The critical path to COVID-19 therapies
Plotting a path to accelerated COVID-19 therapy development
The path to safe, effective COVID-19 therapies is crowded and poorly marked. There is an urgent need to fast-track a manageable number of therapies, illuminate every step along the route, and create signposts for research, development and manufacturing strategies to maximize the chances of success.
The life sciences community is rising to the challenge. Biopharmaceutical companies and academic teams are scouring their pipelines and lab books for agents that could help COVID-19 patients. Over 500 countermeasures are in progress, and at least 202 are in clinical development, according to BioCentury’s COVID-19 Resource Center.
Ironically, the scale of the response is part of the problem, due largely to the lack of organization.
There is a danger that a plethora of underpowered trials with uninformative endpoints may crowd out research on the most promising compounds. Competition for access to strained supply chains and limited manufacturing capacity is creating the potential for shortages. And the prospect of emergency authorizations and approvals of numerous products makes sense only if systems that do not yet exist are put in place for postmarket real-world data collection and analysis.
A report, "Building the Critical Path for COVID-19 Therapeutics,” from a team of some of the U.S.’s most prominent medical policy experts convened by the Duke-Margolis Center for Health Policy proposes solutions to these and other problems that threaten to delay progress on COVID-19 therapies.
A central tenet of the report is the need for transparency from drug developers, government agencies and academic researchers across the spectrum of COVID-19 therapeutics development, from discovery to manufacturing, including about topics that are traditionally treated as closely guarded secrets.
The report’s title and its themes echo the critical path initiative Janet Woodcock launched in 2004 that marked the beginning of the modern era of medical product regulation.
Woodcock has temporarily stepped away from her position as director of FDA’s Center for Drug Evaluation and Research (CDER) to take charge of the therapeutics team at Operation Warp Speed (see "Woodcock Steps Away from CDER for Operation Warp Speed"). She has been a champion of many of the approaches that will be needed to get COVID-19 therapies on track: master protocols and adaptive trials; and advanced manufacturing technologies.
The Duke-Margolis report makes the case that while the world is holding its breath in anticipation of vaccines, clearing the path for COVID-19 therapies is essential. Therapies may offer solutions sooner than vaccines can be deployed, and will be cornerstones of treatment in the event no vaccine offers full protection.
In particular, the report sees transparency as a crucial element on the critical path.
While the level of collaboration around COVID-19 among companies that are usually fiercely competitive has been impressive, the report makes the case for industry and government to move even further away from their comfort zones by publicly disclosing detailed data about their capabilities, progress and plans.
A hyper-parallel framework
The Duke report was written by two former FDA Commissioners, Mark McClellan and Scott Gottlieb; a former acting FDA Chief Scientist, Luciana Borio; Jeff Allen, president and CEO for Friends of Cancer Research; and Pamela Tenaerts, executive director of the Clinical Trials Transformation Initiative.
Its fundamental recommendation is to replace the traditional sequential paradigm of drug development with a parallel one.
“We need to adopt a hyper parallel framework for discovery, development, manufacturing, and effective use - a process in which the usually highly sequential process for developing therapeutics is compressed, and activities are done in an overlapping fashion,” states the report.
This can be accomplished through seamless trial designs that “allow rapid transition from the early evaluation of a product’s safety in small series to the large-scale evaluation of its efficacy in pivotal trials,” while simultaneously launching commercial-scale manufacturing that will ensure that, once safety and efficacy have been established, products are available.
The report envisions large-scale postmarket real-world data collection that will tell clinicians and healthcare systems how to use therapies effectively for specific patients and populations. It is particularly important to put continuous learning systems in place because therapies developed at pandemic speed will be put into physicians’ hands before researchers have had opportunities to study their optimal use, and at a time when the disease itself is changing (see “The Pivotal Role of Real-World Data in a Pandemic”).
Implementing the recommendations will take concerted and rapid action from regulators, product developers, payers and healthcare systems. Many of the necessary activities have never been done before; none of them have been accomplished at the necessary speed and scale; and some of the data necessary to make them work doesn’t exist and must be quickly assembled.
The report breaks the critical path into four steps: creating a clear pathway for promising therapeutic candidates; increasing clinical trial effectiveness and capacity; anticipating capacity for rapid access without shortages; and conducting real-world data collection and studies.
A clear pathway
Creating the clear pathway proposed in the Duke-Margolis report requires not only providing financial and regulatory support to propel potential COVID-19 therapies down the path, but also triaging them to focus limited resources on the most promising ones.
“Prioritization starting with preclinical assessment should be based on publicly available criteria that can be shared across industry and refined to enable more effective early assessment, including better informed funding decisions by foundations and private investors,” the report recommends.
“Moreover, all priority products should have information made available - to the extent feasible - on each product’s expected development milestones, including expectations about the start and duration of clinical testing and manufacturing needs.”
The report makes the case that transparency about product development needs is required to minimize delays and shortages. It also calls on the public-private partnerships that are selecting therapies for testing to make their selection criteria public.
