STRATIfy-PK SCORE: A predictive tool for identifying pharmacokinetic failure risk in anti-TNF-treated patient

The purpose of this study is to identify risk factors associated with pharmacokinetic failure, defined as immunogenicity or subtherapeutic drug concentrations, in patients treated with infliximab (IFX) or adalimumab (ADA), and to develop a predictive model (STRATIfy-PK SCORE) to support proactive therapeutic drug monitoring (TDM).

A retrospective cohort study was conducted in adult patients with immune-mediated inflammatory diseases receiving IFX or ADA. Pharmacokinetic failure was defined by anti-drug antibodies or subtherapeutic serum concentrations, based on indication-specific therapeutic thresholds. Multivariate logistic regression models were developed separately for each drug and outcome. Model performance was assessed using area under the receiver operating characteristic curve (AUC) and calibration plots.

Of 457 patients (298 ADA, 159 IFX), 1035 TDM samples were analysed. Subtherapeutic drug concentrations occurred in 59.7% of patients, and immunogenicity was observed in 18.8%. In IFX-treated patients, predictors of immunogenicity included prior low induction or maintenance levels, ankylosing spondylitis and rheumatoid arthritis. For ADA, predictors were non-adherence, prior low maintenance levels, early TDM, monotherapy and rheumatic diagnoses. Predictors of subtherapeutic concentrations included prior low maintenance levels, early TDM, high inflammatory burden, and rheumatoid arthritis for IFX, and prior low maintenance levels, ankylosing spondylitis and early TDM for ADA. Predictive models demonstrated good discriminative performance (AUC 0.795-0.800 for immunogenicity; 0.686-0.700 for subtherapeutic levels) and satisfactory calibration.

STRATIfy-PK SCORE allows early identification of patients at high risk of pharmacokinetic failure and may support proactive, personalized TDM strategies to improve clinical outcomes.