WHO/BS/2013.2225 ENGLISH ONLY EXPERT COMMITTEE ON BIOLOGICAL STANDARDIZATION Geneva, 21 to 25 October 2013 Collaborative Study to Evaluate the Proposed 2nd WHO International Standard for Hepatitis A Virus (HAV) for Nucleic Acid Amplification Technology (NAT)Based Assays Jacqueline F. Fryer 1,3, Alan B. Heath 2, Clare L Morris 1 and the Collaborative Study Group * 1
Division of Virology and 2 Biostatistics, National Institute for Biological Standards and Control, South Mimms, Potters Bar, Herts, EN6 3QG, UK 3 Study Coordinator; Tel +44 1707 641000, Fax +44 1707 641050, E-mail Jacqueline.Fryer@nibsc.org * See Appendix 1
Note: This document has been prepared for the purpose of inviting comments and suggestions on the proposals contained therein, which will then be considered by the Expert Committee on Biological Standardization (ECBS). Comments MUST be received by 4 October 2013 and should be addressed to the World Health Organization, 1211 Geneva 27, Switzerland, attention: Department of Essential Medicines and Health Products (EMP). Comments may also be submitted electronically to the Responsible Officer: Dr Ana Padilla at email: padillaa@who.int with a copy to David Wood at email: woodd@who.int © World Health Organization 2013 All rights reserved. Publications of the World Health Organization are available on the WHO web site (www.who.int) or can be purchased from WHO Press, World Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland (tel.: +41 22 791 3264; fax: +41 22 791 4857; e-mail: bookorders@who.int). Requests for permission to reproduce or translate WHO publications – whether for sale or for noncommercial distribution – should be addressed to WHO Press through the WHO web site (http://www.who.int/about/licensing/copyright_form/en/index.html). The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the World Health Organization concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by the World Health Organization in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by the World Health Organization to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall the World Health Organization be liable for damages arising from its use. The named authors alone are responsible for the views expressed in this publication. This publication contains the collective views of an international group of experts and does not necessarily represent the decisions or the polices of the World Health Organization.
WHO/BS/2013.2225 Page 2
Summary This report describes the preparation and collaborative study evaluation of the replacement 2nd WHO International Standard for hepatitis A virus (HAV) for use in the standardization of nucleic acid amplification techniques (NAT). Two freeze-dried candidates were evaluated, each comprising HAV RNA-positive plasma diluted in pooled human plasma. One candidate was a new preparation of lyophilized HAV plasma. The other candidate was an older batch of HAV plasma (prepared from the same bulk as the 1st WHO International Standard for HAV, NIBSC code 00/560), lyophilized in 2001, and evaluated alongside 00/560 in the original WHO collaborative study in 2001. In the present study, fifteen laboratories from twelve countries participated in a collaborative study to evaluate the fitness for purpose and potency of each candidate using their routine NAT assay for HAV. The freeze-dried candidates (samples 1 and 3) were evaluated alongside the 1st WHO International Standard for HAV, 00/560 (sample 2), and two HAV RNA-positive plasma samples. A range of HAV NAT assays were used in the evaluation, the majority of which were commercial qualitative assays based on real-time PCR technology. The variability in individual laboratory mean estimates for samples 1-3 was greater for qualitative assays (~2 log10) than for quantitative assays (~0.4-0.5 log10 IU/mL). The agreement between laboratories for each sample was improved when the potencies were expressed relative to the 1st WHO International Standard for HAV (sample 2). The overall mean potency estimates for the candidate samples 1 and 3, relative to the 1st WHO International Standard for HAV (sample 2), from qualitative and quantitative assays combined, and were 4.79 and 4.73 log10 IU/mL, respectively. For candidate 12/234 (sample 1), there was a discrepancy between the overall mean estimates determined by qualitative vs. quantitative assays which needs further investigation. The results obtained from accelerated thermal degradation studies at 3 months indicate that there may be some slight loss of potency when candidate 12/234 is stored at -20 °C. However, the results are from a limited number of variable assays, and further data from extended stability studies will be required before a reliable estimate of the long term stability of 12/234 can be obtained. The results from a transport study carried out in 2007 5suggest that candidate 00/562 is stable when stored at 20 °C. However, there was some evidence for degradation at elevated temperatures. Overall, the results of this study indicate the suitability of the candidate 00/562 as the replacement 2nd WHO International Standard for HAV, when maintained at or below -20 °C. It is therefore proposed that the candidate sample 3 (NIBSC code 00/562) is established as the 2nd WHO International Standard for HAV for NAT, with an assigned potency of 54,000 IU/mL (~4.73 log10 IU/mL), when reconstituted in 0.5 mL of nuclease-free water. It is also proposed that when established, the standard should be shipped to customers on dry ice, with specific instructions for storage of the material at -20 °C or below.
Introduction Stocks of the 1st WHO International Standard for HAV are diminishing and need to be replaced. The HAV International Standard is used by manufacturers of in vitro diagnostic devices (IVDs), blood product manufacturers, control authorities, and clinical laboratories, to calibrate secondary reference materials and in the validation of HAV NAT assays.
WHO/BS/2013.2225 Page 3 The need to standardize NAT-based assays for HAV and maintain the availability of the HAV International Standard is ongoing. Detection of HAV contamination of plasma pools used in the manufacture of blood derivatives is required as a measure to reduce the risk of transmission of infections. In Europe, NAT is the recommended method for the detection of HAV RNA in human plasma (pooled and treated for virus inactivation) 1, and assays are required to detect 100 IU HAV RNA. NAT is also widely used for the detection of HAV in environmental and clinical samples. A range of both commercial and laboratory-developed NAT-based assays are currently in use. International Standards are prepared in accordance with published WHO recommendations 2. The 1st WHO International Standard for HAV was prepared from a wild-type isolate (Genotype IA), derived from human plasma, and diluted in HAV RNA-negative pooled human plasma (positive for antibodies to HAV). At the same time, a potential replacement candidate was also prepared (NIBSC code 00/562). Both materials were prepared from the same bulk, but filled and freeze-dried on two separate occasions in 2001. The two materials were evaluated in parallel in a worldwide collaborative study using a range of NAT-based assays for HAV 3,4. The first candidate (NIBSC code 00/560) was established as the 1st WHO International Standard for HAV RNA in 2003, with an assigned potency of 100,000 International Units (IU)/mL, when reconstituted in 0.5 mL nucleasefree water. In the ECBS report on the collaborative study for these materials it was suggested that the second candidate 00/562 could be considered as a replacement candidate standard subject to ongoing stability testing. However, continued stability assessment of accelerated thermal degradation samples from this batch suggested that there was some loss of potency (~1 log10) upon storage at +4 °C for 5 years and 10.5 months 5. The corresponding loss of potency for 00/560 was less than 0.1 log10. This report describes the preparation of a new replacement candidate, and the collaborative study evaluation of this and the older candidate, 00/562, as the replacement 2nd WHO International Standard for HAV for NAT. The new candidate has been prepared from a different HAV RNA genotype IA-positive plasma sample to the 1st WHO International Standard and candidate 00/562, diluted in HAV RNA-negative, anti-HAV-positive (total antibodies), pooled human plasma. The proposal to replace the 1st WHO International Standard for HAV for NAT was endorsed by the WHO ECBS in October 2011. The proposal and update was also presented at the joint 24th Blood Virology and 4th Clinical Diagnostics meeting of the Scientific Working Group on the Standardization of Genome Amplification Techniques (SoGAT), in Slovenia in May 2013. The proposed standard is intended to be used in the in vitro diagnostics field and it relates to ISO 17511:2003 Section 5.5.
