The pharmaceutical industry has been criticized for being slow to adopt new technology, with a tendency to continue to use antiquated methods and processes when newer ones are available. Many industry leaders attribute this lack of forward thinking to financial and product launch risk models and the burdens of being heavily regulated.

Other industries in similar circumstances have successfully modernized and continue to deploy new technologies. How is pharma different?

At the 2021 PDA Annual Meeting held virtually in March, FDA Center for Drug Evaluation and Research (CDER) Office of Biotechnology Products (OBP) Deputy Director Jeffrey Baker reviewed the current state of new technology adoption by the pharma industry based on what he has seen at FDA. He provided his perspective on why deployment of new technologies in biopharmaceutical manufacturing takes place so infrequently—and why a new paradigm is needed to realize the benefits that can be derived.

[Related: Biopharmaceutical manufacturers can test-drive our platform for free. Click here to get started.]

Get Started with Redica for Free

Below are Dr. Baker’s paraphrased remarks from the PDA conference. His bio is at the end of this article.

Setting the Stage: Little Has Changed in the Last Decade

2020 was a bad year. But 2020 was also a year where there was an almost organic uprising of the biotech community, many times in the absence of formal or centralized leadership to exhibit unprecedented sharing of data, unprecedented collaboration in the manufacturing sciences, and unprecedented sharing capability and capacity, all to get us to a [much] better place.

In cell and gene therapy, we are at the time of CRISPR/Cas [genome editing technique for the treatment of cancer cells]. We are at a time of explosion of drug antibody conjugates and biosimilars. We are going through a time of extraordinary engineering of proteins to provide real healthcare solutions.

What I want to talk to you about today are the challenges that are still in front of us as we enter the second Biotech Renaissance and whether we are seeing the same energy, the same innovation, the same creativity, the same willingness to get it done in manufacturing that we have seen in discovery and development, and target identification, and in the rapid moves to market.

State of the Art in Biotech Manufacturing

At the FDA, I look across a very broad landscape, but also frequently just see ripples. I do not necessarily see the pebble hitting the pond. When we look at the state of the art in the manufacturing sciences—particularly biotech—it is very interesting.

For instance, we can look at the state of the art comparing 351(a) original BLA submissions and 351(k) biosimilar submissions—submissions that are separated in development by years and years and years. But frankly, the processes are not that different, the technologies are not that different, in this head-to-head comparison.

We are going through a time of extraordinary engineering of proteins to provide real healthcare solutions

We can look at the state of the art in analytics, their use for characterization, process control, and lot release. Historically, think about isoelectric focusing. We started using these things for characterization information only, then we moved them into a control setting, process characterization, then we used them for lot release. Really, we are not seeing that big of a difference in the state of the art across the spectrum.

What about the state of art for analytics demonstrating suitability for use and time of use? For example—clean-in-place, PAT [process analytical technology], RTR [real-time release], and stability studies—look at what we have now and look at what we had ten or 15 years ago. It is not that different. And then the frequency or infrequency of supplements that we get here at FDA, which include modernization of assays or modernization and recapitalization of manufacturing.

The Biotech Product Submission Landscape: Little Innovation

If you look at total submissions that come into our office, only about 5% of them are original BLAs, and about half of those are biosimilars. All the rest are supplements (Figure 1).

Figure1 Comparing Original BLAs and BLA Supplements
FIGURE 1 | Comparing Original BLAs and BLA Supplements

In fact, half of those are preapproval supplements. These are just the CMCs (Chemistry, Manufacturing, and Control). Huge load. And recall that for biotech, there is not a separate office that does legacy products.

When you go in and you look at these, what you do not see are transformational technological advances. What you do not necessarily see is continuous improvement and continuous learning. What we see is an adjustment of a buffer or label or concentration or site change.

Why is that? Clearly, it is not about the molecule because we can see with looking at biosimilars that the molecule is the same, but we are still using 15, 20-year-old technology.

It is not about access to technology, equipment, or talent.  It is not about the geography or the size of the company. It is not about the complexity of assays or manufacturing technology.

Is FDA the Problem?

It must be the FDA. It must be the regulators. But if you go back at least in terms of intent, FDA has been trying to encourage modernization of pharmaceutical manufacturing—biologics manufacturing—for over 20 years (Figure 2). There are priority plans from the Commissioner’s Office. If you have not seen the 2021 update, I encourage you to look at it (link provided below).

Figure 2 Advanced Manufacturing Technology Development
FIGURE 2 | Advanced Manufacturing Technology Development

If a problem is on our plate—the FDA’s plate—it is not for lack of trying. We went to NIIMBL. For those of you that do not know what NIIMBL is, it is the National Institute for Innovation and Manufacturing Biopharmaceuticals. It is part of the Manufacturing USA network. The whole Manufacturing USA network is good stuff. It has over 150 members.

We went to NIIMBL and requested them to partner with FDA in talking to 11 companies about new manufacturing technologies and asked the companies this question: ‘With respect to the regulatory landscape, what changes would you like to see implemented that would enable your company to deploy innovative technology for manufacturing or continuous improvement?’ In other words, to engage the same innovation engine that is driving target validation and high throughput screening and discovery—engage that in the actual manufacturing of drugs for the American public. What would you like to see?

