Agnes Shanley is senior editor of Pharmaceutical Technology.

Agnes Shanley

The COVID-19 pandemic, and the need to accelerate development and distribution of vaccines, has required companies and governments to take on considerable risk in the hope that vaccines now under development prove to be safe and effective. Currently, Pfizer and BioNTech’s, Moderna’s, and AstraZeneca’s vaccine candidates are in advanced stages, with more than 150 other COVID-19 vaccines in development. Developers are ramping up just-in-time manufacturing and distribution efforts for their candidate vaccines even before they know their relative effectiveness.

As analysts from Charles River Associates (CRA) point out, vaccines developed during other pandemics have shown what can go wrong with traditional supply agreements. During the 2009 H1N1 pandemic, for instance, proactive governments that had arranged early purchase agreements cornered the supply of vaccines, leaving out countries that were slower to act. However, once that pandemic proved to be milder than anticipated, the proactive buyers were left with excess supplies to sell, donate, or dispose of.

In response to COVID-19, the United States government established Operation Warp Speed, which, so far, has coordinated just-in-time distribution of Moderna’s vaccine to begin the minute FDA issues emergency use authorization. In the United Kingdom, the government has entered into supply agreements with individual companies, and launched such an agreement with Moderna on November 17, 2020 (1).

Meanwhile, in the European Union, the European Commission’s (EC) Joint Procurement Agreement (JPA) has been more conservative about entering into supply agreements with individual manufacturers. Only time will show which approach proves the better one.

However, for future vaccine supply arrangements during pandemics, both approaches will ultimately prove inadequate, CRA researchers suggest. Instead, risk-sharing contracts will offer the best opportunity for balancing risks and benefits for developers, governments, and patients. London-based Eva Marchese, PhD, vice-president, and Abhishek 

Kumar, associate principal with Charles River Associates, shared insights from current CRA research with 

What is the greatest weakness in traditional procurement and distribution agreements for vaccines?

: Looking beyond the US from a holistic point of view, we have seen two approaches to procuring vaccines:that of the US and UK, which have been keen to enter into agreements with manufacturers and developers, and have invested in companies and these programs. As a result, they have secured vaccine supply for their populations earlier in the delivery schedule.

On the other hand, the collective approach shown by [countries that signed] the EC’s Joint Procurement Agreement has been more circumspect. They may be engaging with a lot of manufacturers, but have been slow to sign actual supply agreements. This fact may not be too surprising, since the EC is made up of many countries. But, in general, this strategy reflects the fact that government authorities are in a tight spot. If they move quickly to make agreements and commit large sums of money to support the development of specific vaccines, they also have to go back to their citizens to justify that spending. There is always the risk that the agreements they have entered into may not bear fruit.

Neither approach is ideal. In a pandemic situation, you need a better model to assume risk for both buyers and developers. Vaccine manufacturers are making large investments in testing and ramping up production capacities and producing vaccines even before they know with full certainty that the vaccines will work.

It is important to emphasize that, during COVID-19, countries and governments reacted to a situation for which they were not prepared. Handling and managing procurement must be done by respecting differing regulations, and it takes negotiation and time to change those regulations.

Given the lack of time available, governments had to react, and their approach is still reactive. In the UK and the US, political and economic pressures have also been in play. The UK, for example, was in the middle of Brexit negotiations when COVID-19 hit.

 Are there any examples of how these risk-sharing agreements would work, and are there any precedents from areas outside of vaccines? What types of “go/no go” gates should be in place to trigger key decisions? 

 This type of agreement is being seen in oncology and innovative therapies (e.g., cell and gene therapies). The challenge in adapting such models for COVID-19 vaccines is the fact that the vaccine development timeframes are so short and steps have had to be vastly compressed. It isn’t possible to have a full and clear understanding of the vaccines’ clinical effects because data available will be limited. As a result, agreements would need to link the risk to specific efficacy outcomes that will be learned once the vaccines are on the market.They would have to be linked to real-world evidence since, at the time of launch, the full standard package of clinical data won’t be available.

 A good idea to consider would be staggered payments triggered by specific milestones (e.g., linking a portion of payment to post-approval, post-launch, and real-world data).Payments could also be linked to approvals at different time points (e.g., getting Emergency Use Authorization or conditional approval and even to positive Phase III data).

 Such agreements would be possible once vaccines are commercialized and are being evaluated using more standard approaches.Different models from oncology could also be considered (e.g., linking payments to the effectiveness of the drugs).Using this approach, patients are divided into responding vs. nonresponding groups.Some countries have built registries to track both groups and have linked results to payments to pharma companies.

Gene therapies are also using innovative approaches since they are believed to be curative, but lack the five or six years of follow-up data that are typically found with traditional therapies.

