The pandemic cast new light on the importance and vulnerabilities of the distributed nature of the pharmaceutical industry’s specialized supply and shipping infrastructure. Pharmaceutical scale permits, like those for semiconductor or surgical hospital consortiums, lower barriers to entry for otherwise fantastically complicated or fit-for-one-purpose-only technologies to gain traction. Scale also enables designs dedicated to eking out the last precious microns of performance or purity, which can determine the usefulness or failure of an approach. But specialization at this scale also proliferates nesting doll dependency. Each successive layered network of interlocking manufacture depends on an exact product or design specification to feed it. Shortfalls in one seemingly survivable area cascade like crashing waves onto the following layer, and so on throughout the system. Subcontractors focused narrowly to a single design task are reliant on yet more specialized subcontractors. By intention, all are equally synchronized in configuration, tolerance, and material conformation. Seeking alternatives is not simple, easy, or at times even possible. It was a growing source of pride to become so finely tuned that pharmaceutical manufacturing migrated to a just-in-time (JIT) mode to maximize these benefits, but this also exacerbated interdependencies.

Multiple industry contacts recently emphasized better outcomes these past two years from having increased frequency and depth of dialogue. At the 2021 AAPS, PharmSci 360 event in Philadelphia, Gabe Longoria, chief commercial officer at Astrea Bioseparations, encapsulated this well, saying, “I think supply chain challenges have forced a lot of organizations to actually align their overall focus on what they are trying to achieve. And because of these challenges, we’ve had to identify better ways of doing things, such as intelligent standardization protocols. And these same challenges have required finding better ways to provide resources not just to our customers but B2B as well. I think it’s important to recognize and note down those challenges, a lot of which we’ll continue to see in the next 9–12 months, and even after we come out of COVID, we’ll need to continue to look to better align resources toward goals. We’ve been forced to help one another, and we’ve been talking with our competitors to make sure we can do the right thing for our customers. Which has been intensely interesting. The end point is asking, what does the customer experience look like? And all organizations are looking to achieve success in this regard.” Direct and frequent conversation became the norm rather than a luxurious exception, driven by dire need more than anything else.

From a logistics and innovation perspective, the pandemic can be considered a smaller scale version of World War II. Development and deployment of radar is a parallel for vaccines in that metaphor. In each case they had been around for many decades, both had to be suddenly optimized and brought to scale in an abbreviated timeline. Each required highly skilled operators working in intensively coordinated interlocking systems. But both were only a key part of the overall response, not the sole basis for defense. Historians and analysts will grasp onto vaccine production when writing accounts of what happened during the pandemic response—just as they did with radar for the Second World War. But that falls short of telling the full story. Calls currently abound for a global commitment to a fully coordinated response next time around. Hence, the need to dig into the details, so that lessons learned are not so easily lost. The obvious is also worth pointing out.Pandemic pressures have transformed pharmaceutical manufacturing in a similar way that Amazon next day delivery became an inviolate rule for the shipping industry, dragging everyone in shipping to deliver at the same accelerated pace. However, vulnerabilities and downsides of JIT manufacturing and specialization were widely recognized well before the pandemic. FDA and the National Institute for Standard and Technology (NIST) came together to draw up plans to implement advanced manufacturing for medical supplies. Other groups became involved. A new manufacturing institute, BioMADE, supported by both industry and the defense department, began to organize.

Kevin Nepveux, vice president, Pfizer Global Supply, recalled, “Development and launch of the Pfizer-BioNTech COVID-19 vaccine was unlike anything I have experienced in my career in Pharma Manufacturing—and without a doubt the highlight. Multiple mRNA vaccine constructs were manufactured and tested in the clinic in parallel versus the normal sequential approach, and the selected strain was immediately scaled, both up and out, to begin manufacturing launch supplies before the pivotal efficacy study even started. In manufacturing alone, Pfizer/BioNTech committed over $1B in CapEx and OpEx before the first sign of efficacy.”

Pfizer went all in, coming out the other side having a changed perspective, as Kevin observes. “Learnings from the vaccine will be enduring, including streamlined internal governance, bringing on multiple supply chains simultaneously compared to normal post-approval expansion, and simplified regulatory engagement.”

FDA’s commitment to conversation and cooperation

Which brings us nicely to FDA, which might otherwise unfairly drop to a footnote to the vaccine success. FDA created a dialogue framework for COVID-19 vaccines to be developed at warp speed. Communication was excellent. In November 2021, FDA updated its resiliency roadmap, pointing out that since April 1, 2021, FDA conducted more than 600 domestic and more than 200 foreign remote regulatory assessments, which included review of records submitted upon request under section 704(a)(4) authority, as well as review of documents and other information voluntarily submitted upon request … more than 100 human and animal food firms participated in voluntary remote regulatory assessments (1).

While these assessments cannot be used to meet FDA Food Safety and Modernization Act (FSMA) mandated surveillance inspections, the assessments provided critical oversight to FDA during the pandemic. Dr. Elizabeth Miller, assistant commissioner for the Office of Regulatory Affairs (ORA) Office of Medical Products and Tobacco Operations, commented, “Throughout the course of this pandemic ORA has had to adjust and change how we work.We have been more agile than ever before.Our regulatory agility is an important cornerstone to delivering the oversight and consumer protection the public deserves from ORA and FDA.”FDA continued with mission critical inspections when they paused routine surveillance. They also developed a new process to assess facilities submitting Emergency Use Authorizations. These investigational activities allow the agency to directly observe operations at manufacturers of COVID-19 therapeutics and assess whether the firm has proper controls in place to ensure the safety, purity, and potency of the COVID-19 product.

It was FDA’s open and frequent communications that helped keep everything rolling forward as smoothly as possible. FDA made strong efforts at transparency in their Resiliency Roadmap, and have energetic plans to continue this as they inaugurate an Inspection Affairs Council to update inspection procedures. This new body within FDA will be headed up by Dr. Judy McMeekin, associate commissioner for regulatory affairs. It is a transformative solution born in part by the elevated contacts and cooperation during the pandemic.

Other lines of defense also needed bolstering, and these are the success stories hidden by the glare of vaccine successes. Hovione CEO Guy Villax said at the 2021 CPhI Worldwide conference in Milan, “In the early days of the pandemic, there were hiccups and delays, but that we didn’t see huge problems points to the resilience of the industry … the industry stepped up to the challenge. Gilead’s remdesivir has a critical excipient which only Hovione manufactures, we massively scaled up to produce a years’ production in only a month. CDMOs [contract development and manufacturing organizations] were doing heavy lifting to support the needs brought on suddenly by COVID-19.”

Susan Darling, senior director of capillary electrophoresis and biopharma product management and marketing at SCIEX, echoes the sentiment but also emphasizes the knock-on effects of increased calls for speed (see page 32 in this issue of BioPharm International). Darling comments, “Companies are looking to get development times down to 1000 days; they want to develop everything at pandemic speed, even when they have diverse portfolios that include highly complex next-generation products such as multispecifics, gene and cell therapies, and RNA-based treatments, all of which require new and different analytical capabilities.” So the trend line is becoming obvious. More complex molecules at faster speeds for development equals an opportunity for analytical tool makers to sell into a growing unmet need. Manufacturers must adapt quickly to new types of technologies more quickly because their competitors are already doing so. This will also have a knock on effect for products for purification and separations in both upstream and downstream processes, such as novel chromatography materials or quicker acting buffers and solvents, and so on. It is a defining moment for future planning requirements. “The industry is really driving to be creative, not just in terms of analytical solutions, but in how they manage data and pull it together to knock the inefficiencies out of every step,” Darling concludes. “Our customers are challenging us to develop new analytical techniques that provide more information more quickly, and when we bring them new solutions, they are adopting them at a much faster rate than could ever be imagined even a decade ago,” Darling says.

