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

Combination Products

  • Print
  • Share
  • E-mail

Section Contents Menu

Meetings, Conferences, & Workshops

July 8, 2003 - Innovative Systems for Delivery of Drugs and Biologics: Scientific, Clinical and Regulatory Challenges: Summary of FDA Workshop

July 8, 2003

Bethesda Marriott

Bethesda, Maryland

Prepared by the

FDA Intercenter working group on
Innovative Systems for Delivery of Drugs and Biologics

Miriam C. Provost, Ph.D., CDRH

Patricia Y. Love, MD, MBA, OC/OCP

Tushar Kokate, Ph.D., CDER

Richard P. Felten, CDRH

Warren Rumble, CDER

Vicki Babb, Pharm.D., CDRH

Nancy S. Chang, M.D., CDER

Sheryl Lard-Whiteford, Ph.D., CBER

Susan Zhou, Ph.D., CDER

Daniel A. Shames, M.D., CDER

Joyce L. Frey, Ph.D., CBER


Innovative and novel medical technologies will have the potential to greatly improve the public health in many different areas.  FDA believes that by addressing and clarifying regulatory uncertainty, development of new technologies will be expedited and the predictability in product development will be increased.  This should allow more of these products to reach the marketplace in a timely manner.  As part of a broad effort to increase the development of novel medical technologies[1], FDA recently conducted a public workshop entitled, "Innovative Systems for Delivery of Drugs and Biologics:  Scientific, Clinical and Regulatory Challenges".  The workshop was held in July 2003 in Bethesda, Marriott.  Speakers included outstanding researchers and product developers from academia and industry, as well as regulatory authorities and experts from the FDA and outside government.  The workshop provided a forum for the academic, clinical and industry communities and the FDA to discuss the latest scientific and clinical developments for drug and biologic delivery systems, as well as the perceived regulatory concerns and challenges.  This article provides a summary of the ideas and information presented by speakers, as well as the issues raised by participants.  A complete transcript of the workshop is available at http://www.fda.gov/ohrms/dockets/dockets/03n0203/03n0203.htm (updated posting available at http://www.regulations.gov/fdmspublic/component/main?main=DocketDetail&d=FDA-2003-N-0051).  A copy of the agenda for the workshop is included in the Appendix to this document.

The workshop broadly defined "drug and biologic delivery systems".  The intent was to include any products that combine drugs and medical devices or biologics and medical devices in a system in which the two components work together to produce the desired outcome for the patient.

The FDA asked speakers and participants to address these specific questions:

1. What are the most critical challenges in developing and bringing to market a novel, innovative technology for delivery of drugs or biologics, especially when there is not a well-established regulatory pathway?

2. Which areas are most important for the agency to provide guidance to developers of these novel products?

3.   How can the Agency best collaborate with industry, academia, other government agencies and other scientific bodies in this area of rapidly evolving technology?

The Agency hopes to use the information from the workshop in the future development of guidance documents, memoranda of understanding or other position papers, as well as to provide information on future scientific training needs for Agency staff.

Dr. David Feigal, Director of the Center for Devices and Radiological Health, welcomed all of the participants.  He gave an overview of the purpose of the workshop and emphasized the challenges posed by products that utilize innovative technologies and represent a combination of medical devices and drugs or biologics.  He indicated that the FDA views the regulatory process as an intensely scientific one.  Therefore, he stated that it was appropriate to begin the session with an overview of some of the exciting scientific developments in important disease areas.  He then introduced the next speaker,

Dr. Mark B. McClellan, Commissioner of the FDA, who presented the opening remarks.  Dr. McClellan explained that the workshop was the first of a series of such efforts, with the aim of focusing on important questions for emerging technologies.  Part of his strategic plan for the FDA is to develop clear guidance and regulatory pathways for developers of innovative medical technologies.  He expressed concern about the challenges facing product developers today.  For example, it is estimated that the cost of developing a new drug is $800 million.  Although he acknowledged that this figure may be somewhat controversial, it is clear that costs have gone up, in large part as a result of extensive preclinical development and testing.  There is also a great deal of uncertainty in drug development, with only a small fraction of candidates making it through Phase II testing resulting in an application to the FDA.  Dr. McClellan expressed concern about the decline in the number of NDAs submitted to the Agency.  Investment in research and development (both private and government funded) is at an all-time high but these R&D expenditures do not appear to have translated into new products reaching the market place.  Dr. McClellan stressed that FDA will maintain its gold standard for the world in medical product approvals and that the Agency will do what it can to make the development process more efficient.  As part of the strategic plan, the FDA is undertaking a number of initiatives aimed at increasing the efficiency of review, including undertaking an internal root cause analysis for determining the cause for multi-cycle reviews, development of more guidance documents, especially in critical areas, and instituting a quality system for product reviews.

Specifically, with regard to combination products (such as drug and biologic delivery systems), Dr. McClellan acknowledged that in the past, the review process has not always gone smoothly.  To address some of these concerns, he instituted the Office of Combination Products (OCP), headed by Mr. Mark Kramer.  The OCP will provide better oversight of combination product reviews and will serve as a focal point for the development of guidance documents about jurisdictional issues and other concerns unique to combination products.  Dr. McClellan also cited implementation of the Medical Device User Fee and Modernization Act (MDUFMA)[2] as an important factor in providing FDA with the resources needed to improve review times and allow for guidance development and early conversations with industry and other stakeholders.  He summarized his talk by stating that this is an important time at the Agency for the developers of drug and biologic delivery systems.  The Agency has new resources, new programs in place and strong commitment from all three of the medical product centers (Center for Drug Evaluation and Research, Center for Biologics Evaluation and Research and the Center for Devices and Radiological Health).  FDA is in a great position to help with the innovation process, with the goal of getting safe and effective products to market for the benefit of the American people.

Keynote Address:  Dr. Robert Langer

The FDA was extremely fortunate to have as the Keynote speaker for the conference, Robert Langer, Sc.D., Kenneth J. Germeshausen Professor of Chemical and Biomedical  Engineering, Massachusetts Institute of Technology.  The title of Professor Langer’s talk was "Innovations in Drug Delivery" and was based on some of his recent publications.[3],[4],[5]

Professor Langer focused his talk for the workshop on four areas: the need for new materials for drug delivery, the role of nanotechnology, the potential for non-invasive delivery of drugs and biologics and the potential for high throughput technology. 

With regard to materials, he explained that the historical use of materials was driven by the familiarity of the users with certain material types.  Professor Langer espouses a scientific approach to materials development – i.e., developers should determine what is needed from a material from a chemical and biological standpoint and then should try to synthesize such a material from first principles.  An example given was the development of Gliadel, a controlled release drug product for the treatment of malignant glioblastomas.  In this product, the material comprising the implanted wafer was specifically synthesized for local controlled release of the chemotherapy agent.  This highly toxic drug could thus be delivered locally, only at places where it is needed, thus minimizing systemic toxicity.

Professor Langer next discussed nanotechnology and the potential that it holds for drug and biologic delivery systems.  He discussed the example of MicroCHIPS, made out of a biomaterial with a gold covering.  These chips contain any number of wells, potentially holding one or many different kinds of drugs.  No drug is released from the chip until current is applied.  The current disintegrates the gold covering, releasing the drug.  The current could be applied externally and the device could be designed to deliver drugs in any combination desired, a "pharmacy on a chip."  Professor Langer also envisioned these types of devices incorporating biosensors, thus allowing for closed loop control of drug dosing and delivery.  According to Professor Langer, nanotechnology might also be used to create "nano-sized" medicines such as polymers, which could be used for DNA delivery, avoiding the use of viral vectors and potentially improving the safety of gene therapy.

Professor Langer explained that non-invasive delivery of complex molecules has long been a goal of product development in drug delivery.  Carrier molecules have been developed, such as Emispheres that permit oral delivery of complex molecules.  However, there are issues with oral delivery, including bioavailability and safety.  The transdermal approach is also an area of significant interest, with 10 products currently on the market.  It is difficult to transport drugs through the skin, given the resistance of the various layers.  Developers are continually looking for ways to safely disrupt the layers including iontopheresis as well as ultrasound energy.  Professor Langer specifically mentioned a device called the Sonoprep that uses ultrasound energy to aid in transdermal drug delivery.  Such a system could also be used for obtaining samples of interstitial fluid.

Lastly, Professor Langer described some of the advances in high throughput technology.  He used the example of gene therapy, and the fact that a better and safer delivery vehicle is needed.  With high throughput technology, a developer could use a robot to make thousands of structurally diverse polymers and then screen them to see which ones are appropriate for delivery of DNA.  He gave the example of the experience of one company, Transform Pharmaceuticals, that used a high throughput system to do 10,000 experiments in 3 iterations in 6 weeks for acetaminophen.  The end result was that they discovered a new crystal structure for acetaminophen. 

Professor Langer concluded his talk by stating that he has enjoyed working in drug delivery for nearly 30 years and that it has been wonderful to see the progress being made to the point where many therapies are marketed and many others are in development that have the potential to relieve suffering and prolong life for many people.

