Addressing the Challenge of Sample Size in Bioequivalence Studies
Helmut Schütz, CEO of BEBAC, joins Pharma IQ to discuss challenges and changes in bioequivalence studies. To listen to the podcast now go to Challenges and Changes in Bioequivalence Studies.
Pharma IQ: Helmut, welcome to the show. How are you?
H Schütz: Fine, thank you.
Pharma IQ: Okay. So to start with, where do you think the main challenges are in bioequivalence and bioavailability studies for biologics at the moment?
H Schütz: For biologics, it’s a little bit difficult, because actually, just to go back to terminology, in Europe, they are called biosimilars and in the US, they are called follow-up proteins. In both regions regulations, I think we are still in a phase of developing and gaining knowledge, so there are published guidelines, but every single product is so different and it’s very important to know that the common concept of bioequivalence, which was used for and still is used for small molecules, for generic products, is not really applicable to these biosimilars. This is because it’s only one part to show that the plasma concentrations of the drug are the same or similar, but this is just one part of the story, and the other thing is it’s actually the entire process which should be similar in some way, and it involves clinical studies, it involves pharmacodynamic studies. You have to show that the new product doesn’t cause inflammation or anything like that, and it’s a very complicated story.
Pharma IQ: What about the issue of reducing the time and cost of studies? Is this another challenge that you’re facing?
H Schütz: Yes, it is, like everywhere in the industry - not the pharmaceutical industry especially, but just for biosimilars, I don’t think that it’s really possible to cut down costs, and especially look at the costs, because these types of products are so complicated. Actually with all these products, in Europe you go for a market authorisation only centralised at the EMEA in London, and you do this application in close contact, working together with EMEA, so it’s always going to and fro with questions and about developing new studies, and you have a lot of scientific advisory meetings. We are trying to do our best to work as cost-effectively as possible, but all the products are different, so I have no single answer to the question.
Pharma IQ: What technologies and solutions do you employ in order to address the different challenges that you’ve outlined - concerning both biosimilars and generics?
H Schütz: Okay. For biosimilars, as I said before, for every product it will be different. For example, you have to assess the product actually by all analytical methods you have, because you have to show that your product performs similarly to the innovator, but you have a different process and you have to qualify and quantify all the components of your API, and the API is not like a small molecule. It’s just a single drug. It’s a mixture, and it may consist of monomers and there may be different glycosylation patterns, so it’s very complicated, and so with biosimilars, it’s nothing you can talk about in a general way. You have to look at every single product and then think about it to tackle all these uses.
For generics, it’s small molecules and the conventional bioequivalence method. I think the most important thing right now is that the new bioequivalence guideline came into force on the 1st August, it was published in January, so companies had a half a year to implement the new recommended procedures for their studies, but still there are open questions, believe it or not. So there are actually rumours that there will be something - and I say it in italics, something will be published by EMEA to clarify some open issues, mainly with the biostatistics for the immediate release products. Right now, the guideline is there for half a year, but many companies are still confused how to evaluate the studies.
Right now, it’s really a little bit confusing, so for example, the guideline itself consists of about 25 pages or so, but there is a commentary document which consists of 250 pages. I can recommend everybody looks at the commentary as well, because sometimes this commentary clarifies the point of view of EMEA. But still there are open issues and hopefully, in the near future, these issues will be clarified by EMEA. We’re all waiting for that.
Pharma IQ: Sure. Now, moving on to another area, obviously there’s a lot of hype in the market about quality by design. Is this something that you think will provide benefit with developing biosimilars, and if yes, is this something that you’ve had experience with, Helmut?
H Schütz: Yes. Quality by design is, as you say, it’s a hype, this is just the right word. Yes, it is. There are a lot of papers published on that and innovators actually claim that they are using it, but if you look at their presentations at conferences, my personal view is that actually, they do it the way they have done it in the last 10 years, and now they have a modern heading for that. This is just my personal point of view, and if you look at this presentation, I think they fix a new tag on it, but I don’t see really a big difference in the developing process. It’s a new label, but this is my very personal point of view. Maybe I’m totally wrong.
Pharma IQ: Sure, and the next thing that I’d like to ask you is: do you believe that you can really approve the bioequivalence of a biologic product or not?