Corning Incorporated also stepped up to the challenge during the pandemic, in part supported by the Department of Defense (DOD), in coordination with Department of Health and Human Services (HHS), who awarded a $15 million Firm Fixed Price contract to Corning Incorporated to expand domestic production of pipette tips for robotic liquid handling instruments (2). But more well known was the Biomedical Advanced Research and Development Authority agreement to give $204 million to expand manufacturing capacity for glass vials to store potential vaccines for COVID-19 (3). Valor Glass vials, first introduced in 2017 and made from aluminosilicate glass, were of great strategic significance. “Velocity Vials” are now being introduced. And what more could a marketing department tell us about the business importance of “Amazoning”, than to use velocity in the product name. Velocity Vials are borosilicate externally coated with Corning’s proprietary technology, to eliminate bottlenecks and speed the delivery of life-saving treatments. These vials have demonstrated a 20–50% improvement in efficiency, up to a 96% reduction in glass particulates, and a 3x reduction in the likelihood of damage that leads to cracks and breaks, according to the marketing materials (4).

Other than the headline vaccines, much of great importance went on or evolved behind the scenes to manufacture excellence out of adversity in the fight against COVID-19. Landmark efforts in communications and cooperation kept the conveyer belts moving. Agilent’s digital solutions group “ramped up to meet the needs of customers by offering cloud-based software for remote instrument access and asset utilization monitoring (to) pulls together advanced IoT sensor technology and data analytics to enable lab-wide visibility,” according to Shane Elliott, vice president, Global Laboratory Solutions Marketing, Agilent.This effort was broad based across multiple vendor platforms, again in a concerted effort to increase and improve manufacturing communications. It was a near universal practice over the past two years. Production line pace is set to accelerate rather than slow down. Pharma production has been industrious in the good old-fashioned sense of the word. Ingenuity and hard work manufactured positive results out of adverse conditions. When serious global policies begin to become clear for the next pandemic planning groups, industrious effort, ingenuity, and improved communications will be the seed crystals for successful outcomes.

 Resiliency Roadmap for FDA Inspectional Oversight (November 2021)

, FDA.gov.
2. DoD, “DOD Awards $15 Million Firm Fixed Price Contract to Corning Incorporated to Increase Domestic Production Capacity of Robotic Pipette Tips,”

, Dec. 23, 2020.
3. BioPharm International Editors, “Corning Expands Glass Vial Capacity,” 

, June 17, 2020.
4. Corning, Velocity Vials, Corning.com, accessed Dec. 14, 2021.


Vol. 35, No. 1
January 2022
Pages: 10–13

When referring to this article, please cite it as C. Spivey, “Industrious Pharmaceuticals,” 

Articles

Cover Story

Warp speed is amazoning pharma practices and protocols.

Industrious Pharmaceuticals

Last year was meant to be a transitory year for the bio/pharma industry. With vaccinations deploying in December 2020, there was hope among many of last year’s respondents to the 

Pharmaceutical Technology/BioPharm International 

annual employment survey that the overall state of the industry would bounce back in 2021. More than 75% of survey respondents expected the industry to improve, while 64.3% expected business at their employer to improve (1).

However, while most of the world has moved past the quarantine mandates that were commonplace at the onset of the COVID-19 pandemic, progress has been hampered by various unfortunate developments. These include highly transmissible variants, inequitable vaccine distribution, and a reluctance to receive a vaccine among large portions of the global population.

This year’s employment survey focused on the state of this transition. More than 300 respondents answered questions about the current state of their workplace, pandemic-related changes, and where they believe the industry will be headed in the future Refer to the 

Last year’s survey observed a dramatic decrease in many core metrics. In 2019, 55.4% of respondents received increased pay; this dropped to 41.7% in 2020. These lost numbers seemed to funnel into the “no changes” bucket, which rose from 33.2% to 44.8%. However, there was an increase in pay cuts, which rose from 11.4% to 13.7%.

The 2021 numbers have recovered to a state comparable to 2019. Approximately 53% of respondents reported a pay increase and 39.0% reported no salary changes, a difference of -2.3% and 5.8%, respectively. Interestingly, pay cuts hit a three-year low, with just 8.0% of respondents reporting a decrease in compensation

A similar trend can be observed in job security metrics. From 2019 to 2020, individuals who felt increased job security (relative to last year’s survey) decreased from 30.0% to 17.8%. Those who felt less secure slightly increased from 30.6% to 34.4%, while those feeling no change jumped from 39.4% to 47.8%.

In 2021, there were improved metrics across the board relative to both 2019 and 2020. Increased job security hit a three-year high of 30.4%, a slight increase on 2019, while those who felt decreased job security hit a deep new low of 17.00%. Those reporting no changes saw the biggest bump, with over half of respondents (52.6%) reporting that they felt no changes in job security.

While these trends are encouraging, it should be noted that one limitation of the survey is that these values are measured relatively, which necessitates additional context in their evaluation. For instance, if 2020 can be viewed as a down year, then returning to pre-pandemic levels of compensation/job security would inherently necessitate higher levels to compensate. If exact quantitative measures (for instance, a job security confidence scale) were implemented, we may receive further industry insights.

Pandemic-Related Workplace Changes May Be Here to Stay

Following the onset of the pandemic, bio/pharma companies began implementing new workplace measures and procedures in order to prioritize the safety of their workforce and mitigate the spread of COVID-19. In our 2020 survey, we asked respondents if their company had implemented any of the following procedures:

A. Changed work schedule to social distance on the job

D. Changed or reduced development program

E. Additional work hours on pandemic-related projects

This year, we asked respondents these same questions to measure what, if anything, had changed in widespread implementation of these practices.

Only four measures —C, A, F, and H—saw a decrease: 10.4%, 9.1 %, 4.7%, and 3.4%, respectively. Of these, measure H indicates no significant changes, which suggests more workers felt their company had implemented critical, pandemic-related changes to their workplace.

The overall headline of this data-set is implementation of work-from-home in bio/pharma workplaces. At the time of last year’s survey, a little over half (53.9%) of respondents stated that their company had implemented this measure. However, in this year’s survey, this number jumped to 70.5% of respondents.

While prior questions helped give a contemporary view of the pandemic’s impact, we also wanted to determine to what extent pandemic-related measures would remain in the long-term. This would allow us to not only get a sense of individual workplaces, but also provide commentary on the industry as a whole. In that vein, we asked respondents predictive questions about two of these processes: remote inspections and abbreviated approval pathways.

Remote inspections have emerged as an alternative to traditional on-site inspections, as they allow sites to maintain quality control compliance while also encouraging social distancing and accommodating workflow changes. When questioned to what degree these would occur once the COVID-19 pandemic “significantly subsided”, most respondents (62.4%) felt that they would continue at a higher-rate than before the pandemic. Conversely, just 26.6% felt that it would return to the previous model, while only 11% felt the current increase would continue.

The results were similar as it concerned abbreviated approval pathways. A majority of respondents (81.1%) felt that either “most” or “some” of these measures would continue to exist once the pandemic significantly subsides, while just 18.8% felt that they would completely/almost completely continue to exist or disappear.