Promising Clinical applications

Following the keynote address, the first session of the day focused on promising applications of drug and biologic delivery systems in specific clinical areas.

The first speaker was David C. Klonoff, M.D., FACP, Medical Director, Mills-Penninsula Diabetes Research Institute and Clinical Professor of Medicine, U.C. San Francisco, who spoke on new technologies for treatment of diabetes.  His talk was entitled, "Novel Technologies for the Treatment of Diabetes:  Metabolic Monitoring, the Artificial Pancreas and Alternative Routes for Administering Insulin."

For patients with diabetes, repeated insulin injections are inconvenient, pose the risk of hypoglycemia and also cause pain and trauma.  This highlights the need for new, non-invasive or minimally invasive technologies for monitoring blood glucose levels and delivering insulin.  With regard to blood glucose monitoring, Dr. Klonoff described the current technologies that are requiring less blood to get a reading in less time with better performance and reliability.  Current monitors require 1/10 of the blood that was needed 10 years ago and provide the reading in 5 seconds (1/10th of the time needed 10 years ago).  Current systems can store as many as 3000 readings.  Sampling at alternative sites (not just the finger tip) is also possible, as is sampling of other fluids besides blood, such as interstitial fluid.  Clearly, significant technological developments have been made.

Dr. Klonoff envisions further developments in non-invasive and minimally invasive methods for blood glucose monitoring.  For example, in minimally invasive monitoring, technology is being developed in which the skin barrier is disrupted but the probe does not actually contact a blood vessel.  Specific technologies under consideration include microporation, ultrasound, abrading the skin surface by chemicals or mechanical force or by using fine needles.  Such techniques also have the advantage of permitting continuous monitoring.  Four different products have been developed for continuous glucose monitoring, with 2 of them on the market in the US.  Neither product is intended to replace standard blood glucose monitoring and instead is to be used for evaluating trends or for indicating when a standard blood test is needed.  Other technologies on the horizon include the use of microdialysis and radiofrequency impedence.

Dr. Klonoff discussed the potential impact of integrated systems, in which monitoring and insulin delivery are linked (the so-called "artificial pancreas").  The ideal artificial pancreas would include a continuous sensor, a pump for insulin delivery, a processor to link the sensor and the pump and a radio link to an external monitor/activator.  Dr. Klonoff described some of the challenges faced by developers of these products.  The sensor would be subjected to calibration drift, fouling, immune rejection and fibrosis.  The lag between blood glucose and interstitial fluid glucose levels would have to be addressed.  Insulin storage in the body would be a problem, since insulin denatures at body temperature.  There could also be local complications of such a system, including systemic and surgical complications.  Finally, the system would have to be designed to avoid hypoglycemia.  One approach is to keep the system operating at a "high" glucose level, but keeping the glucose levels in the normal range is an important reason for using such a system.  These challenges must be addressed before an "artificial pancreas" device could become a reality.

Dr. Klonoff  also discussed "alternative" routes for insulin administration, including inhaled and oral insulin.  Inhaled insulin, in particular, offers the advantage of rapid delivery with long lasting effects.  New technologies for inhalers and new protective coatings for insulin, described by Dr. Klonoff, should help ensure proper delivery of insulin to the alveoli.  With regard to oral delivery, Dr. Klonoff described the challenge of avoiding acidic degradation in the stomach and enzymatic degradation in the intestines.  Buccal and nasal delivery and transdermal delivery systems are under development and testing.

Dr. Klonoff concluded his presentation by stating improved metabolic monitoring now permits improved bolus dosing.  Continuous monitoring will allow improved basal dose adjustments.  Closed loop artificial and bioartifical pancreas systems are under development and new routes of administration will remove barriers to the use of insulin.  The end result will be better treatment for patients with diabetes.

The next speaker in the clinical session was Jonathan Kruskal, M.D., Ph.D., Director, Abdominal Imaging, Beth Israel Deaconess Medical Center and Associate Professor of Radiology, Harvard Medical School, who discussed "Emerging Techniques and Technologies for Treatment of Solid Tumors."  As an interventional radiologist, Dr. Kruskal is involved in developing and studying innovative methods for treatment of tumors.  One challenge faced by researchers in this field is how to deliver drugs or genes to a tumor.  Delivery is needed deep within the body. In addition, there are challenges in limiting efflux from the tumor, so that the drug or biologic stays in contact with the tumor long enough to ensure effectiveness.

New catheters and delivery systems are being developed to enhance delivery of chemotherapeutic drugs directly to tumors without blocking or destroying blood vessels. Immunoconjugates of monoclonal antibodies are another technique under study, in which the antibody-drug complex binds to tumor cells expressing specific antigens. 

Assuming that the tumor can be accessed with a catheter or other device, the next challenge lies in getting the drug or biologic inside the tumor.  Dr. Kruskal discussed several innovative technologies for this purpose, including radiofrequency ablation, ultrasound and UV light, that are intended to enhance the uptake and internalization of the drug.

The use of growth factor receptors, for both imaging and treatment, was also discussed.  Probes and imaging can be used to determine where the receptors are and to establish that they are being expressed.  With this information, receptors can then be blocked with new factors.  The example was given of delivering vascular endothelial growth factor (VEGF) to tumors to make them "leaky".  VEGF could therefore be combined with drugs to enhance their effectiveness. 

In the area of cell transplantation, Dr. Kruskal referred to successful studies showing transplantation of liver cells to the spleen.  This opens up new possibilities for ectopic location of functioning cells.  In terms of an oncology treatment, this may allow injection of fibroblasts and dendritic cells into the peritoneum that could have been transformed to produce proteins, such as human growth factors.

Dr. Kruskal described many other possibilities for combined drug/device or biologic/device treatment of solid tumors.  For example, studies have shown that following radiofrequency ablation, macrophages are recruited.  If the macrophages could be transformed to carry drugs or cells, this could be used for targeting.  Magnetic targeting is also a possibility.  Drugs could be bound to magnetic beads and an external magnetic field could be used to focus the beads at the tumor.  Another example of cooperative therapies is the insertion of P53 genes into tumors.  This could not only affect cell death but also permit the use of lower radiation doses.  The use of ultrasound or sonoporation (for tumor penetration) and thermally activated vectors was also discussed.

Dr. Kruskal concluded by stating that these exciting emerging techniques and technologies need to be optimized.  Good randomized, controlled trials are needed to establish safety and effectiveness of these new technologies and to provide a basis for comparison to standard treatments.

The final speaker in the clinical session was Richard Kuntz, M.D., M.Sc., Associate Professor, Harvard Medical School, Brigham and Women’s Hospital, who spoke on the drug eluting stent, an innovative drug/device combination product that was recently approved by the FDA.  Dr. Kuntz explained that bare metal stents were initially tested in "simple" patients, non-diabetics with relatively large vessels.  They were considered a breakthrough technology and quickly became the standard of care.  After approval, the use of bare metal stents was expanded to patients with diabetes and/or patients with longer lesions and small vessels.  This led to an increased restenosis rate over that observed in the trials with the "simple" patients.  Dr. Kuntz explained that this is the typical cycle of new technology.  When new technology is first introduced, good success is shown until clinicians begin to broaden the patient population to such a point that it begins to fail again.  With regard to prevention of restenosis following stenting, systemic drugs have been tried but have failed, potentially because not enough drug was delivered to the tissue site where it was needed.   Addressing this problem requires not only an effective drug but also effective and biocompatible polymers that can retain and elute the drug as needed.   For the drug eluting stent, identifying the appropriate biopolymer was the biggest challenge.  Many of the initial polymers that were tried were themselves toxic and caused vascular responses.  The approved drug eluting stent contains a polymer that retains the drug (sirolimus)and releases it slowly, over 30 days.  The device is indicated for improving coronary luminal diameter in patients with symptomatic ischemic disease due to discrete de novo lesions as defined in the product labeling.

Dr. Kuntz summarized the recently reported clinical trials of the first drug eluting stent approved by the FDA, the Cypher stent manufactured by Cordis.  There were two clinical studies performed with the stent.  The first study, called RAVEL, used a surrogate endpoint of restenosis.  It was a small study of only 200 patients and it showed a 26% restenosis in the control (bare metal stent) group, with a 0% restenosis rate in the study (Cypher stent) group.  The US study, called SIRIUS, was a larger study that used a clinical endpoint of target vessel revascularization.  The results showed that the use of the drug eluting stent resulted in a four fold reduction in target lesion revascularization.  Dr. Kuntz stated that there was a low incidence of adverse events and almost all subgroups benefited to some degree. 

Dr. Kuntz next considered the use of other drugs and asked the question of whether all drug/polymer combinations would be as effective.  He reiterated that the polymer technology is critical and that, with the right polymer, older drugs that were thought not to be effective in preventing restenosis, might be effective if combined with a polymer and delivered via a stent.  Other stent designs are also possible, including those with small holes or "wells" for drugs and polymers.  He also discussed loading a stent with a variety of drugs having different release kinetics and potentially different pharmaceutical effects.