In some cases, even the promise of an individual cure has been linked to payments.For COVID-19 vaccines, the main considerations are how long the vaccine will work and how long the immune response to the vaccine will last. Payment could be linked to these properties, particularly if vaccines on the market differ in their performance in each of these criteria.

 Do you see industry fully embracing the risk-sharing concept in payments? How could such a model be possible in planning for future pandemics and other emergencies?

 In the COVID-19 pandemic, we have seen non-government organizations, such as the Coalition for Epidemic Preparedness Innovation and the Wellcome Trust, come together with a coordinated response to get vaccines approved as soon as possible. These stakeholders will be key to driving better preparedness for future pandemics. We expect stakeholders to learn from the COVID-19 pandemic experience, and to see more flexibility from both pharma companies and governments and more willingness to engage in innovative approaches to sourcing and supply.

We are already seeing this for innovative breakthroughs such as cell and gene therapies. But we haven’t seen it yet for preventive treatments. However, as companies and governments leverage lessons from the current pandemic, we can expect to see more acceptance.

With vaccines, pharma companies could also be open to pricing vaccines more affordably earlier in the cycle, when there is a need to immunize a large portion of the population. In return, the governments could agree to make longer-term purchasing commitments with those manufacturers who have reduced prices.

With risk-based approaches, a major challenge is monitoring, given the number of patients who would receive treatment. With gene therapy, the number of patients involved is very limited, so results would be relatively easy to monitor. Vaccines present a completely different story, however. An infrastructure will be needed to monitor the number of diverse patients that will be vaccinated. How it takes shape will depend on the specific situation within each specific country.

Moderna, “Moderna Announces Supply Agreement with the UK,” Press Release, November 17, 2020.

Articles

Charles River Associates analysts see risk-sharing as the key to improving vaccine availability and access.



New Risk-Sharing Models Are Needed for Future Vaccine Supply Agreements 

As winter approaches, a second wave of COVID-19 infections has hit Western Europe and the southern and western United States with brutal force.In a keynote presentation at the annual Galien Foundation Forum on Oct. 30, 2020, Dr. Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases (NIAID), noted that SARS CoV-2 had exploded in an “unprecedented manner” not seen since the Spanish flu of 1918. On the day of the Forum, 8.9 million cases of COVID-19 were reported in the US, and the death toll had reached 227,2000, 20% of the global total, noted Lynne Zydowsky, PhD, president and co-founder of Alexandria Summit, who moderated the session on COVID-19 vaccine development.

The increasingly urgent need for vaccines and treatments has slashed development timelines dramatically. Clinical research projects launched in late July are already in Phase III. Currently, according to the 

, 42 COVID-19 vaccine candidates are in clinical development and another 151 are at the pre-clinical stage. So far, Moderna, AstraZeneca, and Pfizer-BioNTech are on track to deliver the first doses of their target vaccines during the first quarter of 2021, and perhaps even sooner. 

Applying knowledge gained from coronaviruses past

They, and other vaccine developers, have drawn from years of past experience studying coronaviruses, starting with outbreaks of severe acute respiratory syndrome (SARS) in 2002 and Middle East respiratory syndrome (MERS) in 2012, noted Fauci. Teams at NIAID have worked closely with companies, notably Moderna. in early clinical work on its target vaccine, mRNA-1273.

At the forum, two pharma CEOs whose teams are now in Phase III, the last lap in the race to develop a COVID-19 vaccine, shared insights on the road ahead as well as challenges they’ve met and are overcoming: Stéphane Bancel, CEO of Moderna, whose target vaccine is under development with the collaboration of the US NIAID, in partnership with the US Operation Warp Speed program, and Paul Stoffels, MD, executive vice president and chief scientific officer for Johnson & Johnson Pharmaceuticals (J&J), whose COVID-19 vaccine candidate is currently in Phase III. J&J had briefly paused its clinical work in October to investigate possible connections to adverse effects after a trial subject became ill, but has resumed its work full speed. 

According to Bancel, whose company was founded 10 years ago to commercialize a non-cell-based platform based on messenger RNA (mRNA) technology, COVID-19 vaccine development has benefitted from previous work on MERS. “We had worked on a few emerging pathogens that we all thought could come back,“ he noted. That knowledge came in handy in early January 2020, after the novel coronavirus was identified in Wuhan, China. After the virus’ genetic sequence was published, Moderna teams studied its structure closely, using 3-D modeling to confirm that the spiked protein, which had also been targeted by the company’s previous MERS vaccine, was the best bet for developmental focus, Bancel told attendees. 

“On January 13, we wrote down the experimental vaccine design; within 42 days we shipped product to the National Institutes of Health; and, in early March, we began to work with Dr Fauci and his team on Phase I,” he told attendees.


Bancel believes that mRNA, which is also being used by Pfizer and BioNCell in their COVID-19 vaccine candidate, will make development of future vaccines easier.

