Animal & Veterinary
CVM Seeks Comments on Virginiamycin Prototype Risk Assessment
by Jon F. Scheid, Editor
FDA Veterinarian Newsletter January/February 2004 Volume XIX, No 6
The Center for Veterinary Medicine (CVM) has completed its virginiamycin risk assessment, which will serve as a prototype for looking at the indirect risks of resistance from the use of antimicrobials in food animals, and is seeking comments on the assessment.
CVM conducted the assessment, which was released in late November, to determine whether pathways exist to link food-animal uses of virginiamycin, a member of the streptogramin class of antimicrobials, with resistance to other streptogramins used in human medicine.
The link is considered to be indirect because virginiamycin is not a drug used in human medicine, but any resistance created through the use of virginiamycin in food-animals potentially could be transferred to Enterococcus faecium bacteria that reside in the human gastrointestinal tract. If the human E. faecium acquire resistance, treating them with other streptogramins could result in a treatment failure.
The human steptogramin drug of concern is Synercid, which is used in patients who acquire, while being hospitalized, a resistant E. faecium infection. According to the risk assessment, as many as 70,000 patients in the United States may acquire a vancomycin-resistant E. faecium infection in the hospital each year.
The assumption behind the risk assessment model is that virginiamycin use in food-producing animals could create resistance in E. faecium bacteria in the animals. The bacteria could transfer that resistance to other E. faecium in the human gastrointestinal system via the consumption of contaminated food products. That resistance would render drugs such as Synercid ineffective.
The risk assessment was designed and developed as a prototype for risk assessments of the indirect transfer of resistance from uses of antimicrobials in food-producing animals. Earlier, CVM conducted a risk assessment of a direct transfer model when it looked at the risk from the use of enrofloxacin in poultry water creating resistant Campylobacter, which was transferred via food to humans. (A revised version of the risk assessment was released in January 2001.)
Virginiamycin has been used for food-producing animals for nearly 30 years.
However, the Food and Drug Administration’s approval of Synercid in 1999 caused concerns about virginiamycin creating a “reservoir” of resistance in animals that might transfer to humans. Synercid and virginiamycin are both steptogramins.
Principles of the risk assessment
In the late 1990s, when CVM first said it would take into account the microbial safety of antimicrobial use for food animals, it adopted a series of principles for determining the risk from the use of any antimicrobial in food animals. The principles are laid out in the Center’s Guidance for Industry #152, “Evaluating the Safety of Antimicrobial New Animal Drugs with Regard to Their Microbiological Effects on Bacteria of Human Health Concern.”
The first step is hazard identification. A hazard is defined as the possibility that some harm develops from an action, in this case, a threat to public health resulting from the use of an antimicrobial in food-producing animals.
If a hazard is identified, the next step is to assess the extent of the risk. That is done through an assessment of the likely pathways for resistant bacteria to reach people, the extent of exposure people are likely to face to the resistant bacteria, and the consequence of that predicted exposure. Once the risk assessment is complete, officials can determine what risk management steps would be needed.
Conducting the risk assessment
The virginiamycin risk assessment took approximately four years to complete. CVM first announced in 2000 that it would conduct the assessment. At that time, it sought comments about how to go about the assessment, and asked for individuals to submit any data they had.
Dr. Gregg Claycamp and Dr. Barry Hooberman, who conducted the risk assessment, collected data from several sources, including the CVM’s National Antimicrobial Resistance Monitoring System, under which CVM is collecting data about bacteria, including resistant bacteria, on retail cuts of meat. Data were also obtained from the Centers for Disease Control and Prevention and from published literature.
Where no data existed, CVM obtained them by commissioning extramural research and buying isolates from hospitals to be tested for resistance. “We went anywhere we could find data,” Dr. Hooberman said.
What the risk assessment found
The draft risk assessment reported that streptogramin-resistant E. faecium have been found in isolates from poultry and swine sources in both the United States and Europe. It also said that the prevalence of resistance appears to be related to the usage pattern of virginiamycin on the farms.
Further, it said, streptogramin-resistant E. faecium was found on food products from animals, and low-level streptogramin resistance occurred at low frequencies in humans outside of hospitals.
The risk assessment concluded that “the transfer of streptogramin resistance determinants from animal E. faecium to human E. faecium through the foodborne pathway is biologically plausible, but the extent of such transfer in vivo cannot be estimated at this time.”
According to the risk assessment model, if 10 percent of the Synercid re-sist-ance can be attributed to the use of virginiamycin in food-producing animals (with the other 90 percent attributed to the use of Synercid in hospitals), then the food-animal use of virginiamycin may cause from 2 to 37 cases of impaired Synercid human treatment annually due to resist-ance. If the assumption is that 100 percent of the E. faecium resistance to Synercid is a result of the use of virginiamycin in food-animals, then the estimated number of cases of impaired Synercid treatment resulting from the use of virginiamycin in food animals increases by a factor of 10, increasing to 20 to 370 cases annually.
However, the risk assessment also said the resistance genes found on E. faecium taken from animal sources appear to be different than those from human isolates.
Comments or additional information
CVM is seeking comments, until February 23, 2005, about the structure of the risk assessment and the data used. Written comments on this draft risk assessment may be sent by February 23, 2005, to the Division of Dockets Management (HFA-305), Food and Drug Administration, 5630 Fishers Lane, Room 1061, Rockville, MD 20852. Electronic comments may be sent to: http://www.accessdata.fda.gov/scripts/oc/dockets/comments/commentdocket.cfm. All comments should be identified with Docket Number 2004N-0479. Comments will be considered part of the public record and will be available for viewing on the Internet at http://www.fda.gov/ohrms/dockets/ and in the FDA Docket room.
The risk assessment was presented in draft form, and as more data become available, CVM can revisit the assessment. Conducting a risk assessment is “an iterative process,” Dr. Hooberman said, and whenever additional relevant data become available, the assessment can be revised.
According to the risk assessment, if new data or information become available that can narrow an information gap, the risk estimation can be improved by incorporating the new data into the assessment.