Dr. Kuntz concluded by describing potential new applications for drug eluting stents, considering their demonstrated effectiveness.  For example, considering their effectiveness, drug eluting stents may potentially become a replacement for coronary artery bypass grafting for patients with multivessel disease.  Or, placement of drug eluting stents may be used to prevent rupture of vulnerable plaque and subsequent acute myocardial infarcation.  Dr. Kuntz concluded by expressing a positive outlook for this drug/device technology for the treatment of patients with cardiovascular disease.

Preclinical Challenges

The next session was devoted to a discussion of the challenges posed by novel drug and biologic delivery systems in preclinical development and testing.  Speakers addressed the topic from different perspectives, including those of the drug industry, the medical device industry and from the viewpoint of the FDA.  Preclinical studies are necessary before a novel drug or biologic delivery system could be used in clinical trials.  Thus, the development of an appropriate preclinical testing program is a critical step on the path to market availability of these novel products.

The first speaker was Chet Leach, Ph.D., DABT, Director, Life Science Research and Development, Nektar Therapeutics, who spoke on "Preclinical Development Considerations for Pulmonary Delivery of Drugs approved for Other Routes of Administration."  This was a very appropriate topic, since new device technologies often permit more effective or more acceptable delivery of "older" drugs.  The drug eluting stent is one example.  Dr. Leach discussed several examples in which well known drugs were reformulated to permit pulmonary delivery using novel device systems.

In developing such a product, Dr. Leach pointed out several factors to consider.  First, the developer should consider whether the drug has ever been delivered via the pulmonary route and second, they should determine whether the expected metabolic pathways are present.  In the lung, there are less metabolic pathways present than in other areas of the body, such as the liver.  Developers must determine whether the lung tissue has the enzymes present to metabolize the drug of interest to its active form.  Another consideration is whether there are new cell types that are more susceptible to side effects in the new area of administration compared to the previously approved route of administration.  For example, insulin is a growth factor and it raises a question as to what its effects might be on lung tissue.  The presence of excipients is also an important issue in pulmonary delivery.  Some excipients, which are benign when present in drugs administered orally, cause bronchospasm in asthmatics or normal individuals. The presence of membrane disrupters, often used in drug formulation, is also a big problem for pulmonary delivery.  Finally, developers must consider the question of whether antibodies will form to peptides or proteins.

Dr. Leach described preclinical testing programs for some specific example products.  The first case was an approved drug (Proventil) with the same particle size that was to be delivered using a new CFC-free propellant.  The preclinical development program included safety pharmacology studies and other studies designed to answer targeted questions.  The next example was a product with the same drug, with a different propellant, different dose, and different particle size.  Because of the differences, the drug had better dosing characteristics (more of the drug went to the lung as opposed to being lost in the mouth) and the drug was also delivered to different places in the lung.  The latter finding raised some preclinical safety issues, since previously the drug was not delivered to all of the airways as well as the alveoli.  For this product, the developers conducted range finding studies, 14-day studies and a 12 month inhalation study.  The 12 month study was deemed necessary because there was concern that the drug delivery could cause endocrine disruption.

Dr. Leach repeatedly stressed the need for a connection between the preclinical and the clinical programs.  If clinical results are found in the early phases that need further examination, additional preclinical studies may be warranted.  The findings from the additional preclinical studies may then impact upon the design of future clinical studies.

Dr. Leach closed his talk with a discussion of some of the new developments in pulmonary delivery of proteins, peptides and insulin.  There are a number of exciting developments in the area of therapeutic proteins and peptides but they pose challenges in delivery since they have to be injected.  Pulmonary delivery offers an attractive alternative.  One specific example is the possibility of pulmonary delivery of antibiotics for treatment of lung infections.  When antibiotics are administered orally or through IV, the actual amount delivered to the lung tissue is low.  In contrast, Dr. Leach showed animal data demonstrating a four orders of magnitude greater antibiotic delivery to the lung tissue when the antibiotic is delivered by pulmonary route.  This presents an attractive opportunity for further design of products for pulmonary delivery.

Inhalation offers a number of other advantages, as outlined by Dr. Leach.  Pulmonary delivery may offer, in addition to higher bioavailability, a faster time for onset, fewer side effects and freedom from injection for the patients.  He closed by stressing that the preclinical development program needs to be tailored to meet the scientific and regulatory questions posed by each new product.  The preclinical program should focus on the exploration of known differences, not unsubstantiated speculation and it should not focus on the types of testing currently "in vogue."  Dr. Leach closed by stating that the fear of the unknown, and real or perceived burdens in preclinical and clinical testing requirements, keeps many new drug developments from happening, especially for drugs in the non-blockbuster category.

The next presentation was given by Bill Van Antwerp, Ph.D., Vice President and Chief Scientific Officer, Medtronic Minimed, whose talk was entitled, "Protein Delivery from Implantable Devices:  Challenges and Opportunities", and he gave the device industry perspective on systems for protein delivery.  He opened his presentation by explaining the new developments in therapeutic proteins for a variety of disease conditions, such as cancer, cardiovascular disease, inflammation, HIV/AIDS and Hepatitis C.  However, proteins pose a number of challenges from a drug delivery perspective:  they are not highly bioavailable, they become denatured, they hydrolyze, they are susceptible to enzymatic degradation and they are poorly absorbed, due to their size or polarity.  Dr. Van Antwerp explained that Medtronic Minimed focuses on delivering such drugs "the old fashioned way" using a subcutaneous or intraperitoneal infusion via a mechanical pump. 

According to Dr. Van Antwerp, the main challenge in developing proteins for delivery via a mechanical pump is "formulation, formulation, formulation."  The usual challenges of formulation stability, chemical stability and clearance issues in the body are compounded by new challenges for delivery via a pump. 

One new challenge is physical stability of the protein as it is delivered by the pump.  Another concern is pharmacokinetics/pharmacodynamics since the proteins are now being delivered continuously.  He described an example in which a dose escalation study turned into a dose de-escalation study since the continuous delivery of drug was much more effective.  Toxicity must be viewed differently.  If the formulation is not right, it is possible to observe localized site reactions.  If, for example, the product is an injection catheter in the subcutaneous tissues that is supposed to deliver drugs for three days, the formulation of the drug must be suitable for 3 day delivery.

Regulatory hurdles are also present, as described by Dr. Van Antwerp.  Drug pump effectiveness has to be reproven for each new drug, despite the fact that nothing about the pump (in its design or operation) has changed.  Drug chemistry and drug packaging also has to be restudied and reproven for each new application.  This can cause delays.

Dr. Van Antwerp described two areas of important focus.  One is pump-drug interactions and drug physical stability.  Dr. Van Antwerp stated, "When God invented insulin, she didn’t design it to be stable for 90 days at body temperature sloshing around in a metal can."  Stability in pumps includes physical stability as well as chemical stability.  Lack of physical stability can lead to soluble aggregates, which in turn can lead to the formation of antibodies, something that is not observed with noncontinous infusion.  A wide variety of measurements are used to assess physical stability.   Factors affecting physical stability include the physics of the device (e.g., presence of sheer), the materials contacting the protein and the presence of agitation.  He described a model developed, in part, by Fink, et al. at U.C. Santa Cruz[6], that describes how proteins become unstable in these systems.  It begins with absorption of the protein onto a surface, followed by denaturation (partial unfolding).  These aggregate on the material surface or fall back into solution and form more aggregates.  Dr. Van Antwerp described a test developed in his laboratory for assessing physical stability in proteins.  This test allows him to look at protein stability in contact with a variety of materials.  When designing pumps, it is important to assess a priori whether the pump materials will lead to protein aggregration.  He described an example in which insulin was contacted with a variety of materials, including Teflon, polyethylene, glass and titanium in order to assess the effects of these materials on the insulin stability.  The results show that Teflon exposure made the insulin susceptible to aggregation, while the glass and titanium did not.  Another example was given of a drug formulated two different ways.  The end result was two products with the exact same formulation based on chemical analysis but that had very different physical stability properties.  This result highlights the importance of assessing physical stability during drug development, if the drug is to be delivered using a pump or similar system. 

Dr. Van Antwerp closed with a discussion of an example drug/device combination in which the drug actually "broke" the device.  The device contained a titanium seal in the pump.  Insulin crystals formed on this seal, causing the material to become brittle and crack, leading to leakage of insulin and eventual failure of the device.  Dr. Van Antwerp emphasized the need for drug and device teams to work together and to consider the potential materials interactions.

The next speaker in this session was Kevin Skinner, V.M.D., Associate Director of Science, Biomaterial and Surgical Products, Genzyme Corporation. Dr. Skinner’s talk was entitled, "Developing a Local Delivery Combination Product for Postoperative Atrial fibrillation: Preclinical Progress and Challenges".

