Andrew Sinclair Offers Insight on Optimizing Manufacturing and Operations

TRANSCRIPT

Announcement: 0:06

Welcome to the Bioprocessing Unfiltered podcast. Each month we host conversations with the researchers and leaders tackling and solving the day-to-day challenges of the bioprocessing industry.

Bill Whitford: 0:19

I'm Bill Whitford, and welcome to this installment of Bioprocessing Unfiltered. Today I have the great pleasure of introducing Andrew Sinclair, a very experienced president and founder of Biopharm Services, and I'm looking forward to reconnecting with him. We've known each other for many years, and I'm looking forward to some answers to specific questions I've prepared previously. One thing I appreciate about Andrew's depth is that in in his consulting work, dealing with many different individual firms, he has he gains the experience of different perspectives. Often when we're interviewing one person from one company, we're getting a filtered, a biased perspective. And what I appreciate about appreciate about it discussing with Andrew is the fact that he in dealing with so many different premier organizations has quite a repertoire and and a depth of perspective that gives a full meaning to what we're looking at. So Andrew, would you like to give a little introduction to yourself and what you're doing currently?

Andrew Sinclair: 1:26

Yes, thank you, Bill. And I've been in the industry over 35 years, so I've mirrored the growth of the biofarms industry. So I started work back in the early 80s on the first biotech product, which was Eli Lilly's insulin. And it was very interesting at that time, the focus was on cost, which was quite surprising. That was because it was a replacement hormone, replacing porcine insulin with recombinant. And if we look at the development of the pharmaceutical industry, as I moved jobs and into the mammalian side, the focus and the emphasis of the industry was more on the science and not on cost. And so in my career, I worked for Celltech Biologics that became Lonza. And after leaving Lonza, I set up biopharm services. So in my time in industry, I'd spent a lot of time looking at manufacturing, designing facilities for manufacturing, and looking at multi-product manufacture, and looking at all the issues of how do we get the best out of our facilities. And it was really that aspect that drove me into setting up biopharm services, was really to look at how do we how do we best use the technology and optimize the processes to get the best out of our manufacturing assets. So back in the 2000s, when we started biopharm services, there was not a lot of focus on cost. But then as the monoclonal antibody market developed cost became more and more of an issue, a bit like the early days of insulin. And so it's really my pleasure to have this discussion with Bill because we can we can sort of mull over some of these the history of the business and and how that we can look at what's happening today from that historical perspective.

Bill Whitford: 3:35

Great. Well, you know, there's so much we could talk about today, and you you've introduced a couple of key factors, but I've got some prepared questions, and I thought we'd start from there, or we can wander off as to what we see interesting. The first question I've got is what do you see are the most significant ways data management and predictive modeling can optimize manufacturing operations and improve efficiency? But you could flush this out into any direction you see.

Andrew Sinclair: 4:06

But I think the the concept of what do we mean by process efficiency is probably a good starting point. And I try to look at these from, I try to stand back and look at these from a holistic perspective. And I think, well, what's important to the business? So the business to the business, process efficiency is a word. And and from the business perspective, how does that translate to cost efficiency or cost of goods?

Bill Whitford: 4:41

Aaron Powell Well, you know, and that's what I see is people I sometimes have a local definition that they assume is is the proper definition. We say efficiency, it could be in the terms of the metrics of of time or personnel or square footage or money. So I like the fact that you're starting here with with defining what what do we mean? What is the goal of the phone?

Andrew Sinclair: 5:00

Yeah. Yeah, and and you see this you see this issue all the time. So it when we look at processes, people often define efficiency from their own perspective is what's the productivity in the bi reactor, what's the productivity of the chromatography column? Whereas actually what's important from economic perspective is how do these changes impact on the cost of goods? And also what is the environmental implication? Because if we look at the corporate level, they have two drivers. One is obviously getting new products to market, but getting them in a cost-effective way. And how do we make sure that we minimize the environmental impact?

