Evaluation of the Accuracy of Immunostaining of Claudin 2 to Detect Active or Inactive Inflammation in Endoscopic Biopsies of IBD

In view of the increasing evidence of the importance of mucosal healing as a combination of endoscopic and histologic healing, and the complexity and lack of complete agreement of the evaluation of histologic healing in IBD, the aim of this study will be to evaluate the accuracy of immunostaining for Claudine 2 as compared to others standard well-accepted scoring for histologic activity in IBD. The immunostaining of Claudin 2 in term of presence/absence will also be evaluated in view of the outcomes of the patients after 1 year of follow-up.

The primary objective of the study will be to compare the feasibility and accuracy of immunostaining with Claudin 2 against others standard frequently reported scoring of histology in IBD. The primary endpoint will be therefore the concordance of the immunostaining of Claudin 2 with the histologic healing as evaluated by the standard histologic scoring for IBD, namely Geboes, Nancy, Robarts, PICASSO, and SHMHS.

For secondary objectives all patient enrolled at baseline will be evaluated for 1 year of follow-up, with reporting of clinical activity, inflammatory biomarkers, need for hospitalization, need for surgery, need of escalation of therapy, in relation to the immunostaining of Claudin 2 (presence/absence) and the presence/absence of histologic activity as evaluated by the traditional histologic scores.

The data analysis will be performed at the Promoter center, by the laboratory of Biostatistics. Demographic, anamnestic histological and clinical data will be collected and analysed. Continuous data will be reported as mean ± standard deviation (SD) or median and interquartile range (IQR), as appropriate. Categorical data will be reported using frequencies and proportions. The normality hypothesis will be tested for continuous variables by visual inspection of the graph qq and with Shapiro-Wilks test. Correlations will be assessed according to Pearson R or Spearman Rho, depending on the distribution. The t-test or Wilcoxon rank sum test will be used to compare unpaired averages in normally or non-normally distributed variables, respectively. Fisher's exact test or chi-square test will be used to compare categorical data between groups.

To evaluate the concordance of histologic evaluation (active/remission) of different scoring system and Claudin 2 immunostaining, the Cohen's kapра (к) will be evaluated. Perfect agreement is evident when Cohen's kappa equals 1; a value of Cohen's kappa equal to zero suggests that the agreement is no better than that which would be obtained by chance alone. Although there is no formal scale, the following levels of agreement are often considered appropriate for judging the extent of the agreement: Poor if к< 0.00; Slight if 0.00 ≤ к≤ 0.20; Fair if 0.21 ≤ k ≤ 0.40; Moderate if 0.41 ≤ k ≤ 0.60; Substantial if 0.61 ≤ к ≤ 0.80; Almost perfect if k > 0.80. То evaluate for a possible systematic difference the McNemar's test will be applied with significance set at P<0.05. The different outcomes at 1-year according to different measures (active disease/relapse, remission, change of therapy, hospitalization and surgery) and as composite evaluation (remission + no change of therapy or hospitalization or surgery vs active/relapse ± change of therapy or hospitalization or surgery) will be evaluated with Kaplan Meier survival analysis. The difference in outcome will be also analyzed with multivariate analysis by considering also the clinical variables.

Using a PASS procedure Kappa Test for Agreement Between Two Raters, an estimated sample size of 100 patients achieves 80% power and a error of 5% to detect a true Cohen' Kappa value of 0,66 in a test of H: Карра = ко (where ko= 0.04) vs. H1: Kappa ≠ к0 when there are 2 categories with frequencies equal to 30% and 70%.