In the driving seat

In this excerpt from an interview with NGP’s sister media channel, MeettheBoss TV, Sanofi Pasteur’s Rene Labatut tells Victoria Newing the secrets to staying in control of your vaccine manufacturing processes.

“We want to drive our processes instead of following them; when you drive you are in control”
-Rene Labatut

You have 22 years of experience in the biotechnology across variety of disciplines. How does this insight help you in your current role?
Rene Labatut. My experience has helped me in two ways. First, to have the right level of humility - there are a lot of things you have to learn and you can be surprised every day; the biotech world is like that. And also to always be looking at the big picture; sometimes I need to go into very detailed things, but I must always have the ability to go back and forth between the different levels of vision, as well as to build a vision of what's next based on experience and extrapolation between things.

My key mission is to be the guardian of industrial knowledge for the company. To be a guardian is a dynamic thing and not a passive one; not keeping things as they are, but moving them into what they should be in the next 20 years. It's transferring new products from R&D to manufacturing: taking them from phase II up to commercial launch.

This also involves being in charge of continuous improvement for any manufacturing process, as well as technological innovation. That's my primary responsibility. Besides that and within that, there is all the techno transfer between internal sites and also between Sanofi Pasteur and our partners.

Sanofi Pasteur is the largest company in the world devoted entirely to human vaccines. How do manufacturing processes differ across the globe?
RL. Sanofi Pasteur's manufacturing processes use the same standards but have different processes for the same specialty for the same therapeutic queues. We have three Tetanus processes, for example: one in Canada, one in the US and one in France, and now because of the recent acquisition we have one in India too.

The difference in these processes is historical: they come from the original companies. Now when we are developing a process, even if we have several manufacturing units, at the end it's notably true in formulation. The processes are the same; the challenge in this field is to keep it the same because along the life of a process in a different unit, you always have circumstances where people make minor changes. So you have to be very strong on your change control policy.

There is a system called a process book, which is an electronically organized reference where people find the detailed description of what they have to do. It is web-based, so any information can be accessible in less than six minutes; and there is exactly the same reference for people who are running the same process in two locations.

It's anticipated that the demand for new and existing vaccines will double by the year 2012. What measures are you taking to ensure that Sanofi can delivery and meet market demands?
RL. Four or five years ago we engaged in a strong investment policy. On a yearly basis we invest between €350 million to €400 million [$480 million to $550 million] a year, and have built a huge inactivated polio vaccine capacity in Lyon, France. We have more than doubled our flu capacity in the US for seasonal flu and the formulation/filling in the US has also been extended. We have invested close to $200 million in this unit, and we have also invested very heavily in Normandy, France, to do formulation, as well as filling and packaging for distribution of vaccines on a worldwide basis.

Now we are also leveraging all the capacity we have at the group level, notably in packaging because it's less specific for vaccines. We have a plan to invest more than €250 million [$34 million] to build a new dengue unit; and there is also an ongoing plan to develop new vaccines for diseases which are not yet covered by vaccination programs.

What new approaches are needed to drive manufacturing processes more efficiently and to overcome the complexities of new vaccines?
RL. New vaccines for new diseases can be more complex to produce because they have processes that are more complicated, such as new conjugates or new formulation types, which are more difficult to manage. The other way around is to design your product approach - this can be more complex for something like the dengue vaccine, which is a chimeric vaccine for a chimera, - but the management of the process by itself is not so complex. The way you manage your cell and the interaction between the cell and the virus is difficult, but in comparison to a classical inactivated polio vaccine, the process itself is simpler because the complexity has moved into the product itself and the construct will simplify after the process.

A vaccine is a preventative medicine; when you inject a vaccine into someone they are not sick yet, and the worst thing you can do is to make them sick using the vaccine. So the level of attention you pay to safety is very high.

This ties in with the trend of moving to a more preventative type of medicine from a curative one, and requires a different way of thinking. The only area where the risk/benefit ratio is very high is in therapeutic vaccines, but mostly these type of products are the last line of defense or therapeutic use where the risk is death within six to nine months, so it's a different story.

You're an advocate of the software solution process. Can you elaborate on the benefits that this can offer a company like Sanofi?
RL. We want to drive our processes instead of following them; when you drive you are in control. When you are following, you are just reporting what's happened, which is not really the mindset we should have. We want to go to a QBD-type approach and have a consistency of output of our process; consistency of results versus the classical consistency of settings. There is a lot of classical approach in secure process control.