To this point, the survey has supported the notion that the COVID-19 pandemic has, is, and will continue to impact the bio/pharma industry for the foreseeable future. However, various metrics that indicate a degree of economic recovery, such as job security and salary changes, can also be observed. It is perhaps because of this that we are seeing an increased degree of optimism among respondents.

 indicate, most respondents felt that their company (47.4%) and the bio/pharma industry (53.2%) exceeded their personal expectations in 2021. Additionally, the majority (58.4% and 60.5%, respectively) also felt that this upward trajectory of both would continue into 2022.

The precise future direction of the bio/pharma industry will be in question for as long the pandemic continues. However, as this year’s survey shows, the prevailing attitude is that this impact will not prevent future growth and development. Changes brought on by the pandemic would not have occurred without it, but that by no means equates to entirely
negatively impacts.

R. Peters, “Compensation Does Not Necessarily Align With Job Satisfaction,” 

Grant Playter is the Assistant Editor at 

When referring to this article, please cite it as G. Playter, “The Long Path to Normal,” 

Special Coverage

Survey respondents reflect on the state and future of the bio/pharma employment.

The Long Path to Normal

As COVID-19 swept across the world, it has impacted millions of individuals and families. During that time, the biopharmaceutical industry came together to create treatments as swiftly as possible. As of January 2022, the main types of COVID-19 vaccines to treat SARS-CoV-2 that are approved or authorized include (1):

, which use genetically engineered mRNA to teach cells how to make a protein that triggers an immune response (ex: Pfizer-BioNTech and Moderna’s COVID-19 vaccines).

 give cells genetic instructions to produce antigens, using a different, harmless virus from the one the vaccine is targeting (e.g., Johnson & Johnson’s [J&J] COVID-19 vaccine).

Telma Lery, Janssen EMEA therapeutic area lead for infectious diseases and vaccines at Johnson & Johnson, explains that the company’s COVID-19 vaccine candidate (also known as Ad26.COV2.S, Ad26COVS1, VAC31518, JNJ-78436735, or Ad26-S.PP) is a monovalent, recombinant, inactivated/non-infective adenovirus vector (similar to a cold virus), which contains a transgene that codes for the coronavirus spike (S) protein.

Other vaccines are in the works as well. Some of the COVID-19 vaccines in clinical trials include—but are not limited to—whole virus, protein subunit, and nucleic acid vaccines (RNA and DNA) as well as more mRNA and viral vector vaccines (2). Vaxart, an American biotechnology company, is developing oral recombinant vaccines, one of which is in clinical trials in the United States as a candidate to prevent COVID-19.

“‘Recombinant vaccine’ means using DNA technology to make a protein target (i.e., antigen), rather than using the native virus to make the antigen,” says Sean Tucker, chief scientific officer of Vaxart. “Most flu vaccines [are] made in eggs by growing a flu virus, deactivating it, and injecting it in a vaccine. We use a dead common cold virus to deliver the DNA encoding the antigen to the person, where the antigen is then made. Our vaccine delivers the DNA via a tablet that breaks open in the intestine. Our vaccine works because we have a molecule that tells the immune system to pay attention to the antigen that is being made.”

“Most recombinant vaccines are made in a liquid that is injected into the arm, like the Hepatitis B vaccine. Our vaccine is dried and tableted with inert ingredients; so, it can be swallowed,” Tucker explains.

J&J has worked with industry partners, governments, and health authorities since the beginning of the COVID-19 pandemic to develop a preventative COVID-19 vaccine candidate, Lery explains. But prior to the pandemic, the company researched vaccine technology, which Lery believes granted J&J a head start in the development of its COVID-19 vaccine. The vaccine technology, AdVac, is based on the development and production of adenovirus vectors (gene carriers).

“There are 57 identified distinct variations (serotypes) of adenovirus, a type of virus that causes the common cold. Adenovirus vectors (gene carriers) are genetically altered forms of an adenovirus and lack the DNA needed to replicate. So, the vaccine cannot cause a cold, and the protein it produces cannot cause harm either,” says Lery, adding that the vaccine technology makes use of adenovirus vector Ad26, specifically.

One of the differentiating factors in the development of J&J’s COVID-19 vaccine is what could be termed “multitasking efforts.” The company combined phases of early-stage trials and had its discovery and manufacturing teams conduct processes in parallel—rather than sequentially.

The process for developing an oral recombinant vaccine, on the other hand, is similar to an injectable vaccine in terms of the clinical trials and the path to approval, according to Tucker.

“We began by identifying six vaccine candidates and winnowed the list to two,” Tucker continues. “Our initial Phase I trial tested a version that made two different antigens from SARS-CoV-2, the S protein and the nucleocapsid (N) protein. That S+N candidate made very strong T cell responses and nasal antibody responses that were cross-reactive against a wide variety of coronaviruses. The current set of trials (Phase II in the US) is evaluating an S-only protein approach, which made better antibody responses in non-human primates.”

But the process of identifying and progressing to clinical trials in a pandemic wasn’t without its challenges. In early 2021, Vaxart had to compete with other biotech companies for manufacturing space. Fortunately, the company had two plants for producing pills for their clinical trials and had enough materials on hand.

Some of the main differences between a COVID-19 oral tablet vaccine from injectable COVID-19 vaccines, according to Tucker, include:

Immune responses are triggered in the mucosa and the serum with a COVID-19 oral tablet vaccine (based on preclinical and clinical testing), unlike the existing vaccines that are serum-only.

Ease of distribution and vaccination with a COVID-19 oral tablet vaccine, which eliminates the need for special accommodations to transport injectable vaccines that require cold chain infrastructure as well as the need for trained professionals to administer vaccines.

The research behind COVID-19 vaccines: antigens

Understandably, the COVID-19 vaccines weren’t created without upfront heavy lifting. One company, The Native Antigen Company, part of LGC Clinical Diagnostics, provides commercially available S1 and S2 antigens for SARS-CoV-2, which include the amino mutations associated with different variants. The company developed more than 100 antigens, antibodies, receptors, and kits for the coronavirus, which have been used by diagnostic, therapeutic, vaccine, and public health organizations to study SARS-CoV-2 and develop countermeasures against it.

Vaccine developers can use S mutant antigens to raise mutant-specific antibodies and develop assays (often enzyme-linked immunosorbent assays) to assess the efficacy of their vaccines, according to James Shore, technical product manager of The Native Antigen Company. In addition, biopharmaceutical companies use The Native Antigen Company’s reagents to develop monoclonal antibody therapies against SARS-CoV-2 and its variants. But how can S antigens be adapted to address the various variants?

“If a variant includes only a couple of point mutations, then a modification of PCR [polymerase chain reaction] amplification can be used to introduce changes to the gene sequence,” says Shore. “More commonly, however, the sequence is computationally redesigned and chemically re-synthesized.”

No matter the type, COVID-19 vaccines and treatments must address new variants and mutations of the virus in some capacity. Some of the variants include Delta, Mu, and Omicron.

“COVID-19 variants arise through mutations in the virus. Mutations may result in the virus obtaining increase in transmissibility, more severe disease, reduction in vaccine or treatment efficacy, etc.,” says Kristina Hsieh, scientific affairs lead, Clear Labs. “There are numerous real-world studies that have shown that getting vaccinated reduces one’s risk of severe illness, hospitalization, and death from the different variants. One of the articles recently published by Bajema et al. indicated a reduction in hospitalization due to vaccination from data gathered from five Veterans Affairs Medical Centers” (3–5).

Hsieh adds that the main difference, genetically, between the variants are different mutations that have accumulated throughout the virus. As the Delta variant has demonstrated, some mutations may lead to increased transmissibility and/or more serious illness. In addition, one of the new variants, Omicron, has been detected in the US.

