The use of therapeutic drug monitoring to assess response to ustekinumab during the first 16 weeks of administration may predict endoscopic and clinical disease outcomes after 1 year in patients with Crohn disease (CD), according to study findings published in Digestive and Liver Disease.
Researchers conducted a monocentric, retrospective study from October 2016 to August 2020 that analyzed the long-term response of 80 patients with CD treated with ustekinumab. The researchers sought to identify predictive factors of response that would determine if reactive therapeutic drug monitoring (TDM) was warranted.
Colonoscopy and histology confirmed the CD diagnosis in these 80 patients, most of whom were women (66%). Overall, 236 blood samples were collected at baseline, prior to first ustekinumab injection and before each subsequent injection until week 52.
The patients received an initial ustekinumab intravenous injection with a customized calculated dose of 6 mg/kg of body weight followed by subsequent subcutaneous injections at a dose of 90 mg every 8 weeks for a minimum of 16 weeks.
The researchers analyzed trough levels (indicating the pharmacokinetics of ustekinumab) along with clinical and biological data using a program that identified predictive factors of clinical and endoscopic outcomes at 1 year.
After 1 year of follow-up, 26.3% of patients discontinued use of ustekinumab secondary to lack of efficacy, adverse side effects, or loss of follow-up. For another 22.5% of patients, a drug-optimization program monitoring loss of response was required after a median of 44 weeks.
After a median of 40 weeks, the 52 remaining patients (65%) underwent repeat endoscopy to determine level of CD activity. While most of these patients showed mild (40%) or quiescent (30%) endoscopic CD activity, some still exhibited moderate (20%) to severe (10%) endoscopic CD activity.
Patients with moderate to severe disease activity after 40 weeks demonstrated significantly lower median trough levels for ustekinumab by week 16 compared with patients with mild or quiescent disease activity (P =.04). Additional factors found in blood samples were associated with endoscopic outcomes, including lymphocyte and monocyte counts at baseline and week 8 and C reactive protein level at week 16.
Moreover, of the 3 predictive models analyzed, the random forest model (sensitivity, 91%; specificity, 75%) outperformed the logistic regression (sensitivity, 83%; specificity, 70%) and gradient boosting decision trees (sensitivity, 91%; specificity, 71%).
Study limitations include the retrospective design with missing data, lack of validation in an independent cohort for the tree-based predictive models, and the inclusion of only patients with refractory CD, which limits generalizability outside of this patient subgroup.
“This real-world experience confirms the relationship between early UST TLs [ustekinumab trough levels] and both endoscopic and clinical disease outcomes,” the study authors wrote. “The absence of immunogenicity suggests that it is not a key driver in the loss of response to UST [ustekinumab].
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.
Liefferinckx C, Hubert A, Thomas D, et al. Predictive models assessing the response to ustekinumab highlight the value of therapeutic drug monitoring in Crohn’s disease. Dig Liver Dis. Published online August 14, 2022. doi:10.1016/j.dld.2022.07.015
This article originally appeared on Gastroenterology Advisor