Postoperative atrial fibrillation is a kind of tachycardia that is seen in 20-30% of patients following CABG surgery. There are prophylactic treatments for this condition, but they are not widely accepted. Amiodarone, a class III antiarrhythmic agent is often used off-label for treating postoperative fibrillation.  Amiodarone can be given orally or by injection.  Both routes may be associated with severe pulmonary toxicity.  Intravenous doses may be associated with bradycardia.  Also, oral doses  require at least a 7 day loading period.  adverse effects. Therefore, in order to decrease the systemic effects, this drug was a good candidate for studying local delivery to the pericardial sac. For this purpose, Genzyme used bioreabsorbable polyethylene glycol (PEG)  based hydrogel to deliver Amiodarone locally. This tissue adherent hydrogel is approved by FDA as a lung sealant for pulmonary leaks. It is compatible with drugs and biologics.  

Before proceeding with this study further, Genzyme made sure that there was no drug-device interference. Up to 5% of Amiodarone could be loaded into the gel without affecting the in situ polymerization properties of the hydrogel. In vitro release tests showed that up to 2-3 weeks of drug could be delivered out of the hydrogel. This Amiodarone loaded PEG-based hydrogel (0.5% & 1%: 16 & 32 mg, respectively) was implanted onto the canine heart and levels of Amiodarone were determined in the cardiac and other tissues 7 days post-implantation. About 30% of the drug was eluted into the cardiac tissue within 7 days and no measurable drug levels in the lung, liver or kidneys were observed. The drug-gel combination was well tolerated and no adverse effects were noted. The subsequent efficacy study in dogs involving atrial refractoriness measures showed >50% lengthening of effective refractory period (ERP) over critical 3-5 days post-implantation and the effect lasted up to 3-6 weeks. The treated atria had drug level (›20 µg/g) in a desired therapeutic range over a 3 week period. Thus, the pre-clinical study showed that the amount of Amiodarone needed for therapeutic effectiveness using targeted local delivery of drug is much lower than oral or intravenous routes (mg instead of grams) resulting in a better adverse effect profile. Furthermore, the drug needs to be delivered only once over a three-week period instead of daily injections or oral administration.

Dr. Skinner concluded by saying that the combination product of Amiodarone and PEG-based hydrogel showed promising results in pre-clinical models in terms of safety and efficacy of the drug. Pre-clinical bridging studies looking at the long-term degradation of the product and toxicity issues of placing a biomaterial on the heart are planned.

The last speaker in the pre-clinical session was Ajaz Hussain, Ph.D., Deputy Director, Office of Pharmaceutical Sciences, Center for Drug Evaluation and Research, FDA. Dr. Hussain’s talk was entitled, "Perspective on quality by design: A Preclinical Opportunity".

Dr. Hussain noted that pharmaceutical manufacturing has evolved from old days of compounding to more science and engineering based.  The industry is moving from dosage forms to targeted drug delivery systems. In terms of quality by design, it is important to consider the impact of quality on the efficiency of development and the time to bring the drug to market. Product quality and performance can be achieved and assured by design of effective and efficient manufacturing processes. 

There are physical and chemical aspects as well as test methodologies that need to be considered when developing a drug formulation that is bioavailable and stable. Chemistry and manufacturing controls (CMC)  and good manufacturing practices (GMP) issues are also very important to ensure consistent quality and performance. Designing and developing specifications for a given product, developing and establishing manufacturing processes and their controls, and developing process validation and test methods are some of the key challenges that need to be considered. The types of studies conducted during development can have significant impact on development time. Usually post-approval changes are not in minds of people who are focused on developing formulations and doing clinical studies. However, proactive thinking in terms of post-approval changes, in other words continuous improvement, is important in reducing the development time and bringing the drug to market.

Dr. Hussain defined the concept of quality by design as achievement of product and process performance characteristics that are adequate for their intended use through scientific understanding and management of sources of variation and other risk factors during manufacture.

Drug delivery systems in the current form are primarily based on pharmacokinetic and pharmacodynamic approaches with the objective of improving safety and efficacy for the patient.  Targeted drug delivery poses additional challenges and questions.  Traditional chemistry testing may not adequately assess pharmaceutical quality and consistency, as clearly explained by the previous speaker, Dr. Van Antwerp.  For example, there are many important questions to ask with regard to the pharmaceutical component of this system, such as how does the polymer coating affect the efficacy of the drug, what is the release rate of the drug from the stent, what is the optimal in vivo release profile, what factors influence the release profile, are there any localized effects, and what are the appropriate in vitro test methods for assessing quality assurance. These are the issues that should be considered in designing a product and during development. 

Dr. Hussain summarized by saying that pharmaceutical manufacturing is evolving    from an art form to one that is more science and engineering based. Product quality regulations should be founded on sound science and engineering principles for assessing and mitigating the risk of poor product and process quality in the context of the intended user of pharmaceutical products. Regulatory policies need to be tailored to recognize the level of scientific knowledge supporting product applications, process validation and process capability.

Industry Perspective

The afternoon began with a session devoted to a discussion of perspectives from industry on regulatory issues facing developers of novel drug and biologic delivery systems.  The first speaker was Jonathan S. Kahan,  Partner at Hogan & Hartson, LLP, who spoke on "Regulatory and legal challenges for the developers of drug delivery systems".  Mr. Kahan has many years of experience in working as a regulatory counsel to manufacturers and developers of combination products and he provided a perspective on several regulatory issues that industry has experienced. This perspective was framed by a legal construct, a summary of FDA’s historic approach to regulation of combination products and drug delivery devices, and the obstacles and challenges posed by new policies and procedures. While providing this regulatory framework, Mr. Kahan intertwined several examples of combination product development experiences as a backdrop for his discussion on whether the 20th Century FDA standards and regulations will be adequate to handle the 21st Century technologies. 

Based on the established definitions of a drug, device[7], and biologic[8], Mr. Kahan noted that under the Safe Medical Devices Act of 1990 enacted by Congress, FDA implemented regulations, under 21 CFR 3, to define a combination product and to implement a designation process to determine how a combination product would be assigned within FDA.  The statute requires that FDA base the assignment on the "primary mode of action" of the combination product, but the term was not further defined in the Safe Medical Devices Act and it is not defined in FDA regulations.   He noted that the FDA Intercenter agreements are useful and should be read, to get an idea of where a combination product might be regulated but he noted that they are out-of-date, outmoded and need to be revised. 

Mr. Kahan noted that current medical technologies are moving towards combination products that are not simple prefilled syringes, but may be combinations of a drug, biologic, and device.  He gave an example of a product, a dopaminergic cell encapsulated in a semi-permeable membrane that elutes dopamine.  This product therefore will include dopamine ( a drug), a dopaminergic cell (a biologic), and a semi-permeable polymer (a device).  Mr. Kahan believes that the Agency will be faced with many products like this in the future.  He divided combination products into first, second, and third generations, based on the technical and regulatory challenges posed by the technology.

  • First generation products include prefilled syringes, infusion pumps, metered dose inhalers, and transdermal patches.  He noted that the FDA had developed pathways for the development of such products.  For example, metered dose inhalers effect droplets and the size of the droplets of the drug, and therefore its efficacy.  These are regulated as drugs.

  • Second generation products include cigarettes, laser activated drugs, lollipops and chewing gums as drug delivery devices, drug-coated catheters and drug eluting stents.  The drug coated catheters and drug eluting stents have primarily been regulated in CDRH and the drug component is usually an already approved drug.  He recommended that the agency’s approach to regulating the drug eluting stent be used as a paradigm for other combination products. 

  •  Third generation products are those that he expects will pose a challenge for the FDA, requiring a great deal of creative thinking.  As an example, he referenced Dr. Langer’s presentation about the microCHIPSdevice that could have 100 wells.  He wondered if there were 20 different drugs in those 100 wells, would the agency require an NDA or NDA supplement for each drug, i.e., 20 NDAs or NDA supplements in all for one combination product?  Mr. Kahan noted that many devices are used with drugs that are already approved for some purpose. Device-drug combinations that incorporated drugs that are new, have a different indication, a different mode of delivery, or a different dosage schedule have presented a number of challenges.  First, to what extent does the modification of the drug formulation to optimize delivery with the device, mean that early phase 1 or phase 2 clinical trials must be repeated while demonstrating the safety and efficacy of the device? Is a new NDA required for the drug if you have a different delivery mechanism than the mechanism that was described in the NDA-approved label, e.g., the drug was approved for IV use but the device will be implanted? What aspects of the combination product predominate in the review process; e.g., the newness of the device, newness of the drug (e.g., the PMA process for novel MicroCHIPS vs. the NDA process)? Must the device labeling and the drug labeling be mutually conforming?

Mr. Kahan noted that without cooperation from the pharmaceutical company who manufactures the drug in question, the device companies have a very difficult time obtaining NDA approval.  He described the difference in motivating forces for a pharmaceutical firm.  For example, if the new device increased the use of a drug, then a pharmaceutical company might be likely to authorize cross reference to their proprietary data.  But, if the device made dosing more efficient, fewer drugs might be sold and that might be a disincentive for the drug company.  Mr. Kahan wondered if the Section 505(b)(2) approach to using literature would be useful in gaining approval of a new route of administration without a drug manufacturer’s involvement.