Aims of the study The aim of this collaborative study was to evaluate the suitability and potency of two candidate freeze-dried preparations of HAV in parallel with the 1st WHO International Standard for HAV (NIBSC code 00/560), and two liquid-frozen HAV RNA-positive plasma samples, using a range of NAT-based assays.
WHO/BS/2013.2225 Page 4
Materials Candidate standards One new candidate material has been prepared, comprising freeze-dried human plasma and HAV at a concentration of approximately 100,000 IU/mL. The HAV was sourced from a HAV RNA genotype IA-positive (anti-HAV-negative) plasma sample. The pooled human plasma diluent was sourced from blood donations and had been tested and found negative for HIV antibody, HCV antibody, HBsAg and syphilis. It was also tested at NIBSC and found negative for B19V DNA, HAV RNA and HCV RNA by NAT. The plasma tested positive for total anti-HAV antibodies using the ETI-AB-HAVK PLUS ELISA assay (DiaSorin S.p.A., Vercelli, Italy). The second candidate material comprises the same HAV plasma bulk as the 1st WHO International Standard for HAV, but was filled and freeze-dried on a separate occasion in 2001. Both preparations are freeze-dried to ensure long-term stability.
Preparation of bulk materials The concentration of the HAV stock for the new candidate was determined at NIBSC using the COBAS® AmpliPrep Total Nucleic Acid Isolation Kit (Roche Molecular Systems, Inc., Branchburg, NJ, USA) in combination with an in-house real-time PCR assay for HAV RNA. Samples were tested alongside a dilution series of the 1st WHO International Standard for HAV (00/560). The bulk preparation was formulated to contain approximately 100,000 IU/mL of HAV in a final volume of 1 L of pooled human plasma, and mixed for a total of 20 minutes using a magnetic stirrer. Aliquots comprising 0.55 mL volumes of the liquid bulk were stored at -80 °C for evaluation against the freeze-dried product. The bulk was stored at -80 °C prior to shipping to an external facility for filling and lyophilization into the final product, NIBSC code 12/234. The preparation of the bulk for candidate 00/562 has been described previously 3,4.
Filling and lyophilization of candidate standards The filling and lyophilization of the new candidate bulk was performed under contract at an external Containment Level 3 facility (eQAD, UK NEQAS, Colindale, UK), and the production summary is detailed in Table 1. On the day of filling, the bulk was thawed in a 37 °C circulating water-bath. The bulk was removed from the water-bath when just thawed and stirred constantly during the filling process. The bulk was dispensed in 0.5 mL volumes into 3 mL screw-cap glass vials using a repeat pipettor. The homogeneity of the fill was determined by performing check-weighing of approximately every fiftieth vial, with vials outside the defined specification being discarded. Filled vials were partially stoppered with 13 mm diameter freeze drying stoppers and lyophilized in a Christ freeze dryer. Vials were loaded onto the shelves at 20 °C and the shelves were held at 4 °C for 30 mins. The freeze dryer was then cooled to -35 °C, over 2 hrs, and held at this temperature for a further 1 hr. A vacuum was applied to 100 bar over 1 hr. The shelf temperature was raised to 12 °C over 1 hr 40 mins, and a 100 bar chamber pressure maintained for 20 hrs for primary drying. The shelf temperature was then ramped to 25 °C, over 5 hrs. Secondary drying was performed for a minimum of 10 hrs at 25 °C, with a chamber pressure of 30 bar, before releasing the vacuum and
WHO/BS/2013.2225 Page 5 back-filling the vials with nitrogen. The vials were then stoppered in the dryer, removed and capped, before decontaminating with formaldehyde. The sealed vials were returned to NIBSC for storage at 20 °C under continuous temperature monitoring for the lifetime of the product (NIBSC to act as custodian and worldwide distributor). The filling and lyophilization of candidate 00/562 was performed under contract at Q1 Biotech Ltd. (Glasgow, UK), in March 2001, and has been described previously 3,4. The production summary is detailed in Table 1.
Post-fill testing Assessments of residual moisture and oxygen content, as an indicator of vial integrity after sealing, were determined for 12 vials of product 12/234. For 00/562, assessments of residual oxygen and moisture were not performed at the time of production because of the infectious nature of these preparations and the lack of non-invasive methods available at NIBSC. Instead, data from stability assessments were taken to give an indication of residual moisture levels which can impact on the stability of freeze-dried preparations. Subsequently, assessments of residual moisture and oxygen content, as an indicator of vial integrity after sealing, have been determined at NIBSC (February 2011) for 6 vials of the 00/562 product. Residual moisture was determined by non-invasive nearinfrared (NIR) spectroscopy (MCT 600P, Process Sensors, Corby, UK). NIR results were then correlated to Karl Fischer (using calibration samples of human plasma excipient, measured using both NIR and Karl Fischer methods) to give % w/w moisture readings. Oxygen content was measured using a Lighthouse Infra-Red Analyzer (FMS-750, Lighthouse Instruments, Charlottesville, USA). Samples of the liquid bulk (n=6) and freeze-dried product (n=10) were tested by HAV NAT assay using the COBAS® TaqScreen DPX Test (Roche Molecular Systems, Inc., Branchburg, NJ, USA), alongside a dilution series of the 1st WHO International Standard for HAV (00/560), in order to determine any potential loss of titre of material upon lyophilization.
Stability of the freeze-dried candidates Accelerated thermal degradation studies are underway at NIBSC in order to predict the stability of 12/234 when stored at the recommended temperature of -20 °C. Vials of freeze-dried product are being held at -70 °C, -20 °C, +4 °C, +20 °C, +37 °C, and +45 °C. At specified time points during the life of the product, vials will be removed from storage at each temperature and HAV RNA quantified by NAT (as described for post-fill testing). The stability of 00/562 and 00/560 during simulated shipping at ambient temperature was also investigated. Vials stored at +4 °C and +20 °C for 1 week were compared with vials stored continually at -20 °C. The HAV RNA concentration was determined by NAT (as described for post-fill testing).
