• Decrease font size
  • Return font size to normal
  • Increase font size
U.S. Department of Health and Human Services

News & Events

  • Print
  • Share
  • E-mail


"This text contains Dr. Henney's prepared remarks. It should be used with the understanding that some of the material may have been added or deleted during actual delivery."

Remarks by:

Jane E. Henney, M.D.

Commissioner of Food and Drugs





APRIL 15, 1999


Good morning. I am pleased to be here today at the National Institutes of Health. We are at a unique time and place, where working together we are poised to reap the benefits of the national investment in biomedical research—to our collective benefit as a research enterprise and for individual patients across the country. To achieve our expectation, it is critical that FDA engage and interact with the expertise from many sectors: Federal research agencies, industry and academia. Through these interactions, we can promote the development of new, better therapies as quickly as possible—moving expeditiously without compromising the high standards for safety and efficacy that patients have come to expect.

First, I would like to discuss what has been the traditional standard for approval of new drugs and biologics upon which the Agency has relied. The Federal Food, Drug, and Cosmetic Act requires that there be substantial evidence that a drug will have the effect that it is claimed to have, based on information from adequate and well-controlled studies, and that it be safe for use under such conditions. Historically, the traditional endpoints relied on by the Agency for measuring safety and effectiveness have been morbidity and mortality. However, if we believe that the information provided by the surrogate addresses the appropriate clinical question, we can make risk/benefit decisions based in part on the demonstrated effects on the surrogate. the risks associated with a therapy cannot be justified merely by an effect on a laboratory marker with no direct benefit to the patient.

Surrogate endpoints were used first in limited settings by the Food and Drug Administration in its review of drugs, mainly in the cardiovascular area. For example, antihypertensive drugs have been approved based on their effectiveness in lowering blood pressure. Similarly, in the past, cholesterol-lowering agents have been approved based on their ability to decrease serum cholesterol, not on the direct evidence that they decrease mortality from atherosclerotic heart disease.

However, this narrowly circumscribed area where such considerations had been applied was to expand dramatically with the AIDS epidemic. Growing frustration over perceived lack of access to therapies, as well as concerted efforts to shorten review times resulted in an important change in FDA’s processes. In 1991, the HIV epidemic was already well into its first decade with only one approved therapy– one to which many patients were intolerant or unresponsive. There was a critical need for new therapies. Much data was accumulating about the significance of CD4 cell counts and their predictive value for outcomes. It was upon this basis that didanosine was approved for the treatment of HIV.

Aside from the changes that it brought to the Agency’s regulatory framework and the evolution of the Agency’s thinking about approval for new therapies, the AIDS example is a compelling demonstration of what can be accomplished through communication and collaboration. When there came a critical need to understand whether HIV-RNA was of use in predicting clinical benefit, the industrial sector, the academic community and scientists from NIH and FDA worked together to analyze the vast amount of available data. One party alone could never have accomplished this; but by sharing data and experience, the maximum amount of information was produced. Antiretroviral agents are now approved on the basis of their ability to produce long term HIV RNA suppression.

Utilizing our experience in the AIDS area, FDA codified the use of surrogates in its accelerated approval regulations in 1992. In essence, for serious and life-threatening diseases, the FDA could approve a drug based on a surrogate endpoint that was "reasonably likely" to predict clinical benefit provided certain conditions were met, including that the presence of a therapy-associated clinical benefit would be established in post marketing clinical trials.

The next milestone in the utilization of surrogate markers occurred in March 1996. FDA issued a report entitled, "Reinventing the Regulation of Cancer Drugs." This report announced the expansion of the use of the accelerated approval process for cancer treatments, based on verified and recognized demonstration of objective tumor shrinkage. That new policy, built upon the experience that the Agency had in the AIDS area, was designed to accelerate the study of cancer therapies and shorten the total time for first and subsequent marketing approvals for a wide range of therapeutics.

