Sparse Modeling Reveals miRNA Signatures for Diagnostics of Inflammatory Bowel Disease

Abstract

The diagnosis of inflammatory bowel disease (IBD) still remains a clinical challenge and the most accurate diagnostic procedure is a combination of clinical tests including invasive endoscopy. In this study we evaluated whether systematic miRNA expression profiling, in conjunction with machine learning techniques, is suitable as a non-invasive test for the major IBD phenotypes (Crohn's disease (CD) and ulcerative colitis (UC)). Based on microarray technology, expression levels of 863 miRNAs were determined for whole blood samples from 40 CD and 36 UC patients and compared to data from 38 healthy controls (HC). To further discriminate between disease-specific and general inflammation we included miRNA expression data from other inflammatory diseases (inflammation controls (IC): 24 chronic obstructive pulmonary disease (COPD), 23 multiple sclerosis, 38 pancreatitis and 45 sarcoidosis cases) as well as 70 healthy controls from previous studies. Classification problems considering 2, 3 or 4 groups were solved using different types of penalized support vector machines (SVMs). The resulting models were assessed regarding sparsity and performance and a subset was selected for further investigation. Measured by the area under the ROC curve (AUC) the corresponding median holdout-validated accuracy was estimated as ranging from 0.75 to 1.00 (including IC) and 0.89 to 0.98 (excluding IC), respectively. In combination, the corresponding models provide tools for the distinction of CD and UC as well as CD, UC and HC with expected classification error rates of 3.1 and 3.3%, respectively. These results were obtained by incorporating not more than 16 distinct miRNAs. Validated target genes of these miRNAs have been previously described as being related to IBD. For others we observed significant enrichment for IBD susceptibility loci identified in earlier GWAS. These results suggest that the proposed miRNA signature is of relevance for the etiology of IBD. Its diagnostic value, however, should be further evaluated in large, independent, clinically well characterized cohorts.

Over expressed miRNA-31 is the Most Consistent and Unique Micro RNA in Ulcerative Colitis

“…Many miRNAs are deregulated significantly even in early stages of many acute inflammatory conditions, exhibiting their diagnostic and therapeutic potential. However, miRNA profiling studies were performed on different platforms and very different processing methods by different investigators, their lacks the common concert among these studies, consequently, which may end with inconsistent miRNAs expression profile (Wu et al, 2008, Fasseu et al, 2010, Takagi et al, 2010, Olaru et al, 2011, Olaru et al, 2013, Coskun et al, 2013, Van der Goten et al, 2014, Lin et al, 2014, Schaefer et al, 2015, Hubenthal et al, 2015. Surprisingly, by systematic analysis, we found that only miRNA-31 is sorted out among all the miRNAs over expressed in UC patients, none of others are either highly consistently expressed (cut off>4), further, among the down-regulated miRNAs in UC, none of them are consistently expressed (cut off>4).The reasons why miRNA-31was the only one selected in our systematic analysis are complicated, except the fact that both environmental factors are important in the pathogenesis of UC, miRNA-31 plays crucial roles in intestinal development and homeostasis, particularly, miRNA-31 can function as both oncogene and tumor suppressor, and pro-inflammatory and antiinflammatory factor.…”

“…Each of the published miRNA expression profiling studies comparing miRNA expression between inflamed and noninflamed control intestinal tissues provided a list of differentially expressed miRNAs (Wu et al, 2008, Fasseu et al, 2010, Takagi et al, 2010, Olaru et al, 2011, Olaru et al, 2013, Coskun et al, 2013, Van der Goten et al, 2014, Lin et al, 2014, Schaefer et al, 2015, Hubenthal et al, 2015. For example, Hubenthal et al (Hubenthal et al, 2015) carried out comprehensive analysis of UCrelated miRNAs based on a large number of miRNA profiling studies.…”

“…For example, Hubenthal et al (Hubenthal et al, 2015) carried out comprehensive analysis of UCrelated miRNAs based on a large number of miRNA profiling studies.…”

“…Beyond these, an important feature of miRNA is their remarkable stability, for example, they can be well preserved in tissue samples even after formalin-fixation and paraffin embedding for years, and can be efficiently extracted from such specimens and variety of cell lines (Finkel et al, 2007). Actually, several endeavors have been imposed to search for biomarkers or therapeutic agents by identifying the differentially expressed miRNAs between UC inflamed tissue and corresponding non-inflamed tissue (Wu et al, 2008, Fasseu et al, 2010, Takagi et al, 2010, Olaru et al, 2011, Olaru et al, 2013, Coskun et al, 2013, Van der Goten et al, 2014, Lin et al, 2014, Schaefer et al, 2015, Hubenthal et al, 2015. These studies generated hundreds of differentially expressed miRNAs, however, many of these are likely limited by different bias and finally only a small number of them may be of clinical significance and act as potential diagnostic agents.…”

Immune biomarkers in the spectrum of childhood noncommunicable diseases

“…Recently there has been a flurry of papers exploring miRNA signatures in IBD using blood, saliva and colonic tissue to reveal differential expression profiles associated with UC versus CD. Importantly these IBDs could be discriminated from other inflammatory disorders using machine learning techniques for expression profiling 127 . There is now much interest in evaluating these in large, independent, clinically well characterized cohorts and it would be good to see such studies extended to include pediatric populations.…”