“If our COVID-19 vaccine gets to the finish line in the next month or so, it will be an incredible de-risking for the entire mRNA platform. The next product will use the same chemistry to make the mRNA, the same manufacturing process in the same rooms operated by the same people and use the same formulation around the mRNA to get it into the cells,” he said.

Calling mRNA an “information molecule,” Bancel noted that the technology being used to make its COVID-19 candidate is basically the same that the company used for its Zika and cytomegalovirus (CMV )vaccines. “You just adapt to the biology of each individual virus,” he said. The end goal, he said, is training the recipient’s immune system so that, if a person is infected, it will create antibodies to neutralize the virus. With CMV, for instance, six different mRNAs were needed to make six different proteins, he noted. 

“What is really hard with this technology is getting it to work safely,” he said, a fact that has driven Moderna to spend “a few billion dollars” on the project so far, and to step up collaboration with Big Pharma and academic labs, including Dr. Fauci’s teams at NIAID.

J&J, meanwhile, has applied knowledge gained developing vaccines for Ebola and Respiratory Syncytial (RSV) viruses, to the new COVID-19 vaccine. “We used human adenoma vector (Ad26) for the Ebola virus,” said Stoffels, noting that the vector has been optimized over the past six years to maximize immunogenicity and antibody and immune response, and has been used to vaccinate more than 100,000 people so far, generating a large volume of safety data.

J&J’s trial is one of the largest, with 60,000 participants, but previous clinical experience with the vector allowed the company to speed up preclinical and clinical work, Stoffels said. “We optimized the vector during the first three months of the year, then did animal work. Phase I and II studies began in mid-July and continued through mid-September, while the Phase III study is now underway, he said.

Pausing the work briefly in October merely reflected best clinical practices, Stoffels said. "You pause the study to learn and discuss with experts before moving forward," Stoffels said, emphasizing the importance of protecting participants in clinical studies. 

On the manufacturing side, cell-line optimization was crucial to improving the vaccine's manufacturability, Stoffels explained. “We can produce a billion doses of vaccine using a cell line that works. In addition, stability has been optimized to levels of 2-8 °C for distribution. This is very important for distribution in countries within the developing world,” he said.

Optimizing clinical trial recruitment has been crucial. In the US, members of minority groups have been disproportionately affected by the virus, Bancel said, which underscores the need to include more people from these groups in clinical trials. Insufficient representation could not only affect results, but also limit public trust in the vaccine, he said.

Both Moderna and J&J have included more ethnically diverse subjects in their clinical trials. For J&J, that meant locating half of its 60,000 subjects in the US, and the other half in Latin America and Southern Africa at 180 centers throughout the world.

Moderna, whose clinical trial involves 30,000 people in the US alone, hit the pause button on its clinical studies in September to ensure that the right demographics were in place. Since then, presence of African American subjects has doubled to 10%, and that of Latin American people has grown to 20%, Bancel said.

Another key step to achieving diversity was including more mature subjects (i.e., aged 65 or older), across all different ethnic groups, in the study. Moderna’s clinical trial makeup is now consistent with the US population and includes people at higher risk of developing the disease (eg , older people with diabetes or heart failure), Bancel said.

Moderna’s decision to limit trials to the US had more to do with the company’s size at the time than anything else, since the company only had 800 employees in the spring (a number that has since grown to 1200).“We’ve discussed this issue with global regulators, and, while we may need to run some small studies in a few Asian countries, other countries are very comfortable with data from the US,” he said.

Dosing: efficacy and manufacturability concerns

Dosing has been another key question for developers to address. Depending on how long the novel coronavirus persists, booster dosages will likely be needed over the next few years. Moderna has opted for a two-dose approach, where J&J has stayed with the single-dose strategy that it used with the Ebola vaccine. So far 25,000 of the Moderna trial’s 30,000 participants have received second shots of the target vaccine.

“When we looked at the data, we always saw a higher antibody level and longer period of protection by giving a second booster shot,” explained Bancel. AstraZeneca also decided to move from one dose to two doses in its Phase III COVID-19 vaccine work, he mentioned.

“It’s not easy to manage the logistics dictated by vaccinating people twice, but we needed to ensure the highest efficacy we could, especially for the elderly,” said Bancel. Results so far have shown the vaccine to result in similar levels of neutralizing antibodies in both elderly and younger age groups. “That consistency is very important. For some vaccines you see a big drop between young and elderly.In this case, with the elderly accounting for 80% of all COVID-19 deaths, it was crucial,” Bancel said.

Manufacturing plans and just-in-time distribution

“We agonized last Spring about whether to go with 50- or 100-mL doses,” Bancel recalled.The smaller size would have been better for manufacturing, and Moderna’s own manufacturing department had advocated that approach. Ultimately, the need to optimize efficacy drove the company to opt for the larger dose, Bancel said.

