WHO/BS/2015.2280 ENGLISH ONLY EXPERT COMMITTEE ON BIOLOGICAL STANDARDIZATION Geneva, 12 to 16 October 2015
Preliminary report WHO collaborative study to assess the suitability of an interim standard for antibodies to Ebola virus Dianna E. Wilkinson1,3, Mark Page1, Neil Almond1, Robert Anderson1, Neil Berry1, Thomas Dougall2, Stacey Efsthatiou1, Ruth Harvey1, Mark Hassall1, Giada Mattiuzzo1, Peter Rigsby2, Nicola Rose1, Silke Schepelmann1, Lindsay Stone1, Philip D. Minor1 and the Collaborative Study Group* Division of Virology and 2Biostatistics National Institute for Biological Standards and Control, South Mimms, Potters Bar, Herts., EN6 3QG, UK 3 1
Study Coordinator; Tel +44 1707 641000, Fax +44 1707 641050, E-mail: Dianna.Wilkinson@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 13 October 2015 and should be addressed to the World Health Organization, 1211 Geneva 27, Switzerland, attention: Technologies, Standards and Norms (TSN). Comments may also be submitted electronically to the Responsible Officer: Dr M Nübling at email:nueblingc@who.int © World Health Organization 2015 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
WHO/BS/2015.2280 Page 2 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.
WHO/BS/2015.2280 Page 3 Introduction In support of the WHO response to the Ebola crisis, NIBSC is undertaking a project to develop an International Standard for use in the calibration and control of Ebola antibody assays. The availability of International Standards (IS) for antibodies would facilitate the standardization of Ebola serological methods used in epidemiological studies to measure past or present Ebola virus disease and in vaccinology studies to measure antibodies elicited by vaccination in humans. In the absence of such standards, individual laboratories apply their own reference standards which are not harmonized with other laboratories and methods and thus cannot serve to improve the reproducibility between laboratories. Recommendations made by participants attending the Technical Workshop on the Standardisation of Serological and PCR assays for the detection of Ebola virus (NIBSC, UK, 5-6 March 2015), included the urgent prioritization of the development of an interim Ebola standard for serology assays while pursuing the longer-term goal of establishing an International Standard according to published WHO guidelines and formally endorsed by the WHO Expert Committee on Biological Standardization (ECBS) [1]. This report describes the development and worldwide collaborative study evaluation of a panel of candidate Ebola antibody preparations for the selection of the most appropriate candidate to serve as the interim WHO standard for Ebola antibody assays.
Aims of study The aims [1] of this WHO international collaborative study are to assess the suitability of different antibody preparations to serve as the interim standard with an assigned unitage per mL for use in the harmonization of Ebola serology assays. There is no international conventional reference measurement procedure for Ebola virus antibodies and the interim unitage will not be traceable to the International System of Units (SI) of quantity. characterise the antibody preparations in terms of reactivity/specificity in different assay systems. assess each preparation’s potency i.e. readout in a range of typical assays performed in different laboratories. assess commutability i.e. to establish the extent to which each preparation is suitable to serve as an interim standard for the variety of different samples and assay types. recommend to the WHO Ebola antibody assay working group, the antibody preparation(s) found to be suitable to serve as the interim standard(s).
Materials and Methods Source materials Purified Transchromosomic (Tc) bovine anti-Ebola IgG preparations The source materials were donated by Dr Eddie J. Sullivan, SAB Biotherapeutics, Inc. USA, and is human anti-Ebola antibody purified from bovine plasma collected from transchromosomic (Tc) cattle [2] immunized with experimental Ebola vaccines (Zaire95+Sudan GP DNA vaccine or rGPZaire2014 vaccine). Upon receipt at NIBSC, the purified Tc Bovine IgG samples were diluted to a target protein concentration of 1 mg/mL in sterile PBS-Ca2+-Mg2+ supplemented with 5% human serum albumin.
WHO/BS/2015.2280 Page 4 Plasma obtained from convalescent patients The source materials are plasma samples obtained from three patients recovered from Ebola virus disease (Provided by Dr Richard W. Olaussen, Oslo University, Norway, Dr Susan L. Stramer, American Red Cross (ARC), USA and Sheila MacLennan, National Health Service Blood and Transplant (NHSBT), Leeds, UK). Anticoagulant citrate dextrose solution— formula A (ACD-A) was used to collect the ARC and NHSBT samples. It is known that the Norwegian patient had received ZMAb (anti-GP), Favipiravin (RNA pol inhibitor) and TKM10000802 (iRNA). The NHSBT patient had received Brincidifovar (anti viral) and convalescent plasma treatments. NIBSC has not received information on the treatments that the ARC patient may have received. The three materials had been tested and found negative for Ebola virus RNA and other blood viral markers. Prior to receipt at NIBSC, the convalescent plasmas were held at Public Health England, Colindale, UK until confirmed by PCR that no Ebola RNA could be detected in the materials. Solvent/detergent (S/D) treatment is an established virus inactivation technology that has been applied in the manufacture of medicinal products derived from human plasma for more than 20 years [2]. At NIBSC, as an added precaution, the PCR-negative plasmas were solvent-detergent-extracted using a method validated at NIBSC for the inactivation of HIV-1 IIIB spiked into plasma (Appendix 2). Plasma samples were treated by the addition of 1% v/v TBP and 1% v/v Triton X-100, and incubated at 30°C for 3 hours, with mixing every 15 minutes. 10% v/v soybean oil was added and samples mixed for 30 minutes at room temperature to emulsify the detergent. This was then centrifuged at 3000 rpm for 30 minutes at room temperature to separate the oil/detergent layer from the plasma. The plasma was removed from below the oil layer without disturbing the interface, and the solvent removed using C18 reverse phase chromatography columns (which are activated using methanol, and rinsed in sterile water first). Using a vacuum manifold, 20 ml of oil-free plasma sample was added to each column and run through at 0.2bar. This was repeated twice for each sample to ensure that the solvent was fully removed. The solvent-detergent treated convalescent and negative human plasmas were tested for cytotoxicity against CHO K1, 293TT, Vero and Vero E6 cell lines which typically are used in Ebola virus neutralisation assays. For the cytotoxicity assay, cells were seeded in a 96-well plate at the concentration of 10,000 cells/well in 150 µL/well of the relevant media supplemented with FCS 10%. After 20-24 hours, 25µL of doubling dilutions of the plasma were added to the cells in duplicate. Starting dilution was 1:20 (final dilution in the well) down to 1:1280 (final dilution). The cells were grown for 7 days alongside control cells without the plasma, and the growth of the cells was checked daily for at least the first 3 days consecutively to ensure the growth was comparable to the normal cells. We found that solvent detergent treated plasma, didn’t show any cytotoxicity, and none of the plasma samples sent out in the collaborative study had any toxic effect on the cell lines. Pooled vaccinee plasma The source material is lithium heparin plasma samples obtained from volunteers participating in the Oxford, UK vaccine trial who had been primed with the monovalent formulation of the chimpanzee adenovirus 3 (ChAd3)–vectored vaccine encoding the surface glycoprotein of ebolavirus Mayinga(GSK/NIH vaccine candidate) [3] and boosted with the same ebolavirus gene in a modified vaccinia Ankara (MVA)-vectored vaccine that also contains Sudan ebolavirus and Marburgvirus glycoprotein and Tai Forest nucleoprotein (Bavarian Nordic vaccine candidate).
WHO/BS/2015.2280 Page 5 Coded study samples Table 1 lists the collaborative study samples. All study samples are liquid, filled in 0.1 mL aliquots into 0.5 mL Sarstedt screw-capped tubes. The study samples were stored at -20°C or below until dispatched on dry ice to participants. Sample sets were provided to participants coded and blinded. Due to limiting amounts of some source materials, some sample sets did not contain EBOV Ab Sample Codes 58 and 64. Samples were shipped under NIBSC dispatch reference CS551. Study protocol The final version of the study protocol is given in Appendix 3. In brief, participants were requested to test the samples using their established method(s) for the detection of antibodies to Ebola. Participants were asked to perform 3 independent assays on different days. An Excel reporting sheet was provided with suggested dilutions for assaying each study sample. For each assay, participants were requested to make 2 independent series of dilutions of the study samples, and assay all samples concurrently if feasible. Participants were requested to record in the reporting sheet all essential information including the raw data from each assay. The dispatch of the study samples commenced on 27 May 2015 and participants were requested to return results within 6 weeks of receipt of materials. Participants Seventeen laboratories from five countries completed the study. The participants were from France (1), Germany (2), Italy (1), UK (3) and USA (10). All laboratories are referred to by code number allocated at random and not representing the order of listing in Appendix 1. Participating organisations include government research, public health, counter-measure and regulatory organisations; university and research organisations; developers of biologics, assays and reagents and providers of laboratory services for vaccine trials. Six laboratories reported that they have BSL4 containment facilities for handling Ebolavirus. Assay methods Assays used by participants are summarised in Table 2. Where laboratories performed multiple assay methods, laboratory codes are followed by a letter indicating the different methods e.g. lab 11a, l1b. The assay methods fall into 3 general categories: neutralization of live Ebola virus (Neut); neutralization of Ebola pseudotypes or virus-like particles (PsN); and enzyme immunoassays (EIA). Statistical methods For Neut and PsN assays, median endpoint titres or reduction neutralization titres (RNT) 50 were calculated when more than half of the reported results indicated a positive response for that sample. Relative potencies were reported against sample 79 (taken as representative of convalescent plasma) and sample 31 (taken as representative of Tc bovine material). Enzyme immunoassay data were analysed using a parallel line or sigmoid curve model with untransformed or log transformed responses. Calculations were performed using the EDQM software CombiStats Version 5.0 [3]. Model fit was assessed visually and non-parallelism was assessed by calculation of the ratio of fitted slopes for the test and reference samples under consideration. The samples were concluded to be non-parallel when the slope ratio was outside of the range 0.80 – 1.25 and no estimates are reported. Relative potency estimates from all valid assays were combined to generate an unweighted geometric mean (GM) for each laboratory and assay type, with these laboratory means being used to calculate overall unweighted geometric means for each analyte. Variability between assays within laboratories
WHO/BS/2015.2280 Page 6 and between laboratories has been expressed using geometric coefficients of variation (GCV = [10^s-1]×100% where s is the standard deviation of the log10 transformed estimates).