Following critical parameters step-by-step independently of each other assumes that if you take a given step the result of this step will not influence the step after. In our process there is a lot of difficulty, and it's not mathematically accurate to say that the steps are independent. We have to consider the process as a rule, with interconnection between steps; and the result of what happened in step one will have an effect on what will happen in step six or eight and can be corrected by what will happen in step 10.

But you need to have the overall behavior; and this type of software is trying to address that, looking at the overall process and how it behaves and interlinks. And instead of having a critical parameter only on one step, you have a cluster that are on different steps. But it's based on a mathematical model like genetic algorithm, which is a learning-type process. And the way we use it, notably on high frequency products, enables us to look at what happened in the early stages of the process for some parameters. We must then predict based on all the models where it will end if we don't correct the settings and diagnose what will be the appropriate corrective setting we have to put in place before the end of the process. Then we can be sure that we are very close to the point where we want to end in terms of product quality and product yield.

How does the manufacturing process help drive the need for innovation?
RL. The innovation I'm focusing on centers on discovery work in vaccines. We now face the challenge of being able to do this in an economically viable way and also in a volume way, in terms of the number of doses produced at a reasonable cost. They must be affordable on a worldwide basis, which is not a niche market for most of these products. We need to innovate to make this possible, to extrapolate directly what's done at the lab bench scale. We cannot do it at the large scale that we need, so there is some innovation there to be driven by us.

We also need to improve processes and assure process performance. We need to think about doing it differently - changing and applying new technology or a new way of thinking, leading into new manufacturing technology even for existing products. The big drive behind that is that you can manufacture a certain way, and put the cost of goods at a value of 10. The number has no real value, it's just to give an order of magnitude. If you want to sell that in Western Europe, that's fine. But if you want to address it on a worldwide basis and encompass developing areas, you have to bring your 10 down to under one; and to do this you have to have a technology breakthrough.

There are different ways to achieve this. First, you think of the classical approach of applying Lean to your manufacturing process or getting rid of things that don't make sense or are just there because of history or habits. The second step is to maximize the value of what you get out of the things that it still makes sense to do.

What drives the performance of the process, in terms of product and making yields? The huge leverage it yields. If you understand very well how your complex biological system is working, then you can apply a method or attune the way you drive your process to increase. Just by managing the biological organism we handle its physiology very closely and see how it behaves when it becomes invasive in the body, then use the defense mechanism as a leverage to obtain the product we want versus when it is in a comfortable situation doing regular things.

How is innovation formalized and managed at Sanofi Pasteur?
RL. In order to manage innovation, we have had to sort out what the fields of innovation are. We are looking at what we have in the pot from discovery down to commercial. What is the overall portfolio of products in the company, the ones that are at the early stages, as well as the ones we are manufacturing? We're working on the type of processes and the technology we are using from the early microorganism down to the final box - looking at where the needs for technology are in the company and also in terms of order of magnitude in the market.

First, we are looking at needs we have not covered by existing technology: where we have developed something, where it applies and what is the value for the company according to the span, focus or depth into a certain product. That's the needs part.

We are also scanning what's going on in different areas of the industry, not only in pharma, but also food and microelectronics. We are scanning what's going on in universities and at the very early stages, looking at where it can apply and where it can make a difference; in doing so, we want to develop technology. Our technical teams are evaluating continuously with biotech teams or with universities, several of which we have contracts with.

What role do issues such as the relationship between development costs and drug prices play in the transfer from R&D to operations?
RL. If you take too long in transferring from R&D to operations costs will increase: any time you take will be costly and this will have an impact on the cost of the goods. If you find ways to speed up this transition or transfer, normally you will reduce the cost at the end and you have more leverage to be able to negotiate or provide the services with a better price.

That said, to be able to transfer quickly from R&D to operations, you need to have a very specific approach on the R&D development level because things that are a little bit fuzzy or not totally developed will induce a lot of rework during the transfer or just after the transfer, and that doesn't help to keep costs at an optimized level.

Sanofi ensures that we're optimizing that process by mostly imposing a quality by design mindset, starting in phase I. The pharmaceutical development of new products is jointly managed by the VP of R&D and myself, and to ensure the team diagnoses in phase I or phase II the critical quality attribute of any product, the critical process barometer associated with the system.

Then there is also preparing the scale-up algorithm: starting to optimize upfront as much as we can yield reliability and the capability of processes to eliminate what can be damaging for the operation afterwards; and also to help minimize the investment needed.

Rene Labatut is Vice President of Global Manufacturing Technology, Sanofi Pasteur.