Study samples The freeze-dried candidates 12/234 and 00/562 were evaluated alongside the 1st WHO International Standard for HAV (00/560), and two HAV RNA-positive plasma samples (study samples 4 and 5). Study sample 4 comprised anti-HAV-negative plasma, while study sample 5 comprised anti-HAV-
WHO/BS/2013.2225 Page 6 positive plasma. Total HAV antibody levels were determined using the ETI-AB-HAVK PLUS ELISA assay (DiaSorin S.p.A., Vercelli, Italy). The HAV genotype of each sample was determined by sequencing of the VP1/2A junction as previously described 6, and was genotype IA. Both samples were individual donations sourced from Swedish blood donors. The concentration of HAV RNA in samples 4 and 5 was determined by NAT (as described for preparation of bulk materials), and was 5.39×103 IU/mL and 1.60×104 IU/mL respectively. Samples 4 and 5 were aliquotted in 0.5 mL volumes into 2 mL Sarstedt screw cap tubes and stored at -80 °C prior to shipping to participants. Study samples shipped to participants were coded as samples 1-5 and were as follows: – Sample 1 (S1) - Lyophilized preparation 12/234 in a 3 mL screw cap glass vial. – Sample 2 (S2) - Lyophilized preparation 00/560 in a 3 mL crimp cap glass vial. – Sample 3 (S3) - Lyophilized preparation 00/562 in a 3 mL crimp cap glass vial. – Sample 4 (S4) – Liquid frozen HAV RNA-positive, anti-HAV-negative plasma in a 2 mL Sarstedt tube. – Sample 5 (S5) – Liquid frozen HAV RNA-positive, anti-HAV-positive plasma in a 2 mL Sarstedt tube.
Study design The aim of this collaborative study was to evaluate the potency of the two candidate freeze-dried preparations in parallel with the 1st WHO International Standard for HAV (00/560), and two liquidfrozen HAV RNA-positive plasma samples, using a range of NAT-based assays. Study samples were sent to participating laboratories by courier on dry ice, with specific instructions for storage and reconstitution. Samples 4 and 5 were only sent to laboratories performing quantitative NAT assays for HAV RNA.
Study protocol Participants were requested to test dilutions of each sample using their routine HAV NAT-based assay on three separate occasions, using a fresh vial of each sample in each independent assay. In accordance with the study protocol (Appendix 2). The lyophilized samples were to be reconstituted with 0.5 mL of deionized, nuclease-free molecular-grade water and left for a minimum of 20 minutes with occasional agitation before use. Samples 4 and 5 were to be tested neat and were therefore only evaluated by laboratories performing quantitative assays. For quantitative assays, participants were requested to test samples 1-5 neat and to test samples 1-3 at a minimum of two serial ten-fold dilutions (10-1 and 10-2). For qualitative assays, participants were requested to test ten-fold serial dilutions of samples 1-3, around the assay end-point (in order to determine the actual assay end-point). For subsequent assays, participants were asked to test the dilution at the predetermined end-point, and a minimum of two half-log10 serial dilutions either side of the end-point (i.e., at least five dilutions in total). Participants were requested to perform dilutions using the sample matrix specific to their individual assay (e.g. HAV RNA-negative human plasma), and to extract samples prior to HAV RNA measurement.
WHO/BS/2013.2225 Page 7 Participants were requested to report the concentration of each sample in IU/mL (positive/negative for qualitative assays) for each dilution of each sample and return results, including details of methodology used, to NIBSC for analysis.
Participants Study samples were sent to 15 participants representing 12 countries (Appendix 1). Participants were selected for their experience in HAV NAT and geographic distribution. They represented IVD manufacturers, control and contract testing laboratories, and plasma manufacturers. All participating laboratories are referred to by a code number, allocated at random, and not representing the order of listing in Appendix 1. Where a laboratory returned data using different assay methods, the results were analyzed separately, as if from different laboratories, and are referred to as, for example, laboratory 02A, 02B, etc.
Statistical methods Qualitative and quantitative assay results were evaluated separately. In the case of qualitative assays, for each laboratory and assay method, data from all assays were pooled to give a number positive out of number tested at each dilution step. A single ‘end-point’ for each dilution series was calculated, to give an estimate of ‘NAT detectable units/mL’, as described previously 7. It should be noted that these estimates are not necessarily directly equivalent to a genuine genome copy number/mL. In the case of quantitative assays, analysis was based on the results supplied by the participants. Results were reported as IU/mL. For each assay run, a single estimate of log10 IU/mL was obtained for each sample, by taking the mean of the log10 estimates of IU/mL across replicates, after correcting for any dilution factor. A single estimate for the laboratory and assay method was then calculated as the mean of the log10 estimates of IU/mL across assay runs. All analysis was based on the log10 estimates of IU/mL or ‘NAT detectable units/mL’. Overall mean estimates were calculated as the means of all individual laboratories. Variation between laboratories (inter-laboratory) was expressed as standard deviations (SD) of the log10 estimates. Potencies relative to sample 2, the current International Standard (00/560), were calculated as the difference in estimated log10 ‘units per mL’ (test sample – standard) plus the value in log10 IU/mL for the International Standard. Therefore, for example, if in an individual assay, the test sample is 0.5 log10 higher than the International Standard, assigned 5.0 log10 IU/mL, the relative potency of the test sample is 5.5 log10 IU/mL. For the quantitative assays, variation within laboratories, and between assays, (intra-laboratory) was expressed as standard deviations of the individual assay mean log10 estimates. These estimates were pooled across each laboratory and across each sample.