Mr. Kahan ended his presentation with several comments and recommendations.  First, he recommended that sponsors of combination products request a meeting with the agency before the investigational application is submitted (pre-IDE or a pre-IND).  Also, Mr. Kahan emphasized that the industry should start the interaction process early and communicate very well with their clinicians, engineers, and regulatory affairs staff. This will assist industry in their meetings and interaction with FDA and its centers (CDRH, CBER and CDER).   In his opinion, the review process will go more smoothly if one center is in the lead, as opposed to a collaborative review in which two or more centers share the lead responsibilities. He also stated that dual approvals (i.e., submission of two applications for one combination product) are not optimal in most cases.  Industry usually prefers one submission.

He recommended that the FDA define and update the regulatory meaning of "primary mode of action."  Guidance documents should be issued on product classes where decisions have been made to provide relevant and useful information to developers (e.g., drug eluting stents, nasal inhalation system).

He mentioned the concept of a uniform, unitary combination product approval mechanism and stated that this would be beneficial, i.e., a new type of regulatory submission that recognizes the unique regulatory issues posed by combination products.  He believes that the approach to labeling that the FDA used with the Cypher drug eluting stent should be a model for other combination products.  The mixture of the device and drug labels appropriately reflected the intended use of the device with the precautions and warnings such that the user of the product will have information that's appropriate for both the drug and the device side.

Mr. Kahan closed by stating that developing guidance documents and new approaches to the regulation of combination products is going to require a great deal of work by the Agency and a lot of coordination among the centers.  He sees the OCP as playing a key role in coordinating and helping to expedite combination product reviews.

The second speaker of this session was Keith Smith, Director of Regulatory Affairs from Becton-Dickinson who presented comments on behalf of AdvaMed (Advanced Medical Technology Association), representing more than 1,100 innovators and manufacturers of medical devices.

AdvaMed noted that the term novel drug delivery systems, as used in this conference, might inadvertently misdirect primary mode of action analysis and misdirect jurisdiction.  Several products such as orthopedic devices with biomaterials, hyperthermia/drug combinations, and drug eluting stents contain drugs or biologics, but were determined to have a device primary mode of action.  AdvaMed recommended reserving the term drug delivery systems for products whose primary intended use was to deliver a drug.

AdvaMed’s presentation focused on the pre-market complexities and uncertainties of innovative delivery systems and how the regulatory processes might be improved to further the Commissioner's goals. The information was provided in response to the three questions identified in the June 5th Federal Register[9]notice.

1.Identify current critical challenges in developing and bringing to the market novel delivery devices.

In addressing this question, AdvaMed identified four challenges.  The first challenge concerns determination of the primary mode of action (PMOA) of a combination product.  They recommend that, in determining the PMOA, the combined product as a whole should be analyzed and the mode of action should be determined based on the primary intended function of the combined product.  They believe that strong deference should be given to a company’s belief that device assignment would serve to foster and advance their technologies.  For a subset of combination products that deliver a drug, (e.g., new aerosolized insulin systems and lasers to deliver topical anesthetics) the agency should continue to use other principles.  For example, the inter-center agreement provides that CDRH will have jurisdiction in device-drug combinations when the novelty of the device technology predominates.  They believe that prior use of this approach has allowed new technology to progress and evolve quickly.

The second challenge concerns cross-labeling.  AdvaMed indicated that many of the previous FDA approaches to labeling facilitated approval of devices.  They noted that the approach to cross-labeling should be flexible and discussed not only during labeling negotiations, but also as early as during jurisdiction discussions.  AdvaMed recommends that if the device labeling is generally consistent with three key parameters of drug labeling, i.e., indication, general mode (route) of delivery, and the drug dosage or dosing schedule equivalence, then, the agency should grant CDRH jurisdiction for the product, and generally should waive additional clinical showing of drug effectiveness. To accommodate evolution of drugs after marketing, device labeling need only be generally consistent with the labeling of the drug intended to be delivered (e.g., continuous delivery devices for insulin are generally consistent with the drug route of administration).  AdvaMed believes that even if the device is not consistent in the indication, dose, rate, or route of administration, the differences should be resolved in the device labeling.

The third challenge involves evidentiary standards or authorities for drug delivery systems.  AdvaMed believes that the Agency needs to create new guidance documents that allow for more creative and flexible approaches to data development for this class of products, with the device authorities clearly and consistently applied for the device and/or the device component parts of these reviews.  AdvaMed noted that in combination products that have separate components, historically the review standards from different centers are different and may not be clear.  The jurisdiction of the lead center may determine what is needed.  They proposed a number of recommendations in this area.  First, delivery systems need not reprove the fundamental efficacy of a drug already approved for the same general mode of administration, dosage, and indication.  The FDA should provide concrete and specific guidance through examples as to how this principle can be more effectively and consistently applied.  Review standards should be set for the composite product.  CDRH product jurisdiction for a combination product therefore must necessarily involve device authorities.  When devices are used with drugs or biologics, the FDA should be clear that device authorities, including least burdensome principles, frame the review for the device aspect of the product.  This is particularly important for more complex device designs (e.g., software electronics, electromagnetic principles, ultrasound energy).  New guidance for novel delivery systems should reflect the Commissioner’s initiative goals and be consistent with the current law to promote creative thinking, development and marketing of the innovative systems.

The fourth challenge involves the requirement for "parallel review" of two marketing applications for a combination product.  AdvaMed noted that industry needs a better understanding and clarification of those circumstances where the parallel review path may or may not be appropriate. They referenced their comments made at the FDA November 25, 2002 public hearing[10]as background. They noted that one application should be required in the majority of the cases.  In some cases, sponsors may desire a separate application or a separate application may be appropriate, especially if the system components are expected to have separate distribution and use/reuse patterns, or if the primary jurisdiction for the combination delivery system has been given to the center other than CDRH and the delivery device component is capable of being separately defined and reviewed.  AdvaMed emphasized that use of dual filings should be at the option of the sponsor. 

2. What areas are most important for the Agency to provide guidance to developers of such novel products? And,3. How can the Agency best collaborate with industry, academia, other government agencies, and other scientific bodies in this area?

AdvaMed addressed these two questions together by making some specific recommendations.  Guidance documents should be developed by the Agency to address the four challenges outlined above.  These guidance documents should be developed using interactive good guidance practices.  They recommend the establishment of a database of submissions for novel delivery systems that is transparent and as informative as possible, consistent with FDA's nondisclosure obligations and the proprietary interests of sponsoring companies. Data on approved products should convey primary jurisdiction, time frame for reviews, available information on consultative or collaborative processesused, the number of review cycles involved, and public summaries for review.  This database should be separate and apart from other databases for approved products to facilitate industry's efficient review of combination precedents.

The next speaker was Christine Allison, MS, RAC, Associate Regulatory Consultant, Global Regulatory Affairs El Lilly and Company.  Ms. Allison discussed regulatory challenges from a pharmaceutical company perspective that reflected some of the experiences of her company during development and marketing applications, post-approval issues with of drug-device combination products, and  the challenges that they experienced working with partners and in global development.  The majority of the discussion focused on CMC issues.  The following recommendations were made in the context of pre-marking, marketing applications and post-market activities.


Ms. Allison noted that for innovative drug-device products, the informal discussions with the center ombudsman instead of a formal RFD resulted in inconsistent jurisdiction decisions. Moreover, generally industry questions occur very early, before they are ready to request lead center designation.  Novel product development is very time consuming, and needs frequent consultation with multiple FDA centers and sometimes multiple divisions within the same center.   Cross communication is difficult.  The traditional pre-IND meetings, or end of Phase II meetings are not sufficient.

When the FDA asks that the commercial product [that proposed for marketing] be used for pivotal studies, the chemistry, manufacturing and controls  "must be locked-in" very early.  For a novel product, that stops process development and design feedback.  Resources [unclear – resources as a word don’t match the verb occur], new equipment and building of new manufacturing sites must occur very early.  Also, it is not clear which manufacturing control regulations (e.g., GMP, QSR, or both) and which compliance guidance will be used during the pre-approval inspection.  Similarly, it is not clear which adverse event reporting requirements should be used during investigational development (e.g., 21 CFR 312 or 21 CFR 812).

Ms. Allison made a number of recommendations for the Agency.  First, before the lead center is designated, a single FDA focal point should be identified to receive early questions, and she recommended that the Office of Combination Products coordinate and facilitate the appropriate centers for technical consultation.  For informal jurisdiction discussions, an internal procedure should be established for each center to route those informal requests to the Office of Combination Products for review to ensure the consistency of lead center designation.  Also, at the time the lead center is assigned, the consulting center should be identified.

Second, there should be more flexibility in granting sponsor requests for meetings and consultations when developing novel combination products.  The FDA review team should include members from all relevant centers and divisions from the very first sponsor meeting.

Third, the FDA should consider strategies to allow product process improvements throughout development and commercialization.  FDA should provide clear, documented expectations on what is needed.  For example, would it be acceptable to use a successful, satisfactory comparability protocol for pre-commercial and commercial systems?

