When the White House proposed a $1 billion startup fund for the National Cancer Moonshot, a largely unexpected directive to reform FDA raised many questions among oncology insiders.
The agency will create a virtual Oncology Center of Excellence, the administration proposals and budget documents state.
Alas, nobody can claim to understand what “virtual” means in this context, and how the $75 million in proposed fiscal 2017 mandatory funds would be used to “leverage the combined skills of regulatory scientists and reviewers with expertise in drugs, biologics, and devices.” (The Cancer Letter, .)
Some perplexed insiders suggest that virtual could translate as “we don’t know exactly what we want right at this moment,” in Washington-speak.
“First of all, there’s nothing wrong, and FDA is doing a great job with what they have,” said Ellen Sigal, chair and founder of Friends of Cancer Research, the advocacy group that convened a panel discussion to explore a way forward for the regulatory agency.
“People at Center for Devices and Radiological Health, Center for Drug Evaluation and Research and Center for Biologics Evaluation and Research are all doing a great job. But, the question is, can it be better?” said Sigal at the Capitol Hill briefing Feb. 24. “Patients get diseases, they don’t get a biologic; they don’t get a device. It all works together. Is there a more efficient way that is better? Because the goal is to get patients better treatments. This is what it’s about.
“It isn’t about institutional structures; it’s not about anyone doing a bad job. Is there an efficient, better way to serve the patient towards the goal of really integrating these centers of people in their particular fields working together?”
Panel members—representing industry, FDA, academic oncology and patient advocacy—agreed that the time has come to reassess the way medical products are regulated at FDA.
The panel included:
• Steve Galson, senior vice president of global regulatory affairs and safety at Amgen Inc., former director of CDER, and former acting Surgeon General.
• Mark McClellan, director of the Duke-Robert J. Margolis Center for Health Policy at Duke University, and former FDA and CMS commissioner.
• George Demetri, director of the Ludwig Center at Harvard Medical School.
• Otis Brawley, chief medical officer of the American Cancer Society.
FDA Structure Needs to Change
“The modern world of drug development actually doesn’t fit the existing structure of FDA’s drug-device-biologics organization structure, that has been—and continues to be—remarkably successful in making the U.S. and the FDA the gold standard globally,” Galson said.
“Working with a global pharmaceutical company, I see this even more than I did when I was at FDA,” he said. “Products typically receive multi-center reviews—many, many products—and I’ve been working closely with the agency on these since I’ve been there. FDA has recognized these product changes, so I don’t mean to imply any criticism of the agency not being aware of all this.
“FDA’s evaluated and made what I call ‘tweaking’ types of organizational changes, made sure that synergistic collaborations were in place. But there’s recognition at FDA and by external collaborators that there’s really a need for further assessment of this. In this world, with exciting products coming at an increasing pace, is this really the right structure?
“It’s time to look at this again.”
FDA’s handling of biologics is a good example of how the agency can improve, Brawley said.
“Science is evolving because things are changing, definitions of cancer are changing. We’re using molecular diagnostics along with drugs. It makes sense that the FDA might become a little more efficient in it’s ability to review by a slight reorganization,” Brawley said.
“I in no way want to criticize what the FDA has done in the last 15 years. My experience on the Oncologic Drug Advisory Committee is that the system is actually working very well. The system can work better, but the system is working very well,” Brawley said. “Biologics integration was incredibly important, and I think that we could move further with even more integration.
“It’s important, whenever you have an organization—the FDA is no different from any other—that every few years you look back and you examine and you say, ‘What can we tweak, what can we change, what can we make more efficient?’ That, in my mind, is effective.
“And in many respects, that’s what we’re doing right now.”
Sigal said everyone at FDA—including the newly confirmed commissioner, Robert Califf—is in agreement that the agency needs to come up with a disease-oriented approach to development of medical products.
“This concept, basically, has everyone in agreement, clearly, from FDA, from Dr. Califf to Dr. [Janet] Woodcock, [director of CDER]. There’s no question that conceptually people agree,” Sigal said. “It’s in the moonshot. We’re very happy about that. It may be in some legislative language, but what does it mean?
“Clearly, FDA ultimately has the last word on it. They have to look at it and figure it out—but a large community is impacted by what they do, and that means the developer community, the patient community, the academic community—everyone.
“So some external input into this is really crucial.”
It’s unclear how structural changes at FDA would be implemented—Congress generally doesn’t intervene in intra-agency matters, which means it could all be up to Califf and FDA leadership.