Results and data analysis Collaborative study data received Seventeen laboratories returned data sets for 27 assay methods (Table 2). Four labs (2, 9, 11a, 12b) returned data for neutralization assays performed under BSL4 containment. Six labs (6, 7, 8, 15, 16b, 17) reported results of pseudotype neutralisation assays. Ten labs (1, 3, 4, 5a, 5b, 5c, 5d, 10, 11b, 12a, 13, 14, 16a, 16c, 16d, 16e, 16f) returned data for 17 EIA-based methods. All laboratories used buffer-based diluents to prepare sample dilutions except for lab 5 which used normal human plasma in initial assays. Examples of diluents used are DMEM, DMEM + 2% or 10% FBS, PBS + 5% non-fat milk + 0.1 or 0.05% Tween and PBS + 2% FCS. Laboratory 5d requested that the competitive EIA data not be submitted for analysis. Due to the pressure on resources for normal human plasma (NHP) as the standard diluent, the laboratory elected to titrate the samples before assaying in a commercial diluent rather than NHP. This caused significant perturbation of the competitive EIA to the extent that the lab will only stand by the raw data of the optical density given by the undiluted sample in tested. Laboratory 10 provided tables representing endpoint titres in units/mL and units/mg IgG (Appendix 4). Laboratory 11a reported endpoint titres for neutralisation. Laboratory 11b tested the different sera at a dilution of 1:200 only and so the data could not be submitted to parallel line analysis. Laboratory 11 provided graphs comparing the results obtained with their neutralisation assay against live Ebola virus Mayinga (11a) and EIA against inactivated Ebola virus Makona (11b) (Appendix 5). Laboratory 14 returned results for an indirect immunofluorescence assay (IFA) of infected cells (BSL4) fixed on slides (BSL2). anti-Ebola IgM and IgG were evaluated. For practical reasons, the test was performed in two steps. First, samples were tested at a fixed dilution (1:20 for IgM and 1:40 for IgG). For the second step, the positive samples were tested at limiting dilution. The IgG and IgM titers are reported as the reciprocal of the highest dilution with positive fluorescence. Laboratory 17 provided a graph depicting RNT50 and RNT80 titres for their PsN (LVV) assay for Zaire GP tc/GIN/14/WPG-C05 (Appendix 6). Scoring study samples as reactive for Ebola virus antibodies Samples 9 and 36 are expected to be negative and convalescent patient samples (28, 43 and 79) to be positive for antibodies to Ebola virus. Vaccinee samples (58 and 64) and vaccinated tc bovine samples (31 and 88) are expected to be positive for anti-EBOV GP IgG. Participants who returned the expected results include laboratories 1, 3, 5a, 7, 8 (except weak reactivity for sample 9 as non-specific), 10, 11a, 11b, 12a, 13, 14, 16b as well as laboratories 4, 5c, 5d, 15 and 16a if sample 36 as a negative control is disregarded (See lab comments below). Anomalous/incongruent results were observed for the following laboratories. Laboratory 2 reported sample 79 only to be positive at low neutralising titre. Laboratory 4 has provided Western blots of the study samples (Appendix 7). Also, laboratory 4 has completed additional assays on the study samples. They report that in all their ELISAs,
WHO/BS/2015.2280 Page 7 both using the Zaire GP antigen and Zaire and Sudan nucleoproteins, the signal in sample 36 is consistently higher than in the negative control which is a commercial pooled human serum from Sigma. This is consistent with the idea that it does not represent a true negative control. Laboratory 6 reported the expected reactivities for samples overall using their criteria of determining the area under the curve. The criteria used by NIBSC for analysis indicate that results were incongruent across dilutions and assays. Laboratory 9 returned incongruent results across assays for some samples. The laboratory also reported likely contamination of some samples. Laboratory 12b identified samples 43 and 88 as indeterminate for neutralising antibody and sample 64 as non-reactive. Additional observations include: Laboratory 5b scored sample 9 as low positive for anti-Ebola virus IgM antibody. Samples 28, 43, 79, 58 and 64 were also scored positive for IgM while samples 31 and 88 were scored negative. Laboratory 8 observed weaker neutralisation titres in assays against Sudan and Tai Forest Ebolaviruses for samples 28, 43 and 79. Laboratory 8 stated that they also tested the samples for anti-VP40 by luciferase immunoprecipitation system (LIPS) but did not report the results as they didn't do the required number of repeats due to expense of the substrate and plates etc. The assay only recognized the recovered patient samples, not the bovine or vaccinated individuals as expected as the vaccines did not have VP40 in the antigen. Laboratory 16c (SUDV EIA), 16d (BDBV EIA), 16e (BDBV EIA) scored samples 28, 43, 58, 64 and 79 positive, but anomalously reported sample 36 as positive. Samples 9, 31 and 88 were scored as negative. Laboratory 16f (VSV negative control assay) observed low positive results for samples 36, 43, 58 and 64. Laboratory 16 has performed retrospective analysis of sample 36 in response to a draft of this report. They have tested hundreds of human and nonhuman primate naïve plasma samples and have never found any reactivity. To double check that this is the case, they tested in parallel samples 36 and plasma samples of 5 blood donors from the NIH blood bank in our EBOV BSL2 virus particle ELISA. The data clearly show that sample 36 reacted with all viruses whereas none of the 5 plasma donor gave any reactivity. The laboratory will continue to investigate the nature of the reactivity of sample 36 but recommends that at this point it will be wise to take sample 36 out of the antibody panel and certainly not use it to determine cutoff values of GP assays. (see Appendix 8) For this reason they suggested to disregard the SEBOV (16c), BEBOV (16d), and MARV16e) data from the report.
Neutralisation assays and pseudotype neutralisation assays Table 3 shows laboratories’ individual assay results for Neut and PsN along with the median RNT50 titre estimates. The median titre values are also expressed relative to sample 79 and sample 31. The Ebola strain, if known, is also given for the challenge virus or antigen used.
WHO/BS/2015.2280 Page 8 Enzyme immunoassays Indirect immunofluorescence assay (IFA) of Ebola Zaire-infected cells fixed on slides Table 4 shows IFA results obtained by lab 14 whose method could not be submitted to parallel line analysis. As expected, samples 9 and 36 scored negative for both anti-Ebola IgM and IgG. The known positive samples 28, 31, 43, 58, 64 and 79 were reported positive for IgG at relatively high endpoint dilutions (1/160-1/1280). Samples 28 and 79 (but not sample 43), which are derived from convalescent patients, also showed low levels of IgM antibodies.
Assay validity of EIA submitted to parallel line analysis Figures 1 & 2 show the EIA slope ratios of samples relative to sample 79 and sample 31, respectively. Samples where the ratio of the slope of the dose-response for the sample to the slope of the dose-response to the reference fell outside an 80-125% range were excluded from further analysis. Method 5b, a capture EIA for IgM, had flat dose response curves for all samples and was not analysed.
EIA potency estimates relative to sample 79 Table 5 lists the individual assay potency estimates along with the geometric mean (GM) potencies for samples relative to sample 79. Table 6 summarises the potency estimates relative to sample 79 for each combination of antibody (analyte) and virus used to produce the target antigen. Overall GM and geometric coefficients of variation (GCV) are also given. The data is also shown in graphical form in Figures 3, for total anti-GP IgG, and Figure 4 for anti-GP total antibody.
EIA potency estimates relative to sample 31 Table 7 summarises the potency estimates relative to sample 31 for each combination of antibody (analyte) and virus used to produce the target antigen. Overall GM and GCV are also given. The data is also shown in graphical form in Figures 5, for total anti-GP IgG, and Figure 6 for anti-GP total antibody.
Summary of potency estimates relative to sample 79 Table 8 summaries the potency estimates for Neut, PsN and EIA assays for each sample.
