- Speech by
Scott Gottlieb, M.D.
Commissioner of Food and Drugs - Food and Drug Administration ( May 2017 - April 2019 )
Remarks by Scott Gottlieb, M.D.
Commissioner of Food and Drugs
American Society of Clinical Oncology (ASCO) Annual Meeting
Good morning. I’m honored to be here, and to share some of the work we’re doing at FDA on behalf of my colleagues.
I want to thank ASCO for the opportunity to speak here. We’ve had a long and productive dialogue with this organization on many issues related to cancer treatment and prevention.
Today, I want to talk about the steps we’re taking at FDA to advance treatments for cancer. And I want to highlight some of the new initiatives we’re pursuing across the continuum of drug development – from clinical trial recruitment, to the regulatory evaluation of data, and to our work in the post market setting.
At every stage of drug development, we’re advancing new policies to make the process more modern, more scientifically rigorous, and more efficient. The scientific opportunities we’re seeing demand that we make sure our policies are as sophisticated as the treatments that are being developed.
This science is bringing forward more novel opportunities to more meaningfully address human disease at a pace that’s faster than ever before. But humankind is still much closer to the beginning than to the end of these endeavors.
And we are at a unique moment, a singular point in time, where we can bring to patients battling cancer -- now or in the future-- advances that’ll quite literally alter the trajectory of their lives.
People try to paint any change we make in our regulatory policy, or to the policy requirements we impose, as a binary choice between speed and safety. That’s a false dichotomy.
This isn’t a zero-sum game. Our goal is to make sure that our regulatory policies are modern and efficient so that we’re getting the most regulatory bang for bucks spent on clinical development in terms of a greater assurance of safety and benefit.
These costs aren’t just the direct costs of the trials.
They’re also the considerable costs placed on patients and providers as part of the development process, as well as the costs of opportunities forgone by where we choose to place our focus, and how much time it takes us to do these things.
If we’re adopting the right policies, if we’re focused on incorporating the best science, and making our own processes more efficient, we can achieve greater certainty around the parameters that matter most to providers and patients.
And yes, we can make the overall process itself less costly, less risky, and less time consuming.
In this way, we can lower the barriers to bringing new science forward with the ultimate goal of making sure more patients can benefit sooner from new advances.
I want to focus my remarks on these efforts, and announce some of the new steps we’re taking. But I want to begin with the basics when it comes to reducing the suffering and death from cancer. And that starts with reducing smoking rates.
The bottom line is this: Short of perhaps discovering a universal cure for cancer, there’s no single intervention or advance I can achieve on my watch at FDA, or over any reasonable period of time, that’ll have more impact on reducing the death and suffering from cancer than if I’m able to sharply reduce peoples’ use of combustible tobacco -- principally cigarettes.
This has become one of our principal obligations at FDA, and a key part of my efforts as Commissioner.
As part of the comprehensive plan we announced last year, we’re advancing a process to regulate the nicotine levels in cigarettes to render them minimally or non-addictive.
If cigarettes can no longer induct and sustain addiction, then our analysis shows that overall smoking rates in this nation will sharply decline. And as a consequence, the generational impact -- in terms of life years gained -- is enormous.
At the same time, we’re taking steps to help foster new technology that may offer adult smokers alternatives to access to satisfying levels of nicotine, but without all the harmful effects of combustion. These products must be put through an appropriate series of regulatory gates. And this includes e-cigarettes.
That said, we know that a lot of kids are using e-cigs. And we’re deeply troubled by these trends.
If by opening a path for e-cigs to be an alternative for adult smokers, all we end up doing is hooking a new generation of kids on nicotine, we’ll have failed in our purpose. We’ll have swapped one public health tragedy for a new one.
I don’t intend to let that happen.
So, we’re taking aggressive steps to bring enforcement actions against those who sell e-cigs to kids. And we’re focusing on companies that market these products in ways that are meant to appeal to teenagers. And I’ve been clear to the e-cig industry that we’re going to hold them accountable at every turn.
The companies that market these products can’t expect to build a sustainable business model that lets large numbers of kids get hooked on their products. These e-cig companies have a chance to do something about it. The window is open.
But it won’t be open for very long.