“The CDC and its public health partners are working closely to monitor this variant,” says Hsieh. “Through Whole Genome Sequencing (WGS), public health laboratories are able to quickly identify, track, and monitor the variant.”

In addition to monitoring variants of the virus and being vigilant in the research and development of new COVID-19 vaccines and treatments, the industry needs to continue to communicate and share resources. Consistent communication and collaboration between all organizations in the biopharmaceutical industry—from governments, to agencies, to developers, to manufacturers, and everyone in between—will play a huge role in combatting the spread and impact of COVID-19.

Gavi, “There are Four Types of COVID-19 Vaccines: Here’s How They Work,” 

M. W. Tenforde et al., “Association Between mRNA Vaccination and COVID-19 Hospitalization and Disease Severity,” 

H. Chung et al., “Effectiveness of BNT162b2 and mRNA-1273 Covid-19 Vaccines Against Symptomatic SARS-CoV-2 Infection and Severe Covid-19 Outcomes in Ontario, Canada: Test Negative Design Study,” 

K. L. Bajema et al., Morbidity and Mortality Weekly Report, 

Effectiveness of COVID-19 mRNA Vaccines Against COVID-19–Associated Hospitalization — Five Veterans Affairs Medical Centers, United States, February 1–August 6, 2021


Volume 35, Number 1
January 2022
Pages: 18–19

When referring to this article, please cite it as M. Rivers, "What’s Next for COVID-19 Vaccines and Variants?", 

Although mRNA and viral vector vaccines have been top of mind for COVID-19 treatments, other technologies and treatments are emerging—along with new variants. 

What’s Next for COVID-19 Vaccines and Variants? 

Upstream process development in biologics has seen several improvements in robustness, productivity, and stability. The wealth of bioprocess data generated can be potentially utilized by modeling and data analytics tools, such as mechanistic modeling, machine learning (ML), and artificial intelligence (AI), to gain process knowledge and perform predictions. However, the industrial applications of these tools still leave a lot to be desired. Though many applications of such techniques are found in industry, their widespread application is still not prevalent. This review revisits the data analytics modeling methodologies for upstream processes, provides a perspective on their potential applications across the upstream process development to validation workflow in the biopharma industry, and presents a value chain peak to use it toward better process robustness, process control, and process monitoring with quality by design (QbD) and process analytical technology (PAT) applications. 

In addition to predictive power and improved process understanding, the application of mathematical modeling based methods in biopharma process development has the benefits of increased visibility and agility, which can potentially improve productivity. The role of mathematical modeling for drug development with respect to QbD has already been exemplified by FDA. Furthermore, there is an increased focus on mathematical modeling approaches due to their robustness, predictive power, and understanding driven by PAT and QbD concepts defined by FDA and the European Medicines Agency.

A typical end-to-end upstream process workflow consists of cell-line development, selection of appropriate clones, process development, scale-up, risk assessment scale down model (SDM) development, process characterization, and technology transfer to manufacturing. This review will focus only on the modeling techniques used across upstream process development for biopharmaceutical therapeutics.

Upstream process development includes conceptualization of the process train—including media and feed development—and optimization of bioreactor parameters for a successful scale-up. It also identifies and assesses critical process parameters (CPPs) that influence the critical quality attributes (CQAs) of the product through activities such as process characterization studies. Different types of bioprocess controls are available, and model predictive control (MPC) seems promising over others. An overview of applications of statistical, mechanistic, machine-learning, and hybrid models in upstream process development, optimization, and characterization is summarized in 

, while the available commercial technologies for upstream process development is presented in 

Because of the presence of a large number of correlated decision variables and objectives, the statistical techniques are best suited for cell-culture processing and are applied for defining the design space; improving cell growth, titer, and glycosylation; performing root cause analysis; predicting CQAs; studying interactions for scale-up parameters; scaling-up/scaling-down from clone to bench-scale; and controlling process parameters across scales. During process development, scale-down models are important because they can replicate results at larger scales, where it is impractical to perform factorial experiments. Pfizer developed a system wherein they used a total of 180 micro-bioreactors to assess scaled-down parameters compared to conventional 3-L bioreactors (1). State-of-the-art statistical modeling tools like SIMCA and MODDE from Sartorius, JMP from SAS, Unscrambler from CAMO, and DesignExpert from Statease have also contributed to the ubiquity of statistical methods in the biologics industry.

The primary purpose of mechanistic models is to develop a mechanism-based relationship between inputs and outputs, and they can be used as a predictive tool, once validated. Flux balance analysis (FBA) was applied at UCB Pharma by describing the evolution of intracellular fluxes for four industrial cell lines through a curated Chinese hamster ovary (CHO) cell genome scale metabolic model (GSM) (2), which exemplifies the utility of GSMs. Another study at Bristol-Myers Squibb demonstrated media optimization and process understanding through FBA (3).

In most situations, detailing every aspect through microscopic cellular models is analytically and computationally expensive. For such scenarios, macroscopic kinetic models can provide enough information for process optimization and to test hypotheses and make predictions. Even today, the simplest unstructured-unsegregated Monod kinetics-based models are most commonly preferred even for multicomponent CHO growth kinetics. Even with the limitations in parameter estimation methods and kinetic data, the abundance of multi-omics data gives the scope to add regulatory, signaling, and product-related information to eventually form large-scale models. These models can be applied to gain process understanding and for process optimization, process design, and process control. The kinetic model equations can also be combined with complex metabolic pathways to describe the dynamics in cell culture trends and help tackle process challenges. Some of the capabilities demonstrated by the kinetic modeling of CHO cell culture include analysis of metabolic behavior, glycosylation prediction through feed development, incorporating temperature shift changes, and explaining glycolysis through a kinetic-metabolic model. Similar methodologies are also applied to explain the dynamics in suspension culture for the other expression systems like baculovirus and Vero cell expression systems.

Because mechanistic models cannot be implemented for processes that are not yet fully understood and data-driven models suffer from being unreliable beyond their input data-set, hybrid modeling is a promising approach that aims to establish a combination of the mechanistic and data-driven framework to obtain better flexibility and robustness. A few comprehensive reviews discuss the hybrid models as an element of process systems engineering (4). Combining GSM, bioreactor kinetics, and artificial neural network analysis (ANN), Insilico Biotechnology has claimed to have a commercial hybrid modeling workflow, also known as a ‘digital twin’ for biopharmaceutical process development (5) and clone selection (6). Datahow is another such organization that provides stand-alone software as well as consulting services to apply deterministic and deterministic knowledge-based hybrid models for process development and process monitoring. Software solutions for hybrid modeling are also provided by Novasign based on mechanistic and statistical models. Novasign has demonstrated the approach of intensified-DoE on an industrial 

To ensure that product heterogeneity of complex and fragile therapeutic proteins remains within predefined specifications, bioprocesses must be monitored and controlled through well-established measurement systems to enable their real-time control. This monitoring is also crucial to ensure that the workflow remains within the recommendations of FDA on PAT (8), which incorporates quality risk management and process understanding. According to the location of the analytical system, the bioreactor monitoring techniques can be classified as off-line, on-line, and at-line. Off-line methods for process monitoring are time-consuming and therefore cannot provide a real-time picture of the ongoing process. Ideally, the sensor should be non-invasive, non-destructive, robust, fast, sensitive, and adaptable to the dynamic conditions of the bioreactor. It should also be able to generate good quality multivariate data for data analysis techniques like chemometrics. Several spectroscopic techniques, such as ultraviolet-visible (UV-vis), near-infrared (NIR), mid-infrared (MIR), dielectric spectroscopy, Raman spectroscopy, and fluorescent spectroscopy have been investigated for their usefulness in bioprocess monitoring. Of these, NIR and Raman spectroscopy are most popular in mammalian cell culture (9,10) and cell therapy (11). A direct comparison between NIR and Raman reveals the superiority of the latter in terms of sensitivity, selectivity, and limit of detection. However, since NIR is more precise but less accurate, it is not affected to a larger extent by minor perturbations in spectra compared to Raman (12). Several monitoring systems, such as ProCellics by Merck KGaA, Raman RXN2 analyzer by Kaiser, Matrix-F FT-NIR by Bruker, MB 3600 analyzer series by ABB, EVO i200 biomass system by Hamilton, InVia confocal Raman microscope by Renishaw, NIRSystems Process Analytics process spectrophotometer by Foss Analytics, and analysis software, such as Bio4C PAT Raman Software by Merck KGaA, iC Raman by Kaiser, and SIMATIC SIPAT by Siemens are available in the market.

