Pursuing Quality While Maintaining Process Control With Richard Moroney, Ph.D.

In this episode of “Better Biopharma,” host Tyler Menichiello is joined by Richard Moroney, Ph.D., managing director at Stat Consulting. Moroney shares his thoughts on how organizations should think about and approach process control. His philosophy is one of continuous process improvement, as he believes there is "no ceiling" on quality control.

Follow Tyler Menichiello on LinkedIn: https://www.linkedin.com/in/tmenichiello/

Keep up with everything biomanufacturing at Bioprocess Online: https://www.bioprocessonline.com/

Subscribe to our newsletters: https://www.bioprocessonline.com/user/edit/subscriptions

Transcript

SPEAKER_02 0:16

Hello everyone and welcome back to Better Biopharma, the official podcast of Bioprocess Online. If it's your first time tuning in, thank you so much for being here. I'm your host, Tyler Menechello, and today's guest is Dr. Richard Moroni, Managing Director at Stat Consulting, here to talk about quality management and process control. Richard, thank you so much for being here.

SPEAKER_00 0:34

Oh, Tyler, thanks for having me. I'm excited to be here and resume our conversations.

SPEAKER_02 0:38

Likewise, let's just dive right into it. You know, when we struck earlier, we were talking about your philosophy of continuous improvement and how that relates to process control and quality. So why don't you tell the audience a little bit about yourself and we can continue on that on that philosophy?

SPEAKER_00 0:54

Sure. Sounds good. So as I mentioned to you, but but people out there don't know, about two years ago I left my role. I'd been at PA Consulting Group, a large international consultancy, uh, for just under 20 years. So I've got a lot of experience in process improvement and process control throughout the industry. Uh, but two years ago, I started my own business, stat consulting here, and I decided to take a bit more narrow focus. So specifically looking at statistical decisions and the numbers that drive your business and make your business work better. And all of that is really, you know, bottom line, process control, you know, you know, build building better business through better processes is what I'm doing here. So and I think the point you were touching on, Tyler, was I was talking about how quality basics never go out of style. And I think that's one of the things we'll talk about here in in terms of what that actually means, I guess.

SPEAKER_02 1:46

Yeah. Yeah. Thank you, Richard. Um, in our brief call, we did speak about how you said there's no ceiling to process or no ceiling to quality or process control. Yeah. And I'm curious if you could uh you said like like driving a car biomanufacturing requires constant attention. And the moment that you kind of stop paying attention, you could you could veer off course. So uh why don't you just kind of pull on that thread a little more and tell us about these basics you're talking about in terms of quality.

SPEAKER_00 2:11

So, you know, at its basics, you know, quality control, quality improvement is all about looking at your process, uh, understanding where you are, um, asking questions about, well, where do I want to go? And then ultimately coming up with plans for how to get from where you are to where you want to go. Pretty simple, right? Three three easy questions we should be asking ourselves all the time. And the other, the other point you made is this doesn't stop. I mean, no matter how good you get, if you look closely at your processes, there's always room for improvement. You know, sometimes it's a blessing, sometimes it's a curse. But in general, um, don't ever be satisfied with where you are. There's always opportunity to improve if that's the right thing for your business. Sometimes it's hard to see, sometimes it's easier than others. But uh, but again, the tools of quality improvement are there at some level. If if you don't take advantage of the opportunities, someone else will. That's the competitive market forces at play there. Um yeah.

SPEAKER_02 3:05

So in your mind, is there any such thing as good enough when it comes to quality? Or is that garbage? Is that nonsense?

SPEAKER_00 3:11

No, it's not nonsense. I think I'm going to surprise you. This is a question we didn't talk about in the pre-call, right? And my answer to that is absolutely. There absolutely is good enough quality for today. And as manufacturers, you know, bioprocessing anywhere, as manufacturers, it's our responsibility to own our products and our product output. And at some point, you know, you do have to say, okay, I'm done with my RD stuff. RD is fun. We do different things, we expand the process, we, you know, do do more interesting and exciting things. But at some point you have to say, stop that. I need to produce something to sell to customers and make money. And when you do that, you have to, you know, you've written down now, okay, this is what my process controls are. I need these results for my HPLC, I need these results for my temperature, whatever your process controls are. And and you run with that. And that's what your you know, marketing authorization uh expects and what you need to do. And of course, um, at that point, you would have to say, this quality is good enough because we made that choice. We we engineered our product, we engineered our process, and we're selling it. We know it's good enough quality. But the point of continuous improvement isn't you aren't doing well today. The point is you can do better tomorrow. And and that's a much different question than um, do I have good enough quality today? You probably do have good enough quality today. Most manufacturers, honestly, do a good job at this. This is not, this is not rocket science, Tyler. You know, we all know that uh you know this stuff is hard. You know, we're um, you know, taking careful temperature measurements, where you know, HPLCs are finicky animals in and amongst themselves. So we put a lot of time and effort to understand how can I make sure this new molecule that I'm putting out in the market is, you know, repeatable, you know, good enough, efficacious, all this other stuff. It's it's hard stuff. And at some point you have to declare good enough and sell the product. Yeah, start with money.

