Animal & Veterinary
CVM Improves Its Cumulative Adverse Drug Experience Summaries Website
by Suzanne Sechen, Ph.D., Office of New Animal Drug Evaluation
FDA Veterinarian Newsletter November/December 2005 Volume XX, No VI
There is a new look to the “Cumulative Adverse Drug Experience Summaries” on the Center for Veterinary Medicine’s (CVM) website. The revised format should improve readers’ interpretation of adverse drug experiences associated with specific animal drugs.
New animal drugs are evaluated for safety and effectiveness before they are approved by the Food and Drug Administration (FDA). However, testing is done in a limited number of animals and in controlled settings. Less common adverse drug events that could arise during real-world usage might not be detected during pre-approval testing. CVM scientists use reports of adverse drug experiences (ADE) to monitor a new animal drug after it is approved. The ADE reports are assembled into an ADE database. The ADE database helps CVM scientists decide whether there should be changes to product labeling or other regulatory action.
The reporting of ADEs has grown substantially in recent years due to greater awareness of the program and the approval of new types of animal drug products. For example, approximately 3,000 reports were submitted to CVM in 1995, compared with about 18,000 in 2000 and 33,500 in 2005.
Information in the ADE database has been summarized and made available to the public in various formats since 1989. Annual ADE summaries initially were included as inserts in the FDA Veterinarian. Beginning in 1999, ADE summaries for animal drugs have been posted on CVM’s website. Most recently the ADE information has been presented on CVM’s website as “Cumulative Adverse Drug Experience (ADE) Summaries,” which include data collected since 1987.
Late in 2005, CVM decided to revamp the Cumulative ADE Summaries to allow readers a more accurate interpretation of ADE information. The Summaries were removed from the CVM website in December 2005, and the revised Summaries were rolled out on March 10, 2006.
How ADE summaries are generated
Over 99 percent of the information in CVM’s ADE database results from an animal owner or veterinarian contacting the drug manufacturer directly and reporting an adverse event. A manufacturer’s phone number typically will be on the drug’s label. The manufacturer must complete and submit an “FDA 1932” form to FDA within 15 days for serious and unexpected adverse events, or in periodic reports to FDA for “expected” adverse events, e.g., where risk of the adverse reaction is described on product labeling. Manufacturers typically submit periodic reports to FDA every 6 months during the first 2 years after an animal drug is approved, and then annually.
Owners or veterinarians may instead report an adverse reaction directly to FDA using an “FDA 1932a” form. A 1932a form may be printed from the “Adverse Drug Reactions” link on CVM’s website or may be requested by calling the Center (1-888-FDA-VETS). The 1932a form is pre-addressed and postage prepaid for mailing to CVM.
The 1932 and 1932a forms ask for detailed information on the drug and its usage, the treated animals, and the adverse event. Any information that owners or veterinarians can provide the drug manufacturer or CVM is useful. However, with more detail, CVM can better characterize the adverse event and determine how likely it was associated with the drug.
CVM reviewers, who are experienced clinical veterinarians, evaluate information in ADE reports and score each “sign,” or clinical manifestation seen in the treated animal, using a modified version of a human ADE algorithm. The scoring system takes into account previous experience with the drug, other possible causes, timing of events, evidence of overdose, whether the problem disappears after withdrawal of the drug, and whether the problem reappears when the drug is reintroduced. A summary score ranging from -9 to +7 is assigned to each clinical sign and corresponds to how likely it is associated with use of the drug. Clinical signs with summary scores of 0 or greater are considered “possibly, probably, or definitely” drug-related. Negative scores denote signs that are considered “remotely” drug-related or without enough evidence to draw a conclusion.
CVM reviewers enter the information and summary scores derived from ADE reports into the Center’s ADE database. The Center also uses this information to update the Cumulative ADE Summaries available to the public, typically on a monthly basis.
Changes to the Cumulative ADE Summaries
The Cumulative ADE Summaries are on CVM’s website at the “Adverse Drug Reactions” link found either in the “Hot Topics” section or under the “CVM A-Z Index” of the home page. The Summaries are separated into alphabetical subsections at the bottom of the Adverse Drug Reactions page. Users click on the subsection associated with the generic name or active ingredient of the drug of interest. The page has a guide that allows users to find the generic name by looking up the brand name. A Summary is provided for a specific drug, species, and route of administration reported to FDA.
Drugs are listed in the Summaries by their generic, or active ingredient name. There may be more than one brand of the generic drug. A complete list of brand names associated with each generic name may be found using CVM’s on-line “Green Book.”
Until December 2005, each Summary listed at the top: “reviews” (number of ADE reports for this use of the drug); “treated” (number of animals in the reports that were treated with the drug); and “reacted” (number of treated animals in the report that had an adverse reaction assigned any summary score, i.e., from -9 to +7). These terms often were confusing unless readers referenced a glossary. Each Summary also listed at the top the number of animals in the ADE reports that “died.” However, this term included incidents that CVM reviewers might determine were not “possibly, probably, or definitely” related to use of the drug, plus animals that were euthanized.
The previous version of each Summary listed by frequency every clinical sign and the number of treated animals observed with the sign “possibly, probably, or definitely” associated with this use of the drug (i.e., a summary score of 0 or higher). Also provided for each sign was the percent of all ADE reports for this use of the drug in which the specific sign was reported. These percentage terms were often misinterpreted as meaning the percentage of all animals in the United States receiving the drug that had this sign.
The new version of the Cumulative ADE Summaries now lists at the top of each Summary the “Number of Animals Evaluated,” which refers to the number of animals in the ADE reports for a specific drug, species, and route of administration that CVM scientists review for adverse reactions. The term places focus on the animals in the ADE reports, rather than the number of ADE reports submitted. CVM reviewers may determine that the adverse experiences in some of the animals evaluated were not “possibly, probably, or definitely” related to the drug.
The Summaries continue to list by frequency every clinical sign and the number of treated animals observed with the sign “possibly, probably, or definitely” associated with this use of the drug. However, the percentage term used previously has been dropped to avoid its misinterpretation as a percentage of all animals receiving the drug. Death is now included among the signs by order of frequency but is separated into appropriate categories, such as “death” and “death by euthanasia,” to more accurately characterize the reaction.
Despite the recent revisions, readers must still keep in mind the limitations of the Cumulative ADE Summaries.
The incidence rate or risk of specific clinical signs associated with a drug cannot be calculated because the total number of animals given the drug in the United States is not known. For this same reason, drugs listed in the ADE reports cannot be compared in terms of the number of clinical signs reported. If one drug is widely used, it may have more ADE reports than another drug used in only a small number of animals. Also, media attention to a specific drug might result in many more ADE reports being submitted.
The accuracy of ADE information in the cumulative reports is dependent on the quality of information CVM receives from veterinarians or animal owners. While valuable in terms of monitoring a drug in a real-world setting, the information can be less precise or specific than data coming from a controlled study.
Although the scoring algorithm helps CVM scientists estimate the likelihood of an association between a drug and a reported clinical sign, it cannot definitely show that anadverse reaction was caused by the drug. The adverse reaction may have been related to underlying disease, use of other drugs at the same time, or other non-drug related causes.