Predicting response to dose escalation in patients with inflammatory bowel disease with secondary loss of response to biological therapy.

Abstract

Background and Aim: Biological agents are the most effective treatments to induce and maintain remission in patients with moderate to severe inflammatory bowel disease (IBD). A significant number of patients experience secondary loss of response (LOR). A subset of these patients can be dose-escalated to recapture response, but it is uncertain which patients will respond. This approach is well established for anti-tumor necrosis factor (TNF) therapy, but it is less clear if this is applicable to other biological agents. The aim of this study was to review the outcomes of dose escalation in a virtual biologic clinic at a tertiary hospital in Melbourne, Victoria, and to assess for predictors of response to dose escalation. Method(s): We conducted a retrospective review of patients with IBD with secondary LOR to biologics (infliximab, adalimumab, ustekinumab, and vedolizumab) between April 2017 and November 2020. Included patients had evidence of active disease at baseline requiring dose escalation. Dose escalation for infliximab included increasing dosage from 5 to 10 mg/kg or increased frequency; and for adalimumab, vedolizumab, and ustekinumab, it was increased frequency of dosing. Patients were reviewed at baseline, 6 months, and 12 months with testing of drug levels and clinical and objective markers of disease activity (fecal calprotectin level, C-reactive protein level, colonoscopy, or imaging). The primary endpoint was objective response to dose escalation. Secondary endpoints included clinical response to dose escalation, defined as an improvement in the Physician Global Assessment. Failure of response was defined as ongoing objective active disease and/or the requirement for further dose escalation, switching within class, switching to a different biologic class, steroid use, hospital admission, or surgery. Result(s): Of 580 patients receiving biologics, 82 patients (14.1%) with secondary LOR received dose escalation via the virtual biologic clinic (51% female; median age, 39 years). Of the patients, 72% had Crohn's disease, 24.4% had ulcerative colitis, and 37.8% had perianal disease. Infliximab was the most common biologic (63.4%), followed by adalimumab (22%), vedolizumab (7.3%), and ustekinumab (7.3%). Combination therapy was used by 79.4% of patients. At the time of dose escalation, all patients had active disease by objective markers and 91.5% by clinical parameters. At baseline, the median level of infliximab was 2.1 mu/mL (3-7 mu/mL) and adalimumab was 5.1 mu/mL (5-8 mu/mL). At 6 months after dose escalation, 53 patients (71.6%) responded. Drug levels increased at 6-month followup (median infliximab level, 5.9 mu/mL; and adalimumab, 10.7 mu/mL), but there was no difference in drug levels between responders and non-responders (median infliximab level, 6.2 mu/mL vs 4.38 mu/mL; P = 0.224; median adalimumab level, 10.4 mu/mL vs 11 mu/mL; P = 0.680). There were no factors at baseline that predicted response to therapy at 6 months. Follow-up data to 12 months were available for 68 patients, with 45 (66.2%) showing objective response to dose escalation and 11 (16.2%) in objective remission. Compared with non-responders, responders were older (45 vs 31.5 years; P < 0.001) and less likely to have perianal disease (26.7% vs 60.9%; P = 0.004), prior biological therapy (15.6% vs 43.5%; P = 0.026), or prior TNF exposure (13.3% vs 34.8%; P = 0.08). There was no statistically significant difference in response rates between the different biologics at 12 months (infliximab, 73.2%; adalimumab, 64.7%; vedolizumab, 66.7%; ustekinumab, 25%; P = 0.431). Furthermore, median drug levels did not differ between responders and non-responders (infliximab, 7.95 mu/mL vs 7.67 mu/mL; P = 0.495; adalimumab, 10.4 mu/mL vs 9 mu/mL; P = 0.670). Of the 45 responders, nine patients on infliximab and four patients on adalimumab were successfully de-escalated. Dose escalation failed in 23 patients, with seven (30.4%) requiring further dose escalation, one patient (4.3%%) switching to a drug within the same class, and seven patients (30.4%) switching to a drug in a different class. Conclusion(s): Secondary LOR to treatment affects a significant number of patients with IBD. Dose escalation is successful in most patients across biologic classes. Predictors of response to dose escalation at 12 months included older age, absence of perianal disease, and being native to biological therapy.