Introduction: Discrepancies have been previously reported for one-stage clotting and chromogenic assays for FVIII activity analysis. Inter-laboratory variations in instruments, method of clot detection, assay set-up, reference standard calibration, reagent source and reagent composition all contribute to assay variability. Aim: To characterise multilaboratory assay variability in measuring ADYNOVATE, OBIZUR and ADVATE FVIII activity in human plasma and survey multinational FVIII activity assay preferences. Methods: As samples from patients treated with either of the FVIII products are not available in the quantities required for a systematic collaborative study, haemophilia A plasma was spiked in vitro with either ADYNOVATE (PEGylated rFVIII), OBIZUR [Porcine Sequence Antihaemophilic Factor (Recombinant)] or ADVATE at high (0.80 IU or U mL ) and low (0.05 IU or U mL À1 ) FVIII concentrations, based on labelled potencies. Clinical laboratories used their routine FVIII activity assay to determine FVIII activity of each product. Thirty-five data sets using one-stage clotting assay and 11 sets using chromogenic assay were obtained. Results: A vast majority of laboratories (98%) prefer and rely on the one-stage clotting assay. Mean recoveries across all concentrations were 113%, 120% and 127% for ADYNOVATE, OBIZUR and ADVATE respectively. Assay variation was comparable between ADVATE, ADYNOVATE and OBIZUR with inter-laboratory percent coefficients of variation (%CV) ranging from 11 to 22%. Mean chromogenic assay results were 116%, 51% and 113% for ADYNOVATE, OBIZUR and ADVATE respectively. Inter-laboratory CV's were similar for ADYNOVATE, OBIZUR and ADVATE. Conclusions: One-stage clotting assays can and will be used with sufficient accuracy and precision for the measurement of ADYNOVATE, OBIZUR and ADVATE in plasma samples from subjects with haemophilia A. Chromogenic assay underestimates OBIZUR potency, particularly at lower concentrations.

We review the following labels used as substitutes for radioisotopes in immunoassay systems: bacteriophages, chemiluminescence precursors, fluorochromes, fluorogens, fluorescence quenchers, enzymes, coenzymes, inhibitors, substrates, various particulates, metal atoms, and stable free radicals. New methods for performing immunoassays with these labels are described where appropriate. Methods that require no separation steps and offer special promise for easy automation are noted.

We describe an immunoassay for thyroxine in serum. In the assay specific antibody covalently bonded to latex particles is used, along with horseradish peroxidase as the label, and o-phenylenediamine as the chromogen. The flexible protocol is designed for manual execution. Performance is similar to that of the highest-sensitivity thyroxine radioimmunoassays. Results correlate well with radioimmunoassay (r = 0.99, slope = 0.93, y-intercept = 2.4 microgram/liter for 201 samples) and an automated enzyme immunoassay (r = 0.97, slope = 0.99, y-intercept = 4.7 coefficients of variation are less than 7.2% over the entire useful range of the assay (20--240 microgram/liter). The limit of detection is less than 94 pg/tube at 20 microgram/liter. Only D-thyroxine is known to interfere with serum assays. This assay has no discernible protein effect from 40 to 80 g of protein per liter, unlike many thyroxine radioimmunoassays. Serum preservatives known to be peroxidase inhibitors do not adversely affect assay performance because of the 56-fold dilution in the final assay mixture. Hemolyzed serum and EDTA-treated plasmas are unsuitable for this assay.

No abstract

No abstract