SPEAKER_02 5:09

Right. You can't let uh you can't let perfect be the individual good, right? That's throwing out a lot. Absolutely. Absolute bar right there. Yeah. And and so, Richard, how do you think leaders should kind of embody that mindset of moving towards quality uh through process control, but how do they do that and avoid the trap of kind of settling for a process that, like we said, works well enough and it's good enough for today? Right do you not get too um I mean, obviously these these leaders and directors are pulled in a lot of different directions, especially at smaller biotech companies. So, what would your advice be to them in terms of like practically balancing both of those?

SPEAKER_00 5:46

Yeah, that's a great point. I think the decision is different depending on the scale of your business. Uh, so it's probably a little harder at the small scale. So let me answer the question directly there. If you're a small manufacturer, I've got whatever, one molecule and a tiny pipeline that someone's gonna buy for me to ramp me up to scale once once I get my approvals, that sort of thing. I think the most important thing here is to keep your eye on the long-term prize, is don't ever pretend you have one product only, right? There is always tomorrow's product you have to think about. Otherwise, your business is is just gonna wither and die. There's no universal product that has lasted forever, right? And and if we as producers forget there's something else out there, there's something else I need to do. It doesn't necessarily have to be a different product, it might just be the improved version of the current product. Um, if if we forget there's something else I'm striving for and an advancement I want to make with my business, then you will not be tempted to look to the opportunities for improvement or diversity in your product line or any of those things. So from a leadership perspective, I think the key thing there is own your strategy and make sure your strategy includes uh seeing a future, which is things you don't necessarily have today. And once you get that mindset, I think the ideas of continuous improvement and quality control will naturally play an important role in figuring out what do I have to do. Now that I see the future, now I can come back to that third question of you know, how do I get from where I am today to where I want to be in the future?

SPEAKER_02 7:16

Yeah. That's great advice. Uh, how nice would it be though if you could uh get one product that lasts forever, right? I'm sure it'd be a lot easier, wouldn't it?

SPEAKER_00 7:27

Hey, you know, party, go go to the Caribbean or something like that, retire and just, you know, take my royalties the rest of my life or something. Great. I don't know. I I think you're an engineer too, or you know, uh as well. Well, you're close enough for friends.

SPEAKER_02 7:40

A verbal engineer. I work with words.

SPEAKER_00 7:44

You know, our our job as engineers is to work ourselves out of a job. So it's like I think uh problem solving is part of the fun we see in in the world. And and certain certainly the consulting I'm doing now on my own with stack consulting, it's it is very much, you know, something's gone wrong with my process, and and uh I just need a fix. I need to understand what I'm measuring wrong, or I need to stand out to measure it better, whatever your specific question is. And and you know, when I'm done there, I walk away. You know, honestly, I've been doing this for you know better than 20 years now, and it's uh it's fun. It's yeah, it's fun to help clients figure out those difficult questions and then let them go forward. Yeah. I'm not sure I would really enjoy just going off to the Caribbean and doing nothing.

SPEAKER_02 8:25

No, yeah. Who wants to go bask in the sunlight in the blue waters of the Caribbean? That sounds awful to me.

SPEAKER_00 8:30

Okay, I'm changing my mind, Tyler. I think I would like to do that. Right. But I like my work. Okay, so let's let's we'll work a little bit more before we go off to the Caribbean.

SPEAKER_02 8:40

We're we're creatures of uh problem solving, like you said. You know, that's kind of just you need you need a reason to get up. And I'm sure, I'm sure we need to let us get old, right? After some time.

SPEAKER_00 8:50

Okay, I'm regretting my previous choices in this conversation. Uh okay, okay. We're just retiring and going to the Caribbean now, Tyler. Let's do it.