On Nov. 16, 2020, Moderna announced that its COVID-19 vaccine candidate, mRNA-1273, met criteria established by the US FDA, with efficacy of 94.5%. The company also announced that the vaccine remains stable at 2° to 8°C (36° to 46°F), the temperature of a standard home or medical refrigerator, for 30 days; new stability testing data support this extension from an earlier estimate of seven days. 

The company plans to submit the data required for an Emergency Use Allowance (EUA) by the end of November, when it expects to have its first interim results, he said.

Bancel also noted that FDA guidance for COVID-19 vaccine development has been clear, setting a goal of at least 50% efficacy and at least 30,000 participants in Phase III, with a 50/50 split between subjects receiving target vaccine and those getting placebo. “This clarity avoided guesses and higher risk,” he said.

The company has already been filing parts of the chemistry, manufacturing, and controls (CMC) application so that the detailed documentation required (a stumbling block for many companies along the "critical path" of drug development) will not delay development. “We expect data for the EUA to be filed by the end of November to early December,” he said.

Moderna has also prepared its vaccine distribution plans. “Working with Operation Warp Speed, we are on track to have at least 20 million doses made before the end of the year, and 100 million doses by the first quarter of 2021,” says Bancel, noting that trucks will stand ready to load and start shipping immediately pending receipt of EUA approval.

When asked whether the novel coronavirus might become like the influenza virus and call for a new vaccine to be developed each year, both Bancel and Stoffels emphasized the need for additional booster capacity. “We hope that this virus won’t be like the flu, but we need to understand that [this pandemic] won’t be over in six to twelve months. We will have to protect people for two to three years, until the virus can be eradicated, Stoffels said, emphasizing the need for stakeholders in the industry to learn together. Based on what happened with Ebola and Zika, he is confident that COVID-19 can be eliminated within three years. 

Role of CDMOs and collaboration with the government

Rapid development has emphasized the need for scalable manufacturing platforms and experienced contract development and manufacturing organization (CDMO) partners, said Bancel. Moderna had the advantage of having a head of manufacturing who had previously led manufacturing at Novartis, he said, and also partnered with the CDMO, Lonza Pharma & Biotech, early on. Last spring, the CDMO started working with Moderna in GMP-scale production suites  in the US and in Switzerland, which will manufacture supply outside of the US. 

Collaboration has also been critical with fill/finish and handling suppliers, Bancel said. Moderna had to pay suppliers more for additional products and equipment up front, to assuage any concerns that they might have about potential development failure. “We have been so thankful for government help, and particularly the help of Operation Warp Speed, which saw to important details, including the shipment of some pieces of equipment by military escort to production sites,” he said.

COVID-19 vaccine development has built on a foundation set by strong ties with government agencies such as the US Defense Department’s Biomedical Advanced Research and Development Agency and the National Institutes of Health, as well as regulators, said Stoffels. Of particular help was the fact that Moncef Slaoui, head of scientific operations at Operation Warp Speed, has extensive experience in vaccine and pharmaceutical research and development. “This has allowed us to streamline decision making,” said Stoffels.

The industry is learning painful lessons from the COVID-19 pandemic that must inform global scientific and government strategies in the future, panelists agreed. “We need to invest in the critical scientific knowledge required to prepare for the future, not just with vaccines, but with therapies, better diagnostics, and improvements in the entire primary healthcare system. This must not happen again,” said Stoffels.

Bancel believes that developers need to come up with “what if” scenarios for the world’s 20 most dangerous pathogens likely to come up in the future.  He also stresses the need for improved manufacturing. “We need to build ‘never again’ plants that will be capable of making human-grade clinical material in a matter of 40 days,” he said. 

Both he and Stoffels agree that the issue of pandemic preparedness needs to be viewed as a top strategic security issue. “If governments can spend billions of dollars on nuclear weapons hoping never to use them, what should keep them from spending a fraction of that to prevent diseases like COVID-19?” he reasoned.

So far, though, the level of collaboration seen in developing vaccines has been unprecedented and deep, said Stoffels. In an industry noted for secrecy and strong competition, this spirit bodes well for the development of future vaccines and therapies at a time when the need for them is growing exponentially. Development timelines may never be the same. 

Clarity from regulators, prior work on other vaccines, and closer collaboration with contract partners and suppliers have allowed Moderna and J&J to slash timeframes. Moderna may be on track to deliver initial doses to the highest risk patient populations before January 2021.

The Last Mile: COVID-19 Vaccine Developers Prepare to File for Emergency Use Authorizations

Machine Learning is being applied in a growing number of areas within pharma, including pre-clinical research, to focus efforts and speed up time to results. Leveraging this tool is Riffyn, whose cloud-based Process Data System, Riffyn Nexus, was designed to incorporate principles from Six Sigma (specifically its define, measure, analyze, improve and control [DMAIC] rubric), as well as pharmaceutical quality by design (QbD) and measurement systems analysis.

