Guidance on the Inactivation or Removal of Select Agents and Toxins for Future Use: Verification of a Validated Procedure

Once an entity has  validated an inactivation procedure, they may have to verify the validated procedure depending upon the type of sample and risk assessment. This involves the entity determining a sampling strategy for viability or infectivity for subsequent inactivation. Verification of inactivation or removal procedures will differ depending on the category of the sample. There are three categories of sample: 1) inactivated agent, 2) extracts, or 3) material (within this section, material refers to sample in which intact select agent has been removed). The portion of samples subjected to a validated inactivation procedure, (such as 10%) used for verification will depend on the risk assessment. Samples with more inherent variability may require verification viability testing of a portion of all subsequent inactivated samples while samples with low inherent variability may only require verification viability testing of process controls while the risk assessment for fixed tissue samples may allow for no verification viability testing.

Verification of Inactivated Agent

Consider highly pathogenic avian influenza virus (HPAIV) and Newcastle disease virus as an example. When testing a small volume or to achieve a lower detection limit of the assay, increase the number of eggs (i.e. replicates) and passages to increase the probability of detecting a viable organism at the assay limit of detection.

Consider verification of tissue samples containing select agents. This would involve inactivation of the select agent present in the tissues, and then grinding up the tissue for use in a viability test along with appropriate process controls, resulting in no sample for research purposes. In these situations, or when there is a limited amount of the sample, verification is not possible. The validation of the inactivation procedure must provide confidence to allow for the risk associated with not verifying subsequent inactivated samples. If the validation data cannot provide this confidence, then an entity can choose to take an immediately adjacent sample of the affected tissue in subsequent inactivation experiments to verify inactivation of the agent.

Consider verification of chemically inactivated Ebola virus using a validated procedure. Verification may involve viability testing 10% of samples (1 out of every 10 inactivated samples or 10% from every sample) or just the process controls. The sampling strategy will depend on the risk associated with not verifying every subsequent inactivated sample.

Verification of extract

The term “extract” is commonly used in conjunction with nucleic acids extracted from a select agent. The term “extract” reflects the application of two processing steps: an inactivation step to destroy the select agent (e.g., lysis of select agent) and then another step (such as filtration), to remove any remaining viable select agents. Extracts from a select agent (nucleic acids, antigens, lysates) would be subject to the performance standard for select agents in the new sections 3(d)(4) and 4(d)(4) of the select agent regulations that requires validation of a procedure by confirming its efficacy with data generated in house by the entity using a viability testing protocol but does not necessarily require verification viability testing on every sample. For nucleic acid extractions (other than Bacillus anthracis and Bacillus cereus biovar anthracis) subsequent to validation, it is not always possible to verify every sample for non-viability. The risk associated with not performing verification viability testing on every subsequent sample can be mitigated by the multi-step procedures used to extract nucleic acids, validation of the procedure with all (100%) of the inactivated sample (extracted nucleic acids), inclusion of process controls, and/or the incorporation of a safety margin. For example, the laboratory could choose to only test the process control, only perform verification viability testing on a portion of the total samples, and/or incorporate a safety margin.

Note: Inactivation of Bacillus anthracis and Bacillus cereus biovar anthracis is subject to the new inactivation requirements and the FSAP policy on Inactivated Bacillus anthracis and Bacillus cereus biovar anthracis. For procedure validation for nucleic acid extracts from Bacillus anthracis or Bacillus cereus biovar anthracis, viability test all (100%) of the sample a sufficient number of times to demonstrate the procedure works. After this validation, verify the procedure by viability testing at least 10% of the sample or production lot of subsequent extracts or production lots to verify the validated procedure.

Verification of material

Material containing select agents, as opposed to extracts (e.g., nucleic acids, antigens, lysates), that is subjected to a process to remove (e.g. filtration) all viable cells, spores, or virus particles would require verification viability testing on every sample prior to treating it as a non-select agent. The distinguishing feature between “material containing a select agent” and an extract from a select agent is that in the former the select agent will only be removed and in the latter the select agent will be destroyed before removal. The more stringent requirement for viability testing of all material containing a select agent where the select agent was removed is warranted because of the lack of select agent destruction which increases the risk of viable select agent remaining in the material.

For example, consider verification of filtered serum from an animal infected with Francisella tularensis. This would require verification of every sample, such as subjecting 10% of every sample to viability testing.