Article Figures & Data
Figures
- Fig 1
Scatter plots illustrating the comparison of pairs of laboratories for pertussis toxin antigen 1 (PT1). The values plotted are the geometric mean values obtained for the replicate values of samples by a given laboratory using the indicated PT antigen. The green line indicates the Deming regression line for the illustrated comparison. The red line represents the line of identity indicating perfect agreement between the indicated pair of laboratories.
- Fig 2
Scatter plots illustrating the comparison of pairs of antigens for laboratory A. The values plotted are the geometric mean values obtained for the replicate values of samples by a given laboratory using the indicated pertussis toxin (PT) antigen. The green line indicates the Deming regression line for the illustrated comparison. The red line represents the line of identity indicating perfect agreement between the indicated pair of laboratories.
- Fig 3
Box plot comparing antigens (1 to 4) for each laboratory. For each laboratory, a consensus value was determined for each sample using the data from that laboratory for all four pertussis toxin (PT) antigens. The relative difference between the consensus and actual values reported by the laboratory was calculated and plotted. This box plot illustrates the distribution of reported results for that laboratory by calculating the log, base 2, of the ratio of each measured value to the consensus value for that sample. For example, a point at 2 on the y axis indicates the value from one antigen within a lab is twice as large as the consensus value. The distributions are plotted separately for each PT antigen. The extremes of the box represent the 25th and 75th percentiles, the middle line represents the median, and the asterisk shows the mean. Vertical lines extend to the most extreme observation that is less than 1.5× the interquartile range (75th to 25th percentiles), and the diamonds and boxes correspond to moderate and severe outlying assay values, respectively. The distance of the mean (asterisk) from the gray dotted line at 0% difference is a direct measure of the mean bias for a given antigen within each laboratory.
- Fig 4
Box plot comparing laboratories (A to D) for each antigen. For each pertussis toxin (PT) antigen, a consensus value was determined for each sample using the data for that antigen from all four laboratories. The relative difference between the consensus and actual values reported by the laboratory was calculated and plotted. This box plot illustrates the distribution of reported results for that antigen by calculating the log, base 2, of the ratio of each measured value to the consensus value for that sample. For example, a point at 2 on the y axis indicates that the value from one lab for an antigen is twice as large as the consensus value. The distributions are plotted separately for each laboratory. The extremes of the box represent the 25th and 75th percentiles, the middle line represents the median, and the asterisk shows the mean. Vertical lines extend to the most extreme observation that is less than 1.5× the interquartile range (75th to 25th percentiles), and the diamonds and boxes correspond to moderate and severe outlying assay values, respectively. The distance of the mean (asterisk) from the gray dotted line at 0% difference is a direct measure of the mean bias for a given laboratory for each antigen.
- Fig 5
Results for standard sera (CBER3). This figure illustrates the results from all four laboratories for the CBER3 reference serum included as a control sample. The means and 95% CIs for the ratios of the measured results to the assigned value (200 ELISA units [EU]/ml) are plotted. A plus sign indicates those PT antigens for which the ratio is significantly different than 1.0. The solid black line indicates an observed/expected ratio of 1.0. The broken red lines mark a ratio of plus and minus 1.25-fold, while the broken blue lines mark a ratio of plus and minus 1.50-fold. N is the number of replicate determinations used to generate the confidence intervals.
- Fig 6
Results for standard sera (IS). This figure illustrates the results from all four laboratories for the IS reference serum included as a control sample or as a blind test sample, either undiluted or diluted 1:2 in negative serum. The means and 95% CIs for the ratios of the measured results to the assigned value (335 IU/ml) are plotted. A plus sign indicates the PR antigens for which the ratio is significantly different than 1.0. The solid black line indicates an observed/expected ratio of 1.0. The broken red lines mark a ratio of plus and minus 1.25-fold, while the broken blue lines mark a ratio of plus and minus 1.50-fold. N is the number of replicate determinations used to generate the confidence intervals.
Tables
- Table 1
Properties of pertussis toxin antigens from the four sourcesa
Property of PT antigen List Biologicals Protein Express GlaxoSmithKline Biologicals Sanofi Pasteur Production strain 165 Tohama I Tohama I 10536 Purity (%) Not available >99 >95 99.5 Bioactivity CHO cell assay CHO cell assay CHO cell assay CHO cell assay Purity assessment method 15% SDS-PAGE 13% SDS-PAGE SDS-PAGE SDS-PAGE/densitometry PT bands Storage buffer components 50% Glycerol 50% Glycerol 50% Glycerol 50% Glycerol 0.05 M Tris (pH 7.5) 0.05 M Tris-Cl (pH 7.6) 25 mM Phosphate buffer (pH 7.6) 37.5 mM Potassium phosphate (pH 8.0) 0.5 M NaCl 0.5 M NaCl 250 mM NaCl 112.5 mM NaCl 0.01 M Glycine Storage temp (°C) −20 −20 −20 2–8 ↵a PT, pertussis toxin; CHO, Chinese hamster ovary; SDS-PAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis.