They better step up and step up soon -- to address these trends along with us. So far, I must say, I’ve mostly been disappointed by the tepid response from companies that know that a meaningful portion of their sales are being derived from kids.
They need to look at this as a matter of urgent public health and not a PR challenge. It’s more than their business model that’s on the line. It’s the lives of American kids.
Even if we’re successful at making more progress against the public health basics -- dramatically lowering smoking rates, improving nutrition, and increasing vaccination rates -- people will still develop cancer. And people will still die from it.
But there’s more reason than ever for more cancer patients to expect to live longer lives, and to have a greater shot at a cure.
Oncology is leading the precision medicine revolution through the rise of rational oncology drug development programs based on advances in systems biology and the redefinition of tumor classification. It’s moving from a field based on tissue of origin towards one based on molecular aberrations such as ALK rearrangements, EGFR mutations and, more recently, microsatellite instability or mismatch repair deficiency.
Our improved understanding of basic cancer biology and the tumor microenvironment has led to tailored therapeutics demonstrating impressive clinical responses in very early stage clinical trials, and in biomarker enriched populations.
I’ve never been more optimistic about our ability to extend and improve the lives of patients facing a cancer diagnosis.
But to harness these opportunities in science and technology, FDA must create structures and processes that are able to keep pace with medical innovation. And it means building a dynamic regulatory environment across the entire cycle of product development.
A little more than a year ago, the FDA established the Oncology Center of Excellence (OCE) to break down outdated silos based on the type of product application, and to bring a disease-focused approach to our oncology work.
The OCE model reflects the networked disease-focused teams that we’re seeing among our industry and academic partners. It brings together multiple disciplines to better understand the biological complexity of cancer.
This new alignment has given us a broader platform to advance the science related to how new drugs are developed.
And those drugs are increasingly complex.
In 2017, we approved 16 new oncology drugs and biologics, including the first two cell based CAR-T cell therapies.
We approved 30 new indications for existing drugs and biologics, the first two oncology biosimilars, and three NGS platforms that can scan for hundreds of tumor gene variants to help match patients with therapies or promising clinical trials.
We also recently approved the first tissue-agnostic cancer treatment. It was based on a biomarker that’s associated with a specific DNA repair pathway. Rather than requiring a separate development program for each disease site -- which may have taken many years and would likely never have happened at all for some of the rarer tumors -- we instead created a single therapeutic approach based on a solid understanding of the underlying biology of microsatellite instability. It was based on an observation of treatment effects in a collection of diverse tumors that each had a common biological feature.
There are critics who say we should hold drugs back from the market, and demand more pre-market studies proving overall survival endpoints, before we consider approving new drug.
I disagree. And I suspect some of the patients who face long odds, for whom available therapy gives them just a slim chance of long-term survival, might also disagree.
I had Hodgkin lymphoma. I had a very curable tumor. At the time of my diagnosis, I was told my odds of a cure were 90 percent or better. For me, available therapy was promising.
But to help me make better decisions on how to use the available drugs, I was searching for pristine studies that could get that 90 percent up a few more points.
So, I understand why demanding large, pristine studies ultimately serves the interest of patients like me.
But my situation was very different than being diagnosed with a cancer and being told your chance of surviving five years is 50 percent, or 30 percent, or just 10. Available therapy isn’t very promising if that’s your circumstance. And the ability to access novel treatments becomes more urgent in these circumstances.
Waiting three more years for another large, prospective, randomized trial to be completed – to confirm highly promising results already observed in an earlier clinical trial -- may not sound as compelling to the patient who faces these long odds.
We need a regulatory system that serves both these types of patients equally well. That’s the type of system we’re striving for. That’s the sort of system we are committed to having.
It means establishing a framework that breaks down artificial silos between clinical research and clinical practices.
It means harnessing real world data at the point of care to help every patient access the best care available.
It means enabling better access to experimental therapies being tested in clinical trials by modernizing eligibility criteria and conducting clinical trials in the communities where patients live.
In my opinion, this transformation isn’t optional.
The dysfunction of current drug pricing models is based upon an anachronistic construct for clinical trials -- where failure is typically expensive and routine, and success is the exception.
