Several participants expressed concern that there was a lack of clarity about FDA regulations regarding combination therapies and clinical trials of investigational agents used in combination that makes industry less willing to participate in collaborations. “The regulations or their interpretations are confusing and creates all sorts of hurdles,” summed up Dr. Perlmutter.
One area of confusion is the difference between a combination product versus using drugs in combination, which Dr. Richard Pazdur, director of FDA’s Office of Oncology Drug Products, clarified. An example of a combination product is Excedrin®, which has aspirin, acetaminophen, and caffeine, all in the same product, he said. This is different from drugs used in combination. Although some participants expressed concern that there was a regulatory requirement that necessitatated showing the contribution of individual agents in drug combinations, Dr. Pazdur said that requirement is only for drug combination products (Woodcock et al., 2011). “Our guidance clearly states that we are not going to demand independent isolation of the effect of each of the components when you are using two unapproved drugs together,” he said. Dr. Robert Temple, deputy center director for clinical science at FDA’s CDER, added that “the whole idea of how you demonstrate the contribution of a combination has historically been more flexible than people imagine,” and said that the agency is currently working on clarifying its rule for combination products so it can be interpreted more flexibly. “If two drugs have no
pharmacologic effect that’s relevant, but do have such an effect when you put them together, that might be enough,” Dr. Temple said.
Suggestions from Various Workshop Participants to Address Regulatory Challenges
• Focusing on combinations with a compelling biological rationale and strong preclinical data
• Seeking dialogue with FDA early in the development process, and frequently as development progresses
• Establishing more dialogue between FDA and the European Medicines Agency, to enhance harmonization of regulations
• Obtaining more clarification from FDA regarding the types and levels of evidence needed for combination therapies
• Getting better guidance from FDA on how sponsors should best interact with multiple FDA offices involved in combination product development
FDA recently released a draft guidance for industry on the codevelopment of two or more unmarketed investigational drugs for use in combination.1 At the conference, Drs. Sherman and Temple discussed the major premises on which this guidance is based. These premises include the need for combination therapy, the agency’s flexibility in ascertaining the contributions of individual agents in a combination, the need to demonstrate the biological rationale for the combination, and the case-specific nature of IND submissions and labeling issues for which FDA encourages sponsors to consult with the agency as early as possible and frequently throughout the development process (Woodcock et al., 2011).
Drs. Saber, Temple, and Sherman all stressed that FDA recognizes the need for combination therapy and has experience with regulating combination therapy with investigational agents, not just for cancer, but for many other diseases, including infectious diseases and hypertension. Dr. Sherman added that there are times when it is even unethical or impossible to study agents as single therapies because the agents are much more likely to be effective in combination due to the rapid development of resistance to single therapy or other factors. “With hepatitis C, resistance develops within days, so it’s only possible to expose patients with
1 See http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM236669.pdf (accessed December 14, 2011).
the disease for literally under 3 days. One can gain a little bit of information about each compound, but not much,” Dr. Sherman said. Dr. Temple added, “If two drugs together do something wonderful, who wants to be randomized to the trial to see which component contributed?” He also noted that often a new drug has to be added to another therapy that could not be omitted in a trial because it is the standard treatment with some documented efficacy, which makes it difficult to assess the contribution of the two therapies used in combination. In those cases, the agency has the label for the new drug state that it should be used in combination with the older drug. “We don’t know whether the older drug is still necessary, but we live with it because what else can you possibly do?”
Dr. Sherman stressed that there should be a compelling biological rationale for the combination for FDA to consider approving it. “That doesn’t necessarily mean there needs to be a greater than additive effect,” she said. However, she cautioned that “perpetuating products that definitely are toxic and don’t do much more than hinder the disease progression a tiny bit is not very interesting and not very patient friendly. Patients are not interested in hope and promises, but in things that work,” Dr. Sherman said. She concluded her presentation by stating that “a clear regulatory pathway [for combination therapy] exists.”
SHOWING THE CONTRIBUTION OF EACH DRUG IN A COMBINATION
Dr. Dagher raised the issue that even though the new FDA draft guidance is not for fixed-dose combination products, for which one would have to show the contribution of each compound used in the combination, it still seemed to suggest that sponsors would have to show the contribution of each drug used in combination to some degree, although perhaps not in a clinical trial. Dr. Pazdur responded by saying that a clinical trial with four separate arms was not going to be needed to show the rationale for combining drugs, but rather compelling preclinical information or results from Phase II trials or related information, such as relevant information about other members of the drug class that are already approved. Dr. Temple added that “there are going to be cancer therapies where most of the action comes from hitting one receptor and you are putting another compound in there to deal with the resistant tumors that grow later. Those are very hard to prove in a clinical trial, but you will have laboratory evidence that shows the resistant cells are now killed off by the added drug.”
