The Generative Brain

EMA guideline on the investigation of manufacturing processes for plasma-derived medicinal products with regard to vCJD risk (EMEA/CPMP/BWP/5136/03, October 2004)

This is the European Medicines Agency method guideline that tells makers of medicines from human plasma (such as clotting factors, immunoglobulins and albumin) how to run laboratory studies estimating how far their manufacturing steps would remove the agent of variant CJD, if it were present in the donor plasma. It was adopted by the CHMP in October 2004 with a legal effective date of 1 October 2004 and published on 21 October 2004, and EMA still lists it as the current effective version. It is a 10 page companion to the CHMP position statement on CJD and plasma-derived products: the position statement requires manufacturers to estimate the prion reduction capacity of their processes step by step, and this guideline explains how to design and read those studies, adapting the long standing virus validation principles to prions. The guideline is open about its limits. It says experience had not reached the point where definitive guidance could be given and other approaches may be acceptable, and that confidence in a prion clearance figure is lower than for model viruses. Twenty years later it has become more central, because revision 3 of the CHMP reflection paper (May 2024) makes the no-recall position for donors later found to have sporadic, genetic or iatrogenic CJD conditional on the manufacturer having shown prion reduction steps. Information only, not medical, legal or regulatory advice.

Key facts

  • The guideline carries the reference CPMP/BWP/5136/03 (printed as CPMP/BWP/CPMP/5136/03 on the cover), is dated London, 21 October 2004, and was issued by the CHMP; the EMA page gives a legal effective date of 1 October 2004, first publication on 21 October 2004, and lists it as the current effective version with no later update.
  • Drafting history printed on the cover: discussed in the Biotechnology Working Party from June to November 2003, released for consultation in November 2003 with comments due by end of March 2004, an EMEA expert workshop in January 2004, further working party discussion in July and September 2004, then CHMP adoption and coming into operation in October 2004.
  • Its stated purpose is to give guidance on how to investigate manufacturing processes with regard to vCJD risk, read together with the June 2004 revised CHMP position statement (EMEA/CPMP/BWP/2879/02/rev 1), whose section 9.2.3 says manufacturers must estimate the potential of their specific processes to reduce infectivity using a step-wise approach; consulting the competent authorities is encouraged.
  • It extends the virus validation note for guidance CPMP/BWP/268/95 (legal effective date 14 August 1996) to TSE agents as far as possible: production material is spiked with no more than 10 percent of its volume of an infectious preparation and studied in an exactly scaled-down laboratory model, kept apart from production, with yield and quality comparable to full-scale batches.
  • Because prions resist the usual virus inactivation methods, it says studies of solvent detergent and heat treatment have little point, and work should focus on removal and partitioning steps: cold ethanol fractionation, PEG precipitation and other precipitation, chromatography, depth filtration and nanofiltration. Reduction factors below 1 log are treated as not significant.
  • On spikes, it says brain tissue is the only practicable source of spiking material because infectivity in blood is at least 10,000 times lower than in brain; vCJD material is not mandatory and rodent-adapted laboratory strains (such as scrapie, familial CJD, BSE or vCJD adapted strains) can be chosen on practical grounds, with the rationale stated. It compares crude brain homogenate, microsomal fraction, caveolae-like domains and purified PrPSc, and asks for the spike expected to give the greatest challenge to each step.
  • On assays, animal infectivity assays are named the gold standard, with the fastest hamster model needing at least 6 to 9 months and non-transgenic mouse models 15 to 18 months; where a biochemical PrPSc assay has been shown to correlate with the bioassay, partitioning shown by the biochemical assay may be sufficient.
  • It lists eight limits of such studies, including imperfect scale-down of ethanol fractionation, the unproven assumption that reductions from two steps add up, effects of earlier detergent treatment on later filtration, unknown physical form of any agent in blood, and the gap between high-titre spikes and the low amounts expected in blood.
  • On equipment cleaning, it notes prions stick to stainless steel, describes 0.1 M sodium hydroxide and harsher treatments, and concludes that no specific sanitisation recommendation could be given until new data on plasma processing equipment became available.
  • The 2004 introduction records the context: 147 definite or probable vCJD cases in the UK by early August 2004, two possible transmissions by red cell transfusion, no vCJD case then linked to plasma products, UK plasma not fractionated as a precaution, and French plasma still fractionated because the benefit-risk balance was judged favourable.
  • The EMA 2009 concept paper on updating the CJD position statement said manufacturers had been required to estimate and report the reduction capacity of their processes to the competent authorities, that recommendations should be revisited in the light of evaluating those data, and that references to this guideline needed updating.
  • Revision 3 of the CHMP reflection paper (May 2024) keeps the duty to estimate process prion reduction capacity, ties the no-recall position for sporadic, genetic or iatrogenic CJD donors to it, and sets no numerical log reduction threshold; in its consultation the European Haemophilia Consortium asked whether this 2004 guideline should be revised, for example to cover human TSE risk generally.
  • An example of the guideline in use: the 2017 open access paper on LFB's IQYMUNE 10 percent intravenous immunoglobulin cites it, and reports TSE removal with a hamster 263K scrapie spike prepared from brain by ultracentrifugation, Western blot and hamster bioassay, at 4.0 or more logs (Western blot) and 5.2 or more logs (bioassay) for caprylic acid fractionation, 3.9 or more logs (bioassay) for anion-exchange chromatography and 3.8 or more logs (Western blot) for 20 nanometre nanofiltration, an overall figure of 12.9 or more logs.
  • An industry pooling of more than 200 prion removal studies from PPTA member companies (Transfusion, 2013) reported good agreement between independent groups, agreement between biochemical assays and bioassays, and similar reduction values across spike types except highly purified prion protein in some cases.
  • The EMA guideline on plasma-derived medicinal products (EMA/CHMP/BWP/706271/2010, published July 2011) lists this guideline as related content but its page does not itself mention prions or TSE.
  • Australia's TGA lists the guideline among the international scientific guidelines it has adopted.

Timeline

  1. 1996Virus validation note for guidance CPMP/BWP/268/95 first published; legal effective date 14 August 1996. Its principles are later extended to prions by this guideline.
  2. 2003Draft released for consultation after Biotechnology Working Party discussion from June 2003.
  3. 2004EMEA expert workshop on the draft.
  4. 2004Consultation comments due by the end of March.
  5. 2004Revised CHMP position statement on CJD and plasma-derived and urine-derived products (rev 1) takes effect; the guideline relies on its section 9.2.3.
  6. 2004Legal effective date given on the EMA page; CHMP adoption and coming into operation in October 2004.
  7. 2004Guideline published by EMEA.
  8. 2009EMA concept paper on updating the CJD position statement says references to this guideline need updating.
  9. 2013PPTA member companies publish a pooled analysis of more than 200 prion removal studies.
  10. 2017LFB IQYMUNE biological safety paper cites the guideline for its TSE clearance work.
  11. 2024CHMP reflection paper revision 3 adopted, making prion reduction evidence a condition of the no-recall policy for non-variant CJD donors.
  12. 2026EMA page still shows the October 2004 text as the current effective version, last updated 21 October 2004.
Sources for this page