P565 Capillary blood based self-collection device for infliximab therapeutic drug monitoring

Abstract

Abstract Background Patient friendly new devices may improve therapeutic drug monitoring (TDM) in inflammatory bowel disease (IBD). Our objective was to evaluate an infliximab (IFX) precision guided dosing tool from capillary blood collected using an upper-arm self-collection device amongst IBD patients and compare its performances to gold standard venous blood collected using venipuncture. Methods Patients under maintenance therapy with IFX were recruited from a single IBD center and enrolled in the BEST (Bayesian Estimates Sustaining Trough) study. BEST is a prospective real world evidence clinical utility study to establish the value of an IFX Bayesian forecasting tool in the optimal management of IBD. Blood specimens were collected prospectively, at midcycle or at trough. Capillary blood specimens were collected using an upper arm self-collection device (Tasso Inc., Seattle, WA) connected to a serum separator tube. Paired venous blood was collected using venipuncture. IFX concentrations and antibodies to IFX (ATI) status were measured using drug tolerant homogenous mobility shift assay, albumin levels were measured using nephelometry. IFX Clearance was calculated using nonlinear mixed effect model with Bayesian priors. Statistical analysis consisted of Deming regression. Results Overall, 27 patients (n=9 with UC, mean age=37 years, weight=79 Kg) were enrolled. Median IFX dosing was 10 mg/Kg, and median dosing frequency every 6 weeks. A total of n=58 serum pairs isolated from capillary and venous blood were tested. PK profiles measured from venous blood were optimal during maintenance with no immune response detected (ATI < 3.1 U/mL) and most patients had IFX concentration at trough above 5 µg/mL (98%, median 15.4 µg/mL IQR: 10.9-28.0 µg/mL) and low Clearance (< 0.294 L/day: 89%; median 0.200 L/day IQR: 0.170-0.248 L/day). IFX concentrations and Clearance levels were comparable between venous and capillary blood with high correlation (R2= 0.975; slope =0.928; and R2=0.978; slope =1.035, respectively) (figure). In capillary blood specimens collected at midcycle (median 14 days before the next infusion, n=12 pairs), median IFX concentration was 38.3 µg/mL (IQR: 31.2-64.7 µg/mL) and declined to 21.9 µg/mL (IQR: 14.1-44.5 µg/mL) at trough. There was a high correlation between forecasted and measured trough concentrations in capillary blood (R2=0.925; Deming’s slope=1.093). Similar results were observed with venous blood pairs (R2=0.983; Deming’s slope=1.024). Conclusion IFX therapeutic drug monitoring using patient self-collection of capillary blood performed as well as standard venous blood measurement. This collection method may facilitate the implementation of point of care TDM in clinical practice.