A Look Back at CVM: During 1970s, Division of Nutritional Sciences Added to Scientific Review Sophistication

by Jon F. Scheid, Editor
FDA Veterinarian Newsletter September/October 2005 Volume XX, No V

For the Food and Drug Administration’s (FDA) Centennial, FDA Veterinarian is taking a look back at some of the events that shaped the Center for Veterinary Medicine (CVM). Several significant changes took place in the early 1970s, when the Center developed its expertise in the use of statistics and “experimental design.” Here’s the story of that change, based on interviews with some of the people who were here at the time.

In the 1970s, the Bureau for Veterinary Medicine (BVM), which in 1984 became CVM, began employing broader use of statistics and “experimental design,” especially for drugs used for livestock production purposes. Before that, BVM relied more on reviewer judgment than probability assessments in deciding whether to approve an animal drug, according to Dr. Woodrow Knight, who was at the Bureau then and was part of the change.

Dr. Knight, who recently retired from his position of Director, Division of Production Drugs, at CVM, said reviewer judgments worked well when reviewing therapeutic drugs. However, he said, “Techniques for reviewing drugs intended to increase the rate of gain, improve marketable milk production, increase egg production, affect carcass quality, or improve feed efficiency in normal, healthy animals were unfamiliar ground” to many on the BVM staff at the time.

Dr. Knight, an animal scientist, was one of the first staff members hired for the Bureau’s fledgling Division of Nutritional Sciences. The Division was formally created on February 1, 1971, and staffed with animal scientists, statisticians, and veterinarians.

“The difficulty was how to evaluate drugs that affected small increments of response, normally less than a 10 percent and often less than a five percent response over the negative control treatment,” Dr. Knight said. “Contrast this with a therapeutic drug, which often gives a 75 percent response or greater and whose response is very obvious in a disease-response situation.”

“It is my understanding” that before the Division of Nutritional Sciences was established “the Center’s technical expertise consisted mainly of persons who had years of practical experience in veterinary medicine and who were not trained in the analytical skills of statistics, probability, and experimental design,” Dr. Knight said.

Dr. William Price, one of the founders of the Division, and currently a special assistant to the Director of CVM’s Division of Animal Feeds, said the changes came about because the BVM Director at the time, Dr. C. D. Van Houweling, had a research background and recognized the need for these additional disciplines in drug review.

After the Division was established, the Bureau hired animal scientists with expertise in statistics and experimental design. Scientists hired at the time included Dr. George Graber, who was hired as an animal nutritionist and is now the Deputy Director of CVM’s Office of Surveillance and Compliance, and Dr. Price.

Dr. Price was the Chief of the Ruminant Branch in the Division of Nutritional Sciences. He came to the Bureau in the late 1960s with a degree in animal science as well as extensive graduate training in mathematical statistics. Among those Dr. Price hired were Dr. Knight and Dr. Norris Alderson, who now serves as FDA Associate Commissioner for Science, and Director, Office of Science and Health Coordination.

The experimental design and statistical tools that the scientists brought to BVM led to the use of tests and experiments that generated more information. The techniques the Division of Nutritional Sciences brought to the Bureau, for instance, called for designing feeding trials that would use small subgroups, rather than an entire poultry house, for the experimental groups, and then replicate the tests for a more “powerful” result, Dr. Price said.

Also, according to Dr. Graber, “Dose response studies introduced for the first time the ability to allow for effectiveness over a range of drug content in feed, which when coupled with regression analysis permitted interpolation between levels tested.”

Growing scientific sophistication

The Division was successful in transferring its techniques throughout the rest of the Bureau. “Our early successes included the incorporation of statistics into the process of evaluating drugs intended for both therapeutic and non-therapeutic uses and for food additive products. We wrote guidelines that became a model for the other units within the Bureau,” Dr. Knight said.

The new ways of scientific review of data allowed BVM to be a significant participant in FDA’s effort to remove Diethylstilbesterol (DES) from the market.

