• 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

Lester M. Crawford, D.V.M., Ph.D. - Mayo Alliance for Clinical Trials

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

Speech before
Mayo Alliance for Clinical Trials Conference

Remarks by
Lester M. Crawford, D.V.M., Ph.D.

Acting Commissioner of the FDA

August 26, 2004

Thank you for the kind introduction. I always thought if I ever had to go to the Mayo Clinic, it would mean that I was really bad off health wise. I am especially delighted to learn that I can indeed come here in good health and participate with all of you in this Clinical Research Conference.

In particular I'd like to thank the Mayo Alliance for Clinical Trials and the Center for Patient Oriented Research for inviting me here to speak to you about such a critical topic in medicine -- how to overcome today's challenges and capitalize on today's opportunities in clinical research.

It seems like every day you read the papers and find some new story that relates to clinical research -- articles about developmental drugs to fight cancer or Alzheimer's, articles about the increasing cost of medicine and of developing medicines, articles about the ethics of clinical trial participation and about the reporting of negative results.

More and more, our country is focusing on the challenges and opportunities we face in clinical research. And we need to -- because this is a critical time in our nation's health.

On the one hand, it is an exciting time for the biomedical sciences. New medical breakthroughs submitted to FDA for review have more to offer patients than ever before.


  • New classes of drugs, such as the statins, are proving so beneficial to so many patients with so many different diseases that the introduction of these and other new drug groups rival many of the established "miracle drugs" of previous decades.
  • New engineering applications of nanotechnology promise to deliver the right treatment to the right place in each individual with far fewer side effects and complications than in the past.
  • And next-generation I.T. can turn the explosion of health information into a life-saving database at the fingertips of health care providers and patients, facilitating better informed, far more effective decisions that improve health, safety and cost-efficiency.

There's more spending on biomedical R&D today than ever before. Total spending has grown exponentially over the past 10 years. The NIH budget has been doubled.

And yet, there's a dark cloud lurking behind this silver lining.

The hard truth is that many of the most dramatic scientific advances in basic research are simply not translating into safe and effective medical treatments for patients.

Instead, a disturbing trend is unfolding. Despite the increase in R&D spending, the number of new innovative drugs and biologics being submitted to the FDA for review is actually decreasing. FDA is now receiving fewer applications for new drugs than in the mid-1990's .

The inescapable result of this decline is that fewer new products are reaching patients who desperately need them, and the ones that do are continuing to get more expensive. Despite all of our wonderful science, we are not yet realizing the promise of the Biomedical Century.

But I'm here to tell you today that FDA is working on a remedy.

We're working on a solution that brings together all the key partners in government, in industry, in academia, and in the consumer and patient communities, to deliver innovative, new medical products to Americans faster and more cost-effectively. Because new medical innovations, used appropriately, save lives and make us a stronger, healthier society.

FDA's part of this solution involves a number of different initiatives -- some internal to the Agency, some external -- that are all directed at improving the pathway along which medical innovation must travel from "bench to bedside."

First let me update you on some of the internal initiatives at FDA to improve innovation.

These include a number of efforts to both speed review times and, more importantly, to improve the processes that researchers and manufacturers use to develop innovative products -- the tools of the translational trade so to speak.

First is the development of "quality systems" for our internal review procedures. Our plan is to build on FDA's professional staff expertise to identify and apply best management practices. We're moving forward not only to reduce the time it takes to review new treatments, but also to help product developers get their applications right the first time -- to do the R&D work that demonstrates safety and effectiveness as quickly and efficiently as possible.

Second, though FDA already has guidances for many therapeutic areas, the Agency is working on updating old guidance documents and developing new ones on how to develop therapies for especially critical medical needs. The idea is to bring together experts in government, in academia, and the private sector to design a better pathway for developing new treatments -- to make sure that the clinical endpoints for these studies, the study designs themselves, the statistical analysis tools, and the whole clinical development process is as clear and predictable as possible for both scientists and clinicians.

Interestingly, we have found that medical devices developed in areas with extant FDA guidance documents are almost twice as likely to be approved after the initial review process and are approved in a third less time. That's powerful testament to the value of clarifying the development pathway through regulatory guidance.

Next, FDA has been conducting a root cause analysis for product approvals that require more than one FDA review cycle and result in many months of additional development time. Evidence shows that upfront, focused communication with product developers about FDA expectations can help developers get the science right in a shorter timeframe. FDA now has well over 2,000 meetings a year with companies regarding their product development plans. We have several new pilot programs underway to pursue even earlier and more frequent communications with product manufacturers, and we expect to see this additional upfront, focused communication bear fruit. Predictability is a team effort.

