Dose escalation of biologics in biologic-naive patients with Crohn's disease: Outcomes from the ODESSA-CD study.

Abstract

Background: Dose escalation of biologics may restore response in patients with Crohn's disease (CD) who experience inadequate response or loss of response, but the rates of dose escalation and subsequent adverse clinical outcomes have not been well characterized.

Objective: To evaluate the rate of dose escalation of biologics and associated adverse clinical outcomes and economic outcomes in biologic-naive patients with CD.

Methods: ODESSA-CD (real wOrld Dose EScalation and outcomeS with biologics in IBD pAtients with Crohn's Disease) was a retrospective cohort study conducted using claims data from IBM MarketScan databases. Adults with CD with at least 1 claim for an index drug (adalimumab, infliximab, ustekinumab, or vedolizumab) between January 1, 2017, and December 31, 2018, and no claims for biologics in the 6 months prior (ie, biologic naive) were included. Follow-up ended on June 30, 2020. Cox proportional hazards models and logistic regression models were used to compare the rate of dose escalation and the likelihood of adverse clinical outcomes and costs after dose escalation, respectively.

Results: Of the 2,664 eligible patients, most (71.4%) were younger than 50 years and 50.5% were male. The rate of dose escalation was higher with the anti-tumor necrosis factor α (TNFα) treatments adalimumab (hazard ratio [HR] = 1.703; P < 0.0001) and infliximab (HR = 1.690; P < 0.0001) compared with vedolizumab, but there was no significant difference between ustekinumab and vedolizumab (HR = 0.842; P = 0.730). After dose escalation, the likelihood of infection, sepsis, and inflammatory bowel disease-related hospitalization did not differ among biologics (anti-TNFα vs vedolizumab: odds ratio [OR] = 1.141, P = 0.599; ustekinumab vs vedolizumab: OR = 0.891; P = 0.836); however, corticosteroid use was more likely with anti-TNFα treatment than with vedolizumab (OR = 1.740, P = 0.002). Among patients whose dose was escalated, index drug costs were likely to be higher with anti-TNFα treatment and ustekinumab than with vedolizumab (anti-TNFα vs vedolizumab: ratio of expected cost = 1.429, P = 0.002; ustekinumab vs vedolizumab: ratio of expected cost = 3.115, P < 0.0001).

Conclusions: Patients who were biologic naive and received ustekinumab or vedolizumab were less likely to undergo dose escalation than those who received anti-TNFα treatment. Adverse clinical outcomes after dose escalation were similar among these biologics but with different costs. These analyses may inform providers and payers of the clinical and economic implications of dose escalation.