Stability studies No stability studies have been performed on the study samples to date. The study samples are frozen liquid samples. It is intended that any interim standard implemented would be available over the short term. Stability studies on the liquid interim standard may include assessment of stability after thawing and subsequent storage at 4⁰C and freeze/thaw effects. Salient points and comments from participants Sample 79 was taken as representative of convalescent plasma and sample 31 taken as representative of Tc bovine material. Laboratory 7: Sample 79 is a suitable as an interim standard because of its potency, and because it is convalescent material from human Ebola infection, provided that there is sufficient material to go into the future with. Sample 31 is also suitable as a large amount has been produced
WHO/BS/2015.2280 Page 9 Sample 79 clearly having the highest antibody concentration of all the samples tested and the fact that parallelism/similarity of dose-responses curves is observed would both be factors to support its suitability as a standard. However, there is poor agreement between relative potencies for some samples and it is possible that some further explanation from the labs may help with that. Many variables exist in this study making it challenging to assess the effects of the candidates on intra-lab and inter-lab variability. Factors to consider include platform, sample matrix, diluent, target cell, virus strain of antigen, assay protocol and detection readout method. Tc bovine serum samples are available for use in panels. Tc bovine materials were diluted for this study. Human IgG levels in the Tc bovine materials used in this study were at 1 mg/ml. However, human IgG levels in EBOV Convalescent Abs and Vaccinees Plasma Pool samples tested in this study were 5 to 12 mg/ml. It would be better to have Tc bovine materials at ≥ 5 mg ml for subsequent studies. Although evaluation of assays per se was not a direct goal of this study, it is notable that only three neutralization assays were accurate in detecting Ebola neutralization activity without false positives or negatives (i.e., scored either sample 9 or 36 positive, or any of the anti-EBOV samples negative). These include a single live-virus neutralization assay (Laboratory 11a), a single non-replicating pseudovirion neutralization assay (Laboratory 7), and a single replication-competent pseudovirion neutralization assay (Laboratory 16a). These three assays all reported sample 79 to have the highest titer, followed by sample 31. The trend of anti-Ebola neutralization activity was identical for the Laboratory 11a and 16b and very similar for Laboratory 7 (Fig. 1). 1 0 ,0 0 0
L a b 1 1 a ( L iv e - v iru s n e u tr a liz a tio n ) L a b 7 ( N o n - r e p lic a tin g p s e u d o v ir io n )
M e d ia n E s tim a te
L a b 1 6 b ( R e p lic a tio n c o m p e te n t p s e u d o v ir io n ) 1 ,0 0 0
* n o t a v a ila b le 100
10
* 9 1 8 3 8 7 3 8 2 4
* * * * 8 4 5 6
<<< <<<
6 3
s a m p le
Fig. 1) Three classes of neutralization assay yield similar results. A live-virus neutralization assay (Lab 11a, black bars), non-replicating pseudovirion neutralization assay (Lab 7, grey bars) and replication competent pseudovirion neutralization assay (Lab 16b, striped bars) all generated positive titer data for samples expected to contain antiEbola antibodies, and negative titers for the two negative control samples (36 and 9). Data were sorted by median titer for the Lab 11a assay and plotted from most to least potent. * indicates data not available. < indicates titer below level of quantification.
Laboratory 4 suggests assigning different unitages to the different assay categories
9
WHO/BS/2015.2280 Page 10 Laboratory 16; Some of the pseudotype neutralization assays based on luciferase readings (7 and 15) resulted in neutralization titers 10-100 time higher than the other pseudotype and EBOV neutralization assays. These luciferase-based assays are likely to measure cell entry delay rather than virus neutralization. Laboratory 15: Lab 15: Given there is conflicting data on whether it is the antibody or cellular response that confers protection do we need to say why standards for antibodies were developed and not for CD8 cytotoxic T-cell assays? Can standards for cytotoxic assays be produced? Is this being done in another study? Proposal It is proposed that American Red Cross EBOV Convalescent sample (EBOV Ab Sample Code 79) serve as the interim standard for use in neutralisation, pseudotype neutralisation and enzyme immune assays with an assigned unitage of 1 unit/mL. This is an arbitrary unitage, which needs to be discussed further. Furthermore, discussions should include whether different unitages should be assigned to the different assay categories- Neut, PsN and EIA. Sample 36 should be taken out of the antibody panel and should not be used to determine cutoff values of GP assays. The availability the other collaborative study samples in the form of a WHO panel will help laboratories develop their assays and compare results. Subsequent studies can include formulation of Tc bovine material These samples are available with different blinded codes. Approximately 500 tubes (100uL/tube) of the SD-treated American Red Cross EBOV Convalescent sample are available for distribution. Another ~90mL of the ARC plasma, which has not been SD-treated, is available for future use. Approximately 500 tubes of Purified EBOV139 Tc Bovine IgG (rGPZaire2014) lot PD1401350EG are available for distribution. It is possible to obtain additional amounts of Tc Bovine materials for future use. Acknowledgements We gratefully acknowledge the important contributions of the collaborative study participants. We would also like to thank NIBSC Standards Production and Development for distribution of the candidate materials. We also thank David Wood, Micha Nuebling and Patricia Fast of the WHO and participants of teleconferences for their support, guidance and advice. Reference panel materials were kindly donated by Hua Wu, Jerry Pommer, Eddie Sullivan (SAB Biotherapeutics, Sioux Falls, South Dakota, USA); Teresa Lambe, Sarah Gilbert, Adrian Hill and Katie Ewer (Jenner Institute, University of Oxford, UK); Annie Winkler (Emory University, USA); Scott Koepsell (University of Nebraska); Arne Brantsaeter, Richard Olaussen, Unni Bergerud (Oslo University Hospital); Sheila MacLennan, Alex Barber (National Health Service Blood and Transplant, Leeds, UK). We also thank colleagues Maria Zambon, Angie Lackenby, Simon Carne, Pamela Saunders, Meera Chand and Kevin Brown at Public Health England, Colindale, UK for PCR testing of plasma samples. We also thank Steven A. Rubin, FDA/CBER, USA for facilitating the sample permits and shipments to laboratories in the USA
WHO/BS/2015.2280 Page 11
Tables Table 1. Collaborative study samples. Shipped under NIBSC dispatch reference CS551. EBOV Ab Sample Code 9 Sample Name Preparation
Purified Tc Bovine IgG (negative) lot PD1402332EG-2 NHSBT EBOV Convalescent Ab Purified EBOV139 Tc Bovine IgG (rGPZaire2014) lot PD1401350EG NHSBT EBOV Ab Negative Plasma Norwegian EBOV Convalescent Ab Vaccinees Plasma Pool (high) Prime: ChAd3–vectored EBOV Mayinga GP Boost: MVA-vectored EBOV Mayinga GP, Sudan ebolavirus GP, Marburgvirus GP and Tai Forest nucleoprotein Vaccinees Plasma Pool (low) Prime: ChAd3–vectored EBOV Mayinga GP Boost: MVA-vectored EBOV Mayinga GP, Sudan ebolavirus GP, Marburgvirus GP and Tai Forest nucleoprotein American Red Cross EBOV Convalescent Ab
1 mg/mL in sterile buffer# SD-extracted 1 mg/mL in sterile buffer# SD-extracted SD-extracted Plasma pool not SDextracted
28 31
36 43 58
64
Plasma pool not SDextracted
79
SD-extracted
88
Purified EBOV132 Tc Bovine IgG (Zaire95+Sudan GP DNA) lot PD1401304ED
1 mg/mL in sterile buffer#
Abbreviations: NHSBT= National Health Service Blood and Transplant; SD= Solventdetergent. # PBS-Ca2+-Mg2+; 5% human serum albumin.
WHO/BS/2015.2280 Page 12 Table 2. Laboratory codes and assay methods Lab Code 1 2 3 4 5a 5b 5c 5d 6 Assay Method Description Assay Method Category EIA Neut EIA EIA EIA EIA EIA EIA PsN (LVV) PsN (VSV) Analyte (anti-) unit
Indirect ELISA Seroneutralisation against infectious Ebola virus Indirect ELISA Direct ELISA IgG capture EIA IgM capture EIA Double antigen sandwich EIA Competitive EIA Neutralization assay with pseudotyped lentiviral vector particles Neutralization of Ebola pseudotypes [VSV-luciferase nonreplicating] Neutralization of filamentous Ebola virus-like particles (VLP) Neutralisation of infectious Ebola virus Direct ELISA
Ebola Zaire GP IgG infectious Ebola Zaire virus EBOV GP IgG Ebola Zaire GP Ebola Zaire GP IgG Ebola Zaire GP IgM Ebola Zaire GP Ebola Zaire GP Ebola GP
OD ELISA units/mL FFU OD and ELISA units/mL OD OD OD OD OD RLU
7
Ebola Zaire 95 (Kikwit) GP Ebola Zaire GP infectious Ebola virus Makona Ebola virus Makona rGP IgG live Ebola virus Zaire Mayinga inactivated Ebola virus Makona Ebola Zaire GP IgG
RLU
8 9 10
PsN(VLP) Neut EIA
RLU pfu OD, units/mL and units/mg IgG CPE OD and ELISA units OD
11a 11b 12a
Neutralisation Assay Direct ELISA ADI Human Anti-Zaire Ebola Virus Glycoprotein IgG ELISA kit (ADI # AE-320620-1) Fluorescence Reduction Neutralization Assay (FRNA) Capture ELISA Indirect Immuno-Fluorescence Assay of infected cells fixed on slides. Neutralization assay with pseudotyped lentiviral vector particles Virus particle ELISA coating with Ebola GP pseudotyped VSV RC Ebola (Pseudo)neut VSV RC-GFP Virus particle ELISA coating with SUDV GP pseudotyped VSV RC Virus particle ELISA coating with BDBV GP pseudotyped VSV RC Virus particle ELISA coating with MARV GP pseudotyped VSV RC Virus particle ELISA coating with wt VSV GP pseudotyped VSV RC
Neut EIA EIA
12b 13 14
Neut EIA EIA
live Ebola virus Makona GP particles Fixed Ebola Zaire
Fluorescence OD Fluorescence
15
PsN (LVV)
16a 16b 16c 16d 16e 16f
EIA PsN (VSV) EIA EIA EIA EIA
Ebola virus Makona H.sapienswt/GIN/2014/GueckedouC07 GP Ebola virus Mayinga GP Total Ab rVSV-EBOVgp-GFP SUDV (Gulu) GP BDBV GP MARV (Musoke) GP VSV (Indiana) GP
RLU
OD Fluorescence OD OD OD OD
WHO/BS/2015.2280 Page 13 17 Neutralization assay with pseudotyped lentiviral vector particles PsN (LVV) Zaire GP tc/GIN/14/WPGC05 RLU
Table 2 continued. Abbreviations: EIA = enzyme immunoassay; GP = glycoprotein; OD = optical density; Neut = neutralisation assay; FFU = focus-forming units; EBOV = Ebola virus; PsN = pseudotype neutralisation assay; LVV = lentiviral vector; RLU = relative light units; VSV = vesicular stomatitis virus vector; VLP = virus-like particticals; pfu = plaqueforming unit; CPE = cytopathic effect; RC = replication-competent; GFP = Green fluorescent protein; SUDV = Sudan virus; BDBV = Bundibugyo virus; MARV = Marburg virus.
