Getting Started with a Combination Product: Part I

Originally Published MDDI March 2003REGULATORY OUTLOOK

March 1, 2003

22 Min Read
Getting Started with a Combination Product: Part I

Originally Published MDDI March 2003


Overcoming the regulatory challenges in this ever-expanding field yields rewards.

Barry S. Sall, Peter Lassoff, and Bruce Babbitt
Parexel International Corp.

When medical product developers blend drug, device, and biologic components into a single product, they have created a combination product. Combination products can also be created when cross-referenced in labeling. 

One reason for creating such products is to deliver drugs or biologics to a well-defined target within the body. The drug or biologic may elute from a solid support that is part of a device, or the device may serve as a conduit for the precise delivery of a tiny volume of liquid. These models provide numerous therapeutic advantages by potentially increasing the effectiveness of both products and reducing systemic side effects by enabling local delivery. 

As biotechnology, genomics, and proteomics yield more information on disease processes, new therapeutic agents are being developed. Some must be targeted to specific tissues or anatomical regions, while others can only be used if patients have certain genetic characteristics. Combining these drugs and biologics with devices offers one avenue for the successful use of such products. 

Creating combination products raises a spectrum of unique business, technical, and regulatory questions, however. These questions need to be evaluated from the perspective of each component of the product. If the developer has previous experience in the device area, there can be a tendency to minimize drug or biologic issues; if the developer is a drug person, then the opposite may be true. Therefore, it is important that input from all perspectives is factored into the project plan. 

Medical product developers must recognize, early in the process, when the product design assumes characteristics of a combination product. These products do require additional regulatory oversight; however, the rewards of increased efficacy and added safety can more than compensate for the additional efforts.

Regulatory History

FDA defines combination products in 21 CFR 3.2(e). They can be a single product with two or more device, drug, or biologic components. They can also be two separate device, drug, or biologic products that are either packaged together or specified for use together in their labeling. Drugs, devices, and biological products are defined in the Federal Food, Drug and Cosmetic (FD&C) Act. Prior to 1991, the relatively small numbers of combination products were regulated on an ad hoc basis. Then, as a result of the Safe Medical Devices Act of 1990, FDA published intercenter agreements between the device, drug, and biologic centers.1

Examples of Combination Products

The current intense interest in drug-coated, coronary-artery stents is an example of a combination product that includes a device—the metallic stent—with a variety of drugs coated on its surface. The drugs are intended to enhance the physical effect of the device by preventing closure of the vessel after treatment. Another example is the recently approved tapered-metallic spinal-fusion cage. This is used in combination with recombinant human bone morphogenic protein, which is placed on a resorbable collagen sponge. The recombinant protein induces formation of new bone. 

Both of these products are regulated by CDRH, via the PMA process, in consultation with CDER and CBER. Other combination products currently in development include living cells encapsulated in a semipermeable membrane that secrete various biological substances. These products are regulated by CBER, with consultation from CDRH. Catheters intended to deliver an angiogenesis gene to heart muscle are also considered combination products, as their clinical utility depends on their use with the biological product. In addition, several photoactivated drugs have been approved by FDA. These drugs, which are regulated by CDER, require a specially designed light source to activate the drug. The photoactivation light sources are cross-labeled with the drug and regulated by CDRH via the PMA path. 

A relatively new approach associated with biotech therapeutics uses in vitro diagnostic assays to predict if a drug or biologic product would be effective in treating a particular patient. Several regulatory approaches have been successful with this product type. Product configuration is dependent on business objectives as well as technical and medical considerations. Data describing assays performed at a central laboratory have been submitted in a separate section of the drug company's new drug application (NDA) or biologic license application (BLA), while data describing assay kits that are sold to commercial laboratories are generally submitted as stand-alone PMAs. 

Combination products such as those listed in Table I offer great potential to enhance therapies; however, these combinations frequently raise multifaceted regulatory issues.

FDA Regulation

Since 1991, when the product jurisdiction regulation (21 CFR 3) came into effect, developers of combination products have been able to use the request for designation (RFD) process to receive a formal jurisdictional determination from FDA. The intercenter agreements offer examples of how FDA will evaluate broad categories of products, and some specific examples are also included. In some cases, the intercenter agreements can be a useful guide when evaluating the appropriateness of an RFD. Due to their age and the fast pace of technological development, they do not provide guidance for many of the most innovative products currently in development. FDA has recognized this and has recently begun posting jurisdictional updates on its Web site.2 These updates describe the jurisdictional decisions for some products that are not mentioned in the intercenter agreements. 