Results and data analysis Validation of study samples and stability assessment Production data for the candidate standard sample 1 (12/234) showed that the coefficient of variation (CV) of the fill mass and mean residual moistures were within acceptable limits for a
WHO/BS/2013.2225 Page 8 WHO International Standard 2 (Table 1). Residual oxygen content for 12/234 was within the NIBSC working limit of 1.1%. Evaluation of multiple aliquots of 12/234 (n=10) at NIBSC prior to dispatch, indicated that the HAV RNA content was homogenous (2SD of <0.06 log10 IU/mL). Comparison of the liquid bulks versus the freeze-dried products indicated that there was no loss in potency for 12/234 upon freeze-drying (data not shown). Production data for the candidate standard sample 3 (00/562) showed that the CV of the fill mass was within acceptable limits for a WHO International Standard 2 (Table 1). The non-invasive moisture measurements were out of range of the calibration curve, therefore, an estimate was made by extrapolating the calibration curve beyond the last calibration point. The residual moisture and oxygen contents were 5.19% and 14.7% respectively. These were higher than expected and may have been due to ingress of atmospheric air during storage of the product at -20 °C. Samples of the candidate standard 12/234 were stored at elevated temperatures, and assayed at NIBSC in parallel with samples stored at -20 °C and -70 °C by HAV NAT (as described for post-fill testing). Nine vials of each sample were evaluated after storage at each temperature for 3 months (tested in four assay runs). The mean estimated log10 IU/mL and differences (log10 IU/mL) from the -70 °C baseline samples are shown in Table 2. A negative value indicates a drop in potency relative to the -70 °C baseline. The initial assay (assay 1) shows a drop in potency for the -20 °C sample relative to the -70 °C sample. However, this is not reflected in the subsequent three assays, and overall there is no observed drop in potency after 3 months at -20 °C. There is a pattern of increasing loss of potency for the higher temperatures of storage. The potencies of the higher temperature samples were expressed relative to the -70 °C sample and combined using weighted geometric means. The Arrhenius model for accelerated degradation studies was applied, to obtain predictions of the expected percentage loss per year or percentage loss per month at the different temperatures. The results are summarized in Table 3, expressed as percentage loss, and also as a drop in log10 IU/mL. These predictions are based on only four assays, with variability in the order of 0.2 log10 between repeat estimates for the same samples. They are not reliable and on-going studies are required to assess the stability of 12/234 more robustly. This limited data does not give adequate assurance that 12/234 is sufficiently stable for long-term storage to serve as an International Standard. The stability of 00/562 during simulated shipping at ambient temperature was also investigated. Vials stored at +4 °C and +20 °C for 1 week were compared with vials stored at -20 °C by HAV NAT (as described for post-fill testing). Six vials stored at each temperature were evaluated in each of three assay runs (total of 18 vials at each temperature). The mean estimated log10 IU/mL and differences (log10 IU/mL) from the -20 °C baseline samples are shown in Table 4, with a negative value indicating a drop in potency relative to the -20 °C baseline. The results show a drop in potency upon storage at both +4 °C (-0.041 log10) and +20 °C (-0.168 log10) for 1 week. This observed drop in potency was not significant (p>0.05), when assessed relative to the intra-assay variability (however, the drop in potency at +20 °C was only marginally not significant). Given the overall trend of increasing degradation at elevated temperatures, it would still be recommended to ship vials of 00/562 on dry ice.
WHO/BS/2013.2225 Page 9 The stability of 00/560 during simulated shipping was also investigated using the same study design as for 00/562. However, only two vials stored at each temperature were evaluated in each of three assays (total of 6 vials at each temperature). The mean estimated log10 IU/mL and differences (log10 IU/mL) from the -20 °C baseline samples are shown in Table 5, with a negative value indicating a drop in potency relative to the -20 °C baseline. The differences from the -20 °C sample are small and within the limits of assay variability. The data are limited, but there is no evidence that exposure of a week at +20 °C would lead to a loss in potency that would be detectable within the limits of the assay variability of the assays used. The stability of both candidates when reconstituted has not been specifically determined. Therefore, it is recommended that the reconstituted material is for single use only.
Data received Data were received from all 15 participating laboratories. Participants performed a variety of different assay methods, with one laboratory performing more than one assay method. In total, 18 data sets were received from 7 quantitative assays and 11 qualitative assays. The results from one laboratory performing a qualitative assay, laboratory 06, were highly variable across dilutions and across repeat assays. The laboratory reported that “the data did not show linearity in the indicated range and also Sample 2 was not amplified well in our system”. The results from this laboratory were not included in subsequent analysis.
Summary of assay methodologies The majority of participants prepared dilutions of study samples 1-3 using HAV RNA-negative human plasma, however, deionized nuclease-free water (laboratory 07) and phosphate buffered saline (laboratory 08), were also used. The range of the dilutions performed varied slightly between each laboratory. Assay methodologies for qualitative and quantitative assays are summarized in Table 6. Laboratory 02 used two extraction methods (the QIAsymphony® DSP Virus/Pathogen Midi Kit and QIAamp® Viral RNA Mini Kit) and two NAT methods (artus® HAV LC RT-PCR Kit and artus® HAV TM RT- PCR Kit), in combination, resulting in four data sets (referred to as; 2A, QS-TM; 2B, QS-LC; 2C, QIAamp-TM; and 2D, QIAamp-LC, respectively). Laboratory 14 did not test samples 4 and 5.
Estimated IU/mL or ‘NAT detectable units/mL’ The laboratory mean estimates of IU/mL (log10) from the quantitative assays and ‘NAT detectable units/mL’ (log10) from the qualitative assays are shown in Tables 7 and 8 respectively. The individual laboratory mean estimates from qualitative and quantitative assays are also shown in histogram form in Figures 1a-1e. Each box represents the mean estimate from one laboratory, and the boxes are labeled with the laboratory code. Qualitative assay results are shaded in grey. It must be remembered that the units are different, and therefore, there is no direct conversion from NAT detectable units to IU. From the Figures 1a-1e, and Tables 7 and 8, there appears to be good agreement between the laboratories performing quantitative assays (range of individual laboratory mean estimates is 0.4-0.8 log10 IU/mL). The results from the qualitative assays are more variable (range of individual laboratory mean estimates is ~2 log10). This is in line with previous experience
WHO/BS/2013.2225 Page 10 of these types of assays. With limited numbers of repeat assays, the estimation of ‘detectable units/mL’ is not precise. Tables 9 and 10 show the overall mean estimates, standard deviations of the log10 IU/mL, and log10 NAT detectable units/mL from quantitative and qualitative assays respectively. The difference in inter-laboratory variability between qualitative and quantitative assays is highlighted by the SD values which range from ±0.16 to ±0.25 log for the quantitative assays, and from ±0.54 to ±0.69 log10 for the qualitative assays. The overall mean estimates from quantitative assays for the current HAV International Standard (00/560) is 5.33 log10 IU/mL, higher than the assigned value of 5.00 log10 IU/mL. This was unexpected since quantitative assays use secondary references that are calibrated in IU by direct comparison with the International Standard, 00/560. The overall mean estimates from quantitative assays for the candidate replacement samples 1 (12/234) and 3 (00/562) are 4.89 and 5.07 log10 IU/mL, respectively. For the qualitative assays, the overall mean for the current International Standard (00/560) is 5.31 log10 NAT detectable units/mL. In the original collaborative study to establish the 1st International Standard for HAV, the corresponding figure was 5.29 log10 detectable units/mL. This shows surprisingly good agreement, given the imprecision of the qualitative assays, the different laboratories taking part, and the potential changes in assay method over the intervening time period. The overall mean estimates from qualitative assays for the candidate replacement samples 1 (12/234) and 3 (00/562) are 5.27 and 5.03 log10 IU/mL, respectively. A comparison of the overall mean estimates for samples 1-3 from qualitative and quantitative assays (Tables 9 and 10), shows that the values for samples 2 (00/560) and 3 (00/562) are similar for both assay types. In contrast, the values for sample 1 (12/234) are ~0.4 log10 lower for quantitative assays compared to those from qualitative assays. The reason for this discrepancy is unclear. For quantitative assays, the inter-laboratory variability for HAV-positive plasma samples 4 and 5 is similar to that of the lyophilized samples 1-3.