WHO/BS/2015.2280 Page 14 Table 3. Individual and median endpoint or reduction neutralisation titre 50 (RNT50) estimates for neutralisation assays (Neut) and pseudoneutralisation (PsN) assays, with RNT50s relative to samples 31 and 79 # Samples not provided for testing. n/a = not assessed. Method Lab Analyte antiAssay 1 Virus Sample 9 28 31 36 43 58# 64# 79 88 (1) < 20 < 50 < 20 < 20 < 20 . . < 80 < 20 (2) < 20 < 50 < 20 < 20 < 20 . . < 80 < 20 (3) . . . . . . . . . (1) < 20 < 50 < 20 < 20 < 20 . . < 80 < 20 Assay 2 (2) < 20 < 50 < 20 < 20 < 20 . . < 80 < 20 (3) . . . . . . . . . (1) < 20 < 50 < 20 < 20 < 20 . . < 80 < 20 Assay 3 (2) < 20 < 50 < 20 < 20 < 20 . . < 320 < 20 (3) . . . . . . . . . Median Estimate
Relative to sample 31 . . . . . . . . .
Relative to sample 79 . . . . . . . 1.00 .
Neut
2
Whole virus
Zaire
Negative Negative Negative Negative Negative . . <80 Negative
Lab 2 comments: Regarding the results with our experience of this neutralisation test, it seems that the sera do not contain Ebola neutralizing antibodies. Only one serum (number 79) seems to have a clear neutralizing activity when not too much diluted. The others ones do not demonstrate clear neutralizing activity from our point of view. 9 28 31 36 43 58 64 79 88 320 < 20 < 20 < 20 < 20 < 20 < 20 < 20 < 20 320 < 20 < 20 < 20 < 20 < 20 < 20 < 20 < 20 . . . . . . . . . < 20 40 < 20 < 20 < 20 160 < 20 < 20 < 20 < 20 40 < 20 < 20 < 20 320 < 20 20 < 20 . . . . . . . . . <5 20 MD MD MD MD MD MD 20 10 20 MD MD MD 20 MD MD <5 . . . . . . . . . n/a 30 Negative Negative Negative 160 Negative n/a n/a . . . . . . . . . . . . . . . . . .
Neut
9
Whole virus
Zaire 2014 Makona
Lab 9 comments: Virus was EBOV Makona with initial titer of 8.25E+06 pfu/mL, diluted to target of 1,500 PFU/mL for target of 150 plaques per well. MD = Monolayer damage We observed significant monolayer damage with many of the sera when used at high concentration (Assay 3). This appeared to be bacterial contamination; we cannot say for certain if the contamination arose from our reagents or the sera, but it seems more likely to be sera based on the inconsistent pattern of contamination.
WHO/BS/2015.2280 Page 15 Table 3. continued Method Lab Analyte antiAssay 1 Virus Sample 9 28 31 36 43 58# 64# 79 88 (1) <8 91 91 <8 27 . . 181 91 (2) (3) . . . . . . . . . (1) <8 27 54 <8 27 . . 256 64 Assay 2 (2) (3) . . . . . . . . . (1) <8 45 102 <8 16 . . 181 64 Assay 3 (2) (3) . . . . . . . . . Median Estimate
Relative to sample 31 . 0.50 1.00 . 0.30 . . 2.00 0.71
Relative to sample 79 . 0.25 0.50 . 0.15 . . 1.00 0.35
Neut
11a
Whole virus
Zaire 1976
. .
. .
. .
Negative 45 91 Negative 27 . . 181 64
Endpoint titres reported. Lab 11a comments: …samples 9 and 36 are considered negative throughout our Neut assays. The other samples gave either high or moderate neutralisation titers (79, 31, 88, 28, 43; in descending order). In addition, we tested the provided samples by ELISA (See graph in Appendix 4). Both assays correlated well with each other. 9 28 31 36 43* 58 64 79 88 < 20 20 20 < 20 < 20 40 < 20 160 < 20 < 20 40 20 < 20 20 20 < 20 160 20 . . . . . . . . . < 20 20 20 < 20 20 20 < 20 80 20 < 20 20 20 < 20 < 20 20 < 20 80 < 20 . . . . . . . . . < 20 40 40 < 20 20 20 < 20 160 20 < 20 80 80 < 20 < 20 20 < 20 320 20 . . . . . . . . . Negative 30 20 Negative n/a 20 Negative 160 20 . 1.50 1.00 . . 1.00 . 8.00 1.00 . 0.19 0.13 . . 0.13 . 1.00 0.13
Neut FRNA
12b
Whole virus
Zaire 2014 Makona
FRNA = Fluorescence Reduction Neutralization Assay Lab 12b selected comment: Samples are considered positive if the FRNA50 was calculable. *Sample titer was indeterminate as the two independent dilutions results were incongruent. This study tested seven samples by ELISA (Assay 12a in EIA results) and nine samples by FRNA. The ELISA was more sensitive in identifying antibody titers in the provided samples. The FRNA assay, which detects neutralizing antibodies, identified two samples (code # 43 and 88) with indeterminate titers. However, these samples were identified as positive by ELISA. The incongruent results may be due to low antibody titers in the sample or a mismatch in the antigen which the antibodies developed against and the live virus EBOV/Mak-C05 used in the neutralization assay.
WHO/BS/2015.2280 Page 16 Table 3 continued Method Lab Analyte antiAssay 1 Virus Sample 9 28 31 36 43 58 64 79 88 (1) < 20 < 20 < 20 117 < 20 281 238 172 < 20 (2) < 20 < 20 < 20 < 20 < 20 187 60 113 < 20 (3) . . . . . . . . . (1) < 20 130 1 75 672 1033 < 20 159 < 20 Assay 2 (2) < 20 99 < 20 63 < 20 < 20 < 20 < 20 < 20 (3) . . . . . . . . . (1) < 20 115 < 20 105 89 5682 < 20 164 < 20 Assay 3 (2) < 20 123 < 20 < 20 265 < 20 < 20 180 < 20 (3) . . . . . . . . . Median Estimate Negative n/a n/a n/a n/a n/a n/a 164 Negative Relative to sample 31 . . . . . . . . . Relative to sample 79 . . . . . . . 1.00 .