Fourth, the FDA should establish a clear policy for selecting which manufacturing controls will be followed, and and how pre-approval inspections will be conducted.  Ms. Allison recommended that the drug regulations should apply to the drug portions of the product, and the device regulations should apply to the device portions.  The investigators should be trained to perform combination product inspections using the appropriate regulation for each component of the combination. 

Finally, Ms. Allison indicated that there is uncertainty in which regulatory provisions apply for the reporting of adverse events that occur during the conduct of investigational clinical trials.  She recommended that both drug (21 CFR 312) and device (21 CFR 812) investigational reporting requirements would be applied, as appropriate.  For example, she recommended that if device unanticipated adverse events would be reportable under the device regulation, then the device adverse events would be submitted to the IND.  Ms. Allison recommended that device investigation results should be included in the report, and this report should be directed to the lead center doing the review.  The timing for submission of the report would not matter, but should be made clear.

Marketing applications:

The critical concern Ms. Allison expressed was the limited guidance in determining the number of applications to submit, the format of the application, and associated user fees.  For example, if dual applications are submitted, how will user fees apply?  If one application is submitted, how should device information be included in an NDA or in a CTD submission?

She believes that in most cases, a single pre-market review mechanism and a single approval of a combination product should be needed. She recommended that the Agency establish standardized submission formats and provide guidance on how to include device information in an NDA or CTD submission. She also recommended that FDA provide guidance on what data are needed for the device at each phase of the IND. FDA should clarify if the device submission should be in device format (e.g., 510(k)) for easier review by the consulting reviewers. 


Ms. Allison described several key challenges in the post-marketing of combination products related to the lack of agency guidance on 1) how to address post market changes for device components of combination products approved under an NDA, 2) which regulations apply to post-market adverse event reporting requirements (21 CFR 314 or 21 CFR 803), and 3) whether to include device investigation results in the expedited device reports.

For cross-labeling of products intended to be used together, often a drug may be used with several different devices, but the device is not cleared for specific drug products.  Guidance is needed on how the drug label can incorporate these devices and meet any conforming labeling requirements.

Ms. Allison noted that the pace of device evolution is generally faster than the drug review process, particularly since labeling submissions do not have set review times.   For some devices (e.g., 510(k) products), a new model is cleared before the drug review is completed. 

In working with other commercial partners, Ms. Allison noted that device companies may take their platform technology and license it to several drug manufacturers. Thus, the drug manufacturer may have limited knowledge of the device information and testing.  For example, if the platform technology is provided in confidential information in a DMF (device master file) the drug firm marketing application can not describe the information in detail.  Also, post market, the device manufacturer may make changes that are not known to the drug firm.

Global marketing of combination products adds additional requirements and complexity for industry document preparations, quality system requirements, post-approval changes, regulatory reporting, labeling, and compliance inspections.  She recommended that the Agency use the 510(k) decision tree as a guidance for  how and when a post-market change to a device component of a combination product approved under an NDA should be submitted to the agency.  For example, if the 510(k) decision tree result is that a 510(k) is not needed, then the device change would be submitted to CDER in the annual report.  If the 510(k) decision tree result is that a 510(k) is needed, then this would be submitted in a NDA Supplement for either major changes requiring prior approval or for moderate changes being effected in 30 days. 

She believes that device malfunction adverse reports should be reportable , and submitted to the NDA.  The device investigation results should be included in the report.  The device adverse event reports should be directed to the lead center that has reviewed the submission and approved the products.  Although there was no preference for the reporting time frame, the reporting time requirement should be made clear.

Ms. Allison recommended that the Agency consider international harmonization efforts underway, since many companies now market devices internationally, and combination products may require different numbers of applications for different regulatory authorities. 

Overall, the presentation concluded with emphasis on the need for continued dialogue.  The Office of Combination Products was recommended as a single focal point to handle issues regarding combination products.  Redundancy should be reduced, especially in the reviewing process.   The entire product life cycle should be considered in establishing post market requirements and global harmonization should be considered.

Regulatory Issues:  FDA Perspective

The last session of the day included a series of presentations by FDA staff who discussed some of the Agency’s experiences with combination products, both in general and for specific products.

The first speaker was Mark Kramer, Director, Office of Combination Products in FDA’s Office of the Commissioner.  The presentation focused on the definition of a combination product, how FDA regulates combination products, the role of the Office, and an explanation of the Office’s activities and current initiatives. Mr. Kramer noted that combination products are combinations of drugs and devices; drugs and biologics; devices and biologics; or drugs, devices and biologics.

As defined in 21 CFR § 3.2(e), the term combination product includes:

(1) A product comprised of two or more regulated components, i.e., drug/device, biologic/device, drug/biologic, or drug/device/biologic, that are physically, chemically, or otherwise combined or mixed and produced as a single entity;

(2) Two or more separate products packaged together in a single package or as a unit and comprised of drug and device products, device and biological products, or biological and drug products;

(3) A drug, device, or biological product packaged separately that according to its investigational plan or proposed labeling is intended for use only with an approved individually specified drug, device, or biological product where both are required to achieve the intended use, indication, or effect and where upon approval of the proposed product the labeling of the approved product would need to be changed, e.g., to reflect a change in intended use, dosage form, strength, route of administration, or significant change in dose; or

(4) Any investigational drug, device, or biological product packaged separately that according to its proposed labeling is for use only with another individually specified investigational drug, device, or biological product where both are required to achieve the intended use, indication, or effect.

Mr. Kramer noted that the cross-labeling issue identified by earlier speakers usually is raised by the separately marketed components that are often provided by or manufactured by different companies.

Examples of combination products include: drug eluting stents, antimicrobial coated catheters, condoms with spermicide, and pre-filled drug or biologic delivery devices such as injector pens, metered dose inhaler, and transdermal patches.  Other examples include radiopharmaceuticals combined with a biologic or monoclonal antibodies combined with a chemotherapeutic drug; a hyperthermia device used with a chemotherapeutic drug; and photodynamic therapy drug and laser light source; a diagnostic device requiring the administration of a particular drug or biologic; or a drug requiring a specific diagnostic device to determine a dosing regimen. 

Most concomitant use of drugs, devices and biologics does not constitute a combination product (e.g., an infusion pump that is not intended for use with a specified drug or biologic product), nor do combinations of two devices, two drugs, or two biologics.  Mr. Kramer noted that some of the products discussed at the meeting may not meet the technical definition of a combination product, but , they may raise many of the same complex issues. 

The Office of Combination Products (OCP) was established in December 2002 and has six statutory roles outlined in the Medical Device User Fee and Modernization Act of 2002.  These are 1) to assign a lead Center for premarket review and regulation of a combination product based on a determination of the product’s primary mode of action, 2) to ensure the timely and effective pre-market review of combination products by coordinating and facilitating the review process when more than one agency center is involved, 3) to ensure the consistent and appropriate post market regulation of combination products. 4) to resolve disputes regarding the timeliness of review of a combination product, 5) to review and update guidance, agreements and practices relative to the assignment of combination products, and 6) to report to Congress on an annual basis on the Office’s activities and impacts.  Mr. Kramer’s presentation on these roles intertwined key activities and initiatives that reflected these functions with responses to or examples of issues raised by earlier speakers.  His comments may be summarized as follows.

Regarding assignment of a combination product, Mr. Kramer explained that the formal jurisdictional assignment (i.e., the response to a request for designation submitted by a sponsor) for combination products must be completed within 60 days by statute.  The OCP is formulating a regulatory definition for primary mode of action that will be made available for public review and comment.  In order to facilitate the review of new jurisdictional requests, the Office is documenting prior jurisdictional determinations to make them more searchable and readily reviewable. 

Mr. Kramer explained and defined what the Agency means by consultative and collaborative review processes.  The goal of both the consultative and collaborative review processes is to apply the appropriate expertise needed to best evaluate and address the safety and effectiveness issues raised by the combination product.  During the consultative process, the consulting center provides recommendations, but the lead center is ultimately responsible for all decision-making on the application.  During the collaborative review process, both centers must reach agreement in order to approve or disapprove the product.

To ensure the timely and effective pre-market review process, OCP is overseeing the timeliness of and coordinating reviews involving more than one center.  A standard operating procedure for the intercenter consultative review process was established in 2002 and updated in February 2003 to monitor the consultative process between the centers.[11]  A copy of each consultation request is sent to OCP for screening to determine that the request is clear, that the second center understands what's expected of them, and that the time frames are reasonable.  Then OCP monitors the process to make sure that the originating center receives quality and timely feedback..  Mr. Kramer noted that the SOP formalizes the agency’s policy that "consults count"; i.e., the centers are accountable for the timeliness and the quality of their consulting review.   Also, as of May 1, 2003, submissions under review in CBER, CDER and CDRH are categorized as whether or not they involve a combination product, and if so, what type. The office also serves as a resource to sponsors and review staff for issues and questions.  OCP is developing reviewer tools and training.  In monitoring the SOP they are seeing real-life practical issues.  The common denominators of these issues and the lessons learned will be provided to the staff.

