INTERNATIONAL WORKING GROUP ON THE STANDARDISATION OF GENOME AMPLIFICATION TECHNIQUES FOR THE SAFETY TESTING OF BLOOD, TISSUES AND ORGANS FOR BLOOD BORNE PATHOGENS

SoGAT XVIII: 24-25 May 2005

Co-sponsored by CBER, FDA, USA and NIBSC, UK

Report of the meeting prepared by S Baylis (NIBSC), C Davis (NIBSC),

H Holmes(NIBSC) and J Saldanha(Roche)

The eighteenth meeting of SoGAT was held in the USA for the first time at the Natcher Facility at NIH campus, Bethesda on the 24th and 25th May 2005. The meeting was opened by Dr. Karen Midthun, the Deputy Director of CBER, FDA, the organisation that was hosting the meeting. Dr. Steven Inglis, the Director of NIBSC, informed the delegates that this was the 10th anniversary of SoGAT and commented on the long history of standardisation associated with NIBSC.

Standardisation – the way forward(Chaired by Professor Wolfram Gerlich)

In the opening presentation,Dr. Elwyn Griffiths(Health Canada, Canada) gave a brief overview of the history of standardisation and the role that the World Health Organisation (WHO) plays in the development of recommendations (previously requirements) and guidelines on the production of specific standards for the control of biologicals. The WHO Expert Committee for Biological Standardisation (ECBS) has been meeting since 1947 and now interacts with numerous organisations. The standards produced includeWHO international Standards which are highly characterised; WHO reference reagents, which are less characterised interim standards with no unitage assigned; International Reference Panels for assay and test kit evaluation. Such standards represent complex biological preparations, by their nature difficult to characterise. With the implementation of the IVD directive, the status of these standards has come into focus with the recognition that biological standards and reference materials cannot be characterised readily by chemical or physico-chemical means. WHO and ISO together with other standard setting bodies are having on-going discussions of these issues. Many of these issues are addressed in the new WHO recommendations for the production of standards. Where appropriate, ISO standards such as 17511 should be used for metrological traceability. The recommendations cover issues such as, quality assurance; processing; stability; characterisation and collaborative studies; intended use.

Dr. Robert Wielgosz (Bureau International des Poids et Mesures (BIPM), France) described the involvement of the Joint Committee for Traceability in Laboratory Medicine (JCTLM) with BIPM, the International Federation for Clinical Chemistry and Laboratory Medicine (IFCC), and the International Laboratory Accreditation Cooperation (ILAC). Dr. Wielgosz highlighted the issues relating to traceability to higher order reference materials, particularly in light of the European IVD Directive and the many ISO standards relating to aspects such as the of nature the measurand, uncertainty of measurement, and commutability. The BIPM website lists higher order reference materials, including those designated in SI units, such as electrolytes, non-peptide hormones and metabolites as well more complex standards expressed in other ways(www1.bipm.org/en/committees/jc/jctlm/jctlm-db). Dr. Ana Padilla noted that following a meeting with WHO and EC representatives, it had been agreed that the HBsAg WHO International Standard calibrated in IUs would now be recognized as a higher order standard. Dr. Wielgosz confirmed that the HBsAg IS has been accepted asan international conventional calibrator in accordance with ISO17511. He further commented that a primary calibrator must be defined in SI units.

Dr. Paul Neuwald (AcroMetrix, USA) discussed NAT extraction methods. He started off by describing the many potential sources of variation in NAT assays. Citing an example of a CMV reference panel and the results generated from 23 laboratories, Dr. Neuwald speculated that the variability observed between participating laboratories may have been due, at leastin part,to extraction differences. In a smaller study comparing two different extraction methods for parvovirus B19 with amplification using the Artus kit, differences in titre were observed suggesting an extraction issue. Hehighlighted that in many previous studies the overall impact of the extraction method had not been fully investigated using appropriate controls and that AcroMetrix would be prepared to coordinate such a study. There was general agreement on the affect of the matrix on extraction efficacy. ProfessorGerlich asked if cultured virus had been spiked into plasma for the CMV study, given that CMV exhibits a high degree of cell association. Dr. Neuwald confirmed that spiking had been carried out. Dr. Kurt Roth emphasized the importance of using optimal extraction and amplification methods to avoid widely varying degrees of sensitivity. ProfessorGerlich raised the specific problems associated with extraction of HBV and on occasion HDV and added that variability in viral lysis added to the complexity. Dr. Dan Colestressed the large differences seen between different batches of plasma.

Dr. Cindy Walker-Peach (Ambion, USA) described the extension of the armored RNA technology to DNA. In the case of DNA viruses, controls have included plasmids, infectious virus including attenuated/ inactivated viruses, patient samples and PCR amplicons. Dr. Walker-Peach proposed that armored DNA was a good alternative overcoming many of the shortfalls of the alternatives. Ambion have packaged HBV and internal control amplicons, these are resistant to DNase treatment. QC analysis includes HPLC of the nucleic acid to confirm the number of nucleotides and NIST-traceable phosphate determination. Functional assays and stability studies were performed using real time quantitative PCR demonstrating the utility of the armored DNAs. ProfessorGerlichquestioned the need for armored DNA,commenting thatfor HBV DNA there were commutability issues that armored DNA could not control for,such as partial single and partial double stranded regions of the genome and the covalent binding of protein to the genome.