H Schütz: In strict terms, and there have been discussions about this, maybe when these issues of biosimilars came up about 10 years ago, that a biosimilar or a follow-up protein in the US is not a biogeneric. In the beginning, the term biogeneric was used to notify the similarity of the generic approval of small molecules, which would mean that you run a bioequivalence study and just show that when the plasma levels are within some boundaries, it’s okay, but this doesn’t work.
You have to show the similarity of the entire process, so a bioequivalence study showing the similarity of plasma profiles is just one part of the show. I have seen approvals for some biosimilars, where the conventional bioequivalence study failed, but still the product got that approval, because by very sophisticated models of pharmacokinetic, pharmacodynamic modelling, for example, it was possible to demonstrate that actually, the blood concentrations were not important for this type of product. This is because it was possible to show that the effects were connected to the pharmacokinetics, not primarily in the plasma, but in a deeper compartment, and by modelling it was possible to show that in this deeper compartment, which was itself connected to the effect compartment, here there was worse bioequivalence, but this is not the conventional bioequivalence approach.
So for biosimilars, it’s a very complicated story, as I said before. So everybody should keep in mind that the concept of bioequivalence is nice to have for a biosimilar, and you have to run these studies, but it doesn’t necessarily mean when you fail with these studies, the product doesn’t get an approval. It depends on many other things as well. So it’s just one part, and not the main part, I would say.
Pharma IQ: Leading on from that, in your opinion, is there a difference between a biosimilar and a biobetter, and if so, how would you say that they differ in terms of development?
H Schütz: Oh. This is an interesting term. I never heard the term biobetter, so I’ll try to improvise. As a consultant I’m independent, but I have contact with regulators and the industry and CEOs, and if I try to switch my brain to a regulatory perspective, I think every regulator will be happy to approve a better product. But that would mean that it’s not a similar product any more, so it’s seen as a new product. That would mean that you have to show efficacy and safety for that product, and that would mean large studies. So if you have a product, which is better – and I’m asking myself, better in what respect – if you find out in the developing process by some studies that your product performs better – the concept of biosimilarity isn’t applicable any more. So that would mean, I would say, it’s a new drug and it would mean large studies. From a regulatory perspective, that’s fine. If you show that this product works and has, let’s say, safety and efficacy, it would mean it can also be a better product. If you show that, fine. Why not? But when you said, how does it differ in terms of development, I would say it would mean large studies, because you step out of the boots of a generic company and become an innovator.
Pharma IQ: Moving on to another area, have you got any experience with the application of biowaivers?
H Schütz: Yes. Biowaivers are a very important point - because in this immediate release guideline, it’s clear that biowaivers are an important point. Everybody, I think, in the industry, when you talk at conferences to people, everybody is eager to go for a biowaiver because they think that when running in-vitro studies based on dissolution, etc, of course, it will cut down costs, because in-vitro studies are costly and the analytics take some time. So biowaivers are really a major point, but what some people forget is that a biowaiver is not created only by showing the similarity of dissolution of two products. It also involves a large literature research. You have to make a risk assessment, so what does it mean? That when you go for a biowaiver, this product was never given to humans, so all your data is based on in-vitro data, so you have to perform an in-depth literature research and do some risk assessment. What it would mean? What is the outcome if the product is not bioequivalent, which you haven’t tested yet? And many people forget that. So I would say biowaivers are really a great thing, but everybody should be aware that it’s not that easy, as they might think.
It’s not just to show similarity of dissolution. So this risk assessment is really important, and sometimes it will be very difficult to get this data, because if no data is published, the cards are stacked against you.
Pharma IQ: Absolutely. Now, to go back now to something that we’ve touched on slightly already, can you elaborate on the main challenges in gaining regulatory approval for biosimilars, and how you avoid the need for bioequivalence data?
H Schütz: To avoid the need - I’m not sure whether it’s possible to avoid this state. Let’s take it from a European perspective. There is one guideline for biosimilars, but there are product specific guidelines, let’s say, for a growth hormone, etc, and all of these product-specific guidelines call for a bio-study, and I’m not sure whether it will be possible to avoid a bio-study. You can try it, but I would say the chances are pretty low. As I said before, it doesn’t mean that you have to show bioequivalence in the strict sense. I have seen some complicated PKPV modelling, which was okay, but still you have to run a PK study, comparing your product with the innovator, and I think when you ask whether it’s possible, how to avoid that, I would say it doesn’t work. I think the chances are pretty low.