These sorts of internal improvements are making a difference, but they are not enough. They are not enough because the biomedical research and development path continues to grow more complicated, more expensive, and more uncertain.

For very innovative and unproven technologies, the probability of an individual product's success is highly uncertain, and perceived risks are extremely high. Historically, 14 percent of products submitted to the Agency eventually cleared the hurdle of FDA approval and entered the marketplace. Today, the success rate of a candidate drug entering Phase 1 trials is estimated at 8 percent. We're currently seeing a 50 percent failure rate among products in late-stage Phase 3 trials. 50 percent! At the same time, the cost of developing new products has soared. By some estimates, it now costs more than $800 million to bring a new drug to market.

This is not only troubling from a scientific and regulatory perspective, this is deleterious to the public health.

So what is behind all this?

The answer: For all that modern science has to offer, developing new treatments is still very much an art -- in which hunches, intuition, and luck play a critical role. The odds are long. For medicine that is affordable and innovative, we need more well-understood science and we need regulatory predictability.

One thing is becoming increasingly clear -- the applied sciences needed for 21st century medical product development have not kept pace with the tremendous advances in the basic sciences.

Earlier this Spring, FDA issued a major report examining these challenges to medical development, focusing on what we call the "critical path" to medical product development.

"Critical path" refers to the series of translational hurdles that medical products such as drugs, biologics, or devices must negotiate as they move from laboratory concept to the actual commercialized treatments that make a real difference in patients' lives. The critical path encompasses not only the science of developing drugs and devices and other medical products, but also the know-how for turning an experimental medicine into a finished product that can be used safely and effectively.

Our critical path thesis is that better, more current and predictable scientific methods and standards must be developed and devoted to streamlining the critical path. Investment in basic research is not enough. Specifically new development tools -- including assays, standards, computer modeling techniques, biomarkers, and validated surrogate clinical trial endpoints -- are needed to improve the predictability of the development cycle and to lower the cost of research by helping industry identify product failures earlier in the clinical trial process and to identify successful candidates more quickly. When Thomas Edison was asked why he was so successful he responded, "Because I fail so much faster than everyone else."

Our critical path report finds that updated development tools are needed especially in three critical areas: Product Safety, Medical Utility, and Manufacturing. The vast majority of development costs are attributable to these three areas, and they are all critical elements to success. And so, a new set of technical tools in these areas will not only make the development process more efficient and effective, it will help us start to see a better return on our biomedical investments.

With improved scientific methods and a new, shared effort by all of us, I believe we can develop and improve standards for product characterization and product efficacy and safety testing, both for traditional and innovative products.

For example, by applying genomic and proteomic techniques, we can develop safety assessment programs for new biomaterials. With better scientific methods, we can develop better animal models, new biomarkers, and surrogate end-points for clinical safety and effectiveness.

Armed with information from these tools, doctors can then individualize therapy, targeting those patients who will respond to medication and avoiding those who are at risk for side effects. Such tools will also allow practitioners to better monitor patient response to different treatments. Targeting therapies in this way will not only be helpful from a health perspective but should also help decrease the number of patients required to actually demonstrate the effect of a drug.

We can also apply modern engineering and cutting-edge scientific knowledge to medical product manufacturing. We can improve standardization and automation of clinical research. We can develop novel and improved clinical trial designs and analytical methods for evaluation of safety and effectiveness that can reduce costs. We have done this with our use of various statistical approaches, and we have opportunities to further build on this concept.

FDA is uniquely suited to take an organizational role in the Critical Path effort because of our unique cross-industry and cross-cutting knowledge of the hurdles companies and products encounter that are causing them to fail in late stage clinical trials. FDA has the technical expertise that can draw together stakeholders, help prioritize research that is most needed, and to partner with others to conduct this research.

Currently academics and private companies collect data and establish correlations, but no one is responsible for organizing this information into the broader knowledge that could lead to more generalized principles that industry and FDA could use for faster and more accurate product development and evaluation.

As an example, think about the millions of dollars that would be saved by all types and sizes of companies if publicly discussed and vetted biomarkers could be used -- and used predictably -- during the product approval process. In fact, we estimate that a mere 10% improvement in predicting products' failures in clinical trials could save $100 million in development costs per drug. It is precisely that kind of cost reduction that will get innovative medicines in the hands of patients sooner and cheaper.

I'd be remiss if I failed to point out that this initiative will complement other existing efforts to strengthen basic and translational research, for example those underway at NIH and in the private sector. FDA's critical path activities and research will complement, not compete with, what our sister agencies and industry are already doing.

What I've just outlined to you is not just a vision, it is quickly becoming reality.