SPEAKER_02 8:56

Yeah, right. Let's hop on a flight. Um, so Richard, I'm curious in your in your experience, you've been in remind me again, how long have you been in the in the in the field of quality and process control?

SPEAKER_00 9:08

Well, uh, I'll call that 20 years. I mean, I had uh some experience beforehand, but I was melted in product development before that time. So I became a statistical engineer about 20 years ago.

SPEAKER_02 9:18

Okay. And Richard, in your opinion, where do most companies or biopharma manufacturers fail in terms of process control? Where do you see the biggest points of failure?

SPEAKER_00 9:28

Let me take the second part. I like the second question better than the first, because truthfully, most people, I think I mentioned this before, I think most manufacturers are actually pretty good. You know, I don't, I've never met, maybe this is just my personality and I and I avoid these people. I've never met someone who's intentionally trying to, you know, scam the system or produce terrible product or something like that. Everyone I've worked with has been, you know, high credibility, very likable, smart people. And a lot of this work, a lot of quality work is not rocket science in some sense. I mean, it it can be some difficult math, and there are some difficult questions that come up. It's not, it's not for everyone, but it isn't um, it isn't unknown. You know, an awful lot of what we know about statistics for quality control was well understood by, I don't know, call it 1965 or something like that, right? So it's it's been around a while. Um, so people who are caring and conscientious are are doing good jobs. But that doesn't mean it's easy and that you don't fall into regular potholes and problems. And that that I think was your second question, which I think is brilliant because it's like driving, right? You get to this uh unconscious competence, you're just driving down the road, and I know what I'm doing, and I don't have to think too much, but then something suddenly something happens and makes you wake up and realize, oh, I've got to steer more carefully to stay in my lane as this truck is drifting over and and I don't want to get hit, something like that. Um, and that leads me to my first thing that that where things go wrong is step one, where am I? That is measuring variation of your process at all the appropriate points within your process. And it can get really easy to be lulled into sleep in your process control by just, you know, the the reports come out with every batch and someone's checking it off and saying, yeah, everything's within spec, yeah, everything's within spec. If you don't have good control charting or some sort of historical review process in place, it's easy to let these little things drift up. You miss them, and and then suddenly they uh trigger some some line and they're now an actual uh you know deviation or or they require some actual response from a quality management system. And if you'd been looking a little bit closer, a little bit earlier on, you might have seen it earlier and be able to be more proactive. Um, industry-wide, this really doesn't apply much in bioprocessing, but but industry-wide, the the big, the best standard in quality control is six sigma. And the reason for that is by design, your six sigma processes. I don't know how much you know about six sigma. The the basic idea of a six sigma process is I'm looking at the variation at all the important control points, and I allow my process to vary by plus or minus one and a half sigma. I allow the mean of my process to drift up and down. One and a half sigma is a lot. Okay. So I allow it to drift that much. Uh, and then I've got four and a half sigma to my specification ranges. That's where you get the one part per million failure out of six sigma. Anyway, the the the the reason this is so valuable is if I can watch my process variation, I can I can monitor and control much, much more proactively. And if I get up to six sigma, I can even do designed experiments in my process and still sell all of the product. Right. And and you can just imagine if I could do that in biological manufacturing, that would be pretty amazing. Honestly, I don't know of any even in conventional pharmaceuticals, it is exceedingly difficult to get to anything past uh, you know, four sigma. I've seen four sigma processes, but um, you know, three sigma is is kind of the standard for getting stuff done well, driving it more than that. What ultimately happens in medical is one of your one of your uh uh specification limits is going to be uh challenged by your customer. So uh ultimately you need to have a pretty wide customer acceptance range to get a good Six Sigma process. And that just doesn't happen in medical. But looking closely at your variation, I'd say that's the number one thing uh to keep your eyes on.

SPEAKER_02 13:28

Yeah. Uh to to stretch your car analogy here, we'll call those the rumple strips, right? Where are the rumple strips in your right?

SPEAKER_00 13:34

Oh no, this is whole booth coming up. Okay, dating myself again. There you go.

SPEAKER_02 13:38

I remember those. I remember those. I'm glad they're gone. I mean, I'm sorry for the toll workers, but it it's sure it's easy to coast right through the the uh the plate readers, right? Yeah, yeah, much better. Um so Richard, where where would you advise? So you're talking about looking at these variations to keeping a tight eye on those, but where else would you advise our audience to look when it comes to improving their process or tightening the control? Like what are I don't know if this if this makes sense, but are there evergreen kind of variables and factors that you could always be improving no matter no matter how good your process is or how tight your control?

