Susan Ellenberg: Statistician with a commitment to product safety and efficacy
Susan Ellenberg was born and grew up in Tucson, Arizona. She graduated from Radcliffe College in Boston, and then received her M.A. in teaching. “I really loved math,” she recalls “and I learned that what I could be most happy teaching was mathematics.” When her husband went to NIH and her first child was born, however, she decided to pursue her Ph.D. in statistics from George Washington University. “I was really very fortunate. I took my courses at night and was able to get in homework during the day when the kids were sleeping, and it just took me longer…eight years.”
While in school, she was also working part-time for renowned NIH biostatistician, Jerome Cornfield. Cornfield had a contract with FDA, through GW, to develop the methodology and write a computer program to do sequential analysis of clinical trials on the Cornfield model. After receiving her degree, Ellenberg joined the Division of AIDS at the Institute of Allergy and Infectious Disease (NIAHD) where she began working closely with FDA in the midst of an epidemic. “This [AIDS] was like nothing anybody had ever seen before…there was so much emotion and so much urgency about AIDS drug development …People were realizing the need to invent new ways to do research and new ways to regulate.”
Formal weekly meetings of the AIDS Clinical Trial Group were soon supplemented by informal, regular meetings of statisticians interested in AIDS research. An informal group of 25 people soon grew to well over one hundred members including clinicians, statisticians, and eventually AIDS activists. Ever inclusive, Ellenberg invited top researchers to attend meetings of the Statistical Working Group. She was pleased that “nobody ever said ‘no.’”
She also found that AIDS activists such as ACT UP, which up to that point had been best known to the FDA for its "tombstone protest" rally in front of FDA headquarters, protesting the agency's AIDS policies, actually had some sound and sensible ideas regarding the conduct of clinical studies, which, although they were unfamiliar to industry, were defensible. “My feeling, and the feeling of a lot of other people, was that you first have to decide what’s best for the patients …treating everybody as effectively as you can treat them. Then you worry about how you are going to sort it out [the statistical design].”
Ellenberg brought a broad experience in statistics across disease entities to bear on the problem of AIDS trials and analysis. “One of the things I’ve learned over my career, “ she shared, “is that people [at NIH and FDA] develop truths based on their own experience, and find it sometimes very hard to see that what’s absolutely true for them may not apply in other areas.” Her commitment to flexibility in the interpretation of evidence encouraged the emergence of a new more balanced regulatory approach to drug trials for drugs to treat seriously ill patients. It also helped ensure the ultimate approval of the first effective anti-retroviral drugs, beginning with AZT, to treat AIDS patients.
When FDA and CBER created its own Division of Biostatistics and Epidemiology, Ellenberg was selected as its first director, a position she held from 1993-2004. She says that coming to FDA taught her the importance of communication by regulators. She found that conclusions which had once puzzled her as a statistician at NIH, made much more sense in a regulatory context. Her work at FDA was characterized by a commitment to vaccine safety accompanied by an insistence on more publications by regulators in professional journals to make clear not only what was known, but also what was not known, about the safety of children’s vaccines. Dr. Ellenberg is currently Professor of Biostatistics and Associate Dean for Clinical Research at the University of Pennsylvania School of Medicine.