Dr. Theo Cuypers (Sanquin, The Netherlands) posed the question as to whether conversion factors are assay related. He presented data comparing International Standards and the PeliCheck reagents for HIV-1 RNA, HCV RNA and HBV DNA using several different extraction and detection methods in multi-centre studies. The data generated in the studies was summarized and highlighted that there are a wide range of conversion factors of IU to genome equivalents (geq). Such differences may in part be due to different extraction techniques used and the various combinations of amplification/detection platforms. Other factors are things such as the physical and molecular composition of the reference material which may impact on the commutability. It was concluded that the conversion from IU to geq may be assay specific. ProfessorGerlich suggested that SoGAT should set up uniform methods to compare conversion factors, as the use of different methods generates the large ranges observed. He emphasized the importance of the use of the International Unit to avoid these conversion factor issues. Dr. Nico Lelie stressed the importance of commutability and that ideally a standard should reflect what is being measured in a donor sample.

Dr. Susan Bromley (Bayer, USA) and Dr. Bruce Phelps (Chiron, USA) spoke about the development and characterization of synthetic standards. Dr. Bromley outlined the ideals of a synthetic standard; of known sequence; readily manufactured; and may be quantified using traceable methods. In vitro transcribed RNA forms the basis of a synthetic HCV RNA standard, encompassing the 5’ non-coding region and the core gene. Characterization and quantification of the RNA transcript is performed using absorbance (A260), hyperchromatic shift and phosphate determination. This synthetic RNA is used at Bayer as a reference standard to evaluate other controls and to perform validation of kit lots. Dr. Bromley concluded that this is a stable method in use at Bayer, and a similar approach has been used for HIV-1 RNA and HBV DNA. Dr. Bruce Phelps described feasibility studies to evaluate synthetic RNA standards for HCV NAT. Phase I of the study was designed by the Industrial Liaison Committee (ILC), using the in vitro transcribed HCV RNA produced by Bayer. The synthetic RNA was quantified using several commercially available assays and there was generally good agreement between these (within 0.6 logs). On the basis of this original study, Phase II is in the process of being designed to look in more detail at the in vitro transcribed RNA and armored RNA with commutability to the biological WHO International Standard. Dr. Phelps proposed that if these studies are successful, then such a synthetic standard could be used in evaluating replacement standards.

Dr. John Saldanha (Roche Molecular Systems Inc, USA) compared and contrasted the merits of using biological versus synthetic standards. The main advantages of biological standards are that the actual agent is quantified in a matrix similar to that of the test sample and the biological standard contains full length sequence of the agent in question. With biological standards, different assays yield similar results, but this is not well defined for synthetic standard materials. The main advantages of synthetic standards include the possibility to produce large amounts, which should be better characterised and more homogeneous than a biological standard. Dr. Saldanha stated that there should be better consistency between replacement lots of a synthetic material which should be easier to calibrate. Both biological and synthetic standards are stable for long periods. Where the 1st WHO International Standard for HCV RNA has been tested by several laboratories in different collaborative studies, quantification has been inconsistent. Indeed, regulatory requirements in terms of replacing standards and actual assay performance call for accurate calibration. Of primary importance is the commutability of replacement standards to maintain the IU. Dr. Saldanha proposed that in future synthetic standards could be used as calibrators. Ideally such standards would be expressed in SI units with traceable reference methods and competent laboratories.

A panel discussion followed this session and topics discussed included the use of synthetic standards, traceability of reference materials and the use of SI units. Concern about commutability was expressed by several delegates in that if a synthetic standard was to be used as a primary standard, then such a preparation would fail to control for the entire extraction, amplification/detection process. Indeed, given the unique physical properties of different virus families (e.g. HBV), biological standards would be more appropriate. If the approach were to be adopted whereby a replacement International Standard were to be calibrated against a synthetic reference preparation, concern was expressed that in accordance with ISO 17511, it would require metrological traceability and at present there are no established reference methods. Dr. Indira Hewlett asked about the situation in Europe with respect to the use of SI units in the context of the IVD Directive. Dr. Wielgosz replied that where feasible SI units should be used, however this was very much dependent upon the availability of appropriate reference methods. An example was cited for HBsAg where ng values assigned to different standards have varied. The HBsAg International Standard is now recognized as the higher order reference material in the IVD context, as no suitable alternative material or reference method has yet been identified. There was some debate concerning the conversion of IU to copy number. Dr. Harvey Holmes and Dr. Saldanha reiterated that that such conversion factors are variable, being assay dependent. Dr. Holmes raised some additional concerns about synthetic standards. These included the use of sub-genomic fragments of synthetic material that exclude other regions of a viral genome for which alternative assays may be available or that may require work up in future.

