OR WAIT null SECS
Agnes Shanley is senior editor of Pharmaceutical Technology.
Systems that discourage continuous improvement threaten pharma innovation, quality, and supply. Can the industry break through this deadlock?
Regulators, manufacturers, and even the media (1) often ask why more pharmaceutical companies don’t use more of the modern manufacturing and quality approaches and tools used in other industries. The answer becomes clear when one looks at the complexity and cost involved in making process improvements after a drug has been approved, and getting those improvements approved by global regulators.
Simplifying the process for post-approval changes is a topic that is being discussed more frequently at industry events; the Parenteral Drug Association (PDA) has launched a working group to share best practices and brainstorm solutions, and International Council for Harmonization (ICH) Q12 (2) is tackling the problem, yet little actual change is being seen. Some in the industry ask whether it will take a crisis to resolve this problem.
“We saw what the industry can do and what regulators can support when the Ebola virus struck, and vaccines, which usually take over a decade to develop and approve, were developed and released in months,” says Maik Jornitz, CEO of G-CON. “Why can’t we bring these approaches for the urgent to the normal, and create a more rapid approval approach?” he asks.
Despite years of talk about regulatory harmonization, and some pilot programs and alliances, each regulatory authority still takes its own approach to looking at process improvements. When regulators disagree, they sometimes force manufacturers to choose between moving forward with an innovation or just staying with the old process to simplify manufacturing and quality and reduce costs, says Melissa Seymour, vice-president of global quality control at Biogen. “Those are the kinds of things that, if we’re not careful as an industry, will shut down innovation,” she says.
Even proven technologies still run into regulatory snags. One example would be moving from high-performance to ultra- high-performance liquid chromatography (HPLC to UHPLC), which should be easy. “UHPLC is a better method and it’s simple to validate, but it can still take years to get it through the process, so we’re still running HPLC because of the regulatory approval structure, not the capabilities and potential of the analytical method,” says Seymour.
Another case in point is isolator technology. When they were first introduced to the pharmaceutical industry, isolators weren’t adopted as rapidly as expected, due to regulatory uncertainty, says consultant Jim Agalloco. Years later, that uncertainty persists. “If I want to replace a curtain-segregated with an isolator-based fill line, that change will still undergo regulatory scrutiny, even when the isolator-based containment is more robust,” says Jornitz.
A pharmaceutical company that sells its products globally must address post-approval change requirements for each of the regions it sells to, and that can mean dealing with more than 100 different regulatory agencies. “Many of our products are marketed in over 80 countries, so a change approval can take more than five years,” explains Seymour. The result, she says, is inventory segmentation and the need to manage several different versions of a drug over its lifecycle.
“In some instances, we’ve had to go back to older manufacturing processes, or older and less efficient test methods to generate enough inventory for countries where those changes haven’t been approved yet,” she says. In other cases, companies may spend resources running two different analytical methods for one product for up to five years, or running under different specifications for three to five years, because of the time it takes to get changes approved globally, she adds.
Adding to the difficulty is the fact that, in the real world, process improvements don’t just happen one at a time, but accumulate, says Anders Vinther, chief quality officer at Sanofi. “You wind up stacking them on top of each other.” As a result, one year, he says, Sanofi produced 83 batches of one of its vaccines according to 55 different configurations/variations, depending on the region involved.
The costs required can be mind-blowing. Roughly 80% of the industry’s regulatory affairs professionals now spend most of their time on post-approval change-related issues, Vinther says.
At the same time, having to maintain and run parallel manufacturing and quality processes adds complexity and can wreak havoc with workflow and production planning, he says. Consider a situation where a country needs to increase the amount of a vaccine it buys, Vinther explains. The manufacturer may have sold out of that particular version of product, so it may offer another version that the country hasn’t approved yet. Typically, the country will then expedite approval, but the manufacturer may not have enough of that version of product left over to meet other customers’ needs, forcing it into firefighting mode.
In other cases, one set of quality tests can be effectively run on the plant floor, while, for other markets, the same tests on that product must be done in the laboratory, says Seymour. The innovation was adopted to improve efficiency, yet the result is the opposite.
There isn’t a regulatory agency in existence that doesn’t support continuous improvement, in concept. As Vinther says, EMA regulations explicitly require it. However, regulators may send mixed messages. The same submission, containing identical data, can receive very different assessment from major regulatory agencies, resulting in the need for fragmented implementation or duplicative effort. It is time for more harmonization amongst regulators worldwide,” says Vinther.
In the worst case, tradition may win over good science when regulators assess new technology. Jornitz recalls a gloveless, robotic isolator-based filling system, which came under regulatory scrutiny because it did not allow settling plates to be installed for environmental monitoring.
As he notes, there is no need for traditional environmental monitoring in a sterile, enclosed system. “You have to be in control of your process, and not rely on dated environmental technology,” he says.
