (variable definitions and methodology) and 42 parallel studies using ADR (homogeneous methodology) as primary outcome were included. Significant differences in favour of the new method were found in 66.7% of tandem studies (8222 patients) but in only 23.8% of parallel studies (28 059 patients), with higher rates of positive studies for mechanical devices than for imaging methods. In a random-effects model, small absolute risk differences were found, but these were double in magnitude for tandem as compared with parallel studies (imaging: tandem 0.04 (0.01, 0.07), parallel 0.02 (0.00, 0.04); mechanical devices: tandem 0.08 (0.00, 0.15), parallel 0.04 (0.01, 0.07)). Nevertheless, 94.2% of missed adenomas in the tandem studies were small (<1 cm) and/or non-advanced. CONCLUSIONS: A tandem study is more likely to yield positive results than a simple parallel trial; this may be due to the use of different parameters, variable definitions and methodology, and perhaps also a higher likelihood of bias. Therefore, we suggest to accept positive results of tandem studies only if accompanied by positive results from parallel trials.

BowelScope: Accuracy of Detection Using Endocuff Optimisation of Mucosal Abnormalities (the B-ADENOMA Study): a multicentre, randomised controlled flexible sigmoidoscopy trial.

OBJECTIVES: Adenoma detection rate (ADR) is an important quality marker at lower GI endoscopy. Higher ADRs are associated with lower postcolonoscopy colorectal cancer rates. The English flexible sigmoidoscopy (FS) screening programme (BowelScope), offers a one-off FS to individuals aged 55 years. However, variation in ADR exists. Large studies have demonstrated improved ADR using Endocuff Vision (EV) within colonoscopy screening, but there are no studies within FS. We sought to test the effect of EV on ADR in a national FS screening population. DESIGN: BowelScope: Accuracy of Detection Using ENdocuff Optimisation of Mucosal Abnormalities was a multicentre, randomised controlled trial involving 16 English BowelScope screening centres. Individuals were randomised to Endocuff Vision-assisted BowelScope (EAB) or Standard BowelScope (SB). ADR, polyp detection rate (PDR), mean adenomas per procedure (MAP), polyp characteristics and location, participant experience, procedural time and adverse events were measured. Comparison of ADR within the trial with national BowelScope ADR was also undertaken. RESULTS: 3222 participants were randomised (53% male) to receive EAB (n=1610) or SB (n=1612). Baseline demographics were comparable between arms. ADR in the EAB arm was 13.3% and that in the SB arm was 12.2% (p=0.353). No statistically significant differences were found in PDR, MAP, polyp characteristics or location, participant experience, complications or procedural characteristics. ADR in the SB control arm was 3.1% higher than the national ADR. CONCLUSION: EV did not improve BowelScope ADR when compared with SB. ADR in both arms was higher than the national ADR. Where detection rates are already high, EV is unable to improve detection further. TRIAL REGISTRATION NUMBERS: NCT03072472, ISRCTN30005319 and CPMS ID 33224.