The OCP also plays a role in ensuring consistent and appropriate post market regulation.  Where possible, the assignment letters now include preliminary determinations of what good manufacturing (GMP) and adverse event reporting requirements will apply for a combination product.  There are several working groups that are developing guidance on the selection of premarket authorities for good manufacturing controls and adverse event reporting requirements.  Also, Mr. Kramer provided an example of how the mixture of authorities was applied to the drug eluting stents; i.e., the drug substance was subject to drug GMPs while the drug-device combination product was subject to the quality system regulation.

Mr. Kramer discussed the situation of submitting one or more applications for a single combination product.  He agreed with several presenters and noted that the ultimate goal was to have one application whenever possible. Also, he recognized that in some cases a company may actually prefer to have two applications, even if FDA would otherwise needs only one application.  In other cases, two applications will be needed to ensure appropriate regulation of a combination product.  These decisions have user fees implications.  OCP has established working groups to develop guidance on these issues.

In conclusion, Mr. Kramer noted that a "one size fits all" approach will not work for the scope and wide variety of combination products.  He noted that industry consultation with FDA is very important.  The dialogue should begin as early as possible so that OCP and the Centers can work with industry on determining the most appropriate regulatory pathway.  He agreed with earlier speakers that the process also works better when there is collaboration between the device, drug or biologic sponsors of a combination product.  Independent development is difficult.  Also, Mr., Kramer reiterated the availability of the Office to be a resource for industry and FDA review staff about combination product issues.

The next series of speakers gave short presentations describing the FDA’s perspective on regulatory review and approval of specific combination products.  The first speaker, Ashley Boam, Chief, Interventional Cardiology Devices Branch, CDRH, gave the FDA’s perspective on the current approach to the review of drug eluting stents.  She listed several challenges that these products posed for the Agency, including:

  • determination of regulatory jurisdiction
  • determination of inspectional authority and inspection site readiness
  • disparity in statutory and regulatory requirements between CDRH and CDER
  • Appropriate everaging of information from investigational new drug applications (IND) and investigational device exemptions (IDE), drug master files (DMF) and device master files (MAF)
  • Designing appropriate preclinical testing and clinical trial design
  • Post-market studies and post-market surveillance

Ms. Boam explained that CDRH was assigned as lead Center for these products, based on the agency’s determination that the primary mode of action of drug-eluting stents was attributable to the device (stent) component.  As a complex, novel product representing breakthrough technology, there was an extensive review team comprised of members from CDRH’s Divisions of Cardiovascular Devices and Mechanics and Materials Science, and CDER’s Division of Cardio-Renal Drug Products, Office of New Drug Chemistry, and Office of Pharmaceutical Science.    Appropriate review of this application required that the agency have expertise in a variety of areas, including animal experimentation and evaluation, mechanical performance and testing, drug chemistry, drug and device manufacturing, pharmacokinetics/pharmacodynamics, and clinical trial design and methodology.   Ms. Boam outlined some of the challenges that will be posed by future submissions, including combinations of unapproved drugs and/or unapproved stent platforms.  These future products will require even more intensive evaluation.

Next, she reviewed the preclinical testing objectives for new drug eluting stents, and outlined some of the preclinical testing deficiencies that her group has observed for these products.  The objective of preclinical testing is to characterize the finished, sterilized product, including evaluation of the coating/drug loading characteristics, in vitro and in vivo elution of the drug, and stability testing of the drug/device combination.  Some of the deficiencies that they have seen in preclinical testing include inadequate stent platform testing (e.g., lacking fatigue and corrosion testing), inadequate analysis of surface modifications (e.g., coating integrity/durability, drug content/uniformity), incomplete in vitro pharmacokinetics, and inadequate information for CMC (e.g., stability and shelf life).  For these products, adequate animal studies are needed to assess safety prior to human studies.  Common animal study deficiencies include lack of evaluation of doses intended for clinical evaluation and/or overdosage at appropriate time points, lack of evaluation of serial sections of myocardium, lack of description of arterial histopathology, and a lack of necropsy reports.

Clinical evaluations of these products are, of course, necessary and Ms. Boam urged developers to ensure that trials are designed to establish reasonable assurance of both safety and effectiveness.  Since these are combination products, it will not be appropriate for sponsors to assume that the same trial design used for bare metal stents (including the number of subjects and length of follow-up) would be appropriate for drug eluting stents.  She closed by urging developers of these innovative products to work with the FDA early in the process to help ensure an efficient and timely review process for their products.

The next speaker was Richard P. Felten, Expert Reviewer, General Surgery Devices Branch, CDRH, who spoke on "The Review Pathway for Photodynamic Therapy Systems."  Photodynamic therapy systems, that combine a drug with a light source for activation, were one of the first combination products reviewed by the Agency.  CDER was assigned as the lead Center for these products based on a determination that the primary mode of action of the combination product was attributable to the drug component.  CDRH has played an important consulting role on the device portion.  Mr. Felten explained that, during the clinical trials stage, the combination product was reviewed under a single IND application, with CDRH providing a consulting review on the device portion.  Difficulties arose once the clinical trial was completed because the developer of the drug did not want to have regulatory responsibility for the device portion.  This is because the device was a commercially available surgical laser, and lasers have special requirements under CDRH regulations for radiation-emitting products.  The solution that was developed at the time was that the Agency split out the device portion of the NDA and made this, for administrative purposes, into several PMAs.  This allowed the drug company to separate themselves from the device requirements and allowed the Agency a mechanism for tracking design changes to the device post-approval. Mr. Felten reviewed several other photodynamic therapy systems that were reviewed under the same structure. He cautioned, though, that such an approach may not be feasible in the future, now that applications to both CDER and CDRH require user fees.  He emphasized that extensive consultation between CDRH and CDER was required, and he echoed the theme of several other speakers when he encouraged developers to work with the Agency early to determine not only jurisdiction, but also the type of regulatory submission the product will be reviewed under, and what particular responsibilities this may require.

The final speaker in the FDA session was Dan Shames, M.D., FACS, Director of the Division of Reproductive and Urologic Drug Products, CDER.  Dr. Shames spoke on contraceptive drug-device combination products that his group has had experience in reviewing over the past five or six years, with emphasis on lessons learned.  He gave some specific examples and explained that, although the clinical trials were designed the same for these products as for an oral contraceptive, there were new issues surrounding the chemistry and manufacturing of the materials and the pharmacokinetics of the drug released from the implant, that presented an opportunity for learning for his group.  He next discussed a product called Lunelle, which is a monthly injectable that delivers a combination of estrogen and progestin.  Dr. Shames explained that the review appeared to be straightforward, since the product was just a prefilled syringe, however, there were issues related to manufacturing the syringe and the vial which made the review challenging.  The other important point is that there were two drugs involved, which means that the sponsor had to address the regulation regarding combination drug products (21 CFR 300.50).  The next example was Mirena, which is an intrauterine system that delivers levonorgestoral both locally and systemically.  The important take home message for this review, according to Dr. Shames, is that although IUDs have been available for some time, there was a challenge to show that the addition of the drug added to the effectiveness and safety of the product itself.  Next, Dr. Shames discussed an intravaginal ring that delivers estrogen and progestin systemically.  Vaginal contraceptive rings have been studied for decades but, according to Dr. Shames, it took some innovation on the part of the manufacturer to develop the right combination of materials to make the product effective.

The last example was a transdermal patch for delivery of estrogen and progestin.  Although transdermal patches are not a new technology, Dr. Shames pointed out that contraception requires a higher dose of drug than patches for treatment of menopausal symptoms, and it required significant innovation for the sponsor to create a system the size of a matchbox that would deliver the correct amount of drug.

Dr. Shames concluded his talk by reviewing the lessons learned. First, he pointed out that the CDER reviewers worked with device reviewers in a productive manner to review these products efficiently and for the most part, the review process worked very well.  For contraceptive products, the process was eased by the fact that the clinical trials were the same as for oral contraceptives, with clear expectations for the results.  He also echoed the recommendation of the other FDA speakers in urging sponsors to communicate and work the Agency as early as possible.

He pointed out that the process does not work as well when things are "a little out of the box" which is happening more and more.  There also are those who see a "culture clash" between CDER and CDRH, both in terms of the reviewers and the review culture. There are also differences for reviewers when dealing with traditionally large pharmaceutical companies versus small device firms.

Dr. Shames concluded by stating that he believes that the future looks bright for combination products.  The OCP has helped improve the communication and coordination across the centers.  He stated that most staff in CDER and CDRH enjoy working on innovative products and are well motivated to assist sponsors and improve FDA’s internal processes.

FDA and Industry Panel Discussion

The workshop concluded with a panel discussion that included representatives from industry and the FDA.  Representatives from industry included Mr. Jonathan Kahan, Ms. Christine Allison and Mr. Keith Smith.  Representatives of FDA included Dr. David Feigal, Director of the Center for Devices and Radiological Health [also: there is no summary of Dr. Feigal’s introductory remarks that morning] Dr. Jesse Goodman, Director of the Center for Biologics Evaluation and Research, Dr. John Jenkins, Director of the Office of New Drugs, Center for Drug Evaluation and Research and Mr. Mark Kramer, head of the FDA OCP.  Audience participation was encouraged.  The session was moderated by Dr. Liz Jacobsen of AdvaMed.