Many advances have been made in the real-time monitoring of CPPs, such as glucose, lactate, cell viability, amino acids, post-translational modifications (e.g., N-glycosylation), and even viral titers. The huge spectra dataset produced by such spectroscopic measurements can be used to extract information through the application of multivariate data analysis (MVDA). Typically, predictive regression models are made through partial least squares (PLS) or principal component analysis (PCA), which are readily available in instrument software. Biogen has patented a PLS-based method of monitoring manufacturing-scale bioreactors up to 4000 L on Raman spectra (13), and another publication describes the product quality control using a feedback loop process automation platform for glucose (14). Similar to the PLS model based on Raman, NIR spectroscopy is also used for online glucose monitoring during scaling-up of bioreactors. In addition to statistical tools, mechanistic models and machine learning methods, such as support vector machines and neural networks, have also been implemented in recent years. One study at Janssen reveals the better performance of Cubist over other statistical and machine learning models (15), which shows that the performance of prediction algorithms are strongly dependent on the characteristics of data and the choice should be made by a holistic approach. Amgen has proposed and validated an automated machine learning-based approach for calibration, assessment, and maintenance of Raman models (16). However, one of the most interesting concepts is the application of hybrid models for bioprocess monitoring and control. Their early applications for mammalian cell cultures can be found in the mid-1990s, and were based on ANN, Monod kinetics, and fuzzy logic. Other research groups have used ANN to compensate for prediction errors of first-principle models.

Surveys have pointed out the impact of laborious data pre-processing steps in the data analytics chain, which become even more tedious in the case of bioprocessing data, where high levels of heterogeneities are present. To improve productivity over paper processes, electronic laboratory notebooks (ELNs) integrated into laboratory information management systems (LIMS) have now become the industry norm to document experiments, find and reuse information, and promote efficient collaboration. Initiatives like Findable, Accessible, Interoperable, Reusable (FAIR) involving major biopharmaceutical players are already in place (17).

Early stage cell-line development is boosted by the rapid clone screening platforms (e.g., ClonePix, Thermo Fisher) and high-throughput productivity and CQA analysis platforms (e.g., LabChip, Perkin Elmer, and Octet, Sartorius), which make the clone screening process more robust and efficient. The high dimensional data generated from such instruments can be used for multivariate/statistical modeling, and, additionally, deep learning models can be used to select high-performance clones from data generated by instruments. Various bottom-up mechanistic approaches like constraint-based modeling (CBM) and omics-based technologies have been proposed. However, obtaining industry-relevant output through modeling techniques is still in its infancy. In this regard, cybernetic modeling approaches (18), which have already shown success for microbial systems, can also be implemented for mammalian systems. More efforts are expected in robust multi-objective modeling and predictive computational frameworks for glycosylation optimization, which is a critical component in biosimilar upstream development.

The data-driven modeling methodologies described in this article for optimization, monitoring, and control that attempt to model the system are manual and require much human intervention. Moreover, high variability persists in the objectives and data generated in every step of the drug development process, which poses a challenge as the default hyperparameters (e.g, number of hidden layers and nodes) of machine learning or hybrid models are often suboptimal for a given problem. The conventional procedure of manually fitting different hyperparameter combinations could be ineffective for complex problems. A modeling platform incorporating data preprocessing, feature extraction, model selection, and hyperparameter optimization can be implemented through automated machine learning in R&D as well as biomanufacturing.

For process monitoring, process development, and process control, some recent unconventional process control strategies, such as a novel glucose control strategy through oxygen transfer rate, a gassing-based pH control strategy, and a lactate-based feeding strategy, could be explored even further. Up-and-coming monitoring technologies like dielectric spectroscopy have shown promise for biomanufacturing. More complex deep learning algorithms can be used for soft sensor modeling to account for perturbations in monitoring bioprocesses. Novel strategies based on hybrid modeling and oxygen transfer flux can be applied alongside standard practices for scaling-up. The most appropriate combination of monitoring technique, instrument sensitivity, and modeling algorithm should be selected for best results. This selection is more crucial for continuous manufacturing, for which measurement, monitoring, and control tools must be highly robust and accurate.

The application of computational fluid dynamics (CFD) and compartment modeling would benefit the organizations during scaling down, scaling up, and technology transfer, which usually gets hampered due to poorly understood fluid dynamics.

Though other industries have readily adapted digital twins, their dearth still persists in the biopharma industry. The application of digital technologies will reduce the capital expenditure of drug development and manufacturing by reducing experimentation and timelines, improving control and knowledge, and overcoming regulatory bottlenecks. The advances in allied fields such as software and monitoring instrumentation will directly impact their successful implementation. As highlighted in 

, the traditional and innovative technologies are categorized based on their present state of applicability. Industry needs to develop synergy between bioprocess subject matter experts, automation engineers, and data scientists for the smooth implementation of these technologies.

 104 (6), 1107–20 (2009).
2. C. Calmels, et al., 

 160 (January) 107638 (2020).
4. S. Badr and H. Sugiyama, 

 27, 121–128 (2020).
5. S. Nargund and K. Mauch, “Hybrid Models for Biopharmaceutical Process Development—Making the Best of Mechanistic Know-How and Data-Driven Insights,” 

 113 (9) 2005–2019 (2016).
7. B. Bayer, G. Striedner, and M. Duerkop, 

Guidance for Industry, PAT—A Framework for Innovative Pharmaceutical Development, Manufacturing, and Quality Assurance

 (CDER, October 2004).
9. M.R. Riley, C.D. Okeson, and B.L. Frazier, 

 15 (6) 1133–1141 (1999).
10. R.C. Rowland-Jones, et al., 

. 33 (2) 337–346 (2017).
11. M.O. Baradez, et al., 

 J. 137, 205–213 (2018).
13. B. Berry and J. Moretto, “Cross-Scale Modeling of Bioreactor Cultures Using Raman Spectroscopy,” US patent 10,563,163, February 2020.
14. B.N. Berry et al., 

 32 (1), 224–234 (2016).
15. C. Rafferty et al., 

 117 (2) 406–416 (2020).
17. J. Wise et al., 

 24 (4) 933–938 (2019).
18. L. Aboulmouna et al., 

Prashant Pokhriyal is associate scientist; Prateek Gupta is senior general manager; Sridevi Khambhampaty is vice-president; Rajesh Ullanat is executive vice-president; and Mili Pathak

, mpathak38@gmail.com, is principal scientist, Process Science, R&D; all at Intas Pharmaceuticals Ltd. (Biopharma Division), Ahmedabad, Gujarat, India.

To whom all correspondence should be addressed.


