Amgen helped to develop the science used in making biologic medicines, and we remain a leader in designing and manufacturing new biologics. This expertise is now being applied to the challenge of making high-quality biosimilars.
When Amgen announced five years ago that it planned to start making biosimilar medicines, reaction from company scientists was mixed. The concern was that biosimilars wouldn’t involve very interesting science, and these programs might distract from Amgen’s focus on novel therapies.
That early perception has given way to a new appreciation for the rigorous science required to make high-quality biosimilars. Instead of distracting from innovation, as some initially feared, biosimilar research has strengthened Amgen’s science and engineering expertise.
The hurdles involved in biosimilar research are helping our scientists gain new insights and capabilities. In fact, the lessons learned in making biosimilars are being applied to enhance and accelerate the development of novel therapeutics.
These scientific challenges and benefits mainly impact Amgen’s Process Development organization; the group that creates the cell lines and designs the processes for manufacturing biologics. Product quality is measured in terms of staying within a defined range for specified attributes and is assessed for every batch of a drug before it is released for use in patients. Product quality attribute ranges are like bull’s-eyes, with targets that can be large or small based on the capabilities of our tests and limits established with regulators.
With a novel biologic, many of the attribute target ranges aren’t established. “What you create is what you create. You’re defining it for the first time as a product,” said Jeff Yant, Operations director in Amgen’s Biosimilars Business Unit (BBU). By contrast, a biosimilar starts with the range defined by the reference medicine’s characteristics. “You have to begin with the end in mind,” said Yant.
The Challenge Of Making Biosimilars
Targeting Hard-to-match Molecules
Amgen aims to make biosimilar versions of medicines that treat serious illness, and many of these medicines have functions that are challenging to replicate. “Our biosimilar programs are focused on some of the most complicated of the monoclonal antibodies,” said Richard Markus, vice president, Biosimilars Global Development. “They are complex from a size and structure perspective, but on top of that many of these antibodies are also active on cells. So, not only do you need to match how the molecule binds to the target, but you also have to match what happens after it binds to the target. That’s where it gets challenging and interesting.”
The target characteristics a biosimilar drug must match are both narrow and numerous. “There are multiple attributes that we need to dial into the innovator’s ranges,” said Yant. “It’s like trying to shoot multiple bull's-eyes with multiple arrows.”
“We have about 100 different attributes that we track,” said Margaret Karow, the BBU’s executive director for Process Development. “Currently, the similarity assessment section for a biosimilar regulatory application is almost 500 pages long. It’s a tiered approach with layers of data, including analytical and bioactivity data. There may be small differences in some of the results, but you have to look at the totality of data, including analytical, pre-clinical and clinical study results, to decide if two biologic molecules are highly similar.”
“There are multiple attributes that we need to dial into the innovator’s ranges, it’s like trying to shoot multiple bull's-eyes with multiple arrows.”
Amgen Biosimilars Business Unit
Biosimilar programs need to quickly replicate progress that might normally require years of effort. “The attribute ranges for a molecule often get tighter and tighter as you get better and better at making it,” said Karow. “With innovative products, you have years of practice and experience in manufacturing before you get to these very narrow limits. But with biosimilars, we have to immediately match very tight characteristics that the innovator companies have been able to perfect over 10 to 15 years.”
Starting from Scratch
When a company changes the way it manufactures one of its own biological medicines, it needs to ensure that there are no significant changes to the product itself. Even with all the knowledge a company has about its own molecules, it’s not easy to maintain a product’s key attributes after changing one major element of the manufacturing process, such as the cell line. This challenge is even greater with biosimilars, because every step of a biosimilar process is different from the process that was used to make the reference product.
Innovations from Biosimilars
Better understanding of how structure impacts drug function.
Refining key nutrients and other subtle development processes to ensure quality.
Genetic engineering of cells to control product quality attributes.
Ability to make major process enhancements and remain within a reference product’s target attributes.
New ways to accelerate drug development timelines.
