Gene expression in combination with transcriptome-wide association studies can be used for the prediction of colectomy in patients with ulcerative colitis, according to a study published in American Journal of Human Genetics.

Roughly 5% to 10% of patients with ulcerative colitis will require colectomy within 5 years of receiving a disease diagnosis. While clinical genomics often focuses on predicting disease onset, there is an arguably greater need for tools to predict disease progression and adverse outcomes. Thus, investigators aimed to show the efficacy of transcriptome-wide association studies in predicting transcriptional risk scores for disease complications requiring colectomy in patients with ulcerative colitis.

Researchers utilized data and samples from participants included in the PROTECT study cohort; the multi-center, open-label study evaluated the safety and effectiveness of mesalamine oral corticosteroids vs intravenous corticosteroids for treatment of ulcerative colitis ( Identifier: NCT01536535). Participants were children aged 4 to 17 years who had been newly diagnosed with ulcerative colitis. RNA was isolated from 340 rectal biopsies taken at baseline and 92 biopsies taken at week 52 of follow-up. Further data were collected from the United Kingdom (UK) Biobank and from surgical specimens from 210 patients with ulcerative colitis undergoing bowel resection for inflammatory bowel disease within the Mount Sinai Health System.

Continue Reading

Based on bulk rectal gene expression, a highly discriminatory signature was observed in children with ulcerative colitis who needed surgery. This led to a transcriptional risk score with a positive predictive value approaching 50% for colectomy.  Single cell profiling revealed the genes are active in multiple cell types, including epithelial and immune compartments. For genes with differential effects at baseline and week 52 of follow-up, differential expression associated with colectomy risk was independent of local genetic regulation.

Using samples from the UK Biobank population cohort studies, a predicted polygenic transcriptional risk score was derived via summation of transcriptome-wide association study effects.  This analysis identified individuals with ulcerative colitis who had a 5-fold elevated risk for colectomy. This finding was then independently replicated in a National Institute of Diabetes and Digestive and Kidney Diseases Inflammatory Bowel Disease Genetics Consortium dataset.

Study limitations included the small sample size of colectomies in the PROTECT cohort and low precision of the predicted polygenic transcriptional risk score, which resulted in limited clinical significance.

Researchers concluded, “Validation of cross-ancestry assessments should be a high priority, and it will be interesting to evaluate to what extent gene expression prediction is consistent across populations.” “Similar strategies might also be developed for other complex diseases for which sampling of the relevant tissue is impractical,” they noted.

Disclosure: Some study authors declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of authors’ disclosures.


Mo A, Nagpal S, Gettler K, et al. Stratification of risk of progression to colectomy in ulcerative colitis via measured and predicted gene expression. Am J Hum Genet. 2021;108(9):1765-1779. doi: 10.1016/j.ajhg.2021.07.013