So far, the consortia developing therapies have fallen short when it comes to disclosing how they are prioritizing therapies.
For example, on May 19 NIH Director Francis Collins said in a livestreamed interview that a working group of the Accelerating COVID-19 Therapeutic Interventions and Vaccines (ACTIV) public-private partnership had identified about six repurposed candidates for testing in a master protocol, and that the ACTIV executive committee would meet on May 20 to consider the proposal (see “NIH-Led ACTIV Partnership Close to Selecting Repurposed Therapies”).
ACTIV hasn’t announced which candidates were proposed, the executive committee’s decisions, or the criteria it is using to select therapeutic candidates.
The COVID R&D industry consortium announced on April 29 it would partner with the Quantum Leap Healthcare Collaborative to launch an I-SPY adaptive trial to evaluate 10-20 agents within a year to treat ARDS, the late-stage respiratory failure that can lead to death in COVID-19. It has published and provided weekly updates on its protocol, but has said nothing about how it is choosing candidates.
The COVID-19 Therapeutics Accelerator has disclosed little about its activities.
There is also little or no publicly available data on manufacturing needs or on what steps companies and governments are taking to ensure they are met.
Clinical trial effectiveness
The Duke-Margolis team advocates the establishment of master protocols to test COVID-19 therapeutic candidates, and states that selection and enrollment of products should be “made highly efficient, so product developers don’t undergo delays owing to the governance features of the master protocols.”
The report notes that several COVID-19 master protocols have been launched, including NIH’s Adaptive COVID-19 Treatment Trial (ACTT), WHO’s Solidarity Trial, the University of Oxford’s RECOVERY and PRINCIPLE Trials, and an addition of COVID-specific arms to UCSF and Quantum Leap Health’s ongoing I-SPY-2.
Additional master protocols will be needed to study therapies in different settings, the report suggests. It also recommends that “protocol developers should publish their work and identify ways to encourage alignment.” This would facilitate the inclusion of sites and patients, as well as the standardization of inclusion/exclusion criteria and endpoints.
McClellan and colleagues urge NIH and other funders to insist that trials select from common design templates and employ best practices that maximize chances of obtaining meaningful results quickly.
“All COVID-19 clinical trials should be well designed reflecting best practices on endpoints and statistical methods, and powered to support rapid enrollment and execution within a several-month time frame. Variants from the design suggested by these best practices may be appropriate but should be justified.”
The report notes the need for new COVID-19 clinical trial networks, and emphasizes that they should be global to enhance the chances of studying interventions in places with active outbreaks.
Especially when confronting a pandemic, pre-approval access blurs distinctions between research and treatment. Acknowledging this, the report urges FDA to “support the timely implementation of a model expanded access program for COVID-19 patients in cases where manufacturers have additional treatment capacity available.”
Ensuring manufacturing capacity
Although the need for end-to-end planning of manufacturing to ensure adequate capacity is clear, it’s difficult to accomplish given the complexities and opaqueness of supply chains.
The Duke-Margolis report recommends that assessments be made of capacity to manufacture different types of therapies, and that bottlenecks be anticipated and alleviated.
Manufacturing constraints could be particularly acute, for example, for mAb cocktails that could be used both as treatments and for prophylaxis.
It calls for transparency here as well. “To assure coordination with product developers (especially smaller companies with less resources), and to assess whether large companies that have already contracted for forward capacity are prepared to redirect it if their compound does not succeed, steps to increase manufacturing capacity and fill gaps should be reported publicly and updated regularly.”
The problem is that manufacturers, including contract manufacturers, closely guard capacity data. As became clear at the start of the pandemic, even regulators like FDA have little insight into where pharmaceutical ingredients are manufactured, and almost no information about capacity.
National governments might be able to mandate disclosure of some capacity data, but countries that are competing to obtain supplies for their own populations are unlikely to do so.
The report suggests the formation of a public-private collaboration to bring FDA, biotechnology and manufacturer associations, and relevant companies together to “identify opportunities for manufacturers to shift existing capacity to produce the most promising therapies quickly, even if the production is for companies that are normally competitors.”
McClellan and colleagues recommend creating financial incentives for companies to make capacity available for manufacturing competitors’ therapies.
They also recommend a series of additional financial incentives and new payment models to incentivize investment in manufacturing of COVID-19 therapies, including advanced payment contracts.
The Duke-Margolis report identifies a number of information gaps that should be filled to improve decision-making around COVID-19 therapy development.
These include disclosures about early-stage products such as key milestones in preclinical development and when Phase I testing is expected to start.
The list expands for products in clinical testing to include: study design, start date, endpoints, use of a master protocol, expected size and power, expected readout dates, significant updates on enrollment and retention; planned manufacturing capacity.
Transparency around COVID-19 products should continue after approvals, the report states, to include data about randomized and observational postmarket studies, along with information about manufacturing capacity.