Bill Whitford: 5:51

Which which illustrates that it's we not only have to come up with these definitions of, for example, efficiency, but that they evolve. And twenty years ago there wasn't, or ten years ago there wasn't the imperative of of including reduction in environmental burden in the definition of efficiency.

Andrew Sinclair: 6:08

Yeah. And so what I wanted to do really was to think about the context. So if we if we take cost of goods as an example, because that's a commonly measured metric that's used by management and process development. Basically, what cost of goods is, is how much does it cost to manufacture that product in that factory? And it's basically that. It's nothing else, it's not corporate overheads, it's so really it's a true measure of you know what resources does it take to make that product. And and that's a useful metric that the accountants use. They obviously have a more complex definition for it, but that that in simple terms is how the lay person would view it. And then we have um the environmental metrics, which a lot of organizations are defining or setting themselves as objectives as they look at um as they look at the impact of climate change and how they respond to it.

Bill Whitford: 7:14

And we know here there's the same problem of defining what's the scope when you're talking about the environmental burden of manufacturing or of producing a product, in a life cycle assessment, how far back into the materials do you go and how far forward into the life cycle do you go when you're talking about this? So it you know, this is an illustration of why it takes the experience of a person like yourself to help a vendor define exactly what questions he's going to answer before he gets started.

Andrew Sinclair: 7:44

Yeah, and I I think that's an important point. So I wanted to try and give a context to to to that point, those points. The first one being is cost of goods. So this is an area that has interested me a long time because I came across it in insulin. So if you think about insulin in the early 80s, I was asked by Lily to do a cost of goods model. And never had to look at cost of goods ever again until 2000. So that's a good illustration of where how the industry's changed.

Bill Whitford: 8:19

It's good for persons like us who can remember the 80s. Yeah. Maybe not for some people in the audience.

Andrew Sinclair: 8:26

Yeah, but that that reflects the nature of what was happening. So because it was replacing an existing product, it had to meet cost of goods targets and they had to have volume for the all the di. Whereas when we started to get into interferons and antibodies, we just didn't know how to make them.

Bill Whitford: 8:46

Yeah, get out the door with it with a a goal enough.

Andrew Sinclair: 8:50

Yeah. And we worry about the economics later. But if we look at the cost of goods or cost of goods sold in the pharmaceutical sector and go back 30 or 40 years, the cost of goods associated with manufacturing as a proportion of sales price was four percent. Four percent. So that and it was largely driven by small molecules. So in those days, people used to say manufacturing costs don't matter, right? And that that was a mantra that existed right up into the 2000s. But the interesting thing is that if you now move forward to 2008 and focus on the biopharm sector, that four percent has grown to fourteen percent. So cost of goods now represent fourteen percent of this is an average figure of the sales price.

Bill Whitford: 9:48

So, you know, my interpretation of that is to to people not in the business side, that doesn't sound like a lot, but it's it's three or four times what it used to be. So from a business perspective, that's significant.

Andrew Sinclair: 10:00

And it's worse now. So today, so if you look at the situation today, the range of cost of goods as a proportion of sales price relay ranges from 19 to around 36 percent.

Bill Whitford: 10:16

Now, this is specifically for biopharmaceuticals in general, or I mean would you include ATMPs or this is more monoclonal antibodies?

Andrew Sinclair: 10:25

What it's it's the it's sort of the range that you would see across the biopharm sector. So that would cover vaccines, ATMPs. Okay. And so so you can understand the range. So but you still, if you hark back to the 30 or 40 years ago, the way it was 4%, now it's 19 to 36%. And then you've got to add on RD and so on and so forth. So the economics of bringing a job product to market have changed significantly. And we can I'll touch on some of the drivers behind that in a minute. I just wanted to now move on and say, well, what are the priorities from an environmental perspective? And here what I'm saying, here what I'm going to talk about is not what I think the priorities are. I'm gonna we did a survey, an analysis of what the pharmaceutical company's priorities are. And we were able to rank them in terms of popularity.