Pharma IQ: Yes, sure. Now, could you possibly explain some of the different approaches that you’ve used in the past to ensure regulatory compliance? Have you got any good examples?
H Schütz: Yes. What I think it is, just to do it the other way around, as many other people do… Many people, in my experience in this industry, they look at guidelines as if they are carved from stone, but it’s like in the poem by Gertrude Stein: “A rose is a rose is a rose”.
Guidelines are guidelines are guidelines. They are not laws, so I always try to understand the pharmacokinetics of a drug and the specificities of the particular drug, and sometimes, the guidelines are good, but guidelines cannot cover all the drugs. It may be important with a specific drug to set up a couple of studies, which are against the guidelines, but they are really suitable for that specific drug. I would recommend to all people in the industry, keep the guidelines in the back of your mind and just think about your specific drug and your specific formulation first, and then try to match science with the guidelines, but keep science as the first point. I think this is very important. If you just try to follow guidelines literally, maybe you will fail for your drug. This is just a plain answer.
Pharma IQ: Yes. Good tips there. Now, again, to go into some more depth on one of our earlier points, obviously at the moment, bioequivalence and bioavailability is an area where there are a lot of cost-cutting pressures. Now, we’ve acknowledged the difficulty in this area of making cuts, but where do you see as places that the industry can address this, and are there any different measures that you’ve taken to make small adjustments?
H Schütz: Okay. I would say it’s very important to understand how large studies should be. The size of a study depends on the variability, so if you have a drug which shows a large variability, you have to use a larger sample size. This is plain statistics, but I have seen in the past many errors in two respects. Companies ran studies and just to be sure they ran really large studies and actually, they were throwing money out of the window, because they didn’t realise that the drug has just low variability, so it would be necessary to run a study, let’s say, in 16 subjects or so, and they ran studies in 32. Of course it’s possible to do, but this is not really helpful. And on the other point, I saw people, they were so sure that their product was perfect that they ran studies, and when you look at the data from a pilot study, it was evident that the pivotal study would fail.
The worst I have seen was a company showing me the data from a pilot study and their test reference ratio was 81.5%, and remember that the limit for equivalence is 80%. And then I made a sample size calculation based on their RCV, which wasn’t too high, but moderate, but still with such a product they would have needed 700 subjects in the bioequivalence study, and they said, well, we’ll do a study in 60. Okay? Obviously, they will fail. The chance to get bioequivalence with such a study is close to zero.
Maybe it will not survive the ethical approval. I don’t know. It’s very important to look at the data and I would suggest to anybody to run reasonable pilot studies on their product first, because then they get an idea of the size of the main study. In the past, people ran studies just looking at literature data or something like that, and that means nothing, because you cannot control where this data come from. Different analytics, different clinical settings, it doesn’t mean anything.
Pharma IQ: Yes. It’s definitely good to get your perspective on this, speaking from experience. Finally, just to round off, Helmut, I understand that you’re going to be speaking at our upcoming Bioequivalence and Bioavailability Studies conference, which will take place in Munich from 25th to 27th October. Now, for anybody that’s interested in attending this event, what would you say is the key learning point to be gained from your presentation?
H Schütz: Hopefully, you got a glimpse in my last answer. My presentation will deal with sample size estimation, so hopefully people will get an idea that it’s really important to set up a bioequivalent study of a reasonable size and not run studies which have no chance of succeeding or studies being too large, just to be sure. It doesn’t make real sense. Both types are just a waste of money and time, in my opinion, and it’s possible to tackle that, so hopefully I can clarify that in my presentation.
Pharma IQ: Sure. Well, we look forward to the presentation, to hear your views on this issue. Well, Helmut, thank you very much for your time today. We’ll have to end it there, but it’s been a real pleasure talking to you about some of these issues for the industry.
H Schütz: You’re welcome.
Please note that we do all we can to ensure accuracy within the translation to word of audio interviews, but that errors may still understandably occur in some cases. If you believe that a serious inaccuracy has been made within the text, please contact +44 (0) 207 368 9425 or email firstname.lastname@example.org.