Analytical variability of the Fibrotest proteins

Objectives: The analytical variability of the Fibrotest (FT) parameters raises the issue of the test's reliability for routine use. Whereas standardization has been proposed by the International Federation of Clinical Chemistry (IFCC) for specific proteins, few data are available concerning the actual transferability of the FT proteins, i.e. haptoglobin, apolipoprotein A, and 02 macroglobulin. The aim of this study was to evaluate the analytical variability of the FT proteins. Design and methods: During the FIBROPACA study, we evaluated 112 sera from patients with hepatitis C infection who underwent liver biopsy. We compared measurements of haptoglobin, apolipoprotein (A1) and alpha(2) macroglobulin by the autoanalyzers Immage (R) (Beckman-Coulter) and the FT reference BNProspec (R) (Dade-Behring). Results: Optimal concordance was found for haptoglobin (correlation: y = 1.05x -0.09; correlation coefficient = 0.98). However, apolipoprotein A, as determined with Immage (R) was globally 12% lower than with BNProspec (R) (correlation: y = 0.88x -0.05; correlation coefficient = 0.91) and alpha(2) macroglobulin values were 40% greater with Immage (R) than with BNProspec (R) (correlation: y = 1.40x -0.46; correlation coefficient = 0.96). Conclusions: Inter-technique analytical variability of the Fibrotest parameters remains a major issue. After IFCC standardization of specific proteins, some discrepancies remain for alpha(2) macroglobulin and, to a lesser extent, for apolipoprotein (A1). National and international quality control programs would be useful to monitor analytical performance of protein assays. (c) 2005 The Canadian Society of Clinical Chemists. All rights reserved.