Without significant gains in productivity, the status quo is a path to financial, scientific, and clinical unsustainability.
While we’ve seen some movement away from traditional constructs in recent years, with efforts like Lung MAP, NCI-Match, GBM Agile, and ASCO’s Tapur trial, oncology trials have become largely artificial constructs -- capturing the experience of fewer than 5% of patients who can participate.
The generalizability of traditional clinical trials to real world patients at the point of care is increasingly hard. There’s a vast amount of biological space in cancer that we’re not probing.
And given the large number of cancer drugs in development -- and the need to optimize protocols incorporating new treatments, radiation, chemo, and surgery -- there literally isn’t enough time, capital, or patients, to test these approaches conventionally.
And so, we need to make the entire process more efficient without compromising scientific rigor.
I’d like to spend the rest of my time today detailing some of the new steps we’re taking to make the entire continuum of drug development more efficient as a way to lower costs and expand opportunities for innovation.
I want to announce some new steps we’re taking at each stage of the drug development and review process – from clinical trial recruitment, to application submission and evaluation, and to modernizing our approach to post-market manufacturing.
Let me start with recruitment and clinical trial design:
We know that traditional eligibility criteria often exclude the very patients most likely to be treated once the drug is on the market - the elderly, patients with poor performance status, organ dysfunction, brain metastasis, or other co- morbidities.
And even when patients are eligible, there may be geographic or financial barriers that prevent their participation.
Or patients may be understandably reluctant to break the therapeutic bond with their treating oncologist.
All of these things limit patients’ access to investigational agents and negatively impact accrual to oncology clinical trials.
We also know that a patient’s race, ethnicity, and geography have all been found to correlate with patient outcomes.
In 2018, a cancer patient’s hope for recovery shouldn’t hinge on their socioeconomic status or a zip code lottery.
The irony is that even with broad exclusion criteria, oncology trials still have a staggering failure rate. One recent study led by researchers at MIT found that in 2015, the overall success in cancer trials was only 8.3%, up from just 3.4% in 2005.
High failure rates contribute to industry’s high cost of drug development, putting upward pressure on drug prices.
It also reduces the amount of competition new drugs face in different drug categories. Too often, being first to market means being the only one on the market for a long time.
And those monopoly products extract monopoly prices.
The days are ending when most drug categories become quickly crowded with alternative products, providing competition that served to enable therapeutic variety and promote lower prices.
This is a public health concern.
It clearly compromises patient access.
One way to cut this Gordian knot is to harness the vast amount of data generated by routine patient interactions through pragmatic clinical trials at the point of care.
This approach can help us systematically ask and answer critical clinical questions and design better patient matching strategies to optimize medical care, trial design, and product development.
One criticism of this strategy is it appears to rationalize untutored patient experimentation.
But the reality is that extrapolating from median or average treatment effects and hazard ratios in highly selected patient groups may mean very little to a doctor trying to decide on the right course of treatment for a 56-year old diabetic with a history of lymphoma and invasive breast cancer. Or a patient with advanced metastatic melanoma and a history of autoimmune disease considering treatment with a checkpoint inhibitor.
What specific guidance can we offer these patients?
We can’t just ask them to wait.
At the FDA, we’re taking steps to advance our policies to embrace these opportunities. One specific step we’re taking is providing additional guidance to sponsors about how to include more under-represented patients in trials.
In fact, yesterday, we released one of those new guidance documents, focused on Considerations for the Inclusion of Adolescent Patients in Adult Oncology Trials.
Although most cancers in children and adults are distinctly different entities, there are some diseases that occur in both and span the adolescent age groups.
If there’s no evidence that an investigational drug might have exaggerated toxicity in younger patients, then we’re encouraging sponsors to enroll adolescents into disease appropriate trials.
Since the pharmacological parameters of adults and adolescents for most agents are comparable, early access to innovative drugs is warranted where there’s proof of principle and adequate dosing information to maximize potential for clinical benefit.
We’re also advancing new efforts to improve how we evaluate new applications as part of the clinical portion of review.
I’d like to announce two new pilots that we’re undertaking.
