P157 Fecal calprotectin correlates to endoscopic and histologic remission in ulcerative colitis: a prospective study
Hart L.*1, Kherad O.1, Lemieux C.1, Laneuville J.1, Chavannes M.2, Marcus V.3, Maedler C.3, Afif W.1, Bitton A.1, Brassard P.4, Bessissow T.1
1McGill University, Gastroenterology, Montreal, Canada 2University of British Columbia, Gastroenterology, Vancouver, Canada 3McGill University, Pathology, Montreal, Canada 4McGill University, Clinical Epidemiology, Montreal, Canada
Endoscopic healing (EH), has become the target endpoint of successful ulcerative colitis (UC) management while the role of histology remains to be clarified. Clinical and biochemical markers are not sufficient to predict which UC patients are in EH. Fecal calprotectin (FC) has been demonstrated to be useful surrogate marker, however, there has not been a consensus for cut off values for FC for endoscopic and histologic remission.
Our prospective cohort study recruited patients with UC in clinical remission who were being followed at the McGill IBD Center between 2013–2016. Patients were recruited if they had clinical remission (partial Mayo score of ≤2) and were undergoing colonoscopy for disease reassessment. At the time of colonoscopy, fecal calprotectin (FC) was collected, as well as full and endoscopic Mayo score, Geboes histology score and record of basal plasmacytosis.
163 patients were recruited (88 males, 75 females; mean age 49 years; IQR 39–59). Based on endoscopic scores, 65% (n=106), 24% (n=39), 11% (n=18) patients were Mayo 0, 1, 2 respectively. There was a statistically significant difference in fecal calprotectin based on endoscopic mayo score (p<0.0001). The area under the curve (AUC) in receiver operator characteristic (ROC) analysis of FC to predict Mayo 0 (from Mayo 1–2) was 0.747 (CI 95% 0.65–0.83, p<0.01) with a cut off value of FC 150mcg/g yielding 65% sensitivity and 72% specificity.
Sensitivity Specificity FC <50 86% 40% FC <100 75.4% 56% FC <150 65% 71% FC <200 60% 78% FC <250 58% 83%
Similarly, there was a statistically significant difference between fecal calprotectin based on histologic Geboes score and presence or absence of basal plasmacytosis (p=0.005). The AUC in ROC analysis of FC to predict Geboes <3.1 (remission) was 0.582 (CI 95% 0.481–0.683, p=0.119). The AUC in ROC analysis of FC to predict basal plasmacytosis was 0.64 (CI 95% 0.51–0.766, p=0.02), with a cut off value of FC 140 mcg/g yielding 60% sensitivity and 72% specificity.
Sensitivity Specificity FC <50 74% 44% FC <100 63% 60% FC <150 57% 72% FC <200 54% 75% FC <250 51% 80%
Our study demonstrates that fecal calprotectin below 150 mcg/g predicts endoscopic Mayo 0. In addition to correlating with the endoscopic score, fecal calprotectin below 140 mcg/g correlates with absence of basal plasmacytosis. Using non-invasive testing, these predictive values have potential to identify patients with ulcerative colitis in remission, but further validation is needed.