“Does it have to be legislative? I don’t know. I assume there may be some statute issues, because of the way they’re regulated at CDRH, but we don’t want Congress telling us how to be or how to organize,” Sigal said. “There may be some issues that have to be done, but I think it’s absolutely critical to have external input, patient input, and expert thought from professional societies and others, so when the decisions are made, people understand it.”
Restructuring FDA would likely require moving desks, which means that there would have to be discussion about whether the oncology center will be virtual or brick-and-mortar, Sigal said.
“This center, this integration, has to be real. It can’t be ‘Let’s all be friends and talk.’ At some point, it has to have some teeth,” Sigal said. “To do it, it has to be thoughtful, and it has to be done in a way that is real. It should not be disruptive. It should have the resources it needs. It can’t be as if we’re going to be talking to one another as we are talking to one another now. We need to go one step further.
“I also will say that it will need resources. This can’t be done without resources. We have to attract people, we have to keep people; there will be the ability to figure out whether it needs bricks and mortar. Probably not, but will some people have to move over? Probably yes, but it’s a decision that should be done quickly but thoughtfully with a lot of input from people. But ultimately, it’s a decision made by the FDA, but not alone.
“We’re only talking about clinical. We’re not talking about manufacturing, legal, or all the complicated things. We’re just talking about the integration in a way that’s more meaningful, that will get to the right result for patients. And that’s what this is about, and it is our core belief that this can be done. It should be done thoughtfully, it shouldn’t be disruptive, it should be done with resources, and it should also recognize that the agency is doing a very good job, but can we do better? My answer is, ‘Yes.’”
Should Oncology Go First?
Is cancer the right place to begin a discussion about integration at FDA? Or should this effort be coordinated across disease types?
Sigal said oncology is the place to start, because it has strong leadership.
“I do want to suggest that when we did approach FDA with this—some trepidation but with conviction—the leadership really thought that cardiovascular, maybe infectious disease, could really work in other diseases,” Sigal said. “Because people get disease and the diagnostics and biologics, they’re all working together with the drugs, so I think the issue is to be thoughtful: Is the leadership there? Are the resources there? How do you do it with the issues of the statute?
“We happen to have extraordinary leadership with Rick Pazdur [director of the FDA Office Hematology and Oncology Drug Products] and what he’s done with any cancer. There are just huge opportunities, so maybe that can go first, but there certainly are other diseases. We’ll leave that to the other experts. At the moment, we’re suggesting that cancer can be the first place.”
McClellan said combination immunotherapies might be an area where promising progress can be made and where broader integration can be achieved.
“I really think it depends,” McClellan said. “I think we need to do a bit more work building on the Friends proposal. It’s always legislative and statutory considerations about what FDA can and can’t do, and I’m familiar enough with FDA law to know that I need good lawyers to make sure that we’re handling that appropriately.
“But most important, from the scientific and medical product development standpoint, where are the biggest real challenges? Let’s get real about moving from the concepts of complex medicine into the practicalities of a regulatory agency with limited resources.
“I think it would be very helpful to focus on some key areas where the challenges that you all have been talking about on the panel really seem to be greatest, clearest, most pressing. Steve mentioned combination immunotherapies—my sense of one area where there actually might be a number of applications, not just in cancer, but in a combination of products more generally, especially companion diagnostic products.
“This is an area that FDA has been thinking about. There’s some cross-center structures to help coordinate action, but given the rapid growth in the development of targeted treatments and how well it fits with issues like the whole focus of the Cancer Moonshot and the president’s Precision Medicine Initiative—on furthering the use of targeted therapies fits within the Breakthrough Designation—that may be a really good place to start, and start asking the questions of ‘What’s the best way forward in the short term, and maybe the longer term?’
“I like some of the models like these cross center councils that have some significant involvement from the commissioner’s office, or that different centers believe that there is something to be gained from working a bit more formally together short of a restructuring, short of a fundamental restructuring. Maybe something like that is stuff that can be explored in the short term while these bigger issues are getting addressed.”
There are tremendous opportunities in cancer, especially since oncology is well developed as a system, Demetri said.
“Not that cancer is so different—I like the idea—I don’t want to have an internecine medical warfare between cancer and every other disease that afflicts humankind,” Demetri said. “I actually think cancer is just a good proof-of-concept place to start exploring the concepts, and where we might reorganize the FDA for the good of everybody, because we already have protocols and treatments that are based on a standard pharmacy, on a cellular pharmacy with the kind of engineered cellular agents that are causing so much excitement in our field, with novel diagnostics, and everything that hits each and every one of those different branches of the FDA right now.
“So bringing them together in a way that works and testing that out in cancer, first, might make a lot of sense, also because we have a history of making advances in cancer, from prevention to cures, through our comprehensive cancer type of way of doing research. This was a novel way of structuring research back in the 60s, of having basic scientists and clinical scientists work together under the umbrella of comprehensive cancer centers. We did that as a country because we felt that cancer lent itself to that kind of a solution.