Results have already been seen in industrial biotechnology companies such as the industrial enzyme developer, Novozymes. Established in 2014, Riffyn has recently been moving into more pharma applications. During the first quarter of 2020, before the COVID-19 pandemic began, CEO Tim Gardner, a specialist in synthetic biology, discussed the company’s origins and goals with 

 What made you develop the Process Data System and start the company?

: Riffyn was born out of an industrial biotech environment, and knowledge gained from the large-scale process development and manufacturing of chemicals produced by genetically engineered species. I had spent years in industrial microbiology and biotech, as vice-president of R&D for a company called Amyris, which focused on large-scale bio-manufacturing of renewable chemicals. We edited and genetically engineered yeast to make products that ranged from detergents to makeup oil.

In industrial biotech, speed to market and low costs of production are key to success. Ideally, you want to minimize capital investment costs while developing core technologies, so you compete not only on performance but on price. Being able to do this required harnessing data in ways that few companies were capable of doing 10 years ago.

At that time, pharma hadn’t yet made the transition to data-driven process development and manufacturing because they weren’t under the same performance pressures with their processes. 

Of course, they were under pressure from regulators and had to have the right regulatory mindset, but, in general, the focus was not on whether the process worked well but on whether it would be approved. In fact, the worst process in the world would still be OK if it received regulatory approval. This type of mindset does not drive innovation.

With Riffyn, we figured that we could take data-driven process improvement and combine it with the principles behind the ICH (International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use) guidelines from FDA and EMEA (the European Medicines Agency) and the QbD initiative. Those are fundamentally brilliant, some of the best technical directions for pharma that I’ve ever seen, and a recipe for data-driven process development.

Initially, few in the industry were really harnessing these concepts. There were too many difficulties entailed with practical issues (e.g., designing processes; integrating different teams of people; working on those processes and evolving them over time; capturing specifications; collecting data; and synthesizing data and specs of various forms of processes and versions). The goal was to put all of that together, to provide a holistic understanding of how manufacturing processes are working, but when QbD was introduced in 2004, it was too difficult to do.

However, it had already been done in other industries. Experiences that I had in industrial biotech showed it was not only possible to do, but that it could have a profound effect on R&D. In industrial biotech, utilizing QbD principles and data science practices led to four-to-five-fold increases in speed to market, cutting the average time from tech transfer to manufacturing down from more than a year, to three months.

We wanted to incorporate these principles into a software system that made it easy to harness them, not only for users in industrial biotech, but for pharma and biopharma development, to help usher in a transformation in speed-to-market and continuous improvement in the development and delivery of those products.

: Since those early days, before you set up the company, have you seen a significant change in the industry’s acceptance of the fundamentals of QbD?

: We have definitely seen a shift of mindset and culture within the industry, and some change in practice. Today, the concepts of digitization and speed to market are much more commonly discussed as key transformations needed to bring biopharmaceutical organizations to the next level of progress.

These ideas are getting embedded more into practice. Recent surveys have shown that more than half of all pharma companies are using ICH guidelines around process validation in their daily work. That’s encouraging. But many of them are still using outdated technologies.

: How would you describe the Process Data System? How does it work?

: It’s really a digitized Process Lifecycle Management environment, built around capturing production and analytical process designs and specifications and then accelerating improvement cycles for those processes until they are ready to transfer to manufacturing. The principles of DMAIC and QbD are embedded into its world view.

We had used these same principles at Amyris, to double the rate of catalyst and strain development without increasing headcount. 

Measurement system analysis is another important principle we used. All those principles aim to define R&D processes explicitly, instead of leaving that as an afterthought. This process of definition begins with the process that generates data, then the next step is measuring performance, analyzing performance, and improving processes to reduce variability.

Reducing variability is so important. It’s an almost unspoken but understood objective in manufacturing, but not in the research and development space.

That is why there is a lot of churn in pharma R&D. As a result, people can end up making bad decisions, chasing phantoms, and going down the wrong road because they respond to noise in the system rather than to significant developments. 

Once we took QbD into R&D, we achieved such fidelity in experimental answers that predictability, speed, and efficiency all improved. As a result, the number of experiments could be reduced, along with the level of overall churn, because we could understand what the critical process parameters were and take them right to the manufacturing floor.

At Novozymes, one of our first customers, we saw dramatic gains in R&D and tech transfer efficiencies (Read 

The company has been using the Riffyn Nexus for two years, during which it has increased capacity by an order of magnitude, and cut time to market by a factor of two, all with half the original number of staff members, Gardner says. Riffyn is currently working with a number of pharmaceutical and biopharmaceutical companies on pre-clinical and other development programs including bioassays, as well as formulation and bioprocess development. 

