Defining Extrapolation

Drs Roy M. Fleischmann and Joel M. Gelfand explain extrapolation and how it will change the field of dermatology and rheumatology.

Roy M. Fleischmann, MD: We talked about extrapolation of biosimilars. What does that mean? Extrapolation is the approval of a biosimilar for use in an indication held by the original biologic, not directly studied in a comparative clinical trial with the biosimilar. Extrapolation is the scientific rationale that bridges all the data collected and the totality of the evidence from 1 indication for a biosimilar product to all the indications that were originally approved for the originator. The approval and marketing authorization of biosimilars in inflammatory indications are made on a case-by-case and agency-by-agency basis after evaluating the totality evidence from the entire development program. This totality of the evidence comprises extensive comparative analytical, functional, nonclinical, and clinical pharmacokinetic [PK]/pharmacodynamic [PD], efficacy, safety, and immunogenicity studies used by a regulator to evaluate whether their product can be considered a biosimilar. Extrapolation reduces or eliminates the need for duplicative clinical studies of the biosimilar but must be justified scientifically with appropriate data.

In plain English, if the biosimilar has been shown with the totality of evidence that it’s virtually the same as the bio-original, and the bio-original is approved for 5 disease states, one would expect that the biosimilar would be effective in each of those disease states. That’s what extrapolation means. Joel, do you have anything that you want to add to that?

Joel M. Gelfand, MD, MSCE, FAAD: No, other than that it always makes me a little nervous as a dermatologist. We know that the efficacy of biologics varies across different disease states. In the field of psoriasis, particularly with TNFs [tumor necrosis factor inhibitors], we often need higher levels of doses to get good responses in the skin compared with rheumatoid arthritis or psoriatic arthritis [PsA]. In these circumstances, when a biosimilar hasn’t been tested specifically in psoriatic disease, to be certain, we’re going to want to be looking to see that we’re getting the expected outcomes we’d hope to see in our patients.

Roy M. Fleischmann, MD: Let me clarify that. You have a biosimilar of adalimumab which was tested in rheumatoid arthritis, and it’s almost exactly the same—within that variation that you said before—with the reference product. Then the biosimilar is extrapolated by the FDA to be approved in PsA at the same dose that would be used in PsA. Would you have concern?

Joel M. Gelfand, MD, MSCE, FAAD: Ultimately, we do studies in humans for a reason, and that should be considered the gold standard. Despite the fact that they’re well studied in these circumstances, and you should get the same result, it would be nice to have actual clinical data to be certain in these circumstances. We’ll see how these things play out. You’re absolutely right. It’s more likely than not that the drugs will perform perfectly fine in these scenarios, but increasingly it’s moot. As you mentioned earlier, a lot of these biosimilar products have been used for many years in Europe, so we’re already working out these issues, and it’s pretty unusual for a biosimilar product to not have data in other indications. Maybe that was the case many years back, but increasingly, most of the biosimilars will have data across indication.

Roy M. Fleischmann, MD: I’m always data driven, except in this particular situation. The data for me is that the PK/PD of the biosimilar is virtually identical to the bio-original, and therefore the molecule should be present whatever the disease state; the same PK with the same PD. I’ve done studies in different disease states. Some haven’t. Some have done studies in only psoriasis or psoriatic arthritis or rheumatoid arthritis. But the data that you’re talking about are registry data, and to me, registry data aren’t really data. It’s like in social media. Not every patient is followed exactly right. They don’t get all the information, but you get a feel if there are enough patients. I don’t think the companies themselves will do studies in multiple diseases because they get extrapolation. Why should they? But it’s an interesting topic.

Transcript Edited for Clarity

Related Videos
Deepak Sambhara, MD | Image Credit: American Society of Retina Specialists
David Brown, MD | Image Credit: Retina Consultants of Texas
© 2024 MJH Life Sciences

All rights reserved.