“With a biosimilar program, you're not just changing part of the process, you’re changing everything,” said Karow. “The cell line is different, the process and plants are different, and you don’t have the innovator’s experience and historical data to guide you. Rising to the challenges involved in advancing biosimilars has been difficult at times, but it has also been very good for us.”
Acquiring New Tools And Insights
The new scientific insights that Amgen has gained through its biosimilar research are substantial. The benefits include a deeper grasp of how a biologic’s structure impacts its function, plus a new understanding of how subtle changes in process conditions can alter that structure.
Many of these new insights revolve around structures known as glycans, or the carbohydrates that get added to proteins inside the medicine-producing cells. Small differences in the amount or type of glycans on an antibody can alter its pharmacokinetics or function. That’s particularly true of the so-called effector functions, which determine the product’s ability to engage the immune system. Closely replicating the way a reference product engages immune cells has required deeper insights into the link between structure and function.
“We know that the effector function of a molecule can be critical for the mechanism of action we’re trying to match,” said Yant. “The key to getting the effector function right is often a matter getting the glycans just right. But when you turn the knob on one of these glycans, you risk inadvertently turning the knob on another glycan.”
Amgen research is revealing surprising “levers”—small changes in process conditions that can exert a significant impact on product quality. For example, key nutrients can be present in very minute amounts in the media used to grow the cells that make these molecules, but changes in these trace amounts can lead to major changes in product attributes. “We’ve uncovered quite a number of things that can swing a process around—things where we didn’t previously appreciate the degree of change they involved,” said Karow. “On a regular basis, we find a new one where we say, ‘Wow! I didn’t think that would have such an impact, but it did.’”
To gain a whole new level of control over product attributes, Amgen scientists are turning to new technologies. These include genetic engineering approaches that are being applied to host cells in order to prompt these cells to produce the desired glycans. “In the past, we’ve used genetic engineering to create cell lines that make more of the product, but now we’re talking about the ability to make designer molecules, if you will,” said Chetan Goudar, director, Process Development. “We’ll be doing much more of that going forward.”
“In the last five years, we’ve seen a raft of new genome engineering techniques for complex mammalian cell lines,” said Trent Munro, scientific director, Process Development. “We’re able to specifically modify individual genes or groups of genes in order to move activity up or down. We've never been able to do that before, but now we can.”
In another innovation from Amgen’s biosimilars research, scientists are using small molecules to regulate the enzymes that regulate glycans. “When you add very tiny amounts of the right small molecule inhibitor into the bioreactor, you can actually fine-tune a glycan to get the level that you want,” said Art Hewig, director, Process Development.
Getting To Phase 1 Faster
“The lessons we’ve learned in overcoming our biosimilar challenges are having a ripple effect across Amgen’s whole pipeline,” Karow observed. The same scientists who work on the company’s pipeline of novel therapies are also developing biosimilars at Amgen. Knowledge acquired in one of these arenas can quickly make its way into the other.
One result is that achieving comparability is easier. When you’ve learned how to match a competitor’s complex product starting from scratch, matching an Amgen molecule after a process change becomes a less time-consuming challenge.
“With each biosimilar project, we’re doing a very high-bar comparability exercise,” said Karow, “That experience is helping us to get even better at controlling the variability in our own processes when we move a product from one site to another or change the process to increase the yield.”
“We have a much better understanding of our platform bioprocesses, and that confidence has been strengthened by our work in biosimilars.”
Director, Process Development
Amgen Biosimilars Business Unit
This new expertise makes it easier to pursue major process improvements because there’s less risk that a change will result in a product that fails to meet its specifications. It is also helping development teams for innovative products to start Phase 1 studies sooner, because they don’t need to finalize the process for making a drug before they test it in humans.
“In our early platforms, we used to be very concerned about making any change between our first-in-human process and our commercial process,” said Hewig. “We needed to take more time to make sure that our first-in-human process was very good. But now, we’re much more comfortable saying that we can continue to make improvements after Phase 1 and still deliver the same product quality attributes. We have a much better understanding of our platform bioprocesses, and that confidence has been strengthened by our work in biosimilars.”