SPEAKER_00 14:15

Um I like that question. Uh I'm not sh I'm not sure there's a good answer to it, but let me let me uh think on it for a bit because it's uh um uh quite interesting. You know, uh so a lot of a lot of biologics use uh HPLC analysis as a as a key metric in their in their output. It makes a lot of sense. It's the primary product, it's the yield, it's the you know, derivative products that are through degradation or something else. A lot of stuff comes in. Um every every situation HPLC is hard, right? It's high performance liquid chromatography, it's a great tool, but it is not easy. It it does it's a little bit finicky and it and it has always yielded to improvement when I've looked at it. Um that ties into sample preparation, I guess. Uh when we've looked at a lot of HPLC processes, the uh care and attention given sample preparation is always one of the big things. That can directly uh you're often diluting samples and whatnot, and those kind of steps, if not done carefully, can quickly uh throw any sort of measurement out of line. At some level, you might call that an evergreen opportunity, Tyler. But the um the the detail of what you're doing, uh at least in my experience, has been kind of unique to each process. You know, uh they may all be HPLC processes, but each time I've looked at that, it's been nuanced in a slightly different way for the specific measurement you're making. For example, measuring the main product and improving that is different than measuring your you know secondary components or uh or other items in in the peaks, which aren't necessarily the products you care about, but they might be the uh degradation or some other impurity products which you want to keep out. Uh it gets harder the further you get away from your main peak, I think the harder it gets to. I have one client right now who's struggling with um, they're actually following my advice earlier. They are tracking their process, and they've got this one uh unidentified impurity actually, which is slowly trending up. And it's still within spec, but uh, we've just finished an analysis. We look back over the last five years of data, and something like three years ago, it was non-existent. And then somewhere around three years ago, something appears to have changed and it started to appear. And it's uh you know getting active attention now, right? Now we're seeing it, we can look for it and try and understand what it is. Has it come in from a you know supplier change? Who knows exactly what this is coming from? Fortunately, it's not uh you know, it's not a significant quality issue yet, but it is if if the trend line continues, it could be a it could be an issue soon. So let's not let it turn into an issue. Let's look more carefully at it and see where we get to.

SPEAKER_02 17:05

Yeah, I think those words, unidentified impurity. I feel like plenty of QA folks probably wake up in cold sweats in the middle of the night. Those those words ringing around in their ears. That just gave me the chills right there. Oh, there you go.

SPEAKER_00 17:17

Well, hey, uh you know, imagine, I mean, you do a lot of that work. I'm sure in your submission, you have whatever, you've got your primary, you've got whatever, half a dozen degradation products from your primary, and you've got maybe a dozen or two identified impurities, right? But that chromatogram has another 50, 100, 200. I don't know how many other observable peaks you've got on your chromatogram, but they're all unidentified.

SPEAKER_02 17:43

You're giving QA folks the chills right now, I'm sure.

SPEAKER_00 17:46

Okay. So so I'm I'm accepting it then, Tyler. So uh HPLC, you can always improve your HPLC processes. Well, they are they're high performance, but they can be even higher performance. Let's go with that.

SPEAKER_02 17:56

Write that down, folks. Thank you, Richard. That was great. Um so I am a trade journalist uh by by training now. And so as as such, I love stories. I love a good story, especially on this show. And so I wanted to ask you, Richard, about some stories and anecdotes in your experience as a consultant that illustrate good and bad process control. Um I liked that anecdote with the unidentified impurity, but what in your experience, what have been the most notable, like, okay, this person has really good or this company has really good process control? This is uh perfect case study of what not to do. And for each one, what do you kind of see as the overarching characteristics of good and bad process control?