The outcome was interesting in its refreshing bluntness: ‘There is rarely a business case for implementing new manufacturing technologies.’ Prelaunch, new technologies pose a risk to timelines. 

Post-launch, global change management, including maintaining separate processes for different markets, is a huge hurdle. There is generally an aversion to being the first to deploy a new technology in manufacturing due to the perception the sponsor may face overly burdensome hurdles from the regulatory filing.

A Tale of Two Chairs

We are fast followers in the biotech industry. Everybody wants to be a quick third to do anything. This is where this came out. They said there is no return on investment—that, if you came up with the best technology in the world but put the timeline at risk by a matter of a few weeks, they are going with the tried and true technology.

Now, somewhere there is an economist who is having a nervous breakdown as I give you my view of return on investment, but basically it is the gain from the investment minus the cost of the investment divided by the cost of the investment. And then you go through and you anticipate a whole bunch of failure modes.

You want to be careful about taking members of our community and putting them at a dry board to list all the things that can go wrong. But you need to modify this by likelihood of technical success, discount rates, opportunity cost—there is a big one—net present value and adjustment of estimated value added, which includes capitalization.

You end up with something that looks like this (see Figure 3, on the left). This is a chair built based upon return on investment. We have minimized the costs. When you have costs, you minimize them. This is a fine chair. It is a 1,000-year-old Chinese chair. It has three legs. It is very, very stable. There is not a lot that can go wrong. There are no moving parts. It can seat people of many different sizes and heights. It is a perfectly good chair and it is really cheap. We turned all our Lean Six Sigma Black Belts on it. We leaned out of this chair.

Figure 3 Two Chairs
FIGURE 3 | Two Chairs

This is the chair that I have in my office (see Figure 3, on the right) and this is not a very fancy chair. But I know which chair I want and I know which chair I want to have at my workplace. I know which chair is going to give me the better results. I know which chair is going to give a different experience. The chair on the left is about minimizing costs. The chair on the right is about optimizing value. The chair on the left is about doing arithmetic. The chair on the right is about providing an experience that gives us outcomes that we want.

It is easy to measure cost. It is arithmetic. It is hard to measure value…functionality is important. Remember that stool is pretty darn functional, right? If it doesn’t work, do not pass go, do not collect $200. But there are a lot of other things involved in providing a positive value-adding experience. There are emotional elements. There are things that change the environment in which you work and change the social impact. All these things are involved in value and they are hard to put in an Excel spreadsheet.

‘There is rarely a business case for implementing new manufacturing technologies’

Now, this is the point in the presentation where you may be rolling your eyes and saying, ‘Come on, big guy, give me a break. We are talking about Protein A columns here.’

But we are surrounded by these kinds of things. Value provides positive experiences. Here is an example: Americans buy new phones roughly every 18 months. You know what? They all make telephone calls the same way.

How many of us are using a laptop at work that is more than two years old? Why? It is not about making phone calls. It is not about storing files and printing them out. It is about what kind of experience you are having in the workplace and what kind of decisions you are making and what kind of outcomes you are going to have. This is about optimizing value, not minimizing cost.

So when we look at these experiences, people do things either because they really want a positive outcome or they really want to avoid certain things.

Fear of Surprises

In R&D, we search for new, unexpected results. The fear that we have is that we really do not know what is going on. Whereas in manufacturing, we have a hunger, a thirst, and a lust for reproducibility, and we fear surprises. We fear the unknown and we fear surprises.

The thing that makes this very awkward is that in the biotech space today with everything that is going on, at the core of almost everything is the management of residual uncertainty and the totality of the evidence.

We are manipulating the very molecules of life to have a positive outcome. Even my arrogance does not extend so far to know that I have perfect knowledge on that…there are no lists. There is no box checking. We look at all the planks in the platform.

We are fast followers in the biotech industry…everybody wants to be a quick third to do anything

This is at the center of all kinds of things going on—from transition products to risk-based inspections, comparability, established conditions, breakthrough products, biosimilarity, risk-based review, PAT, RTR, and continuous manufacturing—at the core of all of this is managing residual uncertainty, which is always there, and the totality of the evidence. Taken together, we believe we are very likely to have the outcomes that we want and we are very unlikely to have negative outcomes.

A Drop in the Ocean

As we move through this knowledge base, we find that there is a continuum of a residual uncertainty and understanding it from descriptive knowledge to correlative knowledge, causal knowledge, and then mechanics and first principles (Figure 4).

Figure 4 Knowledge Driven Decisions
FIGURE 4 |  Knowledge Driven Decisions

From this knowledge, we make different decisions. If you are at the bottom of the pyramid, you are making authority-driven decisions. Why do we do it a certain way? Because the ticket says so. Because QC/QA says so. Because our SOP says so or the FDA says so or the USP says so. Why? Because they say so. That is descriptive knowledge and it is driven by authority.