FDA has indicated that it will eventually update the agreements; however, the agency recognizes that, due to confidentiality concerns, it is not always possible to prepare a jurisdictional update. As a result, the regulatory status of many innovative technologies remains unknown to industry for extended periods of time.

The decision on which center will have primary jurisdiction over a product has enormous business and regulatory significance. When considering a potential combination product, a developer must evaluate the regulatory situation and first determine if an RFD is necessary at all. If it is, the developer must select which center to propose for primary jurisdiction and when to make the request. Business considerations frequently drive this decision. Investors in development-stage companies want to know if they are investing in a medical device or a drug company, and when the U.S. revenue stream is likely to start. Product reimbursement also factors into the decision process. Management at larger, well-established companies has similar concerns. 

Research into previous FDA decisions for similar products, when they are publicly available, can provide useful information. It is important to remember, however, that FDA considers each decision on a case-by-case basis. Preliminary telephone consultations with management at the center of choice should be useful for further guidance. If the sponsor has been dealing with a particular reviewer for previous, related products, that reviewer would be a good point of first contact. CDRH, CDER, and CBER also have product jurisdiction officers who can provide useful insights. 

If center management is confident that the product review can be handled entirely within that center, that is usually a good indication of jurisdiction. It is not an official determination, however. The sponsor may choose to move forward on a development plan assuming a particular regulatory path, but the picture could change if, over time, different people within the center become involved with the project, additional technical information regarding the product becomes known, or decisions regarding a similar product are made at another center. 

Table I: Selected combination product examples. (Click to enlarge).

RFDs. Official product jurisdiction determinations are made through the request for designation process. These determinations provide a high level of assurance that the specified regulatory path will be followed through the commercialization phase. It should be kept in mind that many sponsors of combination products determine that the regulatory path for their product is so well-defined that an RFD is not necessary.

The RFD process is described in 21 CFR 3.7. Determining when to file the RFD is usually driven by the availability of product data. There must be enough reliable data available for FDA to understand the product and determine its primary mode of action. The primary mode of action is the determining factor FDA uses to assign primary review responsibility to a particular center. Three components of the RFD, in particular, raise significant strategic issues: the mechanism or mode of action, the physical descriptions of the product, and the indications for use. The mode of action must be carefully described and supporting data cited. The indication for use is vital for the decision process because it permits FDA to determine the primary intended purpose of the product. The physical description of the product can change frequently over the course of the development process. The RFD can be submitted with a preliminary description, but the sponsor must keep in mind that significant design changes have the potential to change the outcome of the RFD process. 

These three items, combined with other information listed in the regulation, make up the RFD. The regulation stipulates that the RFD must be no more than 15 pages long, including attachments. Recently, FDA has been encouraging sponsors to submit RFDs both on paper and electronically to facilitate internal distribution within FDA. Each RFD is sent for review to the product jurisdiction officer at the Office of the Ombudsman. 

Letter of Designation. The product jurisdiction officer has 60 days in which to review the information, consult with the relevant centers, and issue a letter of designation. If the letter of designation is not issued within 60 days, the sponsor's recommended center will receive primary jurisdiction for the product. In a practical sense, if the review process is extending toward 60 days, the product jurisdiction officer can suggest that the sponsor agree to an extension of the review period. The sponsor does not have to agree to an extension; however, in that case, the product jurisdiction officer may be forced to render a premature decision that does not take into account all the sponsor's data. And, more importantly, it may be a decision that could handicap the sponsor's future development efforts. Extensions do not carry time limits.

The letter of designation from FDA will briefly describe the product, the product designation, and the rationale for the designation. It can also specify the type of marketing application necessary in order to commercialize the product. FDA encourages sponsors to request that other issues be discussed in the letter, such as which manufacturing regulation will apply to the product or its components (the drug GMP or device quality system regulation). At this early stage it is not possible to clarify all such issues; furthermore, decisions regarding these additional items are not officially binding on FDA. These other jurisdictional issues are frequently resolved during discussions with the primary reviewing center after the letter of designation is received. 