SPEAKER_00 18:39

All right. Well, let me start with the bad one, which um we didn't know is bad until you got into the product, but it's uh uh I kind of alluded to this on the last one. One of the um one of the potential paths we're looking at is was there a change in the supply chain of an excipient product or something like that? Um we we don't know. We we just don't know what's going on with that product right now. But um imagine you know you're in your supply chain and you receive all these incoming ingredients. Uh most of them have some sort of specification, of course, and you've got some sort of you know COA or something that you receive with each batch and it says, Yep, I've tested it for all the things we agreed are important and everything's good. And I had this one client who had this problem where we were trying to understand, uh, we were looking for a source of variability, and we actually we we were quite successful. We found a number of key sources, but there was still enough evidence that something else was going wrong out there for us. And so we ended up uh digging into this one particular supply component that was coming in. The COAs looked fine. Um, it it turned out in this case though, the COA didn't cover the proper uh the proper needs. We looked a little bit closer at the incoming material, and it turned out it was not in spec for a particular way this was being used for this process. It wasn't a standard uh straight-on application. They were doing a little bit of uh uh pre-work on the incoming material. And it turned out in their pre-work process it would not, it was not forming properly the the new ingredient they were trying to make from it. So they so they ended up recognizing they needed a new constraint on the supplier, or at least they actually did not go that way. They basically said, okay, I need a new incoming inspection. I accept the COA, but I need to test this product on this other metric to make sure that it's going to work in my process. And uh that was easy enough for them. So they just did that. And it resulted in not large uh loss of the incoming batches. It's not like this was economically, they decided this was the best way to go. They talked to the supplier about changing the spec on their end, but that was going to be a noticeable cost increase, and they were losing maybe 10% of the batches coming in. They just ate the 10% as opposed to designing a new product custom for them. And well, a bit more than 10% because of the cost of testing on the incoming now. Uh, but in that way, it identified a couple of key things. Uh, best practices, and everyone knows this, are uh receiving inspection and making sure the materials going into your product are fit for purpose for what you want them to do. Um, when you get really good with your suppliers, as everyone is trying to do in supply chains, you want your suppliers to know what's important to you, to agree to it up front, put it on their COA, and share it with you. And so that that is how it works well. It works best when you can take something from your supplier, check the COA, you trust them. We we trust each other. I can then put it on my line and just go with it. And then I don't have to do uh a substantial amount of more additional testing. I'm probably doing some, you know, sampling testing periodically just to make sure everything's okay, but I don't have to regularly test something all the time when it comes in. And that worked great. That I that's an ideal way to go, but uh now and then you find yourself tripping up when uh you missed something. Um, it's actually not that unusual to miss something in those receiving inspections, especially as you do this continuous improvement process over. Over five or ten years of improvements on some of these things, uh, the the platforms changed enough that was at one point uh an excellent uh incoming check and inspection on a on a part is no longer good, you know, five, ten years later because of all the other improvements that have been made on the system, be that a part or a excipient or whatever into your product.

SPEAKER_02 22:23

Yeah. That's a really great um that's a great point. You can't get too I don't know about comfort comfortable or complacent is a better word, but it's good to it's good to trust your suppliers, but you do want to be uh I I want to hear your yeah, I want to hear your anecdote of of good process control, but really quick to take a side route, if because this kind of fed into a question I was planning to ask you, which is on how supplier variation and and changes in materials and equipment can impact your control. Um and you alluded to this, but what other ways can companies kind of mitigate those risks as it relates to material and equipment quality coming in? That's really no fault of your own.

SPEAKER_00 23:04

Oh, I'm sorry, Tyler. That's still all your fault. You can you can assign responsibility, but you can never delegate accountability. So um in our supply chain, we always have to remember it's our product, it's our responsibility. So, yes, get those suppliers up to speed. By all means, you know, get this wonderful world traveling stuff flowing all around and making sure we can get everything done. Uh, but never, never expect it's kind of like our careers, Tyler. You know, we we grow our career and we work for companies. And um, you know, may maybe when I was a kid, you know, my dad's generation or something like that, you could count on your company to help you manage your career and go forward, but certainly not in my lifetime and definitely not in your lifetime. It's it's no longer, you know, we're responsible for our own careers. Products are like that too. You know, you can you can delegate, you know, in the case of excipients and stuff, there's so many things that that come in which are not critical to your product that uh maybe you don't feel that much burden outsourcing it and whatnot. But today there are a lot of companies that I mean, we outsource contract manufacturing, we outsource contract research. You know, there's a there's a much more uh networked approach to developing product in particular for biologics, right? Uh getting a biological manufacturing facility up is expensive, difficult, hard to do. You can imagine not every um you know individual startup company is gonna do that all on their own. So so using a contract manufacturing organization, for example, makes an awful lot of sense. Uh, but again, you can assign them the responsibility to make your product to follow your quality requirements and to go forward with it. But the accountability is still gonna lie on you, assuming you haven't, you know, sold, sold the product to someone else. So, so when you see their testing reports, make sure they're testing what you wanted them to do, make sure they are staying in spec. You know, there's a lot of basic stuff like your testing instruments need to be calibrated. You need to periodically, you know, recheck your gauge RRs on any sort of test. Uh, you know, make sure all your HPLC runs have embedded standards so that you're seeing the correct uh, you know, the correct relative values and whatnot relative to your standards. Um, are your standards drifting over time? You know, that this can become a real rat race when you start looking at uh standard uh stability and whatnot. It's um uh there are a lot of questions there. And and and we can't delegate those questions away. We have we have to keep looking at them. Yeah I I'm afraid I might have drifted off a little bit. Did that get your question, Tyler?