Over the past several months, FDA has been reaching out to all of our stakeholders to obtain vital input regarding specific challenges and opportunities along the critical path. Later this fall, we will be announcing the first National Critical Path Opportunities List, which will pinpoint those areas of product development that could benefit most from innovative approaches and emerging technological advances -- opportunities that have the potential to fundamentally change and modernize the 21 st century critical path for medical product testing and manufacture. With this national opportunities list in hand, the Agency can start to rally our various partners to build the path to innovation in these key areas.

Now, our public docket recently closed for Critical Path input, however we are continuing to aggressively solicit ideas and suggestions from all of our partners, and so I'd encourage you to put forth any ideas you might have. Let's blaze the critical path forward together.

NCI Partnership
As the Critical Path initiative has made clear, collaboration is key to improving clinical research and medical innovation.

Nowhere is this truer than in the area of oncology research. Cancer is one of the top killers of Americans today, and it continues to take a devastating toll on our public health and our society.

However, FDA is presently engaged in a number of interesting collaborations with the National Cancer Institute (NCI) to help combat this deadly disease and hopefully turn the tide.

Last May, my predecessor Dr. Mark McClellan and NCI Director Andy von Eschenbach announced an unprecedented partnership between our two agencies. Drs. McClellan and von Eschenbach established a new Interagency Oncology Task Force, aimed at enhancing the efficiency of clinical research and scientific evaluation of new cancer medications through better knowledge- and resource-sharing.

Over the past year, this task force has been hard at work identifying opportunities for research collaboration and I'm proud to announce today that we are making considerable progress across a number of different areas.

The first area involves bioinformatics. This is an area that is particularly exciting these days, as we are starting to see the immediate positive impact 21st century information technology can have on our public health.

FDA is working closely with NCI on efforts to share clinical research data through a secure infrastructure, as part of the Cancer Biomedical Informatics Grid (caBIG) project. Specifically, we are helping to deploy a standards-based, electronic clinical research exchange to support regulatory submissions and facilitate a more cost efficient, secure flow of data among sponsors, researchers, and regulatory authorities.

Inherent in this effort is a push towards the routine electronic submission of investigational new drug (IND) applications. We are creating a new system for submitting INDs electronically in order to reduce the time it takes for promising new drugs to be reviewed for testing in clinical trials.

Any electronic system requires standards, and FDA recently took a major step forward in the adoption of a standard format that can be used to submit human drug clinical trial data. Specifically, the Agency is supporting a format called the Study Data Tabulation Model (SDTM), which was developed by the Clinical Data Interchange Standards Consortium (CDISC), a recognized group involving more than 40 pharmaceutical companies and clinical research organizations.

The plain truth is that FDA reviewers spend far too much valuable time simply reorganizing large amounts of data submitted in varying formats. Having this clinical trial data presented in a standard structure will improve FDA's ability to evaluate the data more quickly . SDTM will facilitate the automation of the largely paper-based clinical research process, which in turn will lead to greater efficiencies in industry and government-sponsored clinical research. This unified standard will also enhance data integration opportunities, thereby helping to reduce data management barriers and costs for sharing the latest clinical trial information.

FDA is currently exploring regulatory approaches to require the use of the STDM standard for regulatory submissions. Most immediately, the standard will be added to other specifications listed under the draft guidance "Providing Regulatory Submissions in Electronic Format--Human Pharmaceutical Applications and Related Submissions." We are also investigating the possibility of expanding these standards for our other regulated products.

On a related note, our interagency taskforce is working on building a web-based system for clinical investigator and financial disclosure reporting with electronic signatures. This is just another step FDA and NCI are taking to leverage the latest technology to streamline internal processes and better enable innovation.

A second area of collaboration involves developing biomarkers for evaluating new cancer medicines. We are working to develop a standard approach for evaluating biomarkers that demonstrate a drug's clinical effectiveness and that can potentially serve in clinical trials as surrogate endpoints.

Our collaboration on biomarkers includes joint NCI/FDA publications as well as various workshops focused on translational research, surrogate endpoints, biomarkers for early cancer detection, and molecular profile guided therapeutic trials, where staff from both NCI and FDA can confer with academic and industry peers.

We see particular promise for the use of clinical proteomics for biomarkers -- namely, in the discovery of protein markers in the blood that can be used to detect and monitor disease course and drug response. NCI and FDA are currently discussing reference standards that will assist in the area of proteomic and genomic technologies. We are also working to advance the use of imaging as a valid biomarker.

Part of our work on biomarkers involves clarifying clinical endpoints for specific cancers. To this end, we've been conducting joint workshops and meetings for several major disease groups -- including lung cancer, colon cancer, prostate cancer, breast cancer and myeloma -- to identify the key issues that need to be addressed in each disease and develop general guidance on endpoints that can be used in different trial designs.