One is aimed at trying to focus submissions more squarely on data that’s most relevant to assessing safety and effectiveness, and not be obscured by excessive documentation or superfluous information that doesn’t inform key decision-making.
The goal is to improve the overall quality of applications and make sure resources and review time are being focused on evaluating data that’s most meaningful to clinicians and patients.
This means making sure that what gets included in applications counts -- and what counts most isn’t missing from submissions.
The worst possible time to discover this is after an application has been filed with FDA. It delays review and adds to the time and cost of the process, including the staffing burdens on FDA.
So, we’re moving more of the review of data up front, earlier in the process, before the application is even filed with the FDA.
The OCE is addressing these goals by piloting what we’re calling a real-time oncology review or RTOR.
Right now, this pilot is going to be focused on new efficacy supplements for already approved cancer drugs. If the pilot shows the efficiencies we expect, it could be expanded to drugs and biologics being considered for initial approval.
Our aim is to make sure applications contain the most relevant information to inform our review of a product’s safety and effectiveness, and that this data is analyzed appropriately.
Sometimes sponsors take an “everything but the kitchen sink” approach to their applications because they don’t have clear guidance on what’s most pertinent. Or the way they analyze data isn’t consistent with how FDA evaluates results. This can extend review times, and obscure the most relevant facts.
Under the new pilot, as soon as a sponsor locks their database, and has decided they want to file for FDA approval, they’d start sharing the bottom line data with the agency. In effect, FDA will pre-review the data and assess it for adequacy and completeness.
This would happen two to four weeks after the sponsor’s database is locked. This partial package submission will include key raw and derived datasets, including safety and efficacy tables and figures, the study protocol and amendments, and a draft of the package insert. The FDA will immediately start evaluating the data for its sufficiency and integrity.
This informed pre-analysis gives reviewers and sponsors an early opportunity to address data quality issues. The FDA can provide early feedback to a sponsor on the most effective way to analyze data to properly address key regulatory questions.
By the time the sponsor files the application with the FDA, the agency’s review team would already be very familiar with the data and the analysis. Review teams will be in a better position to conduct a more efficient, timelier, and thorough review.
And sponsors will have benefited from feedback on how to best analyze data to effectively evaluate key regulatory questions.
Based on our analysis, we believe that taking this approach can free up 10 to 30% of the reviewer’s time. This leaves more opportunity for engagement with product developers, and will lead to a more efficient review of drug applications.
Common pitfalls, like requests for the submission of additional data or analysis, can also be resolved more efficiently by including a planned scoping meeting and earlier opportunities for discussions on data. It can make sure applications are more thorough and complete at the time of their initial submission.
As we gain experience with this model, we believe that we can use this approach to create a dynamic data submission system that aligns the process around data quality based on early feedback and engagement with review teams. Sponsors and reviewers will share real-time feedback, where issues of data quality and integrity can be addressed early in the process.
This pilot is already underway. And a number of sponsors have already agreed to enter it. We’re in the process of evaluating multiple supplemental applications under this voluntary pilot.
These submissions cross multiple sponsors and cancer types.
As part of this real-time oncology review, we’re also piloting a new assessment aid. This tool is a voluntary submission form that the applicant uses to facilitate FDA’s assessment of the drug application. Right now, we’re using this assessment aid to also focus on the review of supplemental applications.
Under this templated approach to receiving applications, we’ll be using the sponsor’s structured file, which follows a new template format that we’re piloting, as the basis for our review.
So instead of writing a separate analysis of the file, we’ll use the sponsor’s own submission as the document for layering in our review. Under this approach, we’ll annotate the sponsor’s drug file with our assessment rather than creating a separate document that recapitulates many of the same data tables.
In practice, the template is divided into two parts: the applicant’s position, and the FDA’s assessment of that position. We’ll make note, right in the sponsor’s file, where we agree or disagree with the applicant, and add any additional findings of our analysis.
Applicants fill in their positions, and then send the document to FDA before or at the time of the submission. The FDA review team then adds the agency’s assessment to the same document.
This creates a more agile platform for reviewing data after the sponsor’s database is locked. The assessment aid reduces the administrative burden on FDA reviewers. This permits them to focus on key results and perform critical analyses that may have been omitted by the company. Importantly, the new format leads to a more dynamic review process where key regulatory questions can be answered more thoroughly and effectively.