Vol. 35, No. 1
January 2022
Pages: 20–25

When referring to this article, please cite it as P. Pokhriyal, et al., “Recent Trends in Data Analytics for Upstream Process Workflows,” 

Upstream Processing

Modeling techniques can improve process control and monitoring in biopharmaceutical production.

Recent Trends in Data Analytics for Upstream Process Workflows

Automation solutions for upstream bioprocessing have proven to be beneficial, saving labor and time. Automating upstream bioprocesses can also provide a tighter feedback loop for process control and can alleviate the challenges traditionally posed in this stage of biomanufacture.

Ongoing challenges in upstream bioprocessing include maintaining quality control (better control of temperature, humidity, pressure, and other factors influencing quality and cell growth), traceability, and labor savings, says Chris Barnes, strategic account manager, Life Sciences, Omron Automation Americas. 

Chris Sandusky, director, Automation Solutions, Cytiva, notes that automation is a large area of study and focus. “If you look at just basic automation of a machine, it allows for the repeatable manufacture of a step or process within defined boundary conditions. When this is applied to upstream bioprocessing, it allows for tighter control of process variables that have an impact on the growth profile of a bioreactor,” he states.

There is also the benefit of knowing that automation ensures that the process is stable and is controlled within the critical process parameter (CPP) boundary limits. “Without automation, the process would need to be manually monitored and controlled,” says Pamela Docherty, life sciences industry manager at Siemens Industry, USA. “Automation is the backbone for data collection. Without automation, data would need to be captured manually, [possibly resulting in] errors and gaps,” she states.

Automation is being more widely adopted in bioprocessing. It allows for more precise control of process variables to within established boundaries based on a particular step in a process, notes Sandusky, who explains that the process can be rather simple, involving control of the temperature of a vessel, or a bit more complex, involving control of a tank level with a simultaneous inflow and outflow from a vessel. “These process variables can be classified as critical quality attributes (CQAs) or CPPs; they define the acceptable manufacturing ranges for automation, allowing for more accurate and repeatable control of these items than manual control would allow,” he says.

Adoption of automation in upstream bioprocessing is also beneficial to the cell culture environment within the bioreactor. The health (including quality, lifespan), growth, and reproduction rate of cells are dependent on their living environment, Barnes emphasizes; therefore, the precise control of temperature, humidity, pressure, and other conditions are vital. These can be consistently achieved through automated process control. In addition, new artificial intelligence (AI) controllers can aid with process optimization, he states.

All upstream unit operations use automation to varying degrees, says Sandusky. He states that the biggest differentiation with automation is how units interact with each other, such as with automated transfers or manual transfers, and how units are started–either with a manual start or with an automated batch/recipe start. Another area where automation is adopted is with a manufacturing execution system that further documents the process for electronic batch recording, Sandusky adds.

Automation is found on almost all manufacturing scales, adds Docherty, who points out that the market is seeing a spike in the adoption of process analytical technology (PAT). “PAT enables the understanding of parameters that cannot be read through traditional instruments. With more data capture and knowledge gathering we are now able to solve more advanced problems,” she states.

However, the current level of adoption of automation in the biopharma industry can be difficult to assess. “It is a moving target, as automation technologies are rapidly evolving, often faster than the technologies can be adopted and incorporated into processes,” says Barnes. “The time delays of regulatory approvals and long design engineering cycles often mean that a newer technology is available by the time one is implemented.”

Automation is not yet ubiquitous in bioprocessing. For instance, the loading of consumables remains a typically manual operation, says Sandusky. In addition, the replacement of tube sets and single-use bags are still performed manually. Looking to the future, however, a robotic system would be capable of loading consumables, he notes.

Meanwhile, the knowledge of the scientist remains an area that may not easily adopt automation, although some may argue that, with the advancement in AI, the industry may be approaching that ability, observes Docherty.

Furthermore, Barnes points out that although new AI controllers can help to identify outliers or abnormalities in the process, they cannot identify the root cause, such as whether the cause is contamination or cell mutation.

Biopharma manufacturers can look to fast-evolving automation technologies for new ways to address the constant goal of better quality and more rapid production.

Feliza Mirasol is the science editor for 

When referring to this article, please cite it as F. Mirasol, “Automating Upstream Processes Helps Sync Workflows,” 

Upstream processing is better optimized with the help of automated workflows.

Automating Upstream Processes Helps Sync Workflows

There is ongoing enhancement of downstream chromatography resins as biomanufacturers continue seeking ways to improve product separation and purification. Resins have been seen as one tool that can address the purification needs of new COVID-19 vaccines as well as the existing challenges in the downstream processing of increasingly complex biologic modalities.

The COVID-19 vaccines comprise several possible and different modalities, such as messenger RNA (mRNA), attenuated viruses, viral vectors, proteins, and virus like particles, notes Henrik Ihre, PhD, director of Strategic Technology Partnerships, Cytiva. Ihre explains that no one unique purification protocol can be applied to all these modalities because the API for each is very different in terms of shape, size, and form.

“However, the high-level needs are similar. Each [protocol] needs to achieve high-yielding purification protocols to bring vaccines to the market in short time and in high purity,” Ihre explains. “For some, but not all modalities, platform processes and protocols may be developed enabling a shorter time to market. The mRNA vaccines in particular are still rather new to the market, which is why no such platform processes seem to be in place.”

One common feature of all mRNA-based APIs is the poly(A) tail, which consists of multiple adenosine monophosphates. These tails reversibly bind to a poly(T) ligand that can be immobilized to a solid support, which would enable an affinity-type of purification. “This is why oligo(dT) support is often used in the purification of mRNA-based vaccines,” says Ihre.

The array of vaccine modalities that have been in development to address the COVID-19 pandemic all rely on different viruses or viral parts as part of their composition. Although recombinant vaccines are manufactured using well-established technology, proven to be both safe and effective, they take longer to develop. DNA or mRNA vaccines, however, use information from the genome of the virus to create a blueprint of select antigens, allowing them to be produced quicker than traditional vaccines, confirms Nandu Deorkar, vice-president of R&D, Biopharma Production, Avantor.

The main critical purification challenge for mRNA-based vaccines, however, is the lack of well-established and cost-effective large-scale process (1). “There are several purification techniques that are in use, such as precipitation, tangential flow filtration, or chromatographic separations. However, precipitation or size-exclusion chromatography (SEC) are not cost-effective when scaled up,” Deorkar says.

Affinity chromatography resins such as oligo(dT) can be used as a capture step, but these resins are not able to separate single-stranded RNA (ssRNA) from double-stranded RNA (dsRNA), and they have a low binding capacity (1,2). Deorkar notes that hydrophobic interaction chromatography (HIC) and anion exchange (AEX) chromatography can be used, but careful optimization of conditions is required to be applicable for large-scale production (1,2). “Additionally, these HIC and AEX resins are known to foul when used as a capture step, and alkaline stability of resins should be considered, such as resins based on polymethacrylate (e.g., Bakerbond Hipropyl or Bakerbond PolyQUAT),” Deorkar states.

Beyond vaccines, biotherapeutic development is seeing the emergence of a variety of other complex modalities, such as bispecific antibodies and protein fragments in addition to nucleic-acid-base APIs.

In general, the inherent physical and chemical complexity of biologic modalities increases the diversity of production conditions and increases the difficulty of purification processes, Deorkar says. There are, however, similarities in certain classes of biologics, such as monoclonal antibodies. “Because traditional monoclonal immunoglobulin G (IgG) products share a similar structural and chemical motif, a platform purification approach can be utilized in the development and production of this class of proteins. Protein A affinity chromatography provided one simple capture step that made screening of clones faster and purification process development simpler,” he states.

