INFLIXIMAB (IFX) is increasingly the prescribed therapy for chronic inflammatory bowel disease in paediatric patients. Successful treatment with IFX is dependent on numerous factors, including sufficient drug levels, as stipulated by modern treatment guidelines. Traditional methods of measuring IFX levels, such as enzyme-linked immunosorbent assay (ELISA) can take several days on average, which may be extended where samples have to be shipped to third-party laboratories. In light of the less-than-ideal duration of traditional test methods, researchers from the University of Cologne, Germany, analysed the efficacy of point-of-care testing (POCT) as an alternative method on the basis of its rapidity, taking a mere 30 minutes to establish IFX levels.
The team conducted comparative analyses of ELISA and IFX-POCT following routine surveillance of outpatients with IBD, via blood samples extracted prior to subsequent IFX infusion. A post hoc comparison of the results from the two techniques was performed using Passing–Bablok regression, with a 95% confidence interval (CI) for the intercept and slope coefficients. Correlation was evaluated using both Pearson and Spearman methods, and the results were subsequently visualised with a Bland–Altman plot.
In a cohort of 22 patients, IFX trough levels were measured using both ELISA and POCT methods. Passing–Bablok regression analysis indicated that systematic error could be excluded, as the 95% CI for the axis intercept included 0 (0.369; 95% CI: 0.030–2.030). However, proportional differences could not be ruled out, as the 95% CI for the slope did not include 1 (0.875; 95% CI: 0.540–0.987). Pearson (0.841; P<0.0001) and Spearman (0.972; P<0.0001) correlations demonstrated a strong association between POCT and ELISA measurements. The interclass correlation coefficient showed substantial agreement with a value of 0.64. The Bland–Altman blot indicated positive bias for the POCT results (+2.659), with no significance at a low n (P=0.064). The deviation was, typically, more pronounced at high IFX trough levels.
In summary, while there is strong overall correlation and substantial agreement between the two methods, there are indications of potential proportional differences and a positive bias in POCT results, particularly at higher IFX levels. Specifically, in cases where large deviations were observed, POCT measurements tended to be higher than those obtained with ELISA. These factors should be considered when interpreting IFX measurements from these methods in clinical practice.