Potencies relative to the 1st WHO International Standard for HAV (Sample 2) The estimated concentrations of the candidate samples 1 and 3 were expressed in IU, by direct comparison (relative potencies) to the current International Standard (sample 2), which has an assigned unitage of 105 IU/mL (5.0 log10), as described in the statistical methods section. The individual laboratory estimates for all assays are shown in Table 11, with the results from qualitative assays shaded in grey. Units are log10 IU/mL in both cases. For samples 4 and 5, only results from quantitative assays are available, as these samples were only to be tested neat. The results are also shown in histogram form in Figures 2a-2d. The overall mean estimates and interlaboratory variability are shown in Table 12 (qualitative assays), Table 13 (quantitative assays) and Table 14 (all assays combined). From Figures 2a-2d, and Tables 11-14, it is clear that there is an improvement in the agreement between laboratories for the qualitative assays. The SD between laboratories has reduced from 0.65 to 0.28 and 0.54 to 0.37, for samples 1 and 3 respectively. For quantitative assays there is also a slight improvement in the agreement between laboratories. The SD has reduced from 0.16 to 0.15 log10 and 0.25 to 0.12 log10, for samples 1 and 3 respectively. For sample 4, the SD between laboratories performing quantitative assays has reduced from 0.21 to 0.17 log10, while for sample 5,
WHO/BS/2013.2225 Page 11 there is no improvement in agreement. The reduction in variability for the quantitative assays does not appear to be substantial from the figures, principally because the quantitative assays were all in good agreement initially. The overall mean relative potencies for the candidate samples 1 (12/234) and 3 (00/562) for all assays combined are 4.79 and 4.73 log10 IU/mL, respectively. For sample 3 (00/562), the results from the qualitative and quantitative assays are in close agreement (4.72 and 4.74 log10 IU/mL respectively). However, for sample 1 (12/234), there is a difference of ~0.4 log10 with overall means of 4.96 and 4.55 log10 IU/mL for the qualitative and quantitative assays respectively. Again, the reason for this discrepancy is unclear.
Potencies of samples 4 and 5 relative to samples 1 and 3) The estimated concentrations of samples 4 and 5 were expressed in IU, by direct comparison (relative potencies) to the candidate standards sample 1 (12/234) and sample 3 (00/562), as well as relative to each other, as described in the statistical methods section. The overall means from all assays were used as hypothetical ‘assigned values’. The improvement in inter-laboratory agreement is summarized in Table 15. The inter-laboratory SDs of the estimated log10 IU/mL using the different samples as standards are shown, along with the values from the uncorrected estimates. These figures are for quantitative assays only. The results show that the use of any of the lyophilized samples 1-3 gives a marginal improvement in inter-laboratory agreement. The two candidates sample 1 (12/234) and sample 3 (00/562), give a marginally better agreement when compared to the current International Standard (00/560). The reduction in variability for the quantitative assays does not appear to be substantial from the figures. This is likely to be because the quantitative assays were all in good agreement initially, and the fact that these quantitative results are from three laboratories only.
Inter and intra-laboratory variation The intra-laboratory (between assay) variability was measured as the SD of the individual assay mean estimates of log10 IU/mL across repeat assays. The SD’s were pooled across samples to give a single SD for each laboratory (Table 16). These indicate reasonable or good repeatability between assays for all laboratories. SD’s were also pooled across laboratories to give a single SD for each sample (Table 17). Equally, intra-laboratory variability was similar for each sample, suggesting that the assay repeatability does not vary for the different sample types. For all samples, the interlaboratory variation was greater than the intra-laboratory variation (p<0.01 for samples 1-4; p<0.05 for sample 5). Since each assay tested multiple replicates of samples at different dilutions, the resulting betweenassay variability is lower than would be expected if only a single replicate was tested in each assay. The ‘NAT detectable units’ from the qualitative assays are obtained by pooling all assay data to give a single series of number positive out of number tested at each dilution. As such, there is no comparable analysis of intra-assay variation for the qualitative assays.
WHO/BS/2013.2225 Page 12
Discussion and conclusions In this study, a range of NAT-based assays for HAV have been used to determine the potency and evaluate the suitability of the candidate standards as the replacement 2nd WHO International Standard for HAV for NAT-based assays. The new candidate (12/234) was prepared from a different virus stock than that used for current HAV WHO International Standard (00/560) and candidate (00/562), but was diluted in a similar pooled human plasma material. The candidates 12/234 and 00/562 were freeze-dried to ensure long-term stability, and the production data suggests that the batches are homogeneous. For 12/234, the comparison of the liquid bulk versus the freezedried product indicates that there was no loss in potency upon freeze-drying (data not shown). In the collaborative study, the freeze-dried candidate preparations (samples 1 and 3) were evaluated alongside the 1st WHO International Standard for HAV (sample 2) and two liquid-frozen HAV RNA-positive plasma samples; one anti-HAV-negative (sample 4) and the other anti-HAV-positive (sample 5). The overall mean estimate for the 1st WHO International Standard for HAV determined by quantitative assays was 5.33 log10 IU/mL. This is higher than expected, particularly since quantitative assays use secondary references that are calibrated in IU by direct comparison with the current International Standard, 00/560. The overall mean estimates for the candidate standards 1 and 3, as determined by quantitative assays, were 4.89 and 5.07 log10 IU/mL, respectively (based on the calibration of quantitative assay kits in IU/mL). The overall mean estimates for qualitative assays were 5.27 and 5.03 log10 IU/mL, respectively. The reason for the (~0.4 log) discrepancy between the overall mean estimates for 12/234 for qualitative vs. quantitative assays is unclear and needs further investigation. The agreement between laboratories for samples 1 and 3-5 was only slightly improved when the potency was expressed relative to the 1st WHO International Standard for HAV (sample 2). This highlights the fact that assays were already in good agreement. The relative potencies for candidate samples 1 (12/234) and 3 (00/562) from qualitative and quantitative assays combined were 4.79 and 4.73 log10 IU/mL, respectively. The difference in potency between sample 2 (00/560) and sample 3 (00/562) is similar to that observed in the original collaborative study 3,4 (0.19 log10 units/mL reported in 2003 vs. 0.27 log10 IU/mL in the present study). This might suggest that the increased loss of potency for 00/562 compared to 00/560 upon storage at +4 °C for ~6 years, observed during a stability assessment in 2007 5, was indicative of instability at elevated temperatures rather than instability upon prolonged storage at -20 °C. Accelerated thermal degradation studies for both candidate replacements indicate that there are stability issues when these materials are stored at elevated temperatures. As such, both candidates would need to be shipped to customers on dry ice. Given that candidate 00/562 is prepared from the same bulk material as the current International Standard, 00/560, we would propose that this material would be most appropriate as a replacement for the 1st International Standard for HAV. For the purposes of this study it has only been possible to include a couple of clinical samples. However, there is an improvement in the agreement between laboratories when the estimated concentrations of the two HAV plasma samples (anti-HAV-negative and anti-HAV-positive samples) were expressed relative to each candidate, compared to the uncorrected values. This suggests that the candidate preparations may be of value when used to standardize clinical samples, and there is
WHO/BS/2013.2225 Page 13 no evidence of non-commutability with the two plasma samples that were included in the study. Should the opportunity arise further commutability studies may be carried out in the future.