We don’t support manufacturing like a manufacturing execution system (MES) or batch-control software system. Instead, we support the development of processes and their transfer to manufacturing (e.g., catalyst and enzyme development, formulation development, analytical methods and assays used to collect data). We can transfer those designs into a manufacturing context. Right now, this is accomplished mainly by sharing documentation. It’s like transferring blueprints.

In the future, though, we hope to incorporate digital transfers so that users can transfer a design file directly from process development to manufacturing, sending the information right to the process control system (much like a CAD/CAM [computer-aided design/manufacturing] system would do in discrete manufacturing).

: When will this capability be introduced? 

: We’re prototyping and developing with pharma partners and biopharma process equipment companies. It’s probably one to two years away.

: What results are you seeing among customers in the biopharma space?

: We helped one pharma company, which had a team of four doing bioassay development, so that they could cut 2400 hours of labor each year from their processes using the improved data system, a 50% reduction in effort to achieve better outcomes from bioassays.

: How has the technology helped reduce false leads?

: There are a few examples that come to mind, although we cannot give client names. In one case, a group was struggling with cell-line development. They would pick what seemed to be the best cell lines, but these would continue to fail at the next developmental stage.

We collected data from 14 of their past experiments, assessed the resulting data in aggregate, and saw that seven of the 14 experiments they had performed had used the same cell lines, so these seven points became internal controls for their assay system’s performance.

Normally you just look at the controls week by week, but we also tested and followed a number of other variables in the data set (e.g., feed lot of culture media and temperature) from week to week. We found variations in these values across the experiments. As it turned out, this variation was driving aggregate decreases in cell line performance, and not just in the cell lines that the teams had selected as the best options.

The team was really chasing artifacts rather than real factors, so they were reacting to higher temperatures or better media rather than better cell lines, so some cell lines were failing downstream.

When we statistically corrected the data to remove all the temperature and lot effects and then renormalized the data across the experiments, the team could see that not one single cell line had been improved. If they had performed this analysis first, they could have prevented significant waste and loss. However, when they took the lessons learned from this work, they were able to improve processes and got real hits almost immediately.

: Why is there such a problem with reproducibility in pharma R&D, especially at the pre-clinical stage?

: One problem is that people are not transparently articulating what they’re doing. The first step in establishing this transparency is defining what you did, unambiguously.

Riffyn’s platform is designed to help with clear process definition, to get organizations to articulate their processes clearly, rather than fragments of the process, the bits and pieces that they may see on data boards. This can lead to huge boosts in reproducibility.

In addition, reproducibility requires a clear understanding of CPPs. In fact, in many ways, the concept of reproducing a process in the lab is exactly like scaling up a process. You need to understand CPPs, and, if you control them, the entire process can be much more efficient.

: Today a growing number of biopharma companies are doing their own work in machine language and artificial intelligence, using open source code and developing inhouse processes. Amgen is one example. Will it be more difficult to penetrate if everyone is doing this type of work independently? Will this make the market more competitive in the future?

: It's actually making it easier. We take an open ecosystem approach, and one that is vendor neutral in terms of access to data and process design. We adhere to, or provide interchangeability with S88 and S95, with the goal of Riffyn becoming part of a modular ecosystem that can be plugged or unplugged. In general, the more open, the better.

At this point, most of our work in tech transfer is partnered.

: How does the software work and what does the IT connect into at a customer facility?

: It’s cloud-based and can be accessed via web browser, but we can integrate it within customer networks so that it is embedded. All that is required is a web browser. As in VIZIO where you can have flowcharts, Riffyn starts by creating a flow diagram that represents the assay, manufacturing, or formulation process to describe ingoing process.

This can be replicated in the lab so that users can collect all data from instruments, from manual entry. What might have taken 40 hours of data integration in the past can be done in 1 to 2 hours.

: Does this data go into a Laboratory Information Management System (LIMS)?

: In fact, many customers are replacing LIMS with our system because it can integrate LIMS type data in the context of data from many other sources.

: Are contract research and development companies a part of your client base now?

 We expect to see that part of the business grow. Because our system is cloud-based, it can provide a means for contract development and manufacturing organizations to gain real- time access to unfolding experiments. 

Branching out from roots in industrial biotechnology, Riffyn Data Systems (and its CEO Tim Gardner) is taking the message of data-driven process improvement to biopharma. 

Bringing Six Sigma to Preclinical Pharma R&D 

For years, pharma’s key opinion leaders have emphasized the need for deep process knowledge to inform new product and process development. The concept may have taken some time to catch on in the industry, but today, more biopharmaceutical development and manufacturing programs are showing what is possible when ideas such as pharmaceutical quality by design (QbD) are combined with better analytics and automation, as well as with equipment that reflects the principles of process intensification.