When one participant asked if sponsors can show the contribution of the agents in separate trials and not within the same trial, Dr. Sherman responded that it depended on the totality of the evidence. “It could be separate trials, separate populations, some pharmacokinetic results in
healthy volunteers, or just preclinical evidence,” she said. Dr. Pazdur added that a lot of the FDA’s decision making in this regard is going to be results driven. “We are looking for big results here. If you have a single-arm trial of drug A that has a 5 percent response rate, a single-arm trial of B that has a 10 percent rate, and you add these drugs together and, with reasonable numbers of patients, you have a 60 percent response rate, we’ll take that,” he said. Dr. Temple concurred that FDA would not require a long-term study of the single agents if they worked so well together.
But building onto Dr. Pazdur’s example, Dr. Temple added that if there is a 10 percent response rate for each of the agents and only a 15 percent response rate for the combination, there is little evidence that the combination is significantly better than the single agents and a factorial study that can separate the effects of each agent in the combination may be required by FDA. “Unless you [dramatically] save people’s lives, we still want to know what the contribution of each is, because there is a downside—there is toxicity from each of them so you are paying something of a price [to combine them],” he said, adding that the new draft guidance discusses the flexibility allowed in the sources of information that show the contributory effects. “We are planning to use our heads, the lab, animal data—a wide variety of sources to determine whether there is really a significant contribution,” he said. When Dr. Ellen Sigal, chair and founder of Friends of Cancer Research, pressed him to define what a “significant” effect is—whether it’s more than 50 percent, for example—Dr. Pazdur responded, “That’s difficult to answer because it depends on the safety of the drug as well as what available therapies are around to treat that condition, and what endpoints they have been approved on or have shown in clinical practice.”
EFFECT OF COMBINED TOXICITY ON SINGLE-AGENT APPROVAL
Dr. Schlom said that a major industry concern is that if two drugs that show minimal toxicity in Phase II or III trials are combined and then elicit major toxicity, neither drug will be approved by FDA as single agents. “There is no guidance about this and if FDA says it will decide this case by case we go back to the same old story—we don’t know what the FDA is going to do,” Dr. Schlom said. Dr. Temple added that for this scenario to happen, it would be due to something that was surprising and unexpected, and that sponsors might be required to assess which of the drugs might be linked to the toxicity. He gave an example of two drugs used in combination that caused hepatotoxicity that only surfaced after 2,000 people were tested. For such a situation, the sponsor would be asked to examine the transaminase levels in response to each of the drugs and
assess whether one of them causes abnormal levels. “It wouldn’t be that hard to figure out which was the troublesome drug. If something really weird and terrible were to emerge only when you use the combination, you would have to sort out what did it. You would be worried about each of them having that side effect unless you had a cogent explanation,” Dr. Temple said. But he added that wouldn’t necessarily mean that approval would be denied for the single agents.
Dr. Saber said that “combinations are no different than single agents in terms of proof-of-concept and toxicology studies.” She noted that FDA is flexible when it comes to life-threatening conditions and is willing to accept some toxicity preclinically. “We do allow compounds to go forward in a clinical trial knowing they are toxic. We balance the toxicity against benefit,” she said. Dr. Sherman added, “With serious and life-threatening conditions, patients and their physicians are willing to take greater risk. This is a basic tenet of how the agency functions, and it applies here.” Dr. Pazdur stressed that for cancer treatments, efficacy is the major hurdle to cross, not toxicity, because FDA accepts a high degree of toxicity for cancer treatments and thus excessive concern about this issue is unwarranted.
There was some discussion about how many INDs are needed for combination therapies. “Probably more than one. We’re working on that,” Dr. Temple said. He added that if the drugs used in the combination are in different divisions of FDA, each division will probably want to review its own IND, but if they are only being studied together and not singly, one IND may suffice. Dr. Pazdur pointed out that frequently the drugs may be initially studied together, but later might be studied individually, and Dr. Temple said that in those cases another IND will be required for the single-use study. Dr. June pointed out that few cell-based immune therapies for cancer are FDA approved, which means multiple INDs are required to test them in combination, and coordinating those multiple INDs is challenging, especially with regard to ascertaining who holds the IND and how indemnification will be conducted because there are different criteria for this at state institutions, universities, and government. “We have had to develop best practices for each of these to try to get trials approved,” Dr. June said, and added that researchers often seek feedback from FDA in pre-IND meetings.
LABELING COMBINATION THERAPIES
Dr. Temple pointed out that the guidance recognizes that there will not be much dose–response information for each drug used in an inves-
tigational combination therapy, but suggests sponsors acquire as much of this information as early as possible. As for labeling requirements, he noted that sponsors can copackage their product if they want, but he assumed most would rather package their own product and have their own label. The label should specify the drugs with which the compound has been tested in combination and what those results were, and probably will say the drug should be used in combination, according to Dr. Temple. He added, however, that although Herceptin (trastuzumab) was designed for use with certain drugs with which it was tested in combination, “that didn’t mean [a physician] couldn’t add it to something else and it doesn’t mean they actually had to use the other drug.”