In 1997, Congress enacted the Food and Drug Administration Modernization Act, which, among many other things, codified FDA’s accelerated approval language into the statute and thus codified our approach to expedited drug development. This "fast track" provision is meant to facilitate the development and expedite the review of drugs that are intended to treat a serious or life-threatening condition and that demonstrate the potential to address a serious unmet medical need. The provision recognizes that our efficacy standard is unchanged – there must be substantial evidence that the drug will have the beneficial effect it purports to have. But the new provision recognizes that clinical research techniques have progressed and that to best meet public health needs, we must take into account a full array of proper evidence –including evidence from studies with surrogate endpoints that will allow us to make our risk/benefit determination both accurately and in a timely manner.

In order to provide clear guidance, FDA has issued a guidance document on the fast track provision. This guidance defines the criteria for qualification in the fast track drug development program, sets out the process for designation as a fast track drug, and describes programs for expediting development and review of fast track products. This document is an excellent resource for sponsors who are interested in receiving fast track designation for their products.

The Agency’s fast track and accelerated approval programs have been very successful. The numbers speak for themselves. Under the accelerated approval procedures, there have been 26 new drug applications approved since 1992. These approvals include new therapies for the treatment of patients with AIDS, cancer, heart disease, and hypertension. All of these products were made available to patients more quickly due to reliance on surrogate endpoints. In addition, we have granted fast track designation for therapies for those with atherosclerotic vascular disease, acute stroke, diabetes, adult respiratory distress syndrome and pancreatitis.

Our past successes foreshadow what is to come as we move forward in new areas. We need to be looking for additional situations where surrogates can be used. We can consider whether surrogates would be useful in early stages of development, enabling smaller, faster studies. We need to assess how we can pull data together to look across studies for information.

However, we must acknowledge that our understanding of a surrogate marker may be imperfect and may lead to an incorrect decision. We should not forget the lessons learned with antiarrythmic agents. Members of the research community and scientists at FDA expected suppression of premature ventricular contractions to accurately predict a clinical benefit. Unfortunately, for encainide and flecainide that hypothesis was overly simplistic and incorrect.

Surrogate endpoints are objective observations that reliably serve as early indicators for clinically meaningful endpoints. They must be strong leads--more than just a notion. Therefore, there needs to be a demonstrable linkage between the surrogate and the ultimate endpoint. We need to have some evidence that these surrogates will lead to a clinically meaningful endpoint, or we cannot rely on them. If it turns out that a therapy that initially seemed promising does not produce a meaningful effect, we need to be prepared to withdraw that drug from the marketplace. While we will be open to new ideas about the use of surrogates in clinical trials, we must employ the best clinical and laboratory science to support our standards for efficacy.

Additionally, we should begin to think about surrogate markers more broadly. Surrogates can not only predict efficacy but can define toxicity. For example, by utilizing knowledge about genetic polymorphism, scientists are developing information needed to better predict who will be responding favorably and with less toxicity to a given therapy. Much of our knowledge at present results from chance observations. By using surrogate endpoints as tools to predict outcomes, we hope to increase the odds of finding effective, but less toxic therapies more accurately and efficiently. Therefore, the research community should consider investigation in the area of surrogate markers for both efficacy and toxicity.

Another critical area as we work to advance this field is to standardize nomenclature. Surrogates clearly serve as valuable tools in late stage clinical trials. While we do not expect validation of a marker in preliminary trials, it can sometimes provide valuable information about drug activity. By clarifying the utility of such markers, using a common nomenclature, we will able to keep a consistent approach to use of these tools.

I fully recognize that as we move ahead, it is critical that there be both active communication and collaboration among those working in the field—in the industry, academic community and federal scientific agencies. Because we at FDA have a critical role in this process, we need to be involved as science is emerging. Ideally, we must interact frequently and meaningfully as the clinical research proceeds. Nothing is more frustrating or unproductive for the applicant or the Agency than waiting to engage in a discussion of clinical trial design or designation at the end of the process.

Our nation is investing heavily in research in both the public and the private sector. In addition to the traditional areas of research, clinical research and research initiatives in clinical trial design will be important to our future success. I know that over the next few days you will be discussing work in this area in detail. I look forward to learning from these deliberations.

In conclusion, product development can often be greatly aided by the use of surrogate endpoints. Their use will help us to better develop the science upon which so many critical public health decisions are based. It is my hope that we can learn from our failures, build on our successes and continue to collaborate in this very important area. I remind you that we are not only ready, but eager to act. Thank you.