Proposal It is proposed that the candidate standard, NIBSC code 00/562, is established as the 2nd WHO International Standard for HAV for use in NAT-based assays, with an assigned potency of 54,000 IU/mL (~4.73 log10 IU/mL), when reconstituted in 0.5 mL of nuclease-free water. The uncertainty can be derived from the variance of the fill and is 0.39 %. It is also proposed that when established, the standard should be shipped to customers on dry ice, with specific instructions for storage of the material at -20 °C or below. After collaborative study evaluation and stability assessments 1737 vials are available to the WHO. The proposed standard is intended to be used by IVD manufacturers, blood product manufacturers, control authorities, and clinical laboratories, to calibrate secondary reference materials and in the validation of HAV NAT assays. Proposed Instructions for Use (IFU) for the product are included in Appendix 3. It is acknowledged that the results seen in this study for sample 12/234 are not fully understood. At the start of the study it was expected that 12/234 would perform satisfactory during both the collaborative study and the accelerated thermal degradation studies to be proposed as the replacement International Standard. However, as this report shows an early reduction in stability of 12/234 makes this candidate an unsuitable choice. The current stocks of the 1st WHO International standard for HAV are running very low and will not last until the meeting of the ECBS in 2014. Therefore we would like to establish standard 00/562 as an interim standard to maintain continuity of supply. Real time stability of this candidate will be monitored at NIBSC. It is our intension to investigate further the stability issues and establish a stable standard within the next two years.
Comments from participants 9 of 15 participants responded to the report. Some comments suggested minor editorial changes and these have been implemented. Colleagues at the Paul Ehrlich Institute (PEI) raised concerns over the stability of either product and the suitability of using 12/234 or 00/562 as a replacement. It have been explained that 00/562 would be used as an interim replacement in order to maintain supply whilst the stability issues were addressed further. No other laboratory raised concerns over the suitability of the candidate standard (NIBSC code 00/562) to serve as the 2nd WHO International Standard for HAV for use in NAT-based assays.
Acknowledgements We gratefully acknowledge the important contributions of the collaborative study participants. We thank Dr. Todd Gierman (Grifols Plasma Operations) for providing the HAV-positive plasma used to prepare candidate 12/234. We thank Dr. Anders Olofsson (Octapharma AB, Sweden) for providing the HAV-positive plasma used to prepare study samples 4 and 5. We thank Graham Prescott (NIBSC) for technical assistance.
References
WHO/BS/2013.2225 Page 14 1. Monograph on ‘Human plasma (pooled and treated for virus inactivation)’. European Pharmacopoeia. 07/2013:1646. 2. Recommendations for the preparation, characterization and establishment of international and other biological reference standards (revised 2004). WHO Technical Report Series 2007. Geneva, Switzerland:WHO 2007; 932,73-131. http://www.who.int/bloodproducts/publications/TRS932Annex2_Inter_biolefstandardsrev20 04.pdf 3. Saldanha J, Heath A, Lelie N, Pisani G, and Yu MW. Report on the WHO Collaborative Study to Establish anInternational Standard for HAV RNA Nucleic Acid Amplification Technology (NAT) Assays. WHO ECBS Report 2003;WHO/BS/03.1959. 4. Saldanha J, Heath A, Lelie N, Pisani G, Yu MW and the Collaborative Study Group. A World Health Organization International Standard for hepatitis A virus RNA nucleic acid amplification technology assays. Vox Sang. 2005;89:52-58. 5. Baylis SA, Shah N, Heath AB. Continuing stability studies on the 1st WHO International Standard for hepatitis A virus RNA nucleic acid amplification technology (NAT)-based assays. WHO ECBS Report 2007;WHO/BS/07.2056. 6. Chudy M, Budek I, Keller-Stanislawski B, McCaustland KA, Neidhold S, Robertson BH, Nübling CM, Seitz R, Löwer J. A new cluster of hepatitis A infection in hemophiliacs traced to a contaminated plasma pool. J Med Virol. 1999;57:91-9. 7. Saldanha J; Lelie N and Heath A B. Establishment of the First International Standard for Nucleic Acid Amplification Technology (NAT) assays for HCV RNA. Vox Sang. 1999;76:149-58.
WHO/BS/2013.2225 Page 15 Table 1. Production summary for the candidate standards (S1: 12/234 and S3: 00/562). 12/234 00/562 Hepatitis A virus Hepatitis A virus Filling; 6 February 2013 Filling; 29 March 2001 Lyophilization; 6-11 Lyophilization; 29 MarchFebruary 2013 3 April 2001 Sealing; 11 February 2013 Sealing; 3 April 2001 Presentation Freeze-dried preparation in 3 Freeze-dried preparation in mL screw-cap glass vial 3 mL crimp-cap glass vial Appearance Robust opaque cake Robust opaque cake No. of vials filled 1869 ~2000 Mean fill weight (g) 0.509 (n=40) 0.512 (n=63) CV of fill weight (%) 0.49 0.39 Mean residual moisture (%) 0.46 NIR units (n=12) 5.19 NIR units (n=6) CV of residual moisture (%) 7.6 37.1 Mean oxygen content (%) 0.26 (n=12) 14.7 (n=6) CV of oxygen content (%) 28.0 51.4 No. of vials available to WHO 1605 1737 NIBSC code Product name Dates of processing
Table 2. Stability of 12/234 at 3 months. Assay Temperature (°C) -70 -20 +4 +20 +37 +45 1 5.193 5.070 5.009 4.957 4.886 4.732 2 5.004 5.086 4.963 4.922 4.812 4.484 3 4.978 5.042 5.026 4.978 4.740 4.486 4 4.910 4.910 4.937 4.777 4.710 4.563 Difference in log10 IU/mL from -70 °C baseline sample 0.006 -0.038 -0.113 -0.234 -0.455
Mean log10 IU/mL 5.021 5.027 4.984 4.909 4.787 4.566
WHO/BS/2013.2225 Page 16 Table 3. Stability of 12/234 – Predicted loss per month or loss per year. Temperature Percentage loss Percentage loss Log10 drop Log10 drop (°C) per year per month per year per month -20 12.3 1.1 -0.057 -0.005 +4 42.5 4.5 -0.240 -0.020 +20 10.0 -0.046 +37 20.6 -0.100
Table 4. Stability of 00/562 at +4 °C and +20 °C for 1 week. Temperature (°C) Mean log10 IU/mL Difference in log10 IU/mL from -20 °C baseline sample -0.041 -0.168
-20 +4 +20
4.715 4.674 4.547
Table 5. Stability of 00/560 at +4 °C and +20 °C for 1 week. Temperature (°C) Mean log10 IU/mL Difference in log10 IU/mL from -20 °C baseline sample 0.021 -0.024
-20 +4 +20
5.021 5.042 4.997
WHO/BS/2013.2225 Page 17 Table 6. Collaborative study assay methods and codes. Sample extraction method NAT method Quantitative assays QIAamp® Viral RNA Mini Kit, QIAsymphony® DSP Virus/Pathogen Midi Kit, Roche High Pure Viral Nucleic Acid Kit NucliSENS® easyMAG® artus® Parvo B19 LC/TM PCR Kit ART 6 Assay code No. of data sets
Laboratory-developed realtime PCR (TaqMan probe) Qualitative assays Laboratory-developed realtime PCR (TaqMan probe)
LD
1
QIAamp® MinElute Virus Spin Kit SMI-TEST EX-R&D (Medical Biological Laboratories Co., Ltd.) KingFisher Flex magnetic particles QIAamp® UltraSens Virus Kit cobas® TaqScreen DPX Test (cobas s 201 system) QIAamp® Viral RNA Mini Kit PROCLEIX® TIGRIS® System MagMAX™ Viral Isolation Kit
LD
4
cobas® TaqScreen DPX Test RealStar® HAV RT-PCR Kit PROCLEIX® Parvo/HAV Assay VR-I-207™ (Shanghai Huirui Bio-Tech Co., Ltd.)
