S1059 Risk of Serious Infections in Patients with Ulcerative Colitis Treated with Tofacitinib Compared to Biologic Treatments in the United States

Introduction: There is limited information on safety events observed for tofacitinib using real-world data that includes an active comparator. In this analysis, we examined the risk of serious infections (SIs) among patients (pts) with ulcerative colitis (UC) initiating tofacitinib or biologics using data from an adjudicated United States (US) medical and pharmacy claims database (Komodo Health). Methods: Pts with UC initiating a new treatment with tofacitinib, ustekinumab, vedolizumab, or TNFi from 31 May 2018–30 September 2022 were selected. Stabilized inverse probability treatment weights (sIPTW) were calculated using 17 covariates in the main analysis to examine the risk of developing SIs, and an additional 53 covariates in a sensitivity analysis to control for additional comorbidities, UC-related measures, and health care utilization variables. Cox proportional hazards models with sIPTW were used to calculate hazard ratios (HRs) and bootstrapping to calculate 95% confidence intervals (CIs). Results: Data from UC pts initiating new treatments were included: n=5,171 tofacitinib; n=10,424 ustekinumab; n=17,129 vedolizumab; n=29,872 TNFi. Mean age at index date was 43.1 (+ 15.1) years, and mean follow-up 359.2 (+/- 323.7) days. A greater proportion of pts initiating tofacitinib had used >3 prior biologics (14.8%) compared to ustekinumab (6.8%), vedolizumab (1.4%), and TNFi (1.6%). The incidence rate of SI (n=1,572 total) per 100 person-years was 2.62 (95% CI 2.17, 3.13) for tofacitinib; 2.73 (95% CI 2.41, 3.08) for ustekinumab; 2.45 (95% CI 2.23, 2.69) for vedolizumab; and 3.25 (95% CI 3.05, 3.46) for TNFi. In analyses adjusting for 70 total covariates, no significant differences in SIs in pts treated with biologics vs tofacitinib were found (Table 1). Conclusion: In this large US-based claims analysis that adjusted for many covariates, there was no significant difference in the risk of SIs among pts with UC receiving tofacitinib compared to biologics. These findings have important implications for IBD management. Table 1. - Adjusted comparative hazard ratios of serious infections in UC patients on advanced treatments compared to tofacitinib Ustekinumab vs Tofacitinibb Vedolizumab vs Tofacitinibb TNFi vs Tofacitinibb Serious Infectionsa Main analysis (17 covariates)c 1.07 (0.84, 1.39) 1.17 (0.93, 1.48) 1.34 (1.07, 1.69) Minimum treatment time ( >= 6 months) 1.29 (0.97, 1.74) 1.13 (0.86, 1.51) 1.49 (1.14, 1.98) Sensitivity analysis (70 covariates)d 0.96 (0.74, 1.24) 1.06 (0.86, 1.33) 1.11 (0.90, 1.39) Abbreviations: CI: confidence interval; TNFi: tumor necrosis factor inhibitor; UC: ulcerative colitis. Inclusion criteria for the UC overall cohort: aged ≥ 18 years, ≥ 2 outpatient (≥ 30 and ≤ 365 days apart) or ≥ 1 inpatient visit(s) with ICD-9/10 UC diagnosis (K51.X), ≥ 12 months enrollment before index (defined as date of first ICD-9/10 code for UC), and receiving UC therapy at index (tofacitinib, ustekinumab, vedolizumab, or TNFi: infliximab; adalimumab; golimumab) therapy. Exclusion criteria: tofacitinib users with prescriptions of approved JAKi other than tofacitinib at or prior to index date; other advanced treatments users with a history of JAKi use; patients with unknown gender. A patient could only be a new user once for each specific drug; however, a patient could be a new user for a second drug class category. (a) Serious infection was defined as an inpatient diagnosis of the event in the primary diagnosis field. Serious infections included 1749 diagnosis codes, including: bacterial infections (e.g. tuberculosis, nocardiosis, Clostridioides difficile infection, pneumococcal infection, listeriosis), fungal infections (e.g. histoplasmosis, cryptococcosis, aspergillosis and candidiasis), and viral infections (e.g. herpes simplex virus, herpes zoster virus, human papilloma virus, influenza virus, Epstein-Barr virus and cytomegalovirus). Patients were followed from index date to the end of the study period, outcome event, treatment switch or discontinuation, or end of enrollment (whichever came first). (b) Tofacitinib was used as the reference group for all HR calculations. A HR greater than 1 suggests the comparator treatment is associated with a higher rate of the safety event compared to tofacitinib. A HR below 1 suggests the comparator treatment is associated with a lower rate of the safety event compared to tofacitinib. (c) IPTW-weighted Cox proportional hazards models were used to compare safety outcomes between tofacitinib and each comparator arm. Stabilized IPTW weights accounted for age, gender, year entered into cohort, number of prior biologics used, baseline glucocorticoid use, conventional treatment use, nonsteroidal anti-inflammatory drug use, anti-platelet use, anti-coagulant use, statin use, oral contraceptive or hormonal therapy use, diabetes, non-alcoholic fatty liver disease, chronic kidney disease/dialysis, other immune deficiencies or immunological conditions, history of myocardial infarction, stroke, or VTE, and history of malignancy. (d) Additional adjustment covariates included Comorbidities: baseline history of atrial fibrillation, coronary artery disease, extra-intestinal manifestations, heart failure, history of C. diff, hyperlipidemia, hypertension, interstitial lung disease or chronic obstructive pulmonary disease or asthma, obesity, peripheral vascular disease, serious infection, Charlson Comorbidity Index, smoking; Medication use: antidepressants, angiotensin II receptor blockers, angiotensin converting enzyme inhibitors, antiarrhythmic drugs, beta blockers, calcium channel blockers, chronic obstructive pulmonary disease maintenance medication, diuretics, nitrates, lipid lower drugs, non-insulin diabetes medications, insulin, Cox-2 inhibitors, opioids, corticosteroid use, 5-ASA, thiopurines; Health Care Utilization: number of UC visits, number of emergency department visits, any hospitalization, recent hospitalizations (60 days), electrocardiogram, echocardiogram, colonoscopy, mammogram, PSA test, Pap smear, pneumococcal vaccine, flu vaccine, insurance type; UC-Related Measures: location of disease (pancolitis, left sided, proctosigmoiditis, proctitis, other, unspecified), anemia, weight loss, primary sclerosing cholangitis, history of colectomy, UC endoscopy, UC imaging, and intestinal polyps. Other outcomes in this protocol included myocardial infarction, stroke, thromboembolic events (presented separately) and malignancy (analyses ongoing).

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.