P164 Mucosal healing (MH) assessed with PICaSSO (Paddington International Virtual ChromoendoScopy ScOre) and probe Confocal Laser Endomicroscopy (pCLE) do not reflect histological normalisation using the ECAP (Extent Chronicity Activity Plus) score
M. Iacucci*1,2,3,4, R. Cannatelli3, S. X. Gui5, B. C. Lethebe6, A. Bazarova3, G. Gkoutos3, G. Kaplan7, R. Panaccione7, R. Kiesslich8, S. Ghosh1,3,4
1University of Birmingham, Institute of Immunology and Immunotherapy, Birmingham, UK, 2University of Calgary, IBD Unit, Calgary, Canada, 3University of Birmingham, Institute of Translational Medicine, Birmingham, UK, 4National Institute for Health Research (NIHR) Birmingham Biomedical Research Centre, Birmingham, UK, 5University of Calgary, Department of Pathology, Calgary, Canada, 6University of Calgary, Research Unit, Calgary, Canada, 7University of Calgary, IBD Unit, Birmingham, UK, 8HSK Hospital, Division of Gastroenterology, Wiesbaden, Germany
Ulcerative colitis (UC) is a chronic disease that requires long-term therapy and the achievement of mucosal healing (MH) is the target of the treatment. The new histological score ECAP (Extent Chronicity Activity Plus) has been developed to detect minimal chronic changes. The electronic Virtual Chromoendoscopy Endoscopy (VCE) score PICaSSO (Paddington International Virtual ChromoendoScopy ScOre)1 and probe Confocal Laser Endomicroscopy (pCLE) scores reflect acute histological changes (Robarts Histological Index-[RHI]) well,2 but it is unknown whether these may reflect chronic histological changes.
This is a prospective study involving 82 UC patients at the Endoscopy Unit, Foothills Medical Center, University of Calgary, Canada. For each patient, clinical data including follow-up up to 12 months, endoscopic scores (Mayo endoscopic score, PiCasso and pCLE score) and histological ECAP score were determined. The details of ECAP score has already been published.1 Receiver-operating characteristics (ROC) curves were plotted to estimate the cut-offs for PICaSSO and pCLE scores best predicting the MH determined by histological ECAP score.
70 patients (85.4%) were in clinical remission. We have compared the endoscopic scores (MES, PICaSSO, and pCLE) with histological healing defined by ECAP ≤ 4. Only 14 patients (28.6%) with Mayo 0 had ECAP ≤ 4. From the ROC curves, the best cut-off for PICaSSO score was 4, with sensitivity and specificity of 88.9% (95% CI 64.3%-98.6%) and 40.6% (95% CI 28.5%-53.6%), respectively with an area under ROC curve (AUROC) of 69.9% (95% CI 57.2%-82.6%). At this value, out of 54 patients with PICaSSO ≤ 4, only 16 (29.6%) have ECAP ≤ 4. The ROC curves for pCLE showed that the best cut-off point to detect MH (ECAP ≤ 4) was 11 with sensitivity of 94.4% (95% CI 72.7–99.9%) specificity of 31.3% (95% CI 20.2%-44.1%) with AUROC of 71.4% (95% CI 57.9%-84.8%). At this value of pCLE, 17 (27.9%) patient amongst 61 had ECAP ≤ 4.
ROC curve for PICaSSO in the prediction of mucosal healing (ECAP ≤ 4)
During the follow-up period, 8.06% of patients had a relapse: 80% had PICaSSO score ≤ 4, 80% had pCLE ≤ 11, and only one relapse 20% had ECAP ≤ 4.
According our results, the endoscopic scores (PICaSSO and pCLE) are not able to predict histological healing calculated using ECAP (both chronic and acute changes) at the value ≤ 4. An ECAP score ≤4 may however predict stable mucosal healing with few relapses over 12 months.
1. Iacucci M, Daperno M, Lazarev M,
2. Buda A, Hatem G, Neumann H,