Announcement: 11:26

Are you enjoying the conversation? We'd love to hear from you. Please subscribe to the podcast and give us a rating. It helps other people find and join the conversation. If you've got speaker or topic ideas, we'd love to hear those too. You can send them in a podcast review.

Bill Whitford: 11:44

You know, Andrew, if I could interrupt you, this is why I love chatting with you, because we all have our opinions, even from the literature we'd read of the answers to some of these questions. But what the power you have is that from your survey and from all of the diverse clients you have, you can get a true consensus answer is of what's really being done, not what I think should be done or is being done, but but it's this is the power I think in in chatting with a a person like yourself. So please.

Andrew Sinclair: 12:12

Yeah, so I I thought that was quite interesting because um when we think about what are the important priorities, that should dictate what we do at the process development level, basically. And that's so that's why I wanted to get a perspective on it. So number one priority that the majority of pharma companies have is this transition to net zero, which is all about what they call scope two emissions. And scope two emissions are basically the energy you use associated with a manufacturing facility that's making the product. So that would be your electrical supply, the level of carbonization of the electrical supply, so how much fossil fuel is used to make the electricity, and the amount of gas or oil used in the boiler. So that's what we call scope two. So when organizations talk about net zero, they're talking about making the operation of the factory carbon neutral.

Bill Whitford: 13:16

And if I could clarify, some people might not be familiar with the different scopes of emissions and that we deal with. So you're defining that you're looking at scope two here, and this is not involved in the retirement of the product or the mining of the raw materials. It's just to clarify and to categorize these things so we can speak more clearly. Yeah.

Andrew Sinclair: 13:36

Yeah, that's exactly yeah. And then priority two from the pharmaceutical company's perspective is scope three. So they call life cycle carbon emissions. And that's really so if you look at it from the factory's perspective, making the product, it's accounting for the carbon associated with the inputs. So if we think about our typical factory, we're using buffer salts, we're using filters, resins, and single-use components, they all have an associated embedded carbon. So that's the carbon associated with the manufacturer of those products up to that point of delivery at the factory. So when the factory talks about its scope three, it's taking the input car embedded carbon, adding on the scope two, and calculating a value associated with its product.

Bill Whitford: 14:34

So it seems, hey, I I've looked at this before and it gets can be confusing because one person's scope two can be the other person's, or one person's scope two can be the other person's scope three, or it and and down the line. So you can't say whether something is a manufacturing emission if one person's manufacturing is the other person's product.

Andrew Sinclair: 14:52

Yeah, you so you have to look at it from the perspective of the manufacturer. And so if you look at it from the business level, they're looking at the supply chain of each factory in the supply line up to the point of delivery to the patient. But we can but if we're looking at a product being manufactured by the factory, we're looking at the output. So it's it's a question of where you are in the organization and how you look at it. Because if I'm in the factory and I've got say a target for scope 3 emissions, and I want to get them down to near to zero as possible, I can't do anything about upstream, but I can influence the downstream. So my target is to generate a product, an antibody bulk drug substance, for example, with zero carbon associated with it, taking into account of all the carbon inputs. So this is where I I I um I would um say to your point, definition and defining boundaries is extremely important to have a meaningful discussion about some of these points.

Bill Whitford: 16:03

And to that, it takes experience. It takes to know is this a to me it seems a bit like law. And then it's not so much what you think might be good or could be good somewhere else, but it's what's being done, how are people handling these definitions. And so that you can, with regulators and with organizations that are rating, you can speak their language and and know what what is actually being done, not what what could be done or what degree would be.