One audience member questioned the higher standards that were being set for combination products like the drug eluting stent and indicated that these "unrealistic agency expectations and requirements" may stifle innovation.  The questioner suggested that the regulatory requirements for combination drug-device products should lie somewhere between the current requirements for traditional pharmaceuticals and the less burdensome requirements for medical devices.  In addressing this question, Dr. Feigal stressed the importance of a science-based approach to regulatory decision-making and noted that the approaches taken by the Agency for products like the drug eluting stent have included flexibility, e.g., the Agency has not required the manufacturer of three complete commercial batches for the drug in the drug eluting stent, as is required for traditional pharmaceuticals.  He stressed that the Agency needs to have information on safety and effectiveness for breakthrough products (as with all products) in order to understand the risk-benefit ratio.

Another audience participant asked whether there will be any changes at the Agency that will make the regulatory path easier for drug delivery companies pursuing new routes of administration for already approved drugs.  Mr. Kramer addressed this question and indicated that it is the role of his office to help manufacturers determine the appropriate regulatory path for their combination product.  Dr. Goodman agreed and stated that master files are often a useful tool in these situations.

A question was asked about the future of a drug delivery device with closed loop feedback, such as might be used for diabetics.  Dr. Feigal used the example of an implanted cardioverter defibrillator (ICD) that both senses abnormal rhythms and delivers a therapeutic shock when needed.  These have a long history of safe use, so clearly closed loop feedback systems can be studied, approved and marketed.  Dr. Feigal noted some of the challenges already discussed by Dr. Klonoff in the morning session, and discussed the benefits of tackling a challenging problem like this in incremental steps.  He pointed out the importance of building public trust in a technology, which includes things like handling recalls and safety alerts responsibly.  Mr. Kahan added that developers of such closed loop systems should expect a rigorous approval process because of the potential for overdosing or underdosing.

A question was asked about how the Agency is going to be prepared to handle the very challenging products that were discussed during this workshop.  Dr. Feigal noted that one of the challenges to be addressed by the Agency is to determine the role for preclinical testing in answering the challenging questions posed by new technology.  He pointed out that there are some questions and issues that are much better addressed through bench testing, e.g., measuring the beam strength of a radiation therapy device is best done by a physicist on a bench rather than by a clinician on a patient.  It will be important for the Agency to identify the characteristics of novel products that are essential to their performance and to establishing safety and effectiveness.  Dr. Jenkins noted that the biggest challenge in CDER is becoming familiar with CDRH’s regulations and statutory provisions.  Most reviewers in CDER have very little knowledge about the CDRH process.  He referred to the drug eluting stent team and noted that it was a very good example of intercenter teamwork.  He cited training and the opportunity for interaction as keys to making intercenter collaborations work well.  Dr. Goodman noted that this area poses some big challenges for the Agency since it involves new technology that crosses regulatory lines.  He believes that the Agency needs to invest in its people to ensure that reviewers are scientifically up-to-date.  Collaboration both inside and outside the Agency is important.  Dr. Jacobsen of AdvaMed also addressed this question and noted that the day’s workshop and the opportunity for interaction between scientists, industry and the FDA was very useful and urged the Agency to hold more workshops of this type.

Summary and Conclusions

The workshop included outstanding scientific presentations on many new technologies that will be confronting the Agency in the future.  The workshop also presented a forum for useful interaction and feedback between FDA and the developers of these new technologies.  The information will be used by the Agency as it moves forward with developing guidance documents in this area.




FDA Workshop on
 Innovative Systems for Delivery of Drugs and Biologics: 

Scientific, Clinical and Regulatory Challenges

July 8, 2003

Bethesda Marriott, Bethesda, MD

8:00 AM – 5:30 PM

7:00 – 8:00 a.m.  REGISTRATION
8:00 – 8:15 a.m.WELCOME: David Feigal, Jr., M.D., M.P.H., Center Director, Center for Devices and Radiological Health (CDRH), FDA
8:15-8:30 a.m.OPENING REMARKS: Mark McClellan, M.D., Ph.D., Commissioner, FDA
8:30 – 9:15 a.m. KEYNOTE ADDRESS: Innovations in Drug Delivery.
Robert Langer, Ph.D., Kenneth J. Germeshausen Professor of Chemical and Biomedical Engineering, Massachusetts Institute of Technology 
9:30-11:00 a.m.

Moderator:  David Feigal, Jr., M.D., M.P.H., CDRH


  • Novel Technologies for Treatment of Diabetes: Metabolic monitoring, the artificial pancreas and alternate routes for administering insulin.
    David C. Klonoff, M.D., FACP, Medical Director, Mills Peninsula Diabetes Research Institute, Clinical Professor of Medicine, University of San Francisco
  • Emerging Techniques and Technologies for Treatment of Solid Tumors.  Jonathan B. Kruskal, M.D., Ph.D. Director, Abdominal Imaging, Beth Israel Deaconess Medical Center, Associate Professor of Radiology, Harvard Medical School
  • Novel Technologies for the Treatment of Cardiovascular Disease. 
    Richard E. Kuntz,, M.D., M.Sc., Associate Professor, Harvard Medical School, Brigham & Woman’s Hospital



11:00 – 11:15 a.m.BREAK
11:15 a.m. – 12:30 p.m.

Moderator:  Ajaz Hussain, Ph.D., Center for Drug Evaluation and Research (CDER)


  • Preclinical development considerations for the pulmonary delivery of drugs approved for other routes of administration.
    Chet Leach, Ph.D., DABT, Director, Life Sciences Research and Development, Nektar Therapeutics
  • Protein Delivery from Implantable Devices:  Challenges and Opportunities.
    Bill Van Antwerp, Ph.D., Vice President and Chief Scientific Officer , Medtronic MiniMed
  • Developing a local drug delivery combination product for postoperative atrial fibrillation:  preclinical challenges.  
    Kevin C. Skinner, V.M.D., Associate Director of Science, Biomaterial and Surgical Products, Genzyme Corporation
  • Considerations for drug chemistry and stability evaluations for novel drug-device combination products.
    Ajaz Hussain, Ph.D., Deputy Director, Office of Pharmaceutical Sciences, CDER, FDA
12:30 – 1:30 p.m.LUNCH
 1:30 – 2:00 p.m.PUBLIC COMMENT
2:00 – 3:15 p.m. 

Moderator:  Liz Jacobsen, Ph.D., AdvaMed


  • Regulatory and legal challenges for developers of drug delivery devices
    Jonathan S. Kahan, Partner, Hogan and Hartson, LLP.
  • Industry perspective on challenges for product developers – devices
    Patricia B. Shrader, Vice President, Corporate Regulatory Affairs,  Becton, Dickinson and Company 
  • Industry perspective on challenges for product developers – drugs
    Christine Allison M.S., RAC , Associate Regulatory Consultant
    Global Regulatory Affairs (CM&C), Eli Lilly and Company 
 3:15 – 3:30 p.m.  BREAK 
Moderator:  Janet Woodcock, M.D., CDER

FDA Initiatives for Combination Products
Mark Kramer, Director, FDA Office of Combination Products

FDA perspective:  Current approach to review
  • Drug eluting stents
    Ashley B. Boam, MSBE
    Chief, Interventional Cardiology Devices Branch, CDRH
  • Photodynamic therapysystems
    Richard P. Felten, General Surgery Devices Branch, CDRH
  • Contraceptive delivery systems
    Dan Shames, MD, FACS
    Director, Division of Reproductive and Urologic Drug Products, CDER
4:15 – 5:00 p.m. 


Panelists: David Feigal, Jr. M.D., M.P.H., Janet Woodcock,  M.D., Jesse Goodman, M.D., Mark Kramer, Patricia B. Shrader, Christine Allison, Jonathan S. Kahan.




 [1] Improving Innovation in Medical Technology: Beyond 2002".  For reference, the white paper describing the entire initiative is available at http://www.fda.gov/bbs/topics/NEWS/2003/NEW00867.html

 [2] MDUFMA, see http://www.fda.gov/oc/mdufma (updated link to MDUFMA)

 [3] Langer, Scientific American, April 2003:33-39.

 [4] Langer, Nature, Vol 392(Supplement), April 1998:5-10.

 [5] Langer, Science, Vol 293, July 2001:58-59.

 [6] Nielsen L, Frokjaer S, Brange J, Uversky VN, Fink AL. Biochemistry. 2001 Jul 27; 40(28): 8397-409

 [7] Food Drug and Cosmetic Act Section 201(g) and (h)

 [8] Public Heath Service Act , Section 351

 [9] See  http://www.fda.gov/OHRMS/DOCKETS/98fr/03-14209.pdf

 [10] November 25, 2002, public hearing on combination products.  See http://www.fda.gov/oc/combination/default.htm for Federal Register notice, agenda, and transcript.

 [11] http://www.fda.gov/oc/ombudsman/intercentersop.pdf (updated posting at www.fda.gov//CombinationProducts/GuidanceRegulatoryInformation/ucm119234.htm