Deorkar also explains that plasmid DNA (pDNA) and recombinant adeno-associated viruses (rAAvs) are two other classes of biologic molecules that are of high interest for the biopharmaceutical industry but that also face many purification challenges. “For example, the large size of pDNA decreases the capacity of AEX resins due to less accessibility to the pores. In addition, critical impurities, such as open circular pDNA, genomic DNA and high molecular weight RNA have similar properties making the separation of these impurities more difficult (3),” he says.

Deorkar points out that a major challenge in rAAV purification is the removal of empty capsid particles, which is an impurity generated during upstream processing. “These empty capsids do not contain the therapeutic gene of interest and can elicit an undesirable patient immune response (4). Therefore, removal of these empty capsid particles is critical during the manufacturing process,” he states.

Use of AEX resins, which rely on the difference in charge between the empty and full capsid particles, is the most effective way of separation; however, the minor difference in charge based on the DNA encapsulated within the capsid requires AEX resins with higher selectivity.

“Although the purification challenges for each modality are different, all downstream processes have the common goal of increasing yields with process efficiency in a cost-effective way while ensuring the final product meets purity standards. Chromatography is often the mainstream purification process where those goals can be achieved through optimization,” Deorkar says.

“The key challenge for any process developer working on a downstream process is to develop a robust, scalable, and cost-efficient process to get to market in a short time,” adds Ihre. For more mature modalities (e.g., mAbs), decades of process development have resulted in more or less “plug and play” solutions based on Protein A chromatography as the key enabler.

For newer and more complex modalities lacking unique binding regions, Ihre notes that new protocols often need to be developed, and, sometimes, new resins need to be invented to meet the downstream purification goals. “To develop a completely new process—perhaps also based on new resin for a new API—is a time-consuming process delaying the ‘time to market’ aspect,” Ihre says. “There needs to be a balance between solving the downstream purification needs ‘good enough’ with existing resins and processes, as compared to developing new and more tailored resins and processes that may allow for novel and beneficial platform protocols.”

For the past decade much of the focus in bioprocessing has been on process intensification—meaning the development of new resins that have allowed for higher productivities and that enable improved process economy and increased resin lifetime, says Ihre. “This has especially been true for the success of mAbs. Now, we start to see a new landscape based on a variety of new modalities such as viral vectors, mRNA, pDNA, cell and gene therapy, and oligos,” he adds.

Ihre notes that a common feature of some of these new modalities lays in the fact that they are very large in terms of hydrodynamic volume. This results in lower capacities if the molecules cannot access the inner porous structure of the resin, which often is the case, he states. Ihre suggests that for such target molecules, a more open pore structure, such as that seen in fiber-based devices, may allow for high binding capacities and productivity gains. “In studies of plasmid purification processes, it was shown that the productivity could be increased more than tenfold from a fiber-based device, as compared to a conventional bead modified with the same ligand,” Ihre says.

Many downstream solutions can be identified in chromatography steps due to the step’s selectivity, versatility, scalability, and cost-effectiveness compared to other steps, says Deorkar. The challenge of pDNA being a large size molecule, for instance, inhibits its ability to diffuse into the pores of the resin (3), so addressing this challenge through optimizing the pore size of the resin will increase the accessibility of the pDNA, which can thus result in higher dynamic binding capacities and increased production efficiency, Deorkar explains.

If the target molecule can enter the porous structure of the resin and bind to an existing ligand, then the conventional based chromatography process may still be the best alternative from a “time to market” and a process economy perspective, says Ihre. He points out that there is currently a large toolbox of commercially available resins to choose from, all compatible with other downstream hardware purification processes.

“These resins offer a broad range of bead and pore sizes as well as choices of ligand types and ligand densities; therefore, the likelihood of developing a robust, scalable, and highly productive process is very good. However, for truly large target molecules or challenging feed streams, other purification methods may offer more competitive solutions,” Ihre explains.

“Another challenge in purifying pDNA is its intrinsic highly negative charge which is also shared by other impurities such as RNA, thus making the separation of those contaminants difficult using traditional anion exchange resins. In such cases, use of multimode AEX resins (e.g., Bakerbond PolyQUAT) can increase selectivity of pDNA over the impurities,” Deorkar adds.

Deorkar also points out that capacity challenges exist for the purification of mRNA, where dynamic binding capacities of reverse phase (RP) and AEX resins are extremely low. Because of this, large amounts of material are required for purifying the desired quantity of in-vitro transcription product. “Resins with higher binding capacities can address those challenges of the current process inefficiencies observed in mRNA purifications,” Deorkar states.

Chromatography, a well-established technique, shows excellent repeatability and scalability, and the resins used in the chromatography process provide an effective and efficient method for purifying most biological molecules. Meanwhile, precipitation, digestion, ultra or microfiltration, and membrane capture can all be used at the lab scale to efficiently test small-scale samples, notes Deorkar. However, scale up of these methods can be challenging due to process efficiency and reproducibility.

The most common large-scale process for purification of mRNA, for example, is affinity chromatography followed by ion exchange (IEX) or reverse phase (RP) chromatography (1). In lab-scale processing, lithium chloride precipitation may work well, but scalability of such a process is poor, Deorkar states. Affinity chromatography, such as oligodT is commonly used for a capture step, and removal of dsRNA can be achieved by RP or IEX resin steps. “Conversely, with production, other methods, such as membranes and monoliths, have been investigated. Although these methods show great promise, further development is still required,” Deorkar says.

“Traditionally for all biologics, a multistep approach using two or three unique chromatography resins enables purification of these biomolecules. Resins have been shown to be effective and efficient for the separation of these molecules over other methods,” Deorkar adds.

“Often the same process goals can be met based on different purification technologies and what is most practical may depend on what is commercially available and what the specific company is used to for other or similar processes,” Ihre includes.

Chromatography resins will continue to be instrumental in the purification of biologic molecules, says Deorkar; however, improvements will need to be made to better serve the industry, especially for the purification of pDNA, RNA, and viral vectors. “As upstream processing continues to increase titers of these biologic molecules, improving on current chromatography resins to increase the dynamic binding capacity while enhancing separation to remove process-related impurities will be critical for manufacturing of these biologic therapeutics,” he concludes.

“A chromatography resin consists of a ligand coupled to porous base matrix. The key design parameters for the porous bead are the construction material, bead, and pore size, whereas the key design parameters for the ligand is the nature of the ligand and immobilized ligand density on the porous bead,” explains Ihre. He says that the types of porous beads, with their variety of construction materials, bead, and pore sizes, are rather well-advanced in meeting key market needs. New base matrix designs could be seen as fine-tuned versions optimized for certain application areas.

“Going forward, the key field of innovation around novel resins may rather be on the ligand design side. If cost-efficient affinity ligands and resins can be developed on short timelines addressing new molecules, it is likely that affinity chromatography will gain even further market share based on its ability to achieve high purity and recovery in just one step,” Ihre says. Another possible game changer, he notes, would be if a universal tag system were developed and approved for large-scale downstream processing, and not only for small scale or academic applications, which is the case today, he says.

 39 (16) 2190–2200 (2021).
2. P. Gagnon, et al., “A New Runway for Purification of Messenger RNA,” 

, Oct. 29, 2020.
3. N. El-Hajjami, “Plasmid DNA Production for Cell and Gene Therapy,” 


Vol. 35, No. 1
January 2022
Pages: 26–29

When referring to this article, please cite it as F. Mirasol, “Enhancing Resins Addresses Purification Concerns,” 

Downstream Processing

Enhancement of resin technology can improve the ability of chromatography to purify the growing landscape of increasingly complex biomolecules.

