DOI: 10.1097/meg.0000000000003274 ISSN: 0954-691X

Early but not serial ustekinumab concentration measurements are associated with clinical outcomes in Crohn’s disease and ulcerative colitis: the Stockholm Ustekinumab study (STOCUSTE)

Haider Sabhan, Francesca Bello, Samer Muhsen, Fredrik Johansson, Alexandra Borin, Charlotte Höög, Ole Forsberg, Christina Wennerström, Sven Almer, Mikael Lördal, Charlotte Söderman

Introduction

Ustekinumab (UST), an anti-interleukin-12/23 antibody, is used to treat moderate-to-severe inflammatory bowel disease (IBD). The STOCUSTE study includes IBD patients treated with UST at four teaching hospitals in Stockholm to provide long-term follow-up data. The utility of therapeutic drug monitoring to optimize treatment and correlate the serum concentration of UST with clinical outcomes was investigated in a real-world setting.

Methods

This retrospective study includes patients diagnosed with Crohn’s disease and ulcerative colitis treated with UST and followed until withdrawal of treatment for any reason or until the end of the study. Concentration data and dosing interval were collected at each follow-up (3, 6, 9, 12, 24, 36, and 48 months) in relation to the presence of remission defined as Physician Global Assessment = 0.

Results

A total of 418 patients were included, 322 with Crohn’s disease (48% female) and 96 with ulcerative colitis (46% female). UST concentrations were obtained in 229 (55%) patients, while 189 had no measurements. The mean serum concentrations for all were above 1 μg/ml. At 3 months, patients in remission had significantly higher concentrations than nonremitters ( P  < 0.001). No significant differences were observed at 6 months or later. Within 12 months of concentration result, approximately one-third of patients underwent treatment adjustments, including interval shortening, intravenous reinduction, or discontinuation/switch.

Conclusion

In this study, an early (≤3 months) UST concentration was associated with remission, whereas concentrations measured at later time points did not differ between remission and nonremission. In this real-world cohort serial monitoring beyond 3 months showed poor correlation with clinical outcomes.