PsN Lentiviral Vector (LVV)based
6
GP Ig
Zaire
Note: Although NIBSC analysis listed n/a for several samples, Lab 6 scored samples 28,36,43,58, 64 and 79 as positive and samples 9,31 and 88 as negative. 9 28 31 36 43 58 64 79 88 9 28 31 36 43 58 64 79 88 < 20 623 1393 < 20 669 581 72 3080 578 < 20 67 21 25 110 < 20 < 20 132 20 < 20 578 1237 < 20 746 469 121 2525 480 < 20 63 23 27 104 < 20 < 20 137 24 < 20 660 1103 < 20 619 284 118 3214 295 . . . . . . . . . < 20 1331 2431 < 20 879 953 169 5697 636 < 20 64 20 24 103 < 20 < 20 123 24 < 20 1088 1964 < 20 1214 853 206 4961 546 < 20 67 23 25 108 < 20 < 20 127 21 <20 118 7 191 0 <20 102 7 998 314 347 5 499 . . . . . . . . . <20 478 1049 < 20 224 294 124 3134 559 < 20 65 22 23 96 < 20 < 20 123 21 < 20 529 938 < 20 229 322 162 4629 458 < 20 76 24 24 114 < 20 < 20 131 27 < 20 460 1365 < 20 229 257 157 3395 442 . . . . . . . . . Negative 623 1365 Negative 669 469 157 3395 499 Negative 66 22 24 106 Negative Negative 129 22 0.46 1.00 . 0.49 0.34 0.11 2.49 0.37 . 2.98 1.00 1.10 4.81 . . 5.84 1.00 0.18 0.40 . 0.20 0.14 0.05 1.00 0.15 . 0.51 0.17 0.19 0.82 . . 1.00 0.17
PsN Vesicular Stomatitis Virus (VSV)based
7
GP Ig
Zaire 1995 Kikwit
PsN filamentous ebolavirus virus-like particles (VLP)based
8
GP Ig
Zaire
WHO/BS/2015.2280 Page 17 Table 3 continued Method Lab Analyte antiAssay 1 Virus Sample 9 28 31 36 43 58# 64# 79 88 9 28 31 36 43 58# 64# 79 88 9 28 31 36 43 58 64 79 88 (1) < 20 51200 1280 80 20480 . . 40960 640 < 50 < 50 < 50 < 50 < 50 . . 223 < 50 < 20 325 < 20 39 123 < 20 < 20 257 56 (2) < 20 25600 1280 80 20480 . . 40960 640 < 50 < 50 < 50 < 50 < 50 . . 181 < 50 . . . . . . . . . (3) . . . . . . . . . . . . . . . . . . . . . . . . . . . (1) < 20 36450 1620 < 20 14580 . . 43740 540 < 50 84 113 < 50 68 . . 404 < 50 < 20 176 65 30 128 < 20 < 20 378 130 Assay 2 (2) < 20 36450 1620 < 20 14580 . . 43740 540 < 50 86 104 < 50 64 . . 427 < 50 . . . . . . . . . (3) . . . . . . . . . . . . . . . . . . . . . . . . . . . (1) < 20 12150 1620 60 14580 . . 131220 540 < 50 91 125 < 50 58 . . 334 72 < 20 197 53 34 90 < 20 < 20 116 24 Assay 3 (2) < 20 36450 1620 60 14580 . . 131220 540 < 50 107 141 < 50 65 . . 307 61 . . . . . . . . . (3) . . . . . . . . . . . . . . . . . . . . . . . . . . . Median Estimate Negative 36450 1620 70 14580 . . 43740 540 Negative 88 119 Negative 65 . . 320 n/a Negative 197 59 34 123 Negative Negative 257 56 Relative to sample 31 . 22.50 1.00 0.04 9.00 . . 27.00 0.33 . 0.74 1.00 . 0.54 . . 2.69 . . 3.34 1.00 0.58 2.08 . . 4.36 0.95 Relative to sample 79 . 0.83 0.04 0.00 0.33 . . 1.00 0.01 . 0.28 0.37 . 0.20 . . 1.00 . . 0.77 0.23 0.13 0.48 . . 1.00 0.22
PsN Lentiviral Vector (LVV)based
15
GP Ig
Ebola virus Makona H.sapiens wt/GIN/2 014/Guec kedouC07 GP
PsN VSV replication competent
16b
GP Ig
rVSVEBOV gp-GFP
PsN Lentiviral Vector (LVV)based
17
GP Ig
Zaire GP tc/GIN/ 14/WPG -C05
WHO/BS/2015.2280 Page 18 Table 4. Indirect immunofluorescence assay (IFA) reported by laboratory 14*. Lab 14 Sample
IFA Assay 1 (fixed dilution)* IgM Reciprocal dilution IgG Reciprocal dilution IgM
IFA Assay 2 (limiting dilution)** Reciprocal dilution IgG Reciprocal dilution
N 20 N 40 N <10 N <20 9 P/N 20 P 40 P/N 10 P 1280 28 N 20 P 40 N <10 P 1280 31 N 20 N 40 N <10 N <20 36 N 20 P 40 N <10 P 1280 43 N 20 P 40 N <10 P 640 58 N 20 P 40 N <10 P 160 64 P/N 20 P 40 P 20 P 1280 79 N 20 P 40 N <10 P 320 88 * For assay 1,reciprocal dilutions reported are single dilution points tested. **For assay 2, reciprocal dilutions are end-point dilutions; the titre of samples resulting N is assigned <10 for IgM and <20 for IgG, depending on the starting dilution. P = Positive; N = Negative.
WHO/BS/2015.2280 Page 19 Table 5. Individual assay potency estimates relative to sample 79. GM = geometric mean. Lab Analyte Virus Sample 28 31 1 GP IgG Zaire 43 58 64 88 28 31 3 GP IgG Zaire 43 58 64 88 28 31 4 GP Ig Zaire 43 58 64 88 28 31 5a GP IgG Zaire 43 58 64 88 28 31 5c GP Ig Zaire 43 58 64 88 10 GP IgG Zaire 2014 (Makona) 58 64 88 28 12a GP IgG Zaire 31 43 88 28 31# 13 GP Ig Zaire 43 58 64 88 Assay 1 0.33 0.60 0.43 0.53 0.14 0.54 0.23 0.50 0.41 0.42 np 0.41 0.41 0.50 0.58 0.64 0.17 0.52 0.27 np 0.51 0.80 0.22 0.65 nl 0.21 np 0.50 0.12 0.27 0.21 0.07 0.14 0.41 0.93 0.67 0.73 0.89 0.03 0.61 0.87 0.65 0.30 Assay 2 79 nl 79 nl 79 nl 79 nl 79 nl 79 nl 0.23 0.54 0.40 0.50 np 0.40 0.44 0.74 0.70 np 0.13 0.47 0.25 np 0.53 1.18 0.26 np 0.59 0.17 0.07 0.34 0.11 np 0.24 0.07 0.16 0.40 1.05 0.54 0.68 0.87 0.04 np 0.75 0.76 0.38 Assay 3 . . . . . . 0.71 1.66 0.70 0.60 0.22 0.67 0.49 0.85 1.18 0.67 0.14 0.56 0.35 1.33 0.57 1.27 0.30 np 1.11 0.26 np . . np 0.27 0.09 0.19 0.42 1.04 0.57 0.64 . . . . . . Assay 4 0.74 1.21 0.93 0.91 0.63 1.20 0.38 0.97 0.48 0.45 0.14 0.43 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Assay 5 . . . . . . 0.42 0.68 0.43 0.38 0.11 0.39 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Assay 6 . . . . . . 0.39 0.70 0.42 0.41 0.12 0.45 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GM 0.49 0.85 0.63 0.69 0.30 0.80 0.36 0.77 0.46 0.45 0.14 0.45 0.45 0.68 0.78 0.65 0.14 0.52 0.29 1.33 0.50 1.06 0.26 0.65 0.81 0.21 0.07 0.41 0.12 0.27 0.24 0.08 0.16 0.41 1.01 0.59 0.69 0.88 0.04 0.61 0.81 0.70 0.34
WHO/BS/2015.2280 Page 20 Appendix Table 5. Continued Lab Analyte Virus Sample 28 31 16a GP Ig Zaire 1976 (Mayinga) 43 58 64 88 28 31 36** 16c* Gp Ig SUDV (Gulu) 43 58 64 88 28 31 16d* Gp Ig MARV (Musoke) 36** 43 58 64 88 28 31 36** 16e* Gp Ig VSV control 43 58 64 88 Assay 1 1.00 0.77 1.04 4.05 0.73 2.60 0.17 0.02 0.45 0.17 0.18 0.05 0.45 0.25 neg 3.24 0.22 . . neg 1.40 neg 8.75 0.91 . . neg Assay 2 0.47 0.67 0.48 4.42 0.35 1.15 0.37 0.05 1.87 0.21 0.50 0.13 0.60 0.64 neg 13.76 0.70 0.34 0.58 neg 0.45 neg 13.00 0.66 0.52 0.74 neg Assay 3 0.48 0.49 0.62 1.66 np 0.98 0.32 0.04 2.01 0.30 0.19 0.09 0.50 0.51 neg 15.34 0.82 0.40 0.45 neg 0.46 neg 18.49 0.66 0.69 0.64 neg Assay 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . Assay 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . Assay 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . GM 0.61 0.63 0.68 3.10 0.51 1.43 0.27 0.03 1.19 0.22 0.26 0.08 0.51 0.43 neg 8.81 0.50 0.37 0.51 neg 0.66 neg 12.81 0.74 0.60 0.69 neg
Notes: 3 plates from lab 10 are not included since they did not have sample 79 on the plate nl = non-linear 79 nl = reference non-linear so no estimate np = non-parallel # = not used in further analysis (unusually low estimate) * = not used in summaries and plots (Non-Ebola assays). ** = Laboratory 16 has provided additional evidence that sample 36 may not be a true negative (See Appendix 8).