Andrew Sinclair: 16:31

And we're looking into it in a lot of detail because um, and we'll touch on this later, but our platform, Barcel Process, is used for techno technoeconomic analysis by a lot of the pharmaceutical companies and the suppliers. So but ever since we launched it in 2008, people have wanting us to add these environmental metrics. Um so we've been looking at how do we how do we predict scope two, and then um how do we then evaluate scope three? And that's when we come to how do we evaluate scope three, that's a very interesting topic in itself, and they can probably do its own podcast, but the complexity is so if I take, for example, a chemical used for making buffers, which is an input to my factory, acetic acid, so people are looking for what's the embedded carbon of that? What's what's that value? Tell me the value. But if you actually analyze it, and we we this is where we're using AI in an energetic way to really scour the different sources and values for the embedded carbon associated with, say, acetic acid, what you find is probably there's 15 academic publications that have done LCAs on acetic acid. So, and they're all different values, right? So then how do you say what is the correct value? Um so you have to then look at, well, what's the what's the common commercial manufacturing process? And what's the impact of making it pharmaceutical grade? So you go down, you have to be very careful not to go down a rabbit hole, but you have to you have to make a judgment at some point and say, well, this is the reasonable value to use, and this is the basis of it. So as long as you can say that, I think that that's the important thing.

Bill Whitford: 18:41

Yeah, I would think for each of these values, you there there was never one specific exact answer, but you can come close on averages. I could see different manufacturing methods from different countries, and if you have multiple sources or tertiary source that you only employ on occasion, that might change the the it's specific exact answer, but you can get an average to go for it.

Andrew Sinclair: 19:05

And I think and you also have to bear in mind that as we decarbonize all the values are gonna change. So anything everything it's a bit like cost.

Bill Whitford: 19:15

It hadn't occurred to me. I'm thinking about all the variables and that variable hadn't occurred to me. Yeah, no, it's a bit like cost. Yeah.

Andrew Sinclair: 19:23

Yeah, so that's what we, you know, again, going back to the Basel process, we populate that with a database of all the resources that are required for manufacturing at any scale. So that could be lab scale, up to 20,000, 30,000 liter map cell. So that means what does that mean? Well, that means that we have all the buffer salts, all the media salts that commonly are used. So we had to have we would have the pharmaceutical grade pricing and molecular weights and whatever. For consumables, we would have all the product ranges. For whole bags, one liter to 5,000 liters, and so on and so on. And so for bags, we would have things like the weight and the cost. And for equipment, we'd we'd have a similar thing. But again, with that sort of data set, you have to update it regularly. And I see the same principle being applied to the scope, the embedded carbon on the input side. There's something that has to be regularly updated for the models to be retain their credibility or relevance.

Bill Whitford: 20:32

And in your discussion, I can I can see that there could be other variables changing dynamically aside from the process of the sourcing. That is if you're kind if you're changing from making many smaller batches and you you you you defined a larger scale, but now you need it. If if there's a dynamic there that from from two or three smaller ones a year to two or three larger ones, that could really change the the final value. And really nothing else is changed in the definition, just your usage of those scales.

Andrew Sinclair: 21:07

Yeah, yeah. And and that's the next point. Because when I was talking to you earlier about cot the the increase in cost of goods, you can look at the trends in the industry and and they start to explain it. So what, so you, what do I mean? Well, if you look at this trend to personalize medicines, what we're actually talking about is smaller addressable populations of patients that we're making our product for. Which means that the scale of manufacturing required to deliver it is smaller. So that that's a direct consequence. And secondly, the process that we're using is much more complex than his than we would have historically. Now, okay, so that's the sort of context. Now, if we come back to cost of goods, we have to say if you look at across any industries and it applies equally to our industry, that you can change cost of goods by changing technology and process, but one of the key drivers to driving our cost of goods is scale, and it's a power law relationship, it's a negative power law relationship. And so what we have as we go from the early 2000s to today, we we're making smaller quantities of individual products, which in itself, if we did nothing else, would would double treble the cost of goods. And the same principle we found applied to say scope two emissions. So as you reduce the scale of your factory and it's making smaller products, it's less efficient from an energy perspective. And so we have the same power law relationship for scope two emissions. And so that's something we've observed directly from the modeling work. Um, and it's quite an interesting observation.