If a drug application makes it through the process, then the completed review, using the Assessment Aid, will be presented at the meeting of the Oncology Drug Advisory Committee.
The annotated application will be presented as one single, combined background document that contains both the applicant’s opinion and the FDA’s analysis of those positions.
Under the pilot program, the Assessment Aid is only being used for supplemental applications. But, if it’s successful, we believe this approach could be expanded to original drugs and biologics.
Finally, we’re also taking new steps after products are approved, to make the manufacturing process more efficient.
For some of the most promising new products – in particular cell based medicines and gene therapies – the most complex issues can turn on product features related to their manufacturing.
So, we’re taking some new steps to help scale up manufacturing technologies for cell-based oncology therapeutics like genetically modified T cells, such as CAR-T.
In contrast to traditional drug review, where 80 percent of the review is focused on the clinical portion of a file, and maybe 20 percent is focused on the product issues, I’d say that this general principle is almost completely inverted when it comes to cell and gene therapy. The clinical efficacy is often initially established early, and sometimes in small series of patients.
It’s the product questions that are more complex and uncertain.
FDA’s Center for Biologics Evaluation and Research is working with the National Cancer Institute and other stakeholders to establish clear requirements for current Good Manufacturing Processes related to cell based manufacturing and how to optimize production processes in a new, standardized process.
One way to think of this new approach is similar to the way Master Protocols (MAPs) are used for clinical trials. Except in this case, manufacturers could rely on a common general set of manufacturing standards for cell based technologies.
Creating more standardization across the FDA’s requirements for manufacturing requirements could enable more facilities – like hospitals or research facilities – to scale up cell-based manufacturing, and expand local treatment options for patients.
It could also allow them to distribute production costs over larger patient populations, and avoid product shortages.
Different treatments could share a common platform for manufacturing and delivering these therapies. This could reduce the amount of novelty as institutions take on the task of delivering multiple different cell and gene therapy products.
It could also make it cheaper and easier for one platform to be used to manufacture and deliver multiple different treatments.
We’re also announcing that FDA will continue to analyze data and intends to expand its development of a CAR-T database.
This database will aggregate more detailed information about long term toxicity potentially related to CRS, neurological side effects, off target effects, and long-term patient outcomes.
Better understanding these toxicities may help us make better recommendations to help prevent them. Also, many of the potential issues associated with gene therapy are long term considerations related to off target effects and the durability of response that these treatments offer.
Adequately answering these questions will be supported by this type of more effective, post-market data collection.
And so, we need to develop new tools to enable this sort of long-term evaluation. We expect that we’ll have more than 1,000 patients in our new database by later this summer. And we’re planning to use this data to conduct research on how to best manage class wide effects and identify potential predictive biomarkers for durable remission approaches that can help improve the benefit and risk of these promising new treatments.
We’ll maintain patient confidentiality and preserve commercial confidential information (CCI) included in this new database.
Eventually, we may develop ways to allow outside investigators to interrogate some of this information. Our aim is to expand scientific opportunities to evaluate key public health questions, while making sure that we safeguard nonpublic information.
These are just some of the new steps we’re taking at each stage of the process. These are steps to bring new efficiencies to our work, and to help unlock the opportunities offered by better science. We’ll have many more announcements to make soon about how we’re embracing science to improve drug regulation.
Next week, for example, we’ll announce a major, proposed modernization of our entire new drug review process.
We’re at a turning point in the history of cancer, and the openings that science offers. A world has been created in many places – but certainly not all – where individual health and wellness is a clear beneficiary of technological progress. And it’s unfolding at a quickening pace as each new advance comes forward. These opportunities have become intense sources of investment along political and financial and emotional domains.
We’re challenging ourselves at the FDA to make sure that we have the best approach to evaluating new products. And we’re doing so in ways that are modern and efficient, so that we’re making the best us of our own resources, we’re facilitating innovation, and we’re fulfilling the needs of patients.
We look forward to working with you on these and many other efforts as we advance our shared goals: To reduce the personal and societal burden of cancer and to bring more patients hope for a cure. Thank you very much.