WHO/BS/2015.2280 Page 21 Table 6. Summary of potency estimates relative to sample 79 for each combination of antibody (analyte) and virus used to produce the target antigen. GM = geometric mean; GCV = geometric coefficients of variation. Analyte Sample 28 31 43 Ebola Zaire 58 64 88 28 31 43 Ebola Zaire 1976 (Mayinga) 58 64 88 28 31 43 Ebola Zaire 58 64 88 58 Ebola Zaire 2014 (Makona) 64 88 Virus GM GCV n 0.68 44 3 0.38 . 2 0.33 269 3 0.60 41 3 0.23 165 3 0.36 40 3 0.61 . 1 0.63 . 1 0.68 . 1 3.10 . 1 0.51 . 1 1.43 . 1 0.38 26 4 0.97 27 4 0.55 14 4 0.69 53 3 0.22 48 3 0.63 28 4 0.26 . 1 0.09 . 1 0.18 . 1
GP Ig
GP Ig
GP IgG
GP IgG
WHO/BS/2015.2280 Page 22 Table 7. Summary of potency estimates relative to sample 31 for each combination of antibody (analyte) and virus used to produce the target antigen. GM = geometric mean; GCV = geometric coefficients of variation. Analyte Virus Sample 28 43 58 Ebola Zaire 64 79 88 28 43 58 Ebola Zaire 1976 (Mayinga) 64 79 88 28 43 58 Ebola Zaire 64 79 88 28 Ebola Zaire 2014 (Makona) 43 GM 1.61 0.71 1.46 0.37 2.66 0.95 0.96 1.07 4.90 0.71 1.58 2.26 0.41 0.58 0.80 0.24 1.03 0.67 0.38 0.38 GCV . . . . . . . . . . . . 47 25 . . 27 38 . . n 2 2 2 2 2 2 1 1 1 1 1 1 4 4 2 2 4 4 1 1
GP Ig
GP Ig
GP IgG
GP IgG
WHO/BS/2015.2280 Page 23 Table 8. Summary for all assays of sample potencies of antibodies to Ebola relative to sample 79. Lab Sample 28 31 43 58 64 88 Neut 11a 12b 0.25 0.19 0.50 0.13 0.15 ND # 0.13 # ND 0.35 0.13 7 0.18 0.40 0.20 0.14 0.05 0.15 8 0.51 0.17 0.82 ND ND 0.17 PsN 15 0.83 0.04 0.33 # # 0.01 16b 0.28 0.37 0.20 # # ND 17 0.77 0.23 0.48 ND ND 0.22 1 0.49 0.85 0.63 0.69 0.30 0.80 3 0.36 0.77 0.46 0.45 0.14 0.45 4 0.45 0.68 0.78 0.65 0.14 0.52 5a 0.29 1.33 0.50 1.06 0.26 0.65 EIA 5c 10 0.81 ND 0.21 ND 0.07 ND 0.41 0.24 0.12 0.08 0.27 0.16 11b* 0.69 0.49 0.80 # # 0.40 12a 0.41 1.01 0.59 # # 0.69 13 0.88 0.04 0.61 0.81 0.70 0.34 16a 0.61 0.63 0.68 3.10 0.51 1.43
ND = not determined; #= sample not supplied; white rows = convalescent plasma; light grey rows = vaccinee samples; dark grey rows = Tc bovine samples. *Lab 11b results are for the 1/200 dilution only.
WHO/BS/2015.2275Ab Page 24
Figures Figure 1. EIA slope ratios of all samples relative to sample 79
Figure 2. EIA slope ratios of all samples relative to sample 31
WHO/BS/2015.2275Ab Page 25 Figure 3. Lab geometric mean (GM) potencies for anti-GP IgG antibody relative to sample 79.
Figure 4. Lab geometric mean (GM) potencies for anti-GP total antibody relative to sample 79. GP Ig (Total) 4 3 2 Ebola Zaire Ebola Zaire 1976 (Mayinga) Ebola Zaire 2014 (Makona)
1
Lab GM
0.5 0.33 0.25
0.1
Sample
0.05
28
31
43
58
64
88
WHO/BS/2015.2275Ab Page 26 Figure 5. Lab geometric mean (GM) potencies for anti-GP IgG antibody relative to sample 31 GP IgG 4 3 2 Ebola Zaire Ebola Zaire 1976 (Mayinga) Ebola Zaire 2014 (Makona)
1
Lab GM
0.5 0.33 0.25
0.1
Sample
0.05
28
43
58
64
79
88
Figure 5. Lab geometric mean (GM) potencies for anti-GP total antibody relative to sample 31 GP Ig (Total) 4 3 2 Ebola Zaire Ebola Zaire 1976 (Mayinga) Ebola Zaire 2014 (Makona)
1
Lab GM
0.5 0.33 0.25
0.1
Sample
0.05
28
43
58
64
79
88
WHO/BS/2015.2275Ab Page 27 Appendix 1 Collaborative study participants (In alphabetical order by country) Name Laboratory Béatrice Labrosse Christophe Léculier Laboratoire P4 Inserm Jean Mérieux Delphine Pannetier Thomas Strecker, Institute of Virology of the PhilippsVerena Krähling, University, Marburg Sarah Katharina Fehling Barbara Schnierle, Christopher Weber, Paul-Ehrlich-Institut Christine von Rhein Maria Capobianchi National Institute for Infectious Diseases Concetta Castilletti Lazzaro Spallanzani - INMI Derek Gatherer, Lisa Bishop, Katharina Hartman, Lancaster University Robert Lauder Edward Wright, University of Westminster Emma Bentley Richard Tedder, Virus Reference Department, Public Dhan Samuel, Steve Dicks Health England, Colindale Thomas Rudge Battelle-BBRC Graham Simmons Blood Systems Research Institute Office of Blood Research and Review, Gerardo Kaplan CBER-FDA Surender Khurana CBER, FDA Carol D. Weiss CBER, FDA Wayne R. Hogrefe, Nicole Rodriguez, Focus Diagnostics, Inc. Allen Hsu Anna N Honko, Richard S. Integrated Research Facility, National Bennett, Jamie Pettitt, Institutes of Allergy and Infectious Krisztina Janosko, Jonathan Diseases Marchand, Elena Postnikova, Hua Wu SAB Biotherapeutics Inc. Anthony Griffiths TXBiomed Jay W. Hooper USAMRIID
Country France
Germany
Germany Italy UK UK UK USA* USA* USA* USA* USA* USA*
USA* USA* USA* USA*
*Steven A. Rubin, FDA/CBER facilitated sample permits and shipments to laboratories in the USA.
WHO/BS/2015.2275Ab Page 28 Appendix 2 NIBSC Division of Virology Short Report Solvent Detergent Treatment for inactivation of HIV-1 in Plasma as an Indicator of Function in the Treatment of convalescent Ebola Patient plasma. Author: Mark Hassall
Introduction Plasma from convalescent patients may contain blood borne viruses capable of replication which can pose a risk to people handling samples. To ensure the inactivation of enveloped viruses which may be in samples they can be treated with a mixture of a solvent and a detergent. This treatment has been shown to be effective against a number of viruses and has been used by blood services for the treatment of plasma packs. This report describes the treatment of plasma spiked with HIV-1 IIIB for the inactivation of the virus and inhibition of infection on a T cell line, to prove the treatment process is effective. For further reading see: Dichtelmüller, H. O., Biesert, L., Fabbrizzi, F., Gajardo, R., Gröner, A., von Hoegen, I., Jorquera, J. I., Kempf, C., Kreil, T. R., Pifat, D., Osheroff, W. and Poelsler, G. (2009), Robustness of solvent/detergent treatment of plasma derivatives: a data collection from Plasma Protein Therapeutics Association member companies. Transfusion, 49: 1931–1943. Method Solvent detergent treatment Plasma samples are treated by the addition of 1% v/v TBP and 1% v/v Triton X-100, and incubated at 30°C for 3 hours, with mixing every 15 minutes. 10% v/v soybean oil is added and samples mixed for 30 minutes at room temperature to emulsify the detergent. This is then centrifuged at 3000rpm for 30 minutes at room temperature to separate the oil/detergent layer from the plasma. The plasma is removed from below the oil layer without disturbing the interface, and the solvent removed using C18 reverse phase chromatography columns (which are activated using methanol, and rinsed in sterile water first). Using a vacuum manifold, 20 ml of oil free plasma sample is added to each column and run through at 0.2bar. This was repeated twice for each sample to ensure that the solvent was fully removed. Infectivity assay Normal human plasma was spiked with HIV-1 IIIB grown on C8166 cells, and then split into two samples. One sample was solvent detergent treated as per the method described, and the second left untreated. The two samples were titrated 10 fold down to a dilution of 10-9 on C8166 cells in duplicate 24 well plates. Each plate had 4 replicates of each titration, as well as 4 wells of negative control cells which were mock infected with media alone. The cells were incubated and grown for 14 days. During the course of the test the cells were checked for cytopathic effect (CPE) visually and the wells showing CPE were recorded. For the detection of gag protein as a marker of active replication when high levels of CPE were seen for this studay this was at 14 days post infection) media was removed and samples of all cell wells were taken and lysed in NP-40 detergent with 1% trypan blue.
WHO/BS/2015.2275Ab Page 29 After 48 hours these samples were tested with an in house antigen capture assay for HIV-1 Gag p24 protein (see appendix 1.3)
Results Both the visual CPE recording and gag antigen detection methods demonstrated that the cells were only infected by the untreated HIV- spiked plasma. Visually the cells from the treated samples were all healthy and growing after 12 days, with no apparent CPE and no evidence of viral infection. By comparison the cells from the treated samples were very damaged and sparse by 12 days with evidence of CPE that gave a calculated TCID50 of approximately 6.5 (see appendix 1.2 for method of calculation) The antigen capture assay appears less sensitive than visual CPE, but gives a definite non biased measure of viral infection. Figure 1 shows the treated samples have no measurable Gag protein after 14 days at any dilution of the virus in any of the 8 eight replicates tested. The untreated samples shown in figure 2 however have p24 which is measurable in some wells down to a dilution of 10-3 in some case. Calculating the TCID50 in the same manner as for the visual method gave a result from the ACA of approximately 3.5.