Another example Jornitz cites is the European regulatory requirement that sterilizing grade filters undergo integrity tests before use and after sterilization. Testing is used to determine whether a flaw occurred, which might not be found in the post-use test due to blocking of the filter. As Jornitz says, the filters are scaled not to block, and are commonly used to filter clean solution. The Committee for Proprietary Medicinal Products guidelines set in April 1996 established a maximum allowable bioburden of 10 colony forming units per 100 mL in front of the sterilizing filter. “There is typically no chance that the filter will block, yet regulators insist on the pre-use and post sterilization test, which increases potential risk to the patient because the test requires manipulation of the filtrate side,” Jornitz says.
Bearing witness to the challenges of adopting more modern technologies is single-use process equipment, which, Jornitz says, took 15-20 years to gain acceptance in pharma, but is now widely used. Rapid microbial monitoring (RMM), which has been around for about as long as single use, is being used by some pharmaceutical companies, and has received support from regulators, but still isn’t widely applied. For many companies, Jornitz says, there is a fear of the double-edged sword, and regulators requiring that the old technology be used along with the new.
FDA and other regulators have increased staff training., FDA has moved to a team-based approach (See Sidebar) to help make reviews and inspections more consistent.
In addition, FDA established an Emerging Technologies Team (ETT) in 2014 to help companies address concerns about new technologies and the impact that they might have on applications or post-approval changes (See Sidebar). The team promotes early engagement with firms, for both new and legacy products, to discuss potential challenges to implementation of new technologies and possible paths to resolution. “With early engagement, the agency can move quickly to review and act on supplements for implementation of newer technology,” says Lawrence Yu, deputy director, Office of Pharmaceutical Quality, Center for Drug Evaluation and Research at FDA.
Seymour’s team has met with ETT in connection with Biogen’s candidate Alzheimer’s therapy, which has been fast tracked for approval. “It has been an enlightening experience. The team is open and collaborative. We’ll have to wait to see how everything plays out when we file for approval, but the interactions have been positive,” she says.
Vinther sees the team’s existence as a positive development, but, as he notes, “It doesn’t help if FDA can review a post-approval change in four months, but it takes the world five years and, in the meantime, we have to deal with all these different technologies and product versions at the same time.”
One of the reasons for the disconnect between manufacturers and regulators may be the huge difference in focus, he says. “Manufacturers, by definition, must think about the whole world of markets they sell to, where regulators must think nationally, and of protecting the health of people in their own countries.”
Blaming the regulators is no solution, he says. “The industry needs to do things better, to ensure that our systems work better and that we are better at forecasting and preventing drug shortages.”
“If we could get all the regulators to work together that would be a great help and would allow us to innovate,” he says, but he senses that some regulatory authorities may not trust each other. “If regulators with a ‘Strict Regulatory Agency’ grade (the highest grade) from the World Health Organization see a process change as being benign or beneficial, why can’t the others?” he asks.
Besides, Vinther says, most post-approval changes eventually go through. “The International Federation of Pharmaceutical Manufacturers and Associations surveyed leading vaccine companies and found the success rate was over 99%. Can’t we at least reduce the number of filings required by 50-70% by doing a thorough risk assessment?” he asks.
Rejecting innovation to save money can be the first step to neglecting facilities, milking processes but failing to maintain them, says Jornitz. That may result in shortages when there is only one manufacturing site supplying a vital drug. Once a process and facility reach a certain level of aging, they can go into a death spiral. An example could be Benvenue Labs, where $300 million in investments proved to be too little too late.
But companies have also created the impasse in pharmaceutical modernization, says Agalloco. “Comparability and crossover studies are costly and difficult, but you cannot blame the regulators, but yourselves for not having enough conviction in the technology itself or proof of its performance,” he says, “because there can be major financial incentives for using new technologies. Failing to use any is just a big excuse. You have to take some risks, just not huge ones.”
Besides, there may be some innovative technologies that remain unknown, even though they’ve matured to a point where they could be used. “Nobody will touch them until FDA says it’s okay, and FDA cannot approve them until someone files a new drug application specifying their use, so it’s a Catch 22,” he says.
Pharmaceutical companies have to incorporate innovative manufacturing practices, to lower cost and produce the highest quality products, says Girish Malhotra, president, EPCOT International. Regulators suggesting the “how and what” of innovation, he says, is putting the cart before the horse.
“Innovative technology implementation is a team effort between regulators, the industry and suppliers. We need to work hand in hand if pharma is to become as technologically advanced as other industries are,” says Jornitz. “Besides,” he says, “a whole new industry, cell and gene therapy, will require rethinking many areas, and the adoption of new technologies for the future.”
“For now, the key to resolving this problem,” Vinther says, “is to take all politics and emotion out of this issue and base discussions purely on science.”
“We need to provide regulators with tools and assurance to prove, with solid data, that we have the capabilities within our organizations to make changes,” says Seymour. “There’s a responsibility to put the tools out there for organizations to use, to be able to manage change consistently and appropriately,” she says.
1. L. Abboud, and S. Hensley, “New Prescription for Drug Makers: Update the Plants,” wsj.com, September 3, 2003.
2. ICH Q12 Position Paper, ich.org.
Vol. 41, No. 8
When referring to this article, please cite it as A. Shanley, “Breaking Through Obstacles to Improve Drug Manufacturing,” Pharmaceutical Technology 41 (8) 2017.