Recent Changes at FDA

FDA has recognized that the current combination-product jurisdiction process does not fully meet the needs of the agency or industry. Recent agency-initiated reform efforts include the appointment of a director of the combination products program, the inclusion of some jurisdictional decisions on the FDA Web site, public stakeholders meetings, a survey of personnel within FDA that are involved with combination products, and plans to update the intercenter agreements.3

Late last year, the Medical Device User Fee and Modernization Act of 2002 was signed into law.4 One section of this law relates to the designation and regulation of combination products. It formally establishes an Office of Combination Products. This office is responsible for designation and for ensuring that product reviews for combination products proceed in a timely manner. The office will also be responsible for ensuring that jurisdictional decisions are consistent for similar product types. In order to accomplish these objectives, the law provides that the office will have medically and scientifically trained staff that can interact with reviewers in CDRH, CDER, and CBER on a technical level; monitor progress; and prevent or resolve issues that cause delays. These technically trained staffers will also provide the product jurisdiction officer with input during the determination process. 

Another important function of the new office will be the maintenance of a combination products–tracking database. Prior to implementation of this law, submission-tracking databases within each of the centers did not identify combination products, so it was not possible to monitor their progress in a systematic manner. This is now remedied. Another benefit of this database will be the ability to notify the public when a combination product has been approved. All of these initiatives are intended to expedite the review of combination products.

Industry must stay informed of the changing regulatory climate and offer feedback to FDA during this period of change.

Strategic Regulatory and Business Considerations

Combination products, by their nature, raise unique challenges. They involve highly innovative technology and frequently set regulatory precedents, and it is difficult to offer broad generalizations that apply to the entire spectrum of possible products. Starting at the beginning of the development process, project team members must continually make an effort to review their decisions from multiple perspectives. For example, even if the team believes that their product will be regulated as a device, with some pharmaceutical consultation, they need to be prepared to justify their preclinical testing, manufacturing, clinical, and regulatory choices from both the device and pharmaceutical perspectives. 

The device approval process is traditionally industry's first choice when presenting a combination product to FDA. Advantages include less burdensome preclinical testing, one pivotal human clinical trial rather than two, considerably lower application filing fees, and less burdensome postapproval requirements. While these advantages are significant and cannot be ignored, other factors must also be considered. First, even though CDRH may have primary jurisdiction for a product, additional preclinical testing, beyond the ISO 10993-1 matrix, can always be requested.5 This testing is entirely dependent on the risks raised by the product, and may draw upon International Conference on Harmonization S1–S7 guidances or ad hoc judgments.6

An evaluation of the product and the testing necessary to fully understand its risks is also important. Such an evaluation must consider both the device and drug components. Significant efficiencies can be gained in some cases where an ISO 10993 test protocol can be used to gather additional samples and generate data that address pharmaceutical concerns. 

Another consideration involves manufacturing regulations. If a combination product is being developed in a device environment where all the quality system regulation (QSR) procedures are already in place, the combination product should be easily integrated into the system. On the other hand, if the combination product is being developed and produced in a more pharmaceutically oriented environment, the added requirements of the QSR, especially compliance with design controls, must be carefully considered. Adding new procedures and training staff can add considerable expense and consume significant time, although the benefits of design controls may outweigh those costs. Product jurisdiction must be considered in light of all pre- and postmarketing activities. 

User Fees. The passage of the user fee act has brought user fees to the device industry for the first time. The pharmaceutical and biologics industries have worked with user fees since 1992. Currently, the filing fee for a new drug application containing clinical data is $533,400. The standard filing fee for a PMA is set at $154,000 for fiscal year 2003. Clearly, these costs must also be considered when selecting the primary reviewing center; however, sponsors should not allow user fee differences to dominate the decision making process. Other pre- and postmarket regulatory requirements must also be considered. The cost of implementing the QSR in even a medium-sized firm that has no experience with it could outweigh user fee savings.

When preparing submissions for a combination product, one must constantly maintain awareness of the review process and the needs of each center that is involved in the process. The combination products program recently completed an internal survey of FDA personnel at all three centers that deal with combination products.3 It noted several areas where intercenter collaboration can be impeded. If an IDE is necessary for a product and CDER will be consulting, it will speed up the review process if the data for the CDER review are placed in a separate section or binder. Likewise, if a device is used to activate or deliver a pharmaceutical, the device information should be segregated. When submissions are prepared, sponsors should be sure that terminology can be understood by reviewers from all involved centers. Glossaries and lists of abbreviations can also be helpful. Discussion with the lead reviewer or project manager can help to identify potential procedural roadblocks before they affect the process. 