0.8 0.6 OD@450nM 0.4 0.2 0 Neat 10-1 10-2 10-3 10-4 10-6 10-7
Sample 1 Sample 2 Sample 3 Sample 4 Sample 5 Sample 6
10-8
Virus titration
Figure 1. Titration of solvent detergent treated HIV-1 IIIB on C8166 cells. Eight samples of each dilution were tested across two 24 well plates. Cells were incubated with virus for 14 days, with media changes on days 7 and 10.
10-9
-0.2
10-5
Sample 7 Sample 8
WHO/BS/2015.2275Ab Page 30
0.7 0.6 OD@450nM 0.5 0.4 0.3 0.2 0.1 Neat 10-1 10-2 10-3 0 -0.1 Sample 1 Sample 2 Sample 3 Sample 4 Sample 5 Sample 6 Sample 7 10-4 10-5 10-6 10-7 10-8 10-9 Sample 8
Titration of virus
Figure 2. Titration of untreated HIV-1 IIIB on C8166 cells. Eight samples of each dilution were tested across two 24 well plates. Cells were incubated with virus for 14 days, with media changes on days 7 and 10.
Conclusions There is no evidence of HIV infection in cells exposed to virus in plasma after solvent detergent treatment either by visual inspection or by gag p24 antigen capture. The level of infection seen in the titration of the non-treated samples shows that the method reduced infectivity by at least 3 logs, and up to possibly 6 logs. 1.1 Solvent detergent SOP Division of Virology Experimental Protocol Title: Solvent Detergent Treatment of Plasma for the inactivation of Enveloped Viruses. Author: Mark Hassall Authoriser: Mark Page Introduction Plasma from convalescent patients may contain viruses capable of replication. To ensure the inactivation of enveloped viruses which may be in samples they can be treated with a mixture of a solvent and a detergent. This treatment has been shown to be effective against a number of viruses and has been used by blood services for the treatment of plasma packs. This protocol described the method used to treat plasma for the inactivation of enveloped viruses, and then to remove any trace of the additives to ensure the processed plasma can be utilised in cell based assays. Materials Plasma samples Tributyl phosphate (SIGMA-00675) Triton X-100 (SIGMA-T8787)
WHO/BS/2015.2275Ab Page 31 Soybean oil (SIGMA-S7381) Methanol (VWR- 20847.320) Distilled water Bond Elut C18 cartridges – 20ml – (Agilent-12256023) Vacuum manifold and pump Rotator 30°C water bath Centrifuge
Method Pre-filtration of Plasma If there appear to be aggregates of protein in the samples they may be pre filtered, through a 0.2µm filter. This is not an essential step and for small samples where loss needs to be kept to a minimum it can be skipped. Solvent detergent treatment To the plasma sample(s) add 1% v/v TBP and1% v/v Triton X-100. Incubate this at 30°C for 3 hours, with mixing every 15 minutes. After incubation add 10% v/v soybean oil and incubate on a rotating mixer for 30 minutes at room temperature. Centrifuge the samples at 3000rpm for 30 minutes at room temperature. Carefully remove the oil layer trying not to disturb the interface. C18 cartridges must be activated with methanol. To each column (two per sample) add 10ml of methanol and let this run through by gravity flow. Wash the columns with 10ml of distilled water. Cartridges are attached to a vacuum manifold, and then 20 ml of oil free plasma samples added to each. This is run through at approximately 0.2bar. Repeat this step with a second new column for each sample. Samples must be aliquoted and stored appropriately for further distribution. 1.2 Calculation of virus titre Viral titration results can be calculated as follows: Example of viral titration table: Dilution: Number of positive wells: 100 10-1 10-2 10-3 10-4 10-5 10-6 10-7 10-8 8/8 8/8 8/8 5/8 1/8 0/8 0/8 0/8 0/8
10-9 0/8 Number of 8/8 positive wells for a given viral dilution (not
WHO/BS/2015.2275Ab Page 32 Counting the neat (100) dilution Number of single well positives Calculation of viral titration is therefore: 2 + 0.5 (dilution adjustment) + (6 x 0.125 (single well positives)) + 1 (100ul viral inoculum) => Giving a TCID50 of 4.25. 1.3 HIV-1 p24 Antigen Capture Assay Method: 1. Coat plates with 50µl per well of polyclonal capture antibody (Aalto Sheep αHIV-p24 – Aalto CFAR ARP410) at 10µg/ml in PBS and leave overnight at 4oC. 2. Wash plate 4 times in washing buffer (PBS/0.05% Tween-20). 3. Add 200µl per well of blocking buffer (PBS/0.05% Tween-20/5% porcine serum) and leave for 1 hour at room temperature. 4. Wash plate 4 times in washing buffer. 5. Add 50µl of each NP40 treated sample onto plate. 6. Seal and incubate plate for 1 hour at room temperature. 7. Wash plate 4 times in washing buffer. 8. Add 100µl of diluted detector antibody (Biotinylated monoclonal mouse anti-p24 gag diluted 1 in 100 in blocking buffer) to each well and incubate at room temperature for 2 hours. 9. Wash plate 4 times in washing buffer. 10. Add 100µl of diluted streptavidin-peroxidase to each well and incubate at room temperature for 1 hour. 11. Wash plate 4 times in washing buffer. 12. Add 100µl of TMB to each well and incubate at room temperature. 13. Stop colour change reaction with 50µl of 1M H2SO4. 14. Read plate at 450nM after 10 minutes. = = 2 6
WHO/BS/2015.2275Ab Page 33 Appendix 3 Collaborative Study Protocol Protocol for the WHO collaborative study to assess the suitability of an interim standard for antibodies to Ebola virus Background In support of the WHO response to the Ebola crisis, NIBSC is undertaking a project to develop an International Standard for use in the calibration and control of Ebola antibody assays. The availability of International Standards (IS) for antibodies would facilitate the standardization of Ebola serological methods used in epidemiological studies to measure past or present Ebola virus disease and in vaccinology studies to measure antibodies elicited by vaccination in humans. In the absence of such standards, individual laboratories apply their own reference standards which are not harmonized with other laboratories and methods and thus cannot serve to improve the reproducibility between laboratories. Recommendations made by participants attending the Technical Workshop on the Standardisation of Serological and PCR assays for the detection of Ebola virus (NIBSC, UK, 5-6 March 2015), included the urgent prioritization of the development of an interim Ebola standard for serology assays while perusing the longer-term goal of establishing an International Standard according to published WHO guidelines and formally endorsed by the WHO Expert Committee on Biological Standardization (ECBS) [1]. Aims The aims of this WHO international collaborative study are to assess the suitability of different antibody preparations to serve as the interim standard with an assigned unitage per mL for use in the harmonization of Ebola serology assays. There is no international conventional reference measurement procedure for Ebola virus antibodies and the interim unitage will not be traceable to the International System of Units (SI) of quantity. characterise the antibody preparations in terms of reactivity/specificity in different assay systems. assess each preparation’s potency i.e. readout in a range of typical assays performed in different laboratories. assess commutability i.e. to establish the extent to which each preparation is suitable to serve as an interim standard for the variety of different samples and assay types. recommend to the WHO Ebola antibody assay working group, the antibody preparation(s) found to be suitable to serve as the interim standard(s). Materials Transchromosomic (Tc) bovine anti-Ebola IgG preparations The source materials were donated by Dr Eddie J. Sullivan, SAB Biotherapeutics, Inc. USA, and is human anti-Ebola antibody purified from bovine plasma collected from transchromosomic (Tc) cattle [4] immunized with experimental Ebola vaccines (Zaire95+Sudan GP DNA vaccine or rGPZaire2014 vaccine). Upon receipt at NIBSC, the purified Tc Bovine IgG samples were diluted to a target protein concentration of 1 mg/mL in sterile PBS-Ca2+-Mg2+ supplemented with 5% human serum albumin. Plasma obtained from convalescent patients The source materials are plasma samples obtained from three patients recovered from Ebola virus disease (Provided by Dr Richard W. Olaussen, Oslo University, Norway, Dr Susan L. Stramer, American Red Cross (ARC), USA and Sheila MacLennan, National Health Service Blood and Transplant (NHSBT), Leeds, UK). The three materials had been tested and found
WHO/BS/2015.2275Ab Page 34 negative for Ebola virus RNA and other blood viral markers. Prior to receipt at NIBSC, the convalescent plasmas were held at Public Health England, Colindale, UK until confirmed by PCR that no Ebola RNA could be detected in the materials. At NIBSC, the plasmas were solvent-detergent-extracted prior to aliquoting for the collaborative study. Pooled vaccinee plasma The source material is plasma samples obtained from volunteers participating in the Oxford, UK vaccine trial who had been primed with the monovalent formulation of the chimpanzee adenovirus 3 (ChAd3)–vectored vaccine encoding the surface glycoprotein of Zaire ebolavirus (GSK/NIH vaccine candidate) [5] and boosted with the same Ebola virus gene in a modified vaccinia Ankara (MVA) vectored Vaccine (Bavarian Nordic vaccine candidate). Coded study samples All study samples are liquid, filled in 0.1 mL aliquots into 0.5 mL Sarstedt screw-capped tubes. They will be provided coded and blinded. The samples are labelled “EBOV Ab Sample Code xx” where xx is 9, 28, 31, 36, 43, 58, 64, 79 or 88. The coded samples may include negative samples as well as reactive samples from convalescent patients, recipients of trial vaccines and Tc cattle. Suggested starting dilutions are provided in the Excel reporting sheet. Laboratories will receive at least 4 sets of study samples which should allow for 3 independent assays (plus 1 spare) by one method. Laboratories with more than one assay method will receive additional sample sets to allow 3 independent assays (plus 1 spare) per method (subject to availability). Due to limiting amounts of some source materials, some sample sets may not contain EBOV Ab Sample Codes 58 and 64. This set has been labelled “Reduced EBOV Ab sample set”. The dispatch of the study samples will commence 27 May 2015. The study samples should be stored at -20C or below. The study samples shall not be administered to humans. Assay Methods For testing the study samples, participants are requested to use the method(s) in routine use in their laboratory for the detection of antibodies to Ebola. Laboratories may use multiple methods to test the study materials provided that the study design is followed for each method. Design of study Participants are requested to: perform 3 independent assays on different days for antibodies against Ebola. Spare vials of most samples are provided (subject to availability). Use a freshly thawed sample for each independent assay. Each sample should be quickly thawed at 37°C and used immediately or placed on ice until used. for each independent assay, prepare and test a series of dilutions from each coded sample. If possible at least two independent replicate series of dilutions (NOT two samplings from a single series) should be prepared and assayed. Suggested dilutions for each sample are provided in the Excel reporting sheet. Participants are requested to dilute the samples using the sample matrix specific to their individual assay(s) (e.g. plasma, serum, buffer). The optimal dilution range should cover at least 5 to 6 steps including one step beyond the endpoint dilution. Adjust dilutions accordingly for subsequent assays if needed. Record in the excel spreadsheet changes to the dilutions tested. use the Excel reporting sheet to record for each dilution the assay readout (e.g. from the spectrophotometer/luminometer/plaque count etc.). Our statistician will use the raw data readouts to perform the statistical analysis.
