Clinical decision tool helps predict Entyvio outcomes in ulcerative colitis

SAN DIEGO — A clinical decision tool could help predict treatment outcomes of patients with ulcerative colitis on Entyvio therapy and identify who might benefit from dose escalation, according to research presented at Digestive Disease Week.

Parambir Dulai, MD, of the division of gastroenterology at the University of California San Diego, said previous research has identified a number of potential predictors of Entyvio (vedolizumab, Takeda) outcomes, but they have not been translated into clinical decision tools.

“It is very unclear how to integrate those predictors into routine practice and how to use them when deciding on therapy for a patient in clinic,” he said in his presentation. “Well-validated, easy to use prediction models that have been transformed into clinical decision support tools could help bridge this gap and optimize the use of this therapy in practice.”

Using data from the GEMINI 1 trial, researchers identified factors associated with corticosteroid-free remission to build their clinical decision support tool. They decided to use absence of previous anti-TNF exposure, disease duration of 2 years or longer, baseline endoscopic activity and baseline albumin concentration in their model and assigned each factor with point values of +3, +3, +2 and +0.65 points per g/L, respectively.

Investigators stratified patients into low ( 26), intermediate (< 26 to 32) or high (> 32) probability of response groups based on their point scores. Patients in the high probability group were able to achieve symptoms activity reduction more rapidly and attained higher rates of corticosteroid-free remission (P < .001).

Dulai and colleagues validated their decision tool’s ability to predict corticosteroid-free remission and colectomy at week 26 using data from the VICTORY consortium. In that analysis, researchers found that the 26-point cutoff had a 93% sensitivity (95% CI, 79%–98%) for identifying corticosteroid-free remission and an 88% sensitivity (95% CI, 83%–92%) for patients who were colectomy-free at week 26. Patients in this group also benefited from vedolizumab interval shortening for lack of response.

Additionally, investigators found that the high probability grouping had a 68% specificity (95% CI, 60%–75%) for identifying corticosteroid-free remission and a 71% specificity (95% CI, 42%–92%) for identifying patients who did not need a colectomy.

“We developed an externally validated decision support tool with good discriminative performance for predicting corticosteroid-free clinical and endoscopic remission with vedolizumab in UC,” Dulai said. - by Alex Young

Reference:

Dulai P, et al. Abstract 334. Presented at: Digestive Disease Week; May 18-21, 2019; San Diego.

Disclosures: Dulai reports financial ties to Janssen, Pfizer and Takeda. Please see the meeting disclosure index for all other authors’ relevant financial disclosures.