Case study - downstream processing of monoclonal antibodies produced in bioreactors

Hi everyone, and welcome here to the second video in Downstream Processing Bioreactors. And today, rather than giving a broad overview, I'm going to give a very specific example in talking about monoclonal antibodies. And why monoclonal antibodies? Well, there's already been over 500 of them that are used for treating diseases. And as you might know, in your body, you can use antibodies as soon as a foreign invader enters your body. So you could also imagine and also use some of them. Not always monoclonal, but monoclonal. They can have various applications in the sense that antibodies are already present. But in particular, you can see that there's a huge rise in the market to find more and more monoclonal antibodies. And you can also see that there's definitely a scale-up happening monoclonally antibodies used for other immune diseases. So a very important aspect in the pharmaceutical industry. Now, one of the things before we talk about the general process is talking about stability because that's very important. So some products are more stable than others. And definitely for antibodies, you need to look at the stability during the harvesting process and in the general process. Things that we might want to consider is the pH they can tolerate. And this becomes important when we look at homogenous techniques, which I'll come back to later. The temperature, but also freeze for stability in case we want to change that. Some of these cells, as you might remember, particularly when you're working with show cells, that can be quite sensitive to sheer stress. So you need to consider that when you're using some certain filtration techniques. And there's also the colloidal and conformational stability in case you're exchanging buckets, which you will have to look at. So you need to consider that as well. Now, it's not just the stability, but there might be some modification happening. And usually, rather than just having an empirical process to determine this, we would use some computational modelings or some in silico to kind of predict what would happen to these products in order to speed up the process. So certain processes are not that applicable if you have antibodies that are not very stable. So I'll talk you through some of the things that you need to consider in that regard. There are some general trends. We will have seen that there's definitely a trend going from bioreactors in batch or half-batch modes towards continuous reactors, mainly because it's much more cost effective. The same applies to the perfect process. Again, here there's a shift towards continuous or semi-continuous downstream processes. Another shift that we have seen happening is mainly because of the pandemic. There's definitely a shift towards smaller production using single-use technologies, such as, for instance, biowave reactors. So that means they are more flexible towards market demand. But you also have to consider that they might be generating more plastic waste. So that's another thing to consider. So generally, we see this trend in smaller reactors, which is continuous. I have done a video introduction to quality and design, so feel free to check that out in case you're not quite sure what it is. But we will see that process analytical technologies, such as TAPs and QBD, are becoming, even though they have been recommended by the FDA, becoming more and more important. Process analytical technologies are all about the sensors that you use for that purpose. There can be sensors that you monitor offline, so you have to take it to a specialist lab. But here you can particularly see that in-line, so directly, either within the reactor itself or that you have a step in between, where you take a little bit out of the reactor and then you monitor within seconds certain things, or in-line flow monitoring proteins, such as IR and Raman. It's definitely very important to consider when you look at an electronic product. There are lots of very specialist techniques that you can use, so you can make it as complicated as possible. Normally, within quality by design, you have a very specific framework where you assess what the quality attributes are. Within sensors, actually, there's a trend to move towards simpler systems, so also using sensors that are very typical, very low cost, such as pH and conductivity, to estimate whether certain things are copied. The pH can be an indication of stability, and conductivity is related to how your product is flowing through the stream. You will see these sensors even though they're very simple and everyone will use them in the lab, they're actually very important here. Now, let's then have a look at the general downstream process. Of course, there's not one specific format. I'm just going to talk about generally what people do. With that, the upstream process is where you actually produce your antibodies, but these monoclonal antibodies will be in a mixture of other things. Now, you will need to harvest that, and then usually this involves either centrifugation or filtration, and that's what we call the clarification step. The next step, we're removing cells and the cell debris. This is very different from lots of other downstream processes, which is why I want to give monoclonal antibodies as an example, and I'll come back to this in the next slide. Here, also very common is to use protein A affinity chromatography. What this protein A does, it interacts with a very specific part of your antibody, so it can make sure that it really binds to the antibody very well, and removes some other compounds from it. Now, this affinity chromatography is normally controlled with a low pH code for viral inactivation, and it also means that you can elute off your antibody from the body. What follows then is some polishing steps, which usually involve using multiple chromatography steps and a filtration step for viral clearance. What you then do is you would change the product in the