Standards for NAT (Chaired by Dr. Stephen Inglis)

Dr. Harvey Holmes (NIBSC, UK) reviewed a recent collaborative study to replace the 1st HIV-1 RNA International Standard (IS). One of the candidate standards from the original collaborative study had been tested in parallel with the 1st IS by eight laboratories. Statistical analysis showed that the estimated number of RNA copies, observed in the new study were in line with those seen previously. There was no evidence of drift or instability suggesting that the candidate standard was a suitable replacement. Dr Holmes proposed various options to the group regarding assigning a value to the proposed candidate, these were as follows:

  1. Use only data from the initial study in 1999.
  2. Use data from the current study, which showed values of 5.44 or 5.46 depending on whether qualitative data was used.
  3. Pool the data from both studies and take an overall mean, giving values of 5.56 or 5.58 with or without qualitative data respectively.
  4. Produce an overall mean for each assay method.

Dr Holmes informed the group that a report would be submitted to the ECBS this summer and proposed the replacement IS should be assigned a unitage based on the third option, using both qualitative and quantitative data. The group was asked foropinions on this proposal. No comments relating directly to this proposal were raised indicating overall acceptance from the group. Dr Holmes also asked for feedback regarding the need for a second HIV Genotype Panel containing more diverse subtypes and groups, circulating recombinant forms (CRFs) and HIV-2 RNA. Dr.Jean-Pierre Allen expressed concern that disproportionate attention is given to HIV-1 Group O, which whilst being a highly diverse group, is in fact rarely encountered. It was agreed that greater emphasis should be placed on the inclusion of CRFs when designing future genotype panels. Dr. Hewlett informed the group that the CBER/FDA was evaluating an HIV-2 NAT panel with some commercial kit manufactures.

Dr. Sally Baylis (NIBSC, UK) talked about the development of a standard for Plasmodium falciparum the parasite causing for malaria. This particular species is the most prevalent one infecting humans with the greatest levels of morbidity and mortality. Although rare, malaria has been transmitted by transfusion with serious consequences in the recipient. Usually donor selection procedures are sufficient to minimize the risk of transfusion transmission, however a recent P. falciparum case in the UK, was traced to a healthy donor previously resident in an endemic area still positive by PCR for parasite DNA despite not visiting an endemic area for ~9 years. For the standardization studies, NIBSC has freeze-dried blood from a patient infected withP. falciparum high levels of parasitaemia. This material has been freeze-dried, showing no loss of titre after freeze-drying with good stability. Accelerated degradation studies are in progress. A collaborative study is underway, looking at the freeze-dried preparation, a high titre frozen liquid preparation, a low titre liquid preparation and the 3D7 P. falciparum strain cultured in vitro to a titre comparable to the freeze-dried material. Laboratories involved in the study include control labs, reference and diagnostic laboratories as well a vaccine groups. Additional participants were encouraged to take part in the study.

Other NAT issues were discussed including the proposal to replace the first WHO HBV DNA International Standard (97/746) with one of the materials characterized in the original collaborative study. The study would be run along the lines for replacement of the first WHO HCV RNA International Standard mentioned earlier by Dr. John Saldanha. As the 2nd WHO International Standard for HCV RNA will require replacement (at current rate of usage this will be exhausted by ~2009) Dr. Sally Baylis reiterated the new WHO recommendations with like for like replacement of standards where possible. As the first and second HCV RNA international standards came from a genotype 1a, anti-HCV positive donation, compliance with the recommendations would support the use of similar material. However, Dr. Sally Baylis pointed out that a window period donation may be more appropriate and Dr. Micha Nübling (PEI, Germany) and ProfessorGerlich agreed with this. ProfessorGerlich emphasized the importance of matching the viral genotype as closely as possible.

Dr. Marie Rios (CBER) described the development ofWest Nile virus (WNV) standards for NAT by CBER. Two reagents were produced using the flamingo NY99 strain and a plasma derived strain FDA-Hu2002. The viruses were cultivated in vitro and examined by 4 laboratories to determine the RNA copy number. The level of WNV RNA was found to be 1010 RNA copies/ml by analysis of the titration of intermediate dilutions of the virus stocks. These preparations contained 107.5PFU/ml, as determined by two of the laboratories. Heat inactivation (600C, 2h) was shown to completely remove virus infectivity, whilst only reducing the NAT reactivity by ~3 log10. Panels prepared from the heat inactivated stock materials showed a high degree of variability of assay performance with low copy numbers, with qualitative assays performing better than quantitative ones. The FDA’s current standard for WNV ID-NAT assays used to screen blood donations is 100 copies/ml per. Dr. Saldanha commented on the variability of the calibration of in-house reference reagents and on the value of comparing different standards with each other.