SPEAKER_02 25:37

Or or yeah, no, that did. That was great. Um ultimately, yeah, it does fall like you said, it's it's your product, and no matter who you're outsourcing, what to, ultimately it falls on your head, right? So take that, take those steps to be on top of quality and process control. So circling back, I want to hear um a good story about a company that maybe really impressed you with their process control and kind of what you think the defining characteristics of companies and organizations with good process control are.

SPEAKER_00 26:08

One of my clients, there's this client who had the trouble with the incoming uh material um inspection that was poorly specified that I mentioned before. I I think I mentioned when I was telling that story how they had several other sources of variation we found. And I think the process for that was very instructive and very helpful. They were uh of the four sources we looked at, two of them ended up being okay as is. They had a couple of in-process controls which were a little wonky, and they said, I don't think this is right. And so the important work we did for them was to redo the gauge, gauge RNR analysis, the uh test method and you know, test method uh applicability and capability measurement, uh, if you're not in that area. Um, so we redid that for them and demonstrated soundly that the uh that the process control at that particular point was being well served by the by the method they were using, and uh, and they should just stick with it. They didn't need to change it. Um, and that that is not the only time it's happened to me, actually. Um I guess I have a harder time with the good examples, Tyler, because you know, right? I'm the firefight direct. I was coming, I wasn't coming in when you said everything's great. I was coming in when people were saying, Hey, I got a real problem here. Uh, not just that. I have a problem here, and it's been six months and I haven't solved it. So that's when I would come in. And now and then we would have these cases where, hey, you know, I know it looks a little weird, and I and they would show me why they think it was not working right. And on this particular test, it was um the the problem they they were worried about in this particular test was they were trying to measure a very small aliquot distribution of liquid. And they were confirming it by a very large measurement. So, you know, imagine uh pipetter, right? And they wanted to make sure their one drop was reproducible, and so they were measuring it by dropping out a hundred drops. And so they were worried about the statistics of that measurement were they actually getting reproducibility on one drop by measuring it with the uh with the 100 drops. So we did a uh a new, we we did a different gauge R and R than they'd done to demonstrate conclusively, yes, they were this is what your variation is in the single drop, and that was good enough for the process, and it was being well confirmed by this uh in-process check, which was basically calibrating it wasn't every run, it's like every other run or something, they would do this, do this check and make sure that and they would adjust the number in their process control to make sure the the droplet was being identified properly.

SPEAKER_01 28:32

Yeah.

SPEAKER_00 28:32

Um so yeah, so they clearly had a good they had a good foundation, a good idea, a good uh backbone to think about it and look at it. Um and you know, sometimes you think you've got a problem somewhere and you don't.

SPEAKER_02 28:48

Yeah, I'm realizing from what you said, I realize like basically what I did was ask the equivalent of asked a firefighter what house they've seen is the most fireproof.

SPEAKER_01 28:57

Yeah, correct. It's just kind of a silly question in retrospect. There are no silly questions. These are good. Thanks, Tyler. I'm having fun. I hope you're oh me too.

SPEAKER_02 29:05

Um and so I guess and forgive me if this is too too much, if we're biting off too much in this question here, but you know, in bioprocessing, what do you see as maybe the common levers for control that I don't know, maybe they give people the most problems, or there's a lot of um maybe unknown, untold variation that can exist that people might not even think can be impacting their process? Like do you find some similarities in terms of levers in bioprocessing that affect control or quality?