“I actually think we’re now seeing our neuroscientists say, ‘Yeah, that makes sense.’ We should develop neuroscience centers and cardiovascular centers. That model has worked. And frankly we see the London Cancer Center—that’s being replicated around the world because it’s effective. So I think we have many reasons to think this kind of organization would be good for the FDA.”
“I do want to highlight the need, when you’re thinking about organizational issues at FDA especially, to keep in mind that the resources are limited,” McClellan said. “There is a tremendous amount of expertise in cancer at the agency, but looking at the scheme of things, it’s really not that many people, compared to the vast and increasingly complex science and expertise involved and so forth. You really want those resources to go as far as possible.
“Not to defend the status quo, but there are some good efficiency reasons for having the same expert regulatory staff dealing with products across a number of areas. I think that’s where this point of really looking at where this bang-for-the-buck is going to be, in terms of a more integrated structure, is very important.
“I’ve heard that, especially in drug and diagnostic combinations, despite the FDA efforts, there really are some challenges in getting the same answer, based on the coordinated sum total of clinical expertise in this increasingly important and complex area. That might be a place to start, as opposed to radiation therapy or surgical activities.
“I hope we can look more closely at managing the benefits and risk and doing it as efficiently as possible. And then maybe, after we take that on, and Friends takes on these issues around cancer at FDA, we can move to efficiency of cancer product development and the academic medical role.”
The melding of elements in oncology is moving very rapidly in academic medicine and in the cancer centers, Brawley said.
“If I can, I’ll back up what George just said,” Brawley said.
“But as we look forward, all these new drugs that we have that end in 'ib' and 'ab,' virtually every one of them has a molecular marker associated with it. I do the test on the cancer, does it have a molecular marker, then I figure out which 'ib' or which 'ab' I prescribe to the patient. And that’s true for a number of drugs. Indeed, when we talked about biologics, we are now starting to talk about CAR. We talk about designer antibodies, used to treat cancers.
“More and more, we need to have the regulators who do the diagnostics and markers working and collaborating with the regulators who do the treatment, because, in my world and in George’s world, they are becoming the same very quickly. This is the new biology, this is the new 21st century definition of cancer.”
McClellan: Don’t Just Tear Everything Up
FDA shouldn’t start to move people and desks without “serious, thoughtful consideration,” McClellan said.
“When you think about managing this organization, you’ve got an enormous range of activities and an increasingly diverse and complex products to regulate with staff that are accordingly stretched in meeting all of these demands,” McClellan said. “On the one hand, you want to have very effective, knowledgeable review. On the other hand, it needs to be predictable so that companies, product developers going through the process know what to expect.”
FDA’s current infrastructure was developed in response to the needs of an era when products fit into distinct, independent regulatory pathways—a system that no longer applies to many products on the market now, said McClellan.
“Historically, that’s meant this division by types of products—that’s the way the legislation is set up—historically, it’s been for products that don’t cross boundaries. That’s a pretty efficient way to have limited staff predictability go across a range of different development areas,” McClellan said. “But as FDA’s found over the years, there are more and more products that don’t fit neatly into that kind of arrangement. On one hand, you want to keep encouraging the kinds of activities that, say, CDRH has undertaken over past year with their strategic plan to make device regulation more efficient and predictable. There are initiatives at CDER and CBER, all of which have made a difference, and which have been backed by legislation, like the 2012 Breakthrough Therapy Designation that Friends and our group was involved in. But no question, these challenges are getting bigger.”
Informal coordination may work for some products, McClellan said, but a systematic relationship is needed for others.
“When FDA tries to recognize and respond to these issues, there are a lot of different levels of response. And for many of these products there can be some kind of informal coordination across the staff of different offices or different centers in the FDA, and that helps,” McClellan said. “But for many of these product areas, it’s better have more of a systematic kind of relationship in such areas as combination products, for example, there is kind of a more formal coordination process in place intended to give the same kinds of answer at the same time for products that need to be developed together.
“Where that’s not sufficient, FDA has taken further steps and set up more commissioner-level initiatives. These would have support, leadership from the office of the commissioner to really help provide more structure and formal process for helping to make sure that the different parts in the agency do work together where they need to.
“So I think there are a range of options that have been used and can have more significant effects around structure and organization in the agency and it’s very important to manage those with statutory and legislative authority issues and with the scientific issues that drive the most efficient way to handle these kinds of cross-cutting topics.