These abstractions are taking shape most notably in process development, which has become inextricably linked with scale-up. More development projects now start off with clear scalability goals in mind. Developers are collaborating more closely with technology vendors, contract development and manufacturing organizations (CDMOs), and suppliers to achieve these goals. Among the greatest challenges that CDMOs and technology developers agree on are:

Reduced project timelines and the need to reduce costs

Increased numbers of more complex biotherapies

Widepread adoption of single-use equipment at all phases of product development

The need for cell line optimization at the earliest phases.

Automation is driving improvement, according to Joe Makowiecki, director of business development for Cytiva’s FlexFactory and KuBio product lines. He sees the benefits becoming more visible as pharma moves to more real-time monitoring and control, and to applying the principles of Pharma 4.0. “Automation will lead to development of processes that are manufacturing-ready in less time,” he says.

This trend, in turn, should feed continuous improvement. “As they are scaled up, processes will yield much more information, especially on the interdependency between process parameters, allowing development to move faster and to focus only on experiments that will add the most value,” Makowiecki adds.

Technology vendors are incorporating automation and adopting new approaches to help companies achieve these goals. However, they must be careful to balance the need for more data with the need to keep systems from becoming too complex, says Loe Cameron, director of analytics and  controls at Pall Biotech.

“Like many others in the industry, Pall is on a Pharma 4.0 journey, assembling solutions and assessing the impact. Our research is focused on how we can provide tools in a way that does not negatively impact the usability of our systems, and we work closely with users of these systems to reach that goal,” Cameron says.

Pall has installed automation systems in its process development labs that mimic many of the functions found in full-scale manufacturing. “This capability allows customers to project the future of their process as it scales more effectively,” says Cameron. “We have connected this system to a centralized historian database, which has allowed us to more deeply explore how solutions such as soft sensors, modeling, and advanced process controls can improve viral vector processes at each scale, and over the full life of the process,” she says. However, Cameron notes, it is important to remember that taking these steps has a cascading effect. “As we install more sensors to support the automation of our processes, so much data are then created that the analysis must be automated as well,” she says.

Those involved in process development say they are approaching projects quite differently from the way they might have tackled them a few years ago. Broadly, the need for speed (if not to market, then to the next milestone or developmental stage) and early-stage robustness are key drivers, says Patrick McMahon, global operations manager for integrated services at Cytiva.

Biopharmaceutical process development teams are linking the drug discovery phase to process development and scalability early on, leveraging platforms and predictive technology, says McMahon. “This allows them to reduce project timelines, drive in quality and robustness, and reduce the need for repeat work and comparability for process or analytical changes,” McMahon explains.

Process development specialists see efforts as becoming increasingly collaborative. “Across the industry, we see more and more companies working together to overcome challenges. CDMOs are performing a wider range of services, and equipment manufacturers and media suppliers are also contributing to process development efforts,” says Jean-Christophe Drugmand, senior bioprocess architect at Univercells Technologies. He recalls working on projects where media optimization studies were subcontracted to the media supplier, and some QbD studies were also handled by specialists in that area to accelerate the overall project execution.

One fairly recent, and fundamental, challenge to bioprocess development has been the increasing complexity of biotherapies, as more companies pursue gene and cell therapies and multi-specific antibodies. “Historically, process development followed the philosophy that more product equates to a better process, but as molecular complexity (and with it, sensitivity) has increased, we must think about the balance between productivity and quality,” says Timothy Morris, group leader of manufacturing process development at Catalent Biologics. In this new environment, Morris says, process development teams must think about cell culture as the product of both the molecule and the bioreactor. For example, he explains, increased antibody sialylation is viewed as being representative of increased residence time in the body.

“Complex bioproducts require a different approach when it comes to developing robust and scalable processes, and we must focus on developing processes and analytical methods based on the particular characteristics of a molecule,” says Atul Mohindra, senior director for biomanufacturing, R&D, Lonza Pharma & Biotech. Companies are moving to reducing the risk of failure as early in process development as possible. Cell line development is becoming a critical area of focus, McMahon says. This had often been a problem in the past, as development efforts sped quickly only to find that the cell line selected simply did not work.

“Optimizing cell lines reduces risk early on. It also forces the focus from the immediate task at hand, whether feasibility or a clinical milestone, to longer-term success for scalability and regulatory compliance with foresight built in. Using the best quality cell culture media helps further improve critical quality attributes,” McMahon says.

“Understanding how the process will impact the product is very powerful,” he says, recalling a biosimilars development project that Cytiva had worked on with a client, which used predictive modeling to understand critical quality attributes (CQAs) such as glycosylation patterns, and better correlate them with critical process parameters (CPPs), earlier in development. As a result of these efforts, the team improved the glycosylation pattern and charge variance profile, yielding a molecule with quality attributes more similar to the innovator molecule.

Currently, early stage automation supports the generation of high producing clones capable of producing optimum quality molecules, says McMahon. In general, the use of automation not only improves the overall development process, but also clinical and commercial manufacturing.