as it is of interest for us to know whether the samples are considered ‘positive’ in each assay, participants are requested to include the cut-off value indicating seroreactivity for each assay and whether each sample dilution tested is considered positive or negative according to their criteria. Include all study samples in each assay so that the concentration of antibodies relative to one another may be calculated. Please note in the reporting sheet if it is not practicable to test all samples concurrently indicating which samples where tested concurrently. Record in the Excel reporting sheet any deviations from the assay protocol. Results and data analysis Participants are requested to return their results to NIBSC within 6 weeks of receipt of the study materials. If it is not practicable to turn around results within 6 weeks, please inform Dianna.wilkinson@nibsc.org . An excel spread sheet is provided so that all essential information can be recorded including details of assay methodology and the raw data obtained from each assay. The use of the reporting spread sheet facilitates the analysis and interpretation of results. If multiple assay methods are undertaken, a separate worksheet for each method should be completed. The final version of the reporting spread sheet will be e-mailed to each participant following shipment of study materials. The confidentiality of each laboratory will be ensured with each participant being anonymous to the other laboratories. Analysis of the study will assess the potencies of each material relative to each other, and the sensitivity of the different assay methods. Assay data will be analysed at NIBSC by an experienced biometrician using standard statistical techniques. A draft study report will be sent to participants for comment. The report will include data analysis, proposed conclusions and recommendations on the selection, use and unitage of the most appropriate antibody preparation to serve as the interim standard. The finalised report will then be submitted to the WHO Ebola antibody assay working group who will decide on the suitability of the preparation to serve as the interim standard for Ebola antibody. Participation in the WHO collaborative study is conducted under the following conditions: The data obtained in the collaborative study should not be published or cited before the formal establishment of the standard by WHO, without the expressed permission of the NIBSC Study organizer. In order to address the immediate need for Ebola serology standards, participants are permitted to use the study samples for purposes that fall outside of the collaborative study. To better inform the subsequent implementation of any standard, participants are encouraged to provide to the study organizer any information gained through their use of the study materials. It is normal practice to acknowledge participants as contributors of data rather than co-authors in publications describing the establishment of the standard. Individual participant’s data will be coded and reported “blind” to other participants during the preparation of the study report, and also in subsequent publications. Participants will receive a copy of the report of the study and proposed conclusions and recommendations for comment before it is further distributed. Participants accept responsibility for safe handling and disposal of the materials provided. Deadline for completed results spread sheets is 6 weeks from receipt of study materials. If it is not practicable to return results within 6 weeks, please inform Dianna Wilkinson.
WHO/BS/2015.2275Ab Page 35
WHO/BS/2015.2275Ab Page 36 All completed results spread sheets should be returned electronically to: Dr Dianna Wilkinson Principal Scientist Viral Vaccines Section Division of Virology National Institute for Biological Standards and Control Blanche Lane South Mimms Hertfordshire EN6 3QG UK Tel. +44(0)1707 641314 dianna.wilkinson@nibsc.org
WHO/BS/2015.2275Ab Page 37 Appendix 4 Direct ELISA reported by laboratory 10. A. anti-EBOV Makona rGP IgG endpoint titers (units/ml*) EIA Assay 1 EIA Assay 2 EIA Assay 3 Average Sample (units/ml) (units/ml) (units/ml) (units/ml) 9 Negative Negative Negative Negative 14766 12292 15845 14301 28 31419 38938 37539 35965 31 36 Negative Negative Negative Negative 15172 13455 14293 14307 43 12595 10921 10089 11202 58 3803 3555 3373 3577 64 50218 41101 37973 43097 79 8092 7795 7138 7675 88 *The titer value (units/ml) is defined as the highest dilution of sample where the OD450 reading was 3-fold higher than blank. B. anti-EBOV Makona rGP IgG endpoint titer activities (units/mg IgG*) IgG EIA Assay 1 EIA Assay 2 EIA Assay 3 Average Concentration (units/mg IgG) (units/mg IgG) (units/mg IgG) (units/mg IgG) Sample (mg/ml) 0.93 9 Negative Negative Negative Negative 6.54 2188 2260 1881 2425 28 1.08 33357 29141 36114 34817 31 5.86 36 Negative Negative Negative Negative 7.06 2027 2149 1906 2025 43 8.64 1296 1457 1264 1167 58 11.28 317 337 315 299 64 8.97 4804 5598 4582 4233 79 0.91 8439 8897 8571 7849 88 * As IgG concentrations vary among 9 samples (Tc bovine materials were previously diluted to 1mg/ml), to see which sample has high anti-Ebola titer activity, the titer data is reported in units/mg IgG. The titer activity value (units/mg) is defined as the highest dilution of 1 mg of antibody where the OD450 reading was 3-fold higher than blank.
WHO/BS/2015.2275Ab Page 38 Appendix 5 Graphs provided by laboratory 11 comparing the results obtained with their neutralisation assay against live Ebola virus Mayinga (11a) and EIA against inactivated Ebola virus Makona (11b). 11a
Neutralisation titer of WHO Standards 1024 256 64 16 4 1 sc 9 sc 28 sc 31 assay 1 sc 36 assay 2 sc 43 assay 3 sc 79 sc 88
11b
Arbitrary ELISA units (AEU) 35000 30000 25000 20000 15000 10000 5000 0 sc 9 sc 28 sc 31 sc 36 sc 43 sc 79 sc 88
WHO/BS/2015.2275Ab Page 39 Appendix 6 Graph provided by laboratory 17 providing RNT50 and RNT80 titres for their PsN (LVV) assay for Zaire GP tc/GIN/14/WPG-C05.
WHO/BS/2015.2275Ab Page 40 Appendix 7 Western analysis of study samples performed by Laboratory 4.
Column decoding for Westerns: EZ = Zaire nucleoprotein ES = Sudan nucleoprotein Z1 = Zaire GP-1 isoform Z12 = Zaire GP-1,2 isoform SE = Sudan GP
WHO/BS/2015.2275Ab Page 41 Appendix 8 Laboratory 16 Retrospective analysis of sample 36
Method: Plasma samples from WHO#36 and 5 blood donors were titrated in parallel on plates coated with viral particles. Result: The WHO#36 sample reacted against WT VSV and all pseudotypes (rVSV-EBOVgp, rVSV-SEBOVgp, rVSV-BEBOVgp, and rVSV-MARVgp) whereas there was no reaction with plasma samples from the 5 blood donors. Conclusion: The WHO#36 is an anomalous plasma sample that reacts non-specifically by ELISA
WHO/BS/2015.2275Ab Page 42 References 1. WHO, Recommendations for the preparation, characterization and establishment of international and other biological reference standards. WHO Technical Report Series, No. 932., in Expert Committee on Biological Standardization. 2006. 2. Dichtelmuller, H.O., et al., Robustness of solvent/detergent treatment of plasma derivatives: a data collection from Plasma Protein Therapeutics Association member companies. Transfusion, 2009. 49(9): p. 1931-43. 3. EDQM. CombiStats v4.0. Available from: www.combistats.eu. 4. Matsushita, H., et al., Triple Immunoglobulin Gene Knockout Transchromosomic Cattle: Bovine Lambda Cluster Deletion and Its Effect on Fully Human Polyclonal Antibody Production. PLoS ONE, 2014. 9(3): p. e90383. 5. Rampling, T., et al., A Monovalent Chimpanzee Adenovirus Ebola Vaccine - Preliminary Report. N Engl J Med, 2015.