Finally, I'd like to mention some of the ways we're improving oncology drug development through better knowledge-sharing and process improvements.

FDA and NCI are embarking on new Cancer Fellowship Training Programs aimed at developing a corps of physicians and scientists who are expert in clinical research, the regulatory approval process, and the translation of research breakthroughs to clinical practice. We believe these programs will help inform and harmonize all phases of cancer drug discovery, development, and regulatory review for the benefit of cancer patients.

Additionally, we're conducting a systematic review of current policies to identify other ways in which FDA-NCI collaborations can enhance the development and regulatory process for cancer technologies. This includes developing new standard operating procedures for issue resolution and developing a more integrated review process to speed the initiation of Phase III trials.

Just this past January, FDA published guidance on IND exemptions for studies using approved drugs. Later this year, FDA intends to issue guidance on ways to perform early proof-of-mechanism, screening, and microdose human studies in a safe, efficient manner. This guidance will help academic researchers and developers identify the most promising drug candidates to bring forward in development -- a key ingredient to speeding innovation. FDA also intends to issue guidance for small scale production of investigational products to delineate the controls and recordkeeping required for laboratory-based manufacturing.

FDA has also undertaken some important organizational changes to ensure the most efficient and consistent review of cancer products. Last month, we announced plans to create a new oncology office and supporting program in the Center for Drug Evaluation and Research (CDER). The new office will be a consolidation of three existing areas within CDER responsible for the review of drugs and therapeutic biologics used to diagnose, treat, and prevent cancer. This office will also include drugs and certain therapeutic biologics used in medical imaging, many of which are used to detect, treat, or monitor cancer. The supporting Oncology Program is designed to facilitate cross-agency expert consultation, provide a forum to discuss and develop regulatory policy and standards, and serve as a focal point for agency interaction and collaboration with important outside stakeholders.

We're excited about these internal changes, and we think they will go a long way towards improving the quality of clinical research and innovation in the U.S.

Improving Clinical Trials through Better Information
The quality of clinical research depends not only on the research process itself, but also on the human volunteers who participate in clinical trials.

Recognizing the value of better clinical trial participation, FDA has worked to link patients to medical research through a variety of education programs and information tools. The Agency has collaborated with government and community partners to increase clinical trial participation among women and minorities. In addition, FDA has encouraged the use of important resources like ClinicalTrials.gov.

ClinicalTrials.gov is the world's largest source of clinical trial information -- currently listing approximately 11,400 federally and privately sponsored trials throughout the United States and abroad. ClinicalTrials.gov offers up-to-date information on clinical trials for a wide range of diseases and conditions. Studies listed in the database are conducted in all 50 States and in more than 90 countries worldwide.

This site has proven a very useful and trusted resource for patients facing life-threatening illnesses and seeking information on clinical trials for their medical conditions. Specifically, it provides information about a trial's purpose, who may participate, locations, and phone numbers for more details. Users of the site can also access links to other online health resources, such as MedlinePlus and PubMed. ClinicalTrials.gov currently receives more than 2.5 million page views per month and approximately 16,000 visitors daily.

Posting of Trial Data
The posting of key clinical trial information online raises the related question of what is or isn't appropriate to post online.

I know you're all aware of the recent media focus on the publication of negative trial results, or lack thereof. The American Medical Association and others have backed the drive for mandatory disclosure of all clinical trials, voicing concerns that trial sponsors tend to flag positive test findings while playing down those with negative or inconclusive results.

Obviously this is a difficult issue when it comes to the question of confidential commercial information. However, the underreporting of negative trial data is both alarming and detrimental to the public health. The complete flow of information regarding clinical trials is essential to maintaining public confidence in the system.

It is also vital to advancing scientific knowledge; publication bias only serves to impede the flow of information.

And so, I am encouraged by recent statements by the pharmaceutical industry recognizing the need for full disclosure of clinical trial results. This is not an issue that will be resolved easily, however with this recent momentum, we can hopefully start to see a more balanced flow of information from all clinical study sponsors. This better information is key to improving clinical research and advancing the public health.

As I said from the outset, the clinical research enterprise faces a number of key challenges today. However, thanks to new collaborations across the public and private sector, we are taking important steps to clarify the road to innovation, and to educate researchers and regulators, health care practitioners and patients on ways to improve clinical research.

I'd like to thank you again for all that you are doing to support biomedical innovation and help bridge the gap between bench and bedside.

We are, together, right now, in the right place at the right time to make a real difference to America's public health in the 21st century.

Thank you.