CTS RS PRO VR
3 1 1 1
WHO/BS/2013.2225 Page 18 Table 7. Laboratory mean estimates from quantitative assays (log10 IU/mL). nd, not determined. Lab 02A 02B 02C 02D 04 14 15 Assay ART ART ART ART ART LD ART S1 4.72 5.10 4.75 5.06 4.82 4.96 4.79 Sample S2 S3 S4 5.23 5.02 3.82 5.60 5.30 4.23 5.30 5.07 3.76 5.60 5.38 4.09 5.24 5.16 4.16 5.07 4.61 nd 5.28 4.94 3.79
S5 4.09 4.32 3.96 4.34 4.34 nd 4.09
Table 8. Laboratory mean estimates from qualitative assays (log10 ‘NAT detectable units/mL’). Lab Assay 01 03 05 07 08 09 10 11 12 13 CTS RS LD VR LD LD CTS PRO LD CTS Sample S1 S2 S3 5.71 5.63 5.52 4.98 5.40 4.91 4.71 4.41 4.71 3.91 3.91 3.91 5.94 5.94 5.77 5.40 5.73 4.75 5.19 5.61 5.18 5.83 5.96 5.59 5.03 4.82 4.90 5.98 5.66 5.06
Table 9. Overall mean estimates and inter-laboratory variation (log10 IU/mL) for quantitative assays only.
No. of Mean Min Max SD data sets S1: 12/234 7 4.89 4.72 5.10 0.16 S2: 00/560 7 5.33 5.07 5.60 0.20 S3: 00/562 7 5.07 4.61 5.38 0.25 S4: anti-HAV-negative 6 3.97 3.76 4.23 0.21 S5: anti-HAV-positive 6 4.19 3.96 4.34 0.16
Sample
WHO/BS/2013.2225 Page 19
Table 10. Overall mean estimates and inter-laboratory variation (NAT detectable units/mL (Log10) for qualitative assays only.
No. of Mean Min Max SD data sets S1: 12/234 10 5.27 3.91 5.98 0.65 S2: 00/560 10 5.31 3.91 5.96 0.69 S3: 00/562 10 5.03 3.91 5.77 0.54
Sample
Table 11. Laboratory estimates of potency relative to the 1st WHO International Standard for HAV, 00/560 (sample 2), from qualitative and quantitative assays. Units are candidate log10 IU/mL, based on assigned unitage of the International Standard of 105 (5.0 log10) IU/mL. Results from qualitative assays shaded in grey. nd, not determined.
Lab 01 02A 02B 02C 02D 03 04 05 07 08 09 10 11 12 13 14 15
Assay CTS ART ART ART ART RS ART LD VR LD LD CTS PRO LD CTS LD ART S1 5.08 4.49 4.51 4.44 4.46 4.58 4.58 5.30 5.00 5.00 4.67 4.59 4.87 5.21 5.32 4.88 4.51
Sample S3 S4 4.88 nd 4.79 3.58 4.70 3.63 4.76 3.45 4.77 3.48 4.51 nd 4.92 3.92 5.30 nd 5.00 nd 4.83 nd 4.02 nd 4.58 nd 4.63 nd 5.08 nd 4.40 nd 4.54 nd 4.66 3.52
S5 nd
3.86 3.72 3.66 3.74 nd
4.11 nd nd nd nd nd nd nd nd nd
3.82
WHO/BS/2013.2225 Page 20 Table 12. Overall mean estimates and inter-laboratory variation for potency relative to the 1st WHO International Standard for HAV, 00/560 (sample 2). Units are candidate log10 IU/mL, based on an assigned unitage of the International Standard of 105 (5.0 log10) IU/mL (quantitative assays only). No. of Mean Min Max SD data sets S1: 12/234 7 4.55 4.44 4.88 0.15 S3: 00/562 7 4.74 4.54 4.92 0.12 S4: anti-HAV-negative 6 3.60 3.45 3.92 0.17 S5: anti-HAV-positive 6 3.82 3.66 4.11 0.16 Sample
Table 13. Overall mean estimates and inter-laboratory variation for potency relative to the 1st WHO International Standard for HAV, 00/560 (sample 2). Units are candidate log10 IU/mL, based on an assigned unitage of the International Standard of 105 (5.0 log10) IU/mL (qualitative assays only). No. of Mean Min Max SD data sets S1: 12/234 10 4.96 4.58 5.32 0.28 S3: 00/562 10 4.72 4.02 5.30 0.37 Sample
Table 14. Overall mean estimates and inter-laboratory variation for potency relative to the 1st WHO International Standard for HAV, 00/560 (sample 2). Units are candidate log10 IU/mL, based on an assigned unitage of the International Standard of 105 (5.0 log10) IU/mL (quantitative and qualitative assays). No. of Mean Min Max SD data sets S1: 12/234 17 4.79 4.44 5.32 0.31 S3: 00/562 17 4.73 4.02 5.30 0.29 S4: anti-HAV-negative 6 3.60 3.45 3.92 0.17 S5: anti-HAV-positive 6 3.82 3.66 4.11 0.16 Sample
WHO/BS/2013.2225 Page 21 Table 15. Summary of inter-laboratory variability for samples 4 and 5, uncorrected and relative to samples 1-3 (SD in log10 IU/mL) (quantitative assays only).