final formulation to get the final result. Now, let's talk a little bit about the protein A affinity chromatography and where it comes from. Now, in this image here, you can see which part of the antibody protein can bind, and there's a couple of other proteins that are important that can bind to other parts of the antibodies, and they can also be used for chromatography. Now, the main advantage of this technology is that it has very high selectivity, very high purity, and it's very fast. So, if you wouldn't use it, you would definitely have to do more steps in order to get the antibody to make sure. Things that it does is remove host cell proteins, host DNA, and impurities that are associated with host cells. Now, there are also some very obvious disadvantages. Later on, you will see distribution of costs, and you will see the resin really has a very high cost and generally outputs at least 50% of your cost. After this affinity chromatography, you usually have a low pH level in order to elute off your antibodies. Now, even though most antibodies with a very short period of time can withstand these conditions, not all of them can. So, it means that this process might not be very suitable for specific antibodies. It can also lead to some lesion problems that other things come up, and for all of this, there is a drive to find a way to burn it to the surface. But I would say the main drive is cost. So, people are looking at membrane filtration, crystallization steps, but what's particularly gaining more traction is using multi-modal chromatography, the different chromatography steps, in order to achieve the same. And these chromatography steps would often do not, well, it would need to have a lower cost relative to a certain amount of cost associated with this particular technique. Now, if we've gone through this whole process, imagine that we have gone through all these steps, and you would have to change your final product. The key thing is how would you achieve quality control? Now, we will have seen in this QBD approach that you will have to determine your critical quality attributes and monitoring. It is an inbuilt quality control, so monitoring will take place throughout the process and not just at the end. But even at the end, there is a certain number of steps that you will need to consider. For instance, you will need to do viral clearance studies, which is why these filtration steps Now, other things in terms of characterization include chemical analysis, biological activity, impurity, obviously, and looking at impurities, contaminants, and stability. Particularly here for these antibodies, we are looking at protein content, which you can very easily do with UV-vis absorbance, and the potency that needs to be referred to in order to make sure that it fits the threshold that you need. Now, even though we talk about monoclonal antibodies, it should be exactly the same. It is known that throughout the process, with different buffers, the differences in your production process can lead to having different sugars expressed. And these differences in glycosylation patterns are certainly something that you will need to consider throughout, and that can lead to having heterogeneity in your product. So you would carefully need to consider that, and this kind of comes back to the stability of your product throughout the process. So certain chromatography steps might be too harsh, or your antibody can't handle this, and then you can see that these changes might occur. So making sure that your batch is absolutely reproducible and that you are using exactly the same conditions is very important. And the final thing, which is really important to consider, is how do we produce these antibodies. I have done a previous video where I have showed how you can use non-animal technologies to produce antibodies. Despite that we have recombinant technologies available and they should be widespread used, we know that in the EU alone still a million animals per year are used for production of antibodies for diagnostic or treatment. And the EU has now issued that if a non-animal technology is available, then you should adhere to it. So this is where this concept of free arms comes into play, which is related to refining production and replacement of animals in research. So it is very important to consider non-animal technologies for this process. But if you are working with animal immunisation, there are a lot more guidelines that you have to adhere to. And then a final summary here. So as I mentioned, these monoclonal antibodies are becoming increasingly more important in therapeutics. So there is definitely a drive in the pharmaceutical industry to have a look at this. Now, in general, this downstream process includes specifically for monoclonal antibodies clarification, whether to remove the cells in the cell debris, protein-A affinity chromatography in order to get the antibodies, even though I have said that there are alternatives being referred to, hypotheses, and then some chromatography steps which can involve, for instance, looking at a cation or anion exchange, because a lot of these proteins would have a different charge. And finally, filtration, which is needed for viral clits. Now, in this short diagram, you can have a look at the expenses that we're dealing with. And you can see that at the moment, the main cost that we run for the protein-A affinity chromatography is the cost of the residue. So that's why there is a drive to look for alternatives. What I also want you to consider, and as I mentioned before, we've done videos on product design and product analytical technologies, the sensors are very, very much needed to monitor the product during the downstream process, and obviously also while the upstream process is produced in the reaction, but also finally afterwards when we make the product contract. Thanks for watching this video, and do keep an eye out for further videos that will appear in this playlist where we'll talk about the downstream process in viral clits.