However, Dr. Pazdur urged sponsors to be cautious about what treatments they test in combination because the drug will be usually be labeled with that combination indicated, even if the new drug was tested in combination with a standard therapy that was not very effective. He gave the example of the new drugs for melanoma being tested with dacarbazine (an alkylating agent), which he labeled “a toxic placebo” because it is a relatively ineffective drug for melanoma. “I really couldn’t understand why people wanted to marry their new drug that had very impressive response rates or potentially a survival advantage with a relatively ineffective drug,” he said.
COORDINATION OF DIAGNOSTIC AND THERAPEUTIC REGULATION
Another regulatory challenge for combination therapy is the coordination between the different divisions in the FDA when diagnostic devices and drugs are used in the same clinical trial. Dr. Papadimitrakopoulou noted that regulation was a major issue for the innovative clinical trial BATTLE 2 because it involved investigational diagnostic biomarker assays in addition to investigational new drug agents. This required oversight from CDER as well as the Center for Devices and Radiological Health (CDRH), and together they required a much larger amount of data than normally would be needed for a proof-of-concept trial, which BATTLE 2 exemplified (see Appendix A). “There’s a lot of novelty in these trials that creates fears about how we can do them,” she said.
Dr. Temple stressed that such trials are becoming increasingly more common in oncology because it is desirable to develop a companion diagnostic that identifies patient responders, thereby preventing patients from being exposed to a drug to which they cannot respond. Dr. Pazdur agreed that companion diagnostics should be an integral part of drug development and should be explored preclinically when developing a new drug. “It’s new ground for pharmaceutical companies because they must have
the interactions with either the in vitro diagnostic companies or develop that in-house expertise in itself,” he said. It has been challenging for FDA as well, he added, noting that FDA staff in its oncology drug division meet with FDA staff in its in vitro diagnostic division for “every application to have a pathway for the in vitro diagnostic either to be approved or an alternative pathway to have the marker tested at the time of approval.” Several pathways could be considered for this, he said, including 510(k) clearance for the diagnostic.
But Dr. Sigal noted that the time frames for developing a drug and companion diagnostic are not always parallel, to which Dr. Pazdur responded that for a life-threatening illness such as cancer, “if there is not a companion diagnostic for the drug, they will find a way around it to approve that drug. It may be using a university-based test while an in vitro diagnostic is being developed.” He added, “This is an issue we have to really work on and develop with the pharmaceutical industry. This is new ground that requires attention because it is an integral part of the drug development program.”
FDA VERSUS EUROPEAN REGULATION
Dr. Canetta pointed out that copackaging is not legal in Europe and in certain other countries. “For people who do global development, that is a big obstacle,” he said. Dr. Dagher added that “whether it’s combinations or not, drug development is a global activity.” He said that the only guidance available from the European Medicines Agency (EMA) is focused on fixed-dose combinations, and he requested that “as you further develop thinking and finalizing this guidance and/or your thoughts on other rules that may be rewritten or written, any discussion with the EMA on the thinking would be helpful. We know that you can’t always get entirely unanimous thinking. But from a global perspective it would be useful.”
Dr. Canetta noted that the regulatory philosophy in Europe differs substantially from that in the United States. With regard to companion diagnostics, for example, Dr. Canetta said that Europe requires diagnostic tests that are reproducible, stable, and valued in itself, and “what the physician does with the test in Europe is not considered an EMA issue. It’s something that the physician can utilize as a tool.” However, in the United States, the diagnostic test has to be proven by validating it in a clinical setting in a particular trial in order for it to appear on the label with a companion drug. Dr. Hohneker added, “In Europe, the industry is not viewed by the investigator community as the evil empire,” and instead there is more interdependence of the investigator community and industry. Because of European regulations, the pharmaceutical industry provides a lot of support to do clinical trials, whereas in the United States,
“Pharma is not wanted—they just want our money or our drugs, but they don’t want us at the table because they claim it taints the collaboration or the investigation and we have to get around that. Pharma does have something to offer, both intellectually and also scientifically to the discussion and collaborations. If we really want to level the playing field between the U.S. and Europe, we have to think about the cultural barriers that exist in the U.S. that don’t in Europe.”
The FDA draft guidance is mainly for small molecules and not for vaccines and other products. But Dr. Dagher pointed out that there are a lot of good principles in the guidance that could apply to other kinds of products such as antibodies combined with drugs or vaccines. Dr. Sherman responded that the guidance was directed toward small molecules used in combination because that was what industry asked FDA to address, but the same principles could be applied to other types of treatments used in combination.