Uncorrected Estimates Potency vs. S1: 00/560 Potency vs. S2: 12/234 Potency vs. S3: 00/562 Potency vs. S4: anti-HAV-negative Potency vs. S5: anti-HAV-positive
S4: anti-HAV-negative S5: anti-HAV-positive 0.21 0.16 0.17 0.16 0.13 0.12 0.12 0.11 0.08 0.08 -
Table 16. Intra-laboratory standard deviation of log10 IU/mL for quantitative assays. Pooled across samples for individual laboratories.
Lab 02A 02B 02C 02D 04 14 15
Assay ART ART ART ART ART LD ART
SD 0.10 0.06 0.14 0.05 0.18 0.20 0.07
Table 17. Intra-laboratory standard deviation of log10 IU/mL for quantitative assays. Pooled across laboratories for individual samples.
Sample S1: 12/234 S2: 00/560 S3: 00/562 S4: anti-HAV-negative S5: anti-HAV-positive
SD 0.12 0.12 0.11 0.11 0.13
WHO/BS/2013.2225 Page 22
Figure legends Figure 1. Individual laboratory mean estimates (log10 IU/mL) for study samples 1-5 obtained using quantitative and qualitative NAT assays. Each box represents the mean estimate from each laboratory assay and is labeled with the laboratory code. The results from the qualitative assays are shaded in grey.
Figure 2. Relative potencies of samples 1, 3, 4 and 5 against sample 2, for each quantitative or qualitative assay. Units are expressed as candidate log10 IU/mL. Each box represents the relative potency for each laboratory assay and is labeled with the laboratory code. The results from the qualitative assays are shaded in grey.
WHO/BS/2013.2225 Page 23
Figure 1. a. Sample 1: 12/234.
Candidate Standard 12/234 10
9
8
7
6
5
15 05 04 02C 07 02A
14 12 03 02D 02B 10 09 11 01 13 08
4
3
2
1
0
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
IU/ml or NAT detectable units/ml(log10)
WHO/BS/2013.2225 Page 24 b. Sample 2: 00/560. Current International Standard 00/560 10
9
8
7
6
5
4
15 04 02C 07 05 12 14 02A
10 03 02D 02B 13 09 01 11 08
3
2
1
0
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
IU/ml or NAT detectable units/ml(log10)
c. Sample 3: 00/562.
Candidate Standard 00/562 10
9
8
7
6
15 13 12 03 09 07 14 05 02C 02A 10 04 02B 11 02D 01 08
5
4
3
2
1
0
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
IU/ml or NAT detectable units/ml(log10)
WHO/BS/2013.2225 Page 25
d. Sample 4: HAV plasma (anti-HAV-negative).
Study Sample S4 10
9
8
7
6
5
4
3
15 02C 02A 02D 04 02B
2
1
0
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
IU/ml or NAT detectable units/ml(log10)
e. Sample 5: HAV plasma (anti-HAV-positive). Study Sample S5 10
9
8
7
6
5
4
3
15 02C 02A
04 02D 02B
2
1
0
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
IU/ml or NAT detectable units/ml(log10)
WHO/BS/2013.2225 Page 26 Figure 2. a. Sample 1: 12/234 vs. sample 2: 00/560. Candidate Standard 12/234 10
9
8
15 10 04 03 02D 02C 02B 02A 11 09 14 08 07 01 13 12 05
7
6
5
4
3
2
1
0
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Potency relative to 00/560 - IU/ml(log10)
b. Sample 3: 00/562 vs. sample 2: 00/560. Candidate Standard 00/562 10
9
8
7
15 11 08 14 13 10 09 03 02D 02C 02B 02A 12 07 04 01 05
6
5
4
3
2
1
0
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Potency relative to 00/560 - IU/ml(log10)
WHO/BS/2013.2225 Page 27 a. Sample 4: HAV plasma (anti-HAV-negative) vs. sample 2: 00/560.
Study Sample S4 10
9
8
7
6
5
4
15 02D 02C 02A 02B 04
3
2
1
0
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Potency relative to 00/560 - IU/ml(log10)
a. Sample 5: HAV plasma (anti-HAV-positive) vs. sample 2: 00/560. Study Sample S5 10
9
8
7
6
5
15 02D 02C 02B 02A 04
4
3
2
1
0
3.0
3.5
4.0
4.5
5.0
5.5
6.0
6.5
7.0
Potency relative to 00/560 - IU/ml(log10)
WHO/BS/2013.2225 Page 28 Appendix 1 Collaborative study participants (In alphabetical order by country) Name Dr. Agnieszka Dryla Maurice Mogg Mao Qunling Zhenglun Liang Dr. Alke Heitmann Sabine Raith Dr. Stefan Kulick Dr. Angeles JuradoJimenez Dr. Michael Chudy Dr. Julia Kreß Dr. Micha Nübling Dr. Antonia Zucchini Dr. Alessia Monti Dr. Barbara Pacini Dr. Francesca Gracci Dr. Giulio Pisani Dr. Francesco Marino Dr. Daniela Adriani Dr. Saeko Mizusawa Dr. Yoshiaki Okada Dr. Denis York Leigh-Anne Edwards Dr. Dolors Xairó Dr. Cristina Alemany Dr. Mats Olsson Dr. Anders Olofsson Thomas Roten Peter Baillod Yi-Chen Yang Dr. Der-Yuan Wang Dr. Boris Hogema Dr. Maarten Koot Dr. Zhuang Wang Hong Wang Laboratory Baxter AG, Plasma Analytics, Vienna National Institutes for Food and Drug Control (NIFDC), Beijing altona Diagnostics GmbH, Hamburg QIAGEN Hamburg GmbH, Hamburg Country Austria China Germany Germany
Paul-Ehrlich-Institut (PEI), Langen
Germany
Kedrion Biopharma, Bolognana-Gallicano
Italy
Biologicals Unit, National Center for Immunobiologicals Research and Evaluation (CRIVIB), Istituto Superiore Di Sanita, Rome Dept. of Safety Research on Blood and Biological Products, National Institute of Infectious Diseases, Tokyo Molecular Diagnostic Services (Pty) Ltd., Westville BIOMAT, S.A. (GRIFOLS), Barcelona Octapharma AB, Stockholm CSL Behring, Bern Food and Drug Administration, Taipei Sanquin Diagnostic Service, Amsterdam Roche Molecular Systems, Branchburg
Italy
Japan
South Africa Spain Sweden Switzerland Taiwan The Netherlands USA
WHO/BS/2013.2225 Page 29
Appendix 2 Study protocol
WHO/BS/2013.2225 Page 30
WHO/BS/2013.2225 Page 31
WHO/BS/2013.2225 Page 32
Appendix 3 Proposed instructions for use
WHO/BS/2013.2225 Page 33