According to a history of CVM, prepared in 1989 by Orland Soave, a veterinarian and consultant, DES was approved in 1954 as a growth stimulant for cattle and sheep. However, scientists were able to detect residues of the drug in food from treated animals. In 1974, FDA initially prohibited the use of DES in cattle.

The work BVM did in connection with DES led to a further exploration of the law’s requirements about substances labeled as carcinogens. The Division of Nutritional Sciences, with encouragement from then-Chief General Counsel Peter Barton Hutt, developed the early concepts that led to the “Sensitivity of Method” (SOM) regulation.

Certain substances that can technically be labeled carcinogens but can still be safely used in food-producing animals are permitted under Federal law through the “DES proviso.” And Federal regulators use the SOM rules to determine when the DES proviso can be applied.

See 21 U.S.C. §360b(d)(1)(I), which presents the DES proviso:

“(I) such drug induces cancer when ingested by man or animal or, after tests which are appropriate for the evaluation of the safety of such drug, induces cancer in man or animal, except that the foregoing provisions of this subparagraph shall not apply with respect to such drug if the Secretary finds that, under the conditions of use specified in proposed labeling and reasonably certain to be followed in practice (i) such drug will not adversely affect the animals for which it is intended, and (ii) no residue of such drug will be found (by methods of examination prescribed or approved by the Secretary by regulations)…in any edible portion of such animals after slaughter or in any food yielded by or derived from the living animals….”

SOM regulations specify procedures for determining the metabolites, the tissues that the metabolites are concentrated in, the safety of the metabolites as determined by animal tests, and the level of the metabolites that would present a minimal cancer risk (specified in the regulation as corresponding “to a maximum lifetime risk to the test animal of one in 1 million”) before the drug can be considered for use in animals. Consequently, the regulation used to assess the one in 1 million risk of cancer to animals is written to require that the concentration of the drug in edible tissues be so low as to “cause no significant increase in the risk of cancer to people.”

According to Dr. Knight, prior to the development of the SOM regulation, FDA did not have a systematic process for determining the safety of carcinogens added to the food supply.

“Our concern initially related to how to interpret the requirement (in the law prohibiting the use of carcinogens) so that the level of sensitivity for the method that the Secretary would establish to measure the residue resulting from the use of a carcinogen in food-producing animals would not be changing just because scientists developed a more sensitive method. The sensitivity requirements of this method were to be established based on the safe level determined by scientists qualified to make such evaluations. Thus, the SOM document was designed to explain how the Secretary made the determinations of safety for such carcinogens that may be added to the food supply,” Dr. Knight said.

The SOM concepts were first developed in 1973 by the members of the Division of Nutritional Science, Dr. Price said. The SOM rule was finalized in 1987 and amended in 2002. The 2002 amendments revised the definition of “no residue” to mean that no residue of the drug is detected with an approved regulatory method under the conditions of use of the drug. The SOM concept is based on metabolism studies and statistical analysis and disciplines, which were the Division of Nutritional Sciences specialty.

Division became part of ONADE

The Division grew to about two-dozen employees by the mid-1970s. Later during the 1970s, Dr. Van Houweling merged the Division of Nutritional Sciences with the Therapeutic Animal Drugs Division to create the Office of New Animal Drug Evaluation (ONADE). The Division of Nutritional Sciences’ branches of ruminant, non-ruminant, and poultry drugs were brought into ONADE and expanded to include anthelmintics and hormones. Since then, the departments were put back into their own Division of Production Drugs as the current Ruminant Drug Team and the Swine and Poultry Drug Team. Also, the food additive responsibility was shifted to the Division of Animal Feeds in the Office of Surveillance and Compliance.

According to Dr. Knight, the “big asset of the Division of Nutritional Sciences was that we looked at the data presented by the sponsor and made recommendations based on the detailed analysis of that data using the principles of statistics and experimental design.”

Although the Division of Nutritional Sciences no longer exists within CVM, its legacy continues, Dr. Knight, Dr. Price, and Dr. Graber said. The use of statistical analysis and experimental design is now a standard part of the review of most animal drug applications, they said.