| Popis: |
INTRODUCTION Differential diagnosis of Inflammatory Bowel Disease Unclassified (IBDU) patients to Crohn’s Disease (CD) or Ulcerative Colitis (UC) at diagnosis is challenging. Personalized disease-specific strategies are becoming state of art and therefore it is of importance for outcome of the patients and treatment decision to be taken by the clinicians to have a precise diagnosis. METHODS Whole transcriptome RNA sequencing was performed using AmpliSeq human gene expression panel on formalin-fixed paraffin-embedded (FFPE) colonic mucosal biopsies of IBD patients (CD, n=16, UC, n=17) and non-IBD controls (n=10) to study differential gene expression between CD and UC samples. Top 100 of the most significant differentially expressed genes were then tested in two independent external GEO datasets. Functional enrichment analysis of 10 successful candidates were performed and 7 genes were selected for further validation. RT-qPCR analyses were performed on an independent patient cohort of 45 CDs and 53 UCs along with 12 non-IBD controls. The cDNA samples were submitted to quality restriction during the statistical analysis resulting in a cut-off of Ct values below 29 for the endogenous control allowing 46 patient samples to enter the analysis. In situ hybridization (ISH) was performed using RNAscope on 11 selected cases for describing tissue localisation and expression. RESULTS Sequencing results showed 308 genes differentially expressed between CD and UC from a total of 14190 genes. The genes MTCL1, SH3PXD2A-AS1, CLCF1, and CD180 were under the limit of detection with RT-qPCR and were excluded from further analysis. Relative expression levels from RT-qPCR for the genes PI3, ANXA1, and VDR, together with the age and sex, resulted in an area under the receiver-operating characteristic curve (AUC) of 0.84 (P = 0.02) in discriminating CD from UC. Significantly higher expression of VDR alone was observed in CD samples with AUCs of 0.77 (P = 0.01). ISH analyses showed expression of PI3, ANXA1, and VDR mRNAs localized to mucosal epithelial cells. CONCLUSION Using whole transcriptome analysis of FFPE sample-derived RNA, we have identified genes that are differentially expressed in CD and UC, and we have developed an RT-qPCR assay including a panel of 3 genes (PI3, ANXA1, VDR) for the differential diagnosis among IBD unclassified (IBDU) patients. We propose that the RT-qPCR assay could be used in a clinical routine setup for distinguishing UC from CD. |