Beyond the purely regulatory concerns, combination product developers must also consider larger business concerns. While the medical device route may present a less burdensome path to market, the device market may offer a more modest commercial potential than the market that can be addressed with a pharmaceutical mechanism of action. Generally, developers must make a choice to work with one center. In some cases, business needs may make a more complex regulatory approach necessary in order to achieve the desired return on investment. Development-stage companies also need to consider differences in how the marketplace and investor communities value medical device and pharmaceutical technology.

Sponsor-Facilitated FDA Interactions

FDA itself is in the process of implementing a variety of programs that are intended to promote effective intercenter communication. Sponsors can also take actions to improve the process on administrative, regulatory, and technical levels. Above all, sponsors must always keep in mind that different centers have different priorities, backgrounds, and levels of technical understanding.

During the RFD process, providing information in both paper and electronic formats will help FDA distribute documents to appropriate reviewers in a timely manner. These documents should include a concise description of the product and its mechanism of action, as well as specific technical information. In some cases, a glossary of terms can also help ensure that the reviewers gain a common level of understanding.

Sponsors can make recommendations that a particular ISO standard or ICH guideline be followed to address a particular aspect of product safety or effectiveness. A justification for following the standard should be included. These justifications can become highly technical, so sponsors should review them to be sure that scientific and medical concerns are fully addressed. For instance, extensive safety data for a marketed drug may have already been submitted to FDA during the NDA process; however, data describing the in vivo interaction of the drug when immobilized on a solid support may not be available in the existing NDA files.

CDER and CBER Perspectives. In conjunction with CDRH, a large percentage of combination products are likely to be reviewed by CDER for two reasons: one, most combination products currently being developed consist of a small-molecule drug attached to the surface of a medical device; and two, based upon the recent announcement by FDA that CDER will soon be assuming CBER's responsibility for the review of therapeutic proteins, even medical devices with recombinant proteins attached to them will likely be reviewed by CDER.

Alternatively, combination products composed of a device coupled to other, less well-defined classes of biologics (e.g., living cells, tissues, organs, nucleic acids/genes such as DNA or RNA) will likely be reviewed by CBER.

Another key factor in determining whether CDER or CBER takes a lead role in the review of a particular combination product will be the level of pre-existing institutional knowledge of product class and therapeutic indication. For example, if either CDER or CBER has previously reviewed the same small-molecule drug or biologic intended for use in combination with a medical device, and is therefore quite familiar with the toxicity profile and related in vivo performance characteristics of the compound, then it makes sense for that center to work closely with CDRH to review the combination product. 

It will be interesting to see whether or not basic differences in how CDER and CBER view preclinical safety testing of conventional small-molecule drugs and biologics translate into differences in how the two centers review combination products. For example, CDER typically requires toxicology studies in at least two animal species (one rodent and one nonrodent) prior to human testing, while CBER often requires data from only one species. In addition, CDER prefers that first-in-man studies testing synthetic drugs be conducted in healthy volunteers, while CBER typically encourages sponsors to conduct their initial clinical studies in diseased patients. 

Another interesting challenge for both sponsors and FDA will be to determine what absolute amount of clinical trial data that supports combination product safety and efficacy will be required for marketing approval. For example, will CDER require that two consecutive Phase III clinical trials be performed for device and small-molecule drug combination products, as they typically do for conventional drugs? FDA's stance on this issue will probably be determined on a case-by-case basis, influenced by any of the following factors: What primary therapeutic claims are being made for the product? Is the presence of the drug on the device simply to enhance a structural function for the device? Is the drug permanently attached to the device, thereby exerting only a local effect, or is it designed to leech off the device into the local environment?

On one hand, sponsors should argue that the use of FDA-approved drugs or biologics in their combination products should pose lower regulatory hurdles because these compounds have been in humans before and have well-defined toxicity profiles. On the other hand, FDA might argue that these compounds must be dealt with as new chemical entities (NCEs), due to the potential for changes in behavior of the drug or biologic in the body because of its attachment to a device. This translates into longer clinical development programs and the need for more data.

Preclinical Challenges

In cases where a drug or biologic being developed in combination with a device has already received approval for use in humans, it might be possible to build efficiencies into the preclinical safety testing program with prior agreement from FDA. Any prior data on the actual toxicities of the drug or biologic when administered as a single agent (not associated with a device) to humans should be included, or cross-referenced, in submissions to FDA, and potentially re-evaluated in additional studies.