SPEAKER_00 29:38

Yeah, I um my answer here is probably gonna be less satisfying than my last answer, so I apologize in advance. Uh, because um I am an electrical engineer by trade and I've gotten into statistical process control through mostly mechanical ends. I I did a lot of fluid control and mechanical engineering in graduate school. The thing I do not do, which is shockingly important for bioprocess engineering, is biology and chemistry. So I know enough of those two things to be dangerous. You know, I can talk to you know the biologists and chemists who are putting together these fermenter reactors and whatnot. I think the the specific um molecule you're trying to make, and in particular the specific uh yeast or whatever host is making your biological model uh biological molecule is an incredible source of variation. And getting the right fermentation conditions is nuanced and specific to whatever your combination is, uh, to say nothing of you know downstream harvesting questions, which are, you know, perhaps more common and similar. You know, how do I how do I filter out the one molecule I care about as opposed to all of the other, you know, fermentation reaction products, which I I don't care about, right? You know, that's that's uh more common, I think. But but I would I would have to say I think the molecule you're going for and the and the host that's making that molecule are the are the key sources of variation because the whole the whole fermentation process starts there. And and as anyone who's looked at fermentation, again, I'm not the biologist, I'm not the chemist doing this firsthand, but I've been looking over a lot of people's shoulders as they do this. And it's so hard. It's so much hard. I think what makes biological engineering much harder than conventional drugs is the scale up problem. It's hard enough to scale up a regular chemistry process, right? That's that's not an easy task. I'm not pretending that that's an easy task, but that's a lot easier than fermenting and scaling up fermenting, at least from everything I've seen looking at both both of those industries. Um, scale up is always hard. You always have to put time and energy in that. And uh every manager doesn't put enough time and energy into it. It's it's like process transfer from RD to manufacturing. No one puts in enough time and energy for that. It's it's it's hard work and it has to be done. And uh and biological scale-up is hard. And part of that, I think, comes down to the uh, you know, nurturing your your uh yeast or whatever it is that's making your making your making your product and getting you know uniform conditions, you know, you get a 2,000-liter reactor or something like that. It's hard to get uniform production out of there. Um it's not like you can stir the heck out of it, you know, you're gonna bust up all of your yeast cells before you get anything out of them, right? That it's oh darn it. That's just hard.

SPEAKER_02 32:25

Yeah, life can have a mind of its own, right? You're contending with uh these microorganisms. Um that raises another question in my mind. Forgive me if it's naive, but do you generally see more problems, or do people come to you more from upstream or downstream uh manufacturing teams? Like I imagine, before you answer, I imagine downstream um is more concerned with process control as it relates to like directly relates to product quality, right? Because they are they are the ones who purify the product to be ready for patients. But you know, upstream, the control looks like optimizing the yield and the levers that they could pull there. So do you see, does it skew one way or the other, your clients generally, or what?

SPEAKER_00 33:10

Right. So we're probably gonna get a you know sample size of one. Uh never trust your sample size of one, Tyler. So ask every other person like me, you bring on this uh podcast, that kind of question. Well do you okay? So I I think there are two things that bias this. Uh the first one is the personal. I in my experience, actually, I am more biased in the upstream stuff, the fermenter work and and pre uh pre the you know post-filtration processes and whatnot of separation. And I think that is mostly just because of my background. You know, my background, I started in conventional drugs on the uh RD side and uh devices as well in combination drug, drug uh device combination products, and then transferring those to manufacturing. That's all been very hardware-oriented, and and that's that's my roots. And so certainly for my time at PA, uh most of my clients ended up being, you know, on the fermenting stage, the scale-up stage, the early stage manufacturing of those sorts of things. Probably just to bias from my my point, because I think your other point was very well made. Uh, the closer you get to money, the more money is available to fix the problem. So, you know, once I've actually finished the fermenting and I'm trying to filter this stuff out, you know, there's a lot more money available to uh and and a lot more recognition of I gotta get this problem solved in the downstream processing end. So in general, I would guess in industry, there's probably more uh questions and whatnot there. Again, I was fortunate to be on, you know, great teams, and the clients I had typically were happy with their downstream processing, and and the problem was somewhere in, you know, I need better yield coming out of my fermenters. Uh, I need to prevent uh contamination in my fermenting because there's nothing worse than scaling up to a 2,000 liter batch and then having some bug you do not want invade and scrapping 2,000 liters now of what you thought was going to be, you know, another whatever, whatever you what your number is gonna be$50,000 or more, whatever that loss is, it's gonna be uh you know reasonable, reasonable loss in terms of money.

SPEAKER_02 35:17

Yeah. At a minimum. Yeah. That's the stuff nightmares are made of, right?

SPEAKER_00 35:21

Definitely. Those that's the number of my number theory. Yeah, yeah. So it's clean carefully, make sure you've got validated cleaning process. Okay, preaching to the choir again.