At the top are the sorts of knowledge-driven, data-driven decisions that we make when we really understand what is going on. One of the challenges we have is getting up to the top requires a greater investment in understanding. Just like capital, you do not get that return in the first four weeks, even in the first year. You have the return over time as you are making different kinds of decisions.

‘What we know is a drop, what we don’t know is an ocean.’ That Isaac Newton guy was pretty smart. We have to quit characterizing the ocean by pulling out five, or six, or seven drops and taking our analysis out to six decimal places. We need to look at the ocean.


If you take away nothing else from this presentation, here is the vocabulary word of the day: thalassophobia. Thalassophobia is the fear of the ocean…Geert Hofstede came up with a thing called Cultural Dimension Theory on how cross-cultural values and communication get in the way or are enabling (reference listed below).

One of the factors is the Uncertainty Avoidance Index, UAI—to what extent a culture programs its members to feel either uncomfortable or comfortable in unstructured situations, those with high residual uncertainty. Decision-makers in high UAI cultures are less comfortable with ambiguous situations and require more certainty, precision, and detail to decide to move forward.

…in manufacturing, we have a hunger, a thirst, and a lust for reproducibility, and we fear surprises

This has nothing to do with science and engineering and everything to do with the UAI index of a particular culture. Deciders in high UAI cultures tend to avoid uncertainty rather than to manage it. I leave it to you to look in the mirror at our own culture.

What does ‘managing uncertainty’ mean? We accept that uncertainty is neither good nor bad. It merely is. We differentiate between data analysis, which is mathematics, and knowledge management, which is understanding. We differentiate statistical thinking and statistical calculation.

Do we really understand that all the numbers below the LOQ are effectively the same? Please don’t trend them. Don’t just go up to the pull-down menu and do a Shewhart analysis on data, which is not time-sequenced. Make relevant central risk assessment. Not can you measure it, but is it relevant? Optimize value rather than minimize costs, and value understanding over specification.

Enabling the Second Biotech Renaissance

When we look at managing uncertainty, we are managing risk. What risks in fact are we managing? Are we managing risk to the product? Are we managing risk to the predictable patient experience? That is the end game. Do we discuss new technology? Do we deploy it or do we defer it? I can tell you right now we are differing. I hope to goodness we are still discussing how we can deploy it, because otherwise we are making decisions based upon fear.

Here is a great quote from Winston Churchill. ‘If you take the most gallant sailor, the most intrepid airman, the most audacious soldier, put them all together, what do you get? The sum of their fears.’

Fears, such as, ‘all this uncertainty could impact the timeline. We might get a question from the FDA. All these horrible things could happen. No, we are going to minimize costs.’

Until we look at manufacturing as a value center, not as a cost center, something to be optimized rather than something to be minimized, we are never going to engage the innovation engine that is driving discovery, that is driving target validation, that is driving new public health solutions in manufacturing. 

 We have to quit characterizing the ocean by pulling out five, or six, or seven drops and taking our analysis out to six decimal places

We are in the second Biotech Renaissance. This is beyond the dream of replacement hormones from E. coli. We are doing so much more. But if we wake up in the night with so much fear that we defer the decisions to move forward with these new technologies in real-world manufacturing, we are going to be waiting forever because the moment just will not come.

We must saddle up and get out there and take the same enthusiasm, and the same passion that we have for solving these public health solutions and keep pushing until we understand what we are doing in the manufacturing sciences, just like we do with assay validation, adaptive clinical trial design, the new analytics, and the new high throughput screening.

We need to get out there and make medicine for sick people quickly, efficiently with the new technologies, not the stuff that we have been doing for the last 10 or 20 years. They are tried and they are true, they are platforms, but the world changes. It moves fast, and we have to move with it.


FDA Commissioner’s Report: Accelerating the Adoption of Advanced Manufacturing Technologies to Strengthen Our Public Health Infrastructure

Hofstede, Geert (2003). Culture’s Consequences: Comparing Values, Behaviors, Institutions, and Organizations Across Nations. (2nd ed.). SAGE Publications, Inc.

Expert Bio

Dr. Jeffrey Baker spent many years in the industry, most notably at Eli Lilly as a respected senior scientist and director in biosynthetic product manufacturing. He has been recognized with several CDER citations for leadership at program development and was a member of the team which conceptualized the Office of Pharmaceutical Quality. He has participated in the development of regulatory guidances and operational policies for FDA and CDER. In 2018, Dr. Baker received an FDA Honors Award in contributions for modernizing the U.S. regulatory system for biotechnology products through sustained creative leadership and collaboration.

He is an FDA liaison to the Advanced Manufacturing National Programs Office at NIST, and the National Institute for Innovation and Manufacturing Biopharmaceuticals, where he engages with academic, federal, and industry stakeholders to advance the national strategic plan for manufacturing in the United States and related initiatives.

[Related: Biopharmaceutical manufacturers can test-drive our platform for free. Click here to get started.]

Get Started with Redica for Free

Get a Demo

We’ll can show you insights into any of your key suppliers, FDA investigators, inspection trends, and much more.

Request a Demo