“My sense from that experience is that the best place to start in these issues is trying to get a sense of the priority list of problems: what are the biggest particular areas around, say, combination or complex product development in the areas of cancer or neurologic conditions or the like where it seems the benefits of new structure most likely outweigh the risks and try to start there and help that guide you in what further steps should be undertaken.”
Demetri: The Devil is in the Details
The challenge in helping FDA become more disease-specific lies in the creation of a system that will be relevant, for instance, to the complexities of cancer, Demetri said.
“We live in an ecosystem where the science has never been better. And the science has shown us the complexity of what we’re dealing with—cancer is not one disease, cancer is thousands of different diseases,” Demetri said. “With thousands of different diseases, you have a risk profile, that goes from something that people can live with forever such as certain carcinoma-type diagnoses that are eminently curable to the most life-threatening acute leukemias that can kill people in six weeks but are also curable to a great extent with the use of modern technology.
“So I think that’s the complexity of this, and since we’re here to talk about organizational issues, I think that’s the big deal. How can you get the kind of expertise to match, as Steve said, the right drug to the right patient, at the right time, in the right disease, with the right risk, at the right cost (eventually), so that the public benefits and so that the scientists are not left to their own devices, developing things that can’t be applied effectively, and so that the regulators can have the tools at their disposal to use the kind of tools that they have to reliably, appropriately protect the public health and really guide the practice of modern medicine. I think that’s what’s really important.
“Separate from drug development, don’t forget pharmacovigilance and effectiveness—we have a responsibility to say if we developed a drug, ‘Is it really working outside of the rarified atmosphere of the clinical trials that we test things in?’ The ability to really link that together in a different kind of a structure seems like a noble goal. I think, organizationally, it’s possible.”
Moving FDA toward a disease-oriented system is aligned with Vice President Joe Biden’s goals of breaking down silos in oncology bioinformatics, Demetri said.
“I think the idea that is aligned with the Moonshot Initiative that’s real, is this data sharing as well,” Demetri. “Academics, traditionally, hoard data. We are trying very hard to break that culture and share data. We have to change our academic credit systems and we have to do an awful lot to do that.
“But in the same way, I am worried about different groups in the FDA. When I’ve talked to different, wonderful people at the FDA or CDER, or CBER, or in the diagnostics division, I do worry that they may be playing with different decks of cards. That’s the concern I have, and where there could be an easy way of merging that, or at least having some sort of organization so we are all playing with the same deck of cards.
“In the same way, this is the perfect time because things like the AACR Project GENIE is getting places like Dana-Farber, Memorial Sloan Kettering and international institutions to share data together on things as simple as genomics—and they’re not simple, but they sound easy and it’s just a couple of syllables. Genomics is tough, but so is linking genomics to clinical outcomes.
“So if our academic places are having trouble with all the computational biology help we have in all the high-powered help of the Broad [Institute] and Harvard and everything, think of how we could then share that with the agency. I think different structures need to be elaborated so that we can actually serve the country and the patients we serve the best possible way.”
Harvard’s consortia for oncology provide a good analogy for how FDA can bring together its separate divisions to work on specific diseases, Demetri said.
“At our cancer center, I’m one of the associate directors for clinical science at the Dana-Farber Harvard Cancer Center, we’re a consortium cancer center because there is no ‘Harvard Hospital,’ Demetri said. There are many big famous teaching hospitals like the Brigham & Women’s, the Dana-Farber, the Beth Israel, the Mass General, and others, all of which do cancer, so our consortium brought us together and said ‘How can working together, in a structure that makes sense, help us do our work?’ and in many ways I use this analogy for where I see a thoughtful way of helping the FDA.
“How can we not break something apart, but how can we bring things together in a rational organization so that we’re not duplicating resources and wasting them? So that we actually have the right expertise at the right time to put the right regulatory framework on top, so that we can predict the outcomes and really help everybody along the line—the scientist, the development people, the clinical trialists, the regulators, everybody could benefit from this.
“I think the devil is going to be in the details about how to do this, and I really appreciate Mark’s comments about that, because I have great respect for our colleagues at the FDA. They have always been very rational, very helpful to me as a clinical investigator, as somebody who goes to them with good data. We can share the good data, or as somebody who comes to them with not-so-good data, we can share the data and discuss what’s the next step.
“That kind of back-and-forth collaboration is what, as practicing physician as well as a clinical investigator, we’d like to see and help with at the FDA as we think about how we can build this structure in a way that makes the most sense. I do feel bad for them because the resources are so darn limited and I think that’s the other important thing. We want to do this in an effective way and have this be efficient for our colleagues there as well. They serve a vital public purpose.”
Sarah Pavlovna Goldberg contributed to this story.