Even if a combination product includes a previously approved drug or biologic, it is entirely possible that the intrinsic properties of the molecule (e.g., pharmacodynamics, pharmacokinetics, ADME) could be altered by physical or chemical attachment to a medical device. For example, since it is likely that the half-life of a small-molecule drug will be greatly increased by attachment to a device, it is possible that either the nature or the time course of the toxicities due to the drug could be altered. Consequently, as a worst-case scenario, the attached molecule should be considered a new chemical entity at least when designing initial preclinical studies.

A very interesting challenge in the development of combination products is to assess the potential change in immunogenicity of a drug or biologic when it is attached to the surface of a device surface. For example, when recombinant proteins are bound to a device surface, it is likely that these molecules will have altered domains; consequently, they may be recognized by the immune system as foreign. In cases where the recombinant protein is identical or very similar in structure to a naturally occurring protein in the body, an actual autoimmune response against the natural protein may occur.

Since most device-drug combination products involve the attachment of small-molecule synthetic drugs to a device, it is typically presumed that immune responses against these molecules are not an issue. Once small-molecule drugs are bound to the surface of medical devices, however, they become multivalent and able to be recognized by antibody-producing B-cells. This is similar to the effects seen when otherwise nonimmunogenic haptens are covalently linked to large-protein carriers, such as albumin. 

Since implanted medical devices are typically associated with modest amounts of local inflammation, it is also possible that small-molecule drugs and biologics that would otherwise have very little or no direct interaction with immune cells will now be able to directly interact (stimulate, inhibit) with these same cells when bound to the surface of a device. Hence, drugs or biologics coated onto medical devices have the potential to affect circulation patterns or release of soluble factors from various immune cells. This has a resultant potential to alter the functioning of the immune system, either locally where the device is implanted, or possibly even systemically.

Identifying Appropriate Testing Facilities

There are many contract research organizations located throughout the world with many years of experience in performing GLP-compliant preclinical safety studies on either medical devices (meeting ISO 10993 standards) or drug-biologics (meeting ICH S1–S7 guidances). However, the challenge is to find testing facilities with the appropriate mix of technical and surgical expertise, animal models, and analytical methods to allow a rigorous assessment of the relative contributions of each component of the combination product toward any toxicities associated with its use. 

When choosing appropriate animal models and protocol designs for safety testing of combination products, manufacturers are faced with a significant challenge. They must perform rigorous toxicity testing of the combination product when it is used as intended, of course. But at the same time they must also allow for parallel assessment of many of the standard safety parameters required for the preclinical evaluation of stand-alone medical devices or drugs and biologics.

It will be interesting to see whether or not CDER will require that conventional ADME studies be carried out for device-associated small-molecule drugs, even though the device is designed with the drug permanently attached. Of course, if any unexpected toxicities are uncovered during these studies, to determine the cause it will be necessary to go back and perform more standard safety testing of each of the individual components of the combination product. 


1.Intercenter Agreements, (Rockville, MD: FDA Office of Ombudsman); http://www.
intercenter, accessed December 6, 2002.
2.FDA Office of Ombudsman, Jurisdictional Updates;
ombudsman/updates.html, accessed December 6, 2002.
3.FDA Office of Ombudsman, “Regulation of Combination Products: FDA Employee Perspectives, Combination Products Program;”
ombudsman/rcpemployee.pdf, accessed December 6, 2002.
4.FDA Center for Devices and Radiological Health, “Information on the Medical Device User Fee and Modernization Act (MDUFMA) of 2002;” http://www.fda. gov/cdrh/mdufma/index.html, accessed December 6, 2002.
5.ISO 10993-1: Biological Evaluation of Medical Devices—Part 1: Evaluation and Testing, 2nd ed. (Geneva: International Organization for Standardization, 1997).
6.FDA Center for Drug Evaluation and Research, “International Conference of Harmonization Safety Guidances,” ICH S1–S7; index.htm accessed December 6, 2002. 

Post your questions and comments for the authors on-line in MD&DI's Author Forums. Select the Author Forums link at

Copyright ©2003 Medical Device & Diagnostic Industry

Sign up for the QMED & MD+DI Daily newsletter.

You May Also Like