Post-Authorisation Active Safety Surveillance Program Among Patients Treated With Tofacitinib for Polyarticular Juvenile Idiopathic Arthritis (pJIA) and Juvenile Psoriatic Arthritis (PsA) Using Nationwide Swedish Healthcare Registers

Study type

• Study topic:
  Human medicinal product
• Study type:
  Non-interventional study
• Scope of the study:
  Safety study (incl. comparative)
• Data collection methods:
  Secondary use of data

Non-interventional study

• Non-interventional study design:
  Cohort

Study drug and medical condition

• Medicinal product name:
  XELJANZ
• Study drug International non-proprietary name (INN) or common name:
  TOFACITINIB CITRATE
• Anatomical Therapeutic Chemical (ATC) code:
  (L04AA29) tofacitinib

  tofacitinib
• Medical condition to be studied:
  Juvenile idiopathic arthritis

  Juvenile psoriatic arthritis

Population studied

• Short description of the study population:
  The study population will comprise all patients with pJIA or juvenile PsA who receive tofacitinib and are identified from nationwide Swedish healthcare registers following product availability in Sweden since 18 March 2022 through 01 July 2029.
• Age groups:
  • Children (2 to < 12 years)
  • Adolescents (12 to < 18 years)

Study design details

• Study design:
  This is an 8-year active surveillance, secondary data collection study of patients with pJIA or juvenile PsA using data from existing nationwide cohorts from the years 2022-2030 in Sweden.
• Main study objective:
  To estimate the incidence rate of the following outcomes of interest among patients with pJIA or juvenile PsA who initiate tofacitinib and patients with pJIA or juvenile PsA treated with approved bDMARDs: venous thromboembolism, serious infections and other important infections, all malignancies combined (excluding NMSC), lymphoma (examined separately), and lung cancer (examined separately).
• Setting:
  The study period will be defined from 18 March 2022 (corresponding to tofacitinib availability in Sweden) to 01 July 2030 (covering the data available through register linkages on this date) Sweden is a Scandinavian country with 10 million inhabitants. A population-based cohort study conducted in the southernmost county of Sweden reported a mean annual incidence rate for JIA to be 12.8/100,000 children < 16 years (95% CI: 11.3–14.5), with the highest age-specific annual incidence at the age of 2 years (36/100,000) during the years 2002-2010.; Another longitudinal, prospective, population based incidence study reported annual incidence rates of JIA according to the ILAR criteria ranging from 7 patients/100,000 in Iceland and 23/100,000 in the Trondheim; region of Norway, with an annual incidence of 15 per 100,000 per year (95% CI: 12-18). Swedish health care is tax funded and offers universal access. Hospital referral is based on geography rather than insurance status. Patients with JIA are typically treated by rheumatologists, the vast majority of whom work in public and hospital based clinics.; ; Health and demographic information is collected in a series of registers with a high degree of completeness resulting from the mandatory and semi-automated registration of their data. Based on each Swedish resident’s unique personal identification number, issued to all Swedish residents alive in 1947 or born/immigrated thereafter, linkage of data from different registers is possible. The registers are maintained by governmental bodies, who may; perform data linkages and provide de-identified data for research purposes.;
• Comparators:
  Patients with pJIA or juvenile PsA treated with approved bDMARDs: ; 1. Diagnosis of pJIA defined as oligoarthritis, polyarthritis (RF+), or polyarthritis (RF-) or a diagnosis of juvenile PsA defined using ICD-10-SE codes listed in protocol Section 9.2.1.1 and as recorded in the NPR.; 2. Patients younger than 16 years at diagnosis of pJIA or juvenile PsA; 3. Patients aged 2-17 years at initiation of any bDMARD approved for pJIA or juvenile PsA treatment in Sweden (e.g., etanercept, adalimumab, abatacept, tocilizumab, golimumab). This is first use of unique bDMARD, not restricted to first bDMARD use (i.e., not restricted to bDMARD naïve patients). For example, a patient starting etanercept for the first time during the period of 18 March 2022 to 01 July 2029 will be eligible regardless of this patient’s prior use of another bDMARD, for example tocilizumab.; 4. Patients initiating a bDMARD as a monotherapy or in combination with MTX and identified by dispensation of bDMARD in the Prescribed Drug Register during 18 March 2022 and 01 July 2029.
• Outcomes:
  The following outcomes of interest will be examined in the interim and final study report. All outcomes will be identified using ICD-10-SE codes. Please see protocol Annex 1 for relevant ICD-10-SE codes.; Venous thromboembolism; Serious infections and other important infections (including opportunistic infection and tuberculosis); All malignancies combined (excluding NMSC); Lymphoma (examined as a separate outcome); Lung cancer (examined as a separate outcome); Gastrointestinal perforations; Major adverse cardiac events (including MI); Fractures; PML; All-cause mortality; HZ reactivation; NMSC; Interstitial lung disease;
• Data analysis plan:
  Crude incidence rates (IRs) of events overall and stratified by baseline characteristics such as disease activity, subtype of JIA, treatment type defined as monotherapy or combination therapy with MTX on the index date, and prior JIA therapy in both tofacitinib cohort and comparator cohort will be calculated. A comparative analysis will examine the risk of the outcomes of interest among patients from the tofacitinib cohort compared to the patients from the comparator cohort adjusting for confounding by baseline characteristics. Propensity scores will be estimated using baseline characteristics described in the study protocol. Propensity score adjusted IR (per 100 PY) and associated 95% CI will be calculated for the outcomes of interest, for which there are adequate data using an exact Poisson method. Where there are adequate data to compare the risk between cohorts, multivariable Cox proportional hazards models will be fit to compare risk of the outcome of interest.