“Improved early small-scale process models have become increasingly important, and tools to improve the insight we get from these and provide data earlier in the development timeline are proving vital to de-risk early decisions,” says Colin Jaques, technical director of R&D, with Lonza Pharma & Biotech.

Catalent uses its proprietary GPEx and GPEx Boost cell line development technologies to generate high yielding clones, Morris explains. “While clone selection is typically based on yield, we have matrixed in processes to evaluate clones for both yield and robustness. This approach combines the cell line technology with the ambr15/ambr250 bioreactors, and our advanced analytics to support a robust scalable strategy,” he says.

Morris recalls one program in which multiple high-yielding clones came off the Berkeley Lights Beacon clonal selection system. “We were able to take many of those clones into our microbioreactors. However, if we had picked candidates solely based on yield, the third-ranked clone would have been missed, even though it increased its yield at the CGMP scale. This clone has been robust in our failure mode process, whereas the top two showed sensitivity to known failure modes that may not occur in the CGMP process,” he says. As this case clearly shows, a robust process relies on a robust clone, so selection is crucial, Morris adds.

High-throughput analytics makes key data (e.g., CQAs) available for decisions at this stage of cell line construction. “Mathematical models are essential in dealing with the volume of data arising from high throughput scale down systems,” Jaques explains. During the early stages of process development, more high-throughput screening (HTS) and design of experiment (DoE) approaches are being integrated, says Drugmand. He sees more process development teams leveraging mathematical modeling and advanced analytical technologies. “This approach helps increase our understanding of the process and reduce costs through automation while contributing to better process control,” he says.

In addition, Drugmand says, process development teams are routinely using statistical analysis and process analytical technologies (PAT) to review processes and results. “It’s now a common component of process development initiatives, which is a change from the past,” he says. Process modeling tools are also being explored for scalable DoE options, and scale-down models are being used to streamline process development and scale-up. 

Catalent’s bioprocess development team has created many product specifications based on historical information on the molecule or type of molecules involved, says Morris. “We take in-process samples to ensure that we understand how process changes affect the product,” he says, noting use of systems such as ambr 15 and ambr 250, which allow the team to perform a DoE to reduce the chances of failure early in the development process.

“The availability of high throughput, high-fidelity, small-scale bioreactor models has facilitated the use of response surface optimization approaches to process and medium development,”says Jaques. “Improvements in high-throughput product characteristics assays have enabled optimization techniques to be broadened out to more process outputs,” he says, “while response surface methodology allows propagation-of-error techniques to be used to improve process robustness in those platforms.”

Lonza’s team routinely uses small-scale ambr systems, Tecan robots, and high throughput analytics to select clones, screen purification resins and to identify optimal conditions, adds Mohindra. In addition, the company is installing more in-line process measurement technology, which he expects to further improve process control.

Process development specialists note a trend toward greater use of single-use technologies. “Biologics manufacturers are looking to these technologies to reduce capital expenditures and utilities consumption, increase flexibility, and reduce changeover times,” says Mohammed Elfar, product innovation manager at Univercells Technologies. Elfar notes that use of continuous processing is also increasing as a way to help increase productivity and reduce of goods sold. Yet the move to single-use unit operations has introduced technical challenges to process developers and scale-up teams. “Many modern sites can’t handle hot feed preparations, which can lead to challenges in developing concentrated feeds,” Jaques explains. “Furthermore, single-use bioreactors are often underpowered and undersized for gas flow, requiring that compromises be adopted for gassing strategies,” he says.

As high titers have become the rule upstream, recent years have brought a shift in focus from upstream to downstream process development. “Whole process considerations (instead of a focus on cell culture only) are helping to streamline efforts,” says Drugmand. “Unit operations are being intensified and integrated to optimize upstream and downstream work in parallel for overall process optimization,” he adds. “Process development scientists and engineers are being challenged to cope with larger harvests. This is especially true in the case of monoclonal antibodies, where dramatic titer increases have been observed over the years,” says Elfar.

There is also a greater emphasis on the interface between upstream and the downstream processes, says Jaques. “Perversely, this trend has led us away from small-scale high-throughput models for platform process development because there are no representative primary recovery or purification tools at that scale,” Jaques says. This shift has involved measuring more biological properties of the harvest materials and the development of models to understand the impact of the parameters on the quality of the feed stream going into and coming out of the downstream process, he explains.

Clearly, bioprocess development, like scale-up, has become more complex and challenging for the scientists and engineers who drive these efforts. However, new technologies, acquired process experience, and better access to relevant historical data are allowing biopharma teams to make the overall operation more systematic and less risky.


Vol. 44, No. 11
November 2020
Pages: 16-22

When citing this article, please refer to it as  A. Shanley, "Bringing Bioprocess Development into Focus," 

As biopharmaceutical process development advances, a more collaborative and knowledge-based approach, driven by better analytics and high-throughput system integration, is reducing the risk of failure.