- Table 2
ELISA procedures for all four laboratoriesa
Assay parameter or reagent CDC Vanderbilt Medical Center Sanofi Pasteur GlaxoSmithKline Biologicals Microtiter plate Immulon 2HB Immulon 2HB Nunc MaxiSorp Nunc MaxiSorp Coating buffer 0.05 M carbonate-bicarbonate (pH 9.6) 0.05 M carbonate-bicarbonate (pH 9.6) 0.05 M carbonate-bicarbonate (pH 9.6) 0.05 M carbonate-bicarbonate (pH 9.6) Coating incubation time 19 h ± 5 h 16–24 h 18 h ± 2 h Overnight to 3 days Coating incubation temp (°C) 5 ± 3 28 (in a humidified chamber) 21 ± 3 5 ± 3 Washing solution PBS–0.05% Tween 20 PBS–0.05% Tween 20 PBS–0.05% Tween 20 0.9% NaCl–0.05% Tween 20 Washing cycle 300 μl three times 250 μl five times 300 μl three times with soak 300 μl three times Diluent PBS–4% BSA–0.05% Tween 20 PBS–0.5% BSA–0.5% Tween 20–0.005% PPG 0.01 M PBS–10.0% goat serum–0.05% Tween 20 PBS–0.05% Tween 20 Blocking solution None None 0.01 M PBS–1.0% goat serum–0.05% Tween 20 1% BSA Blocking incubation time None None 45 min ± 5 min 3 h Blocking incubation temp (°C) None None 21 ± 3 37 ± 2 Serum dilution Single point 1:100 Eight 2-fold serial dilutions Eight 2-fold serial dilutions Eight 2-fold serial dilutions Serum incubation time 2 h ± 5 min 2 h ± 5 min 1 h ± 5 min 30 min Serum incubation temp (°C) 22 ± 4 28 (in a humidified chamber) 37 ± 2 RT with shaking Antibody Mouse anti-human IgG Fc Goat anti-human IgG Fcγ Goat anti-human IgG Fcγ Goat anti-human IgG Fc Conjugate Peroxidase Alkaline phosphatase Peroxidase Peroxidase Conjugate incubation time 2 h ± 5 min 16–24 h 1 h ± 5 min 30 min Conjugate incubation temp (°C) 22 ± 4 28 (in a humidified chamber) 37 ± 2 RT with shaking Substrate TMB pNPP TMB OPD Substrate incubation time 10 min 1 h 35 min ± 5 min 20 min Substrate incubation temp (°C) 22 ± 4 28 (in a humidified chamber) 21 ± 3 RT Wavelength read (nm) 450 405/590 450/540 490/620 Reference standard In-house pooled sera In-house pooled sera In-house pooled sera In-house pooled sera Reference standard calibration WHO international reference CBER lot 3 CBER lot 3 Japanese reference Yoken-H-3 Controls Negative, medium, high positive Low, high positive Negative, high positive Medium positive Lower limit of quantitation (LOQ) 15 IU/ml 10 EU/ml 4 EU/ml 5 EU/ml Software Gen5/Bio-Tek Instruments, Inc. SoftMax Pro/Molecular Devices SoftMax Pro/Molecular Devices SoftMax Pro/Molecular Devices Analysis (curve fit) 4PL 4PL Parallel line analysis 4PL ↵a PBS, phosphate-buffered saline; BSA, bovine serum albumin; PPG, polypropylene glycol; RT, room temperature; TMB, 3,3′,5,5′-tetramethylbenzidine; pNPP, para-nitrophenylphosphate; OPD, ortho-phenylenediamine dihydrochloride; WHO, World Health Organization; CBER, Center for Biologics Research and Review; IU, international units; EU, ELISA units; 4PL, four-parameter logistic function.
- Table 4
Concordance correlation coefficients (rc) for all four laboratories
Laboratory rc between pairs of antigens for laboratory A 0.99 B 0.99–1.00 C 1.00 D 0.97–1.00