Enhancing Resins Addresses Purification Concerns

Artificial intelligence (AI) and machine learning (ML) are transforming industries as diverse as retail and energy, using powerful algorithms and rich data analytics to improve business agility in a constantly changing, global marketplace. But some highly-regulated industries—such as pharmaceutical manufacturing and life sciences—are missing out by not fully using AI and ML. While there is currently no specific guidance describing how AI must be used, it is generally accepted that AI algorithms fall into the same bucket of regulation as multivariate tools as an analytical math system. FDA and the 

 have clearly indicated their support of AI as a tool to increase quality and efficiency and have published some content backing this technology. For example, the ninth edition of the 

 published in 2017 announced two AI algorithms as valid chemometric methods in chapter 5.21 (1). In 2021, FDA published an action plan regarding the use of AI and ML as software for medical devices (2). Additionally, the Process Validation guidance published by FDA in 2011 described standard methods to calculate the validity of any analytic model (including AI as a multivariate tool) (3).

Another roadblock to AI adoption is the perception that machine learning models are developed in a black box. These models rely on AI algorithms and their unique capacity of self-tuning themselves by learning from data; different data mean different tuning needs. Therefore, qualification of the AI algorithm is a prerequisite before a machine learning model can even be validated. Only explainable and reproducible multivariate mechanisms can be used for good practice (GxP) purposes; AI must be managed in this way to open the perceived black box.

There are some industry initiatives to help overcome this hurdle. For example, Xavier University has several AI working teams that have been working with industry leaders in coordination with FDA to develop Good Machine Learning Practices (GMLP), an AI Maturity Model, show real-use cases, and drive other initiatives related to AI implementation. The Parenteral Drug Association has also published an AI Algorithm Qualification study (4). The goal of this research initiative was to provide a methodology based on a design of experiment (DoE) approach to agnostically qualify the Isolation Forest algorithm. This strategy can be applied to real datasets containing different sources of variability or using synthetic datasets containing several variables with different controlled features that are the object of study, such as correlated variables, various standard deviations, orders of magnitude, etc. An experimental engine was specifically built as part of the project. The strategy was focused on detecting defects in a motor using an AI algorithm for outlier detection. The experiments carried out have been defined using a DoE to cover all operational conditions. The resulting guidelines can be applied to other AI algorithms for regulated drug and medical device manufacturing environments.

Following this blueprint, pharmaceutical and biotech companies can use AI/ML to optimize their highly regulated manufacturing processes. The use of batch data to streamline production, reduce costs, improve yield, and keep up with today’s increasingly complex regulatory landscape will enable Pharma 4.0 and the business transformation benefits that come with it.

AI/ML allows pharmaceutical manufacturers and life sciences companies to interrogate, analyze, and manage huge data sets generated by complex systems. Using qualified AI production chain algorithms provides new ways to scale operations and production. A qualified algorithm presumes a ready-to-use state, making the model validation process easier.

The concept of qualification is associated with systems, facilities, or equipment and is a process to ensure that these things are achieving the expected acceptance criteria defined in quality attributes and parameters. In a regulatory context, “qualification” is the status of quality assurance for systems, facilities, or equipment supported by documented evidence that proves the right use of them. Qualification is different from validation. “Validation” is the documented evidence to ensure that a specific system, process, or facility is generating an output with predefined quality attributes and specifications consistently (3). In summary, validation acts over processes and procedures while qualification affects assets. Generalization is possible for qualification, but not for validation. Usually, the validation of a system or equipment implies its previous qualification. Following this reasoning, the validity of AI applications should be studied by means of the algorithm qualification perspective, proposing a set of standard procedures to establish the acceptance criteria of its implementation based on a set of expected conditions (4).

Now, it’s up to the industry to leverage and optimize this capability. The following are four key points to consider when abstracting this generalized qualification method to qualify other AI algorithms.

AI is only as good as the data you feed into it

AI systems designed to interpret multivariable scenarios subject to safety, quality, and efficacy regulatory requirements must be trained with high quality data. Data must be prepared, including the sufficient variety of use cases to cover process experts, regulators, and patient expectations. If data does not properly cover all the potential scenarios, an incomplete AI model will be generated that is unable to provide the expected accuracy. Conversely, when data contain all the potential situations, the model will present robust outputs.

Global regulatory bodies are increasingly evaluating the benefits of AI as related to pharmaceutical environment processes. In 2019, FDA published a discussion paper to request feedback regarding AI/ML usage for medical devices (5). The European Medicines Agency (EMA) and Heads of Medicines Agencies (HMA) expanded the scope of the Joint Big Data Task Force to include AI, reflected in the Big Data Steering Group Plan. They published a report in 2020 on 

 (6), and in 2021 held a workshop to gather opinions from main pharmaceutical stakeholders regarding big data and AI-related technologies and tools. It’s clear that regulatory bodies are seeing the benefits of AI/ML and are fast-tracking the acceptance of these valuable tools in a safe and responsible way.

AI will improve response to future pandemics and health threats

Research and development are only part of the equation for discovering and developing new therapies, such as the COVID-19 vaccines. Manufacturing and the ability to scale up production and delivery of these life-saving products was key. AI and ML have already helped manufacturers increase yield, identify issues before they occur, and speed the identification of root causes of deviations. These examples demonstrate how AI accelerates current operations by means of a multivariate approach. While speed is critical, all tools, including AI/ML, must be appropriately vetted and proven.

The promise of Pharma 4.0 has never been greater. Powered by qualified AI/ML, pharmaceutical and life sciences companies will be able to fully digitize their manufacturing processes through data-driven decision-making. The smart factory, robotic manufacturing, and the Industrial Internet of Things, for example, will graduate from concepts on a white board to fully realized business tools. Most importantly, these companies will be able to unleash the power of their data to optimize the manufacturing process, improve yield, and guarantee product consistency and quality.

Opening the door to new innovations and efficiencies

The ability to apply AI governance to other algorithms in highly regulated drug and medical device manufacturing environments opens the door for all sorts of innovation. Companies will be able to take advantage of AI/ML to find new ways to scale operations and production while ensuring the quality of delivered therapeutic and medical device products and services. This ability will result in an economy of scale that leads to safer, more efficient, and effective patient outcomes; make us more resilient in the face of future health threats; and open the door to finally realizing the promise of a fully digitalized Pharma 4.0.

, General Text 5.21 (EDQM, Strasbourg, France, 2017)

Artificial Intelligence/Machine Learning (AI/ML)-Based Software as a Medical Device (SaMD) Action Plan

Guidance for Industry, Process Validation: General Principles and Practices

 online DOI:10.5731/pdajpst.2019.011338 (August 2020).

Proposed Regulatory Framework for Modifications to Artificial Intelligence/Machine Learning (AI/ML)-Based Software as a Medical Device (SaMD): Discussion Paper and Request for Feedback

Qualified algorithms enable validation of machine learning models that can be used for process optimization. 

BioPharm International, January 2022

Industrious Pharmaceuticals

The Long Path to Normal

What’s Next for COVID-19 Vaccines and Variants?

Recent Trends in Data Analytics for Upstream Process Workflows

Automating Upstream Processes Helps Sync Workflows

Enhancing Resins Addresses Purification Concerns

Qualifying AI Algorithms in Pharmaceutical Manufacturing

Quality Still a Priority

Finding Solutions to Cold Chain Supply

Outsourcing’s Race Toward Novelty

Multiple Challenges to Multispecific Analysis

Good Manufacturing Practice on Demand?

Major Challenges Ahead for FDA and Pharma