SPEAKER_02 35:28

So yeah. My uh my friends out of the industry would be like, what are you even talking about? But those who get it, those those our audience knows that's that's the bigger fuel for bioprocessors. Absolutely. Um well, Richard, before I ask you our Hallmark question of the show, I want to ask you what your general advice to our audience would be, what you want them to take away in terms of how should they be thinking or rethinking about quality and process control? Whether it's in line with conventional wisdom, or if you want to push back and kind of flip that on its head, what would your advice be to folks how they can reframe their thinking about quality and process control?

SPEAKER_00 36:03

How to think about quality and process control. Don't ever stop asking the hard questions. Um, and I'm not sure that's really re-rethinking or reframing, but sometimes you forget. You you you get in this unconscious competence, you know, I've been making product great for 10 years. What can possibly go wrong? Oh, as soon as you ask what can possibly go wrong, you know there's a lot of stuff that can come up. So he's just all right. Murphy's absolutely Murphy was an optimist. Uh so again, I think it goes back to those three questions I started our chat with. You know, where are you? Know where you are, keep measuring where you are, and look at the variation in where you are. Where do you want to get to? Think strategically about how you want to grow your business and how does your product portfolio go into that? And do I want to just get more efficient at this line, or do I want to make a spin out of this particular product? Who knows where I want to go? And then once you know where you are and where you want to get to, uh, work diligently to find out how I can get there. And often that last one might come down to process control. You know, if I want to take my current yield from 85 to 90 percent or whatever you might have, often that is a process control question. Not necessarily an easy one, but uh but if you believe you can make that kind of a jump, then often it comes down to better process controls. Maybe even scaling down your batch size and just doing that even more efficiently than uh than something else. Differ different issues to be determined.

SPEAKER_02 37:24

But uh great advice for sure.

SPEAKER_00 37:26

Keep asking good questions.

SPEAKER_02 37:28

Very good advice, very good advice. Uh now, Richard, I want to ask you the question that I ask every guest that comes on the show, and that is how can we, Richard, better biopharma? All right. How can we improve the way we design, develop, manufacture these medicines?

SPEAKER_00 37:43

All right. So um I have a bit of a different take on this. You did warn me that I might have to answer this Hallmark question. And and I'm enough outside of the field, I'm not sure I'm the best person to answer it for you, but I've got an opinion, and I'm willing to share my opinion with you, Tyler. So let's hear it. And we want to hear it. All right. I'm gonna go to the stretch strategy question. And I think, you know, biologicals in in my lifetime, we have seen an amazing revolution in biological pharmaceuticals. And it's I you you look back, this goes back to, you know, we only got the double helix in like the mid-50s. You know, we only got the human genome in 2000. Uh, and you know, genomics, proteomics, all these things are the understanding of pathways in our body, understanding of you know, key sites to target through biological keys and whatnot. Amazing. It's just amazing what we've learned. Our cancer therapies are better, our you know, treatments for diabetes are better, so many great things. What I would like, if I could get one thing out of the bioprocessing industry and manufacturers in general, is let's focus on two things for our strategic product growth. I'd like to focus on more diversity in products. There are a lot of Me Too products that come out there. Someone comes up with a target and they say, ooh, you know, I can come up with a uh not necessarily biosimilar, maybe it's their own product, but I come up with a similar drug to go after a similar pathway process and and uh because someone else has validated the market is out there, I would like to see uh even more push for diversity on different uh touch points into the pathways, because ultimately I think physicians, as we get more and more personalized in medicine, physicians and patients need more and more diversity in pharmaceuticals so that they can get better outcomes. Right. The second thing I would ask for, that that one I think is an easy ask. It's just uh, you know, think about how you target things and don't be happy with the one product you've got. Look for other things that can impact the same pathway and have potentially different impacts. The second thing is probably harder because of the commercial incentives against it. But I would love, love, and this is not just for biopharmaceutical manufacturers, but all drug manufacturers, to focus more on acute interventions as opposed to chronic interventions. Um it'd be nice if I could, you know, even if it's whatever, take a drug for six months and then I'm cured of something. That would be much more preferable to take this drug for the rest of your life or when you get off your GLP one, it's just all going to go back to normal, something like that, right? And I I appreciate that as a hard ask, both commercially and technically. Uh, but if I could wave my wand higher, that's what I would get.

SPEAKER_02 40:27

Thank you so much, Richard. I really appreciate that, and I'm sure our audience will too. Um, thank you for joining me on this episode of Better Biopharma, the official podcast of Bioprocess Online. And thank you out there for tuning in. We'll see you next time.