Charles Bennett, MD, PhD, Discusses the FDA’s Biosimilar Approval Timeline

The Center for Biosimilars® (CfB): I'm Tony Hagen, senior editor for The Center for Biosimilars®. Today, we're interviewing Charles Bennett, MD, PhD, a hematologist and oncologist and also a professor in the Clinical Pharmacy and Outcome Sciences Department of the College of Pharmacy at the University of South Carolina. Dr Bennett served as the lead author on a recent paper about the progress of oncology biosimilars and how well accepted they've become.

In part 1 of this interview, we talked about how the FDA measures median time spent on biosimilar application reviews and whether that's an accurate way to do it. We also talked about the rigorousness of FDA reviews in comparison with those in the European Union and Japan. Multiregional regulatory reviews were a key element of Dr Bennett's paper. We also talked about doctor acceptance of therapeutic biosimilars vs supportive care biosimilars.

Bennett: One important point [I want to mention] before we get to the first question on the table, is the time to approval. I know this number is very controversial. One of the things that the FDA says is that they approve these biosimilars within 12 months. So, why is my mean time 1.3 years in this paper? The FDA told me, "No, we have not done a single 1 in less than 12 months because we got the User Fee Act and we've adhered to the User Fee Act.”

I only give you credit for when you pass the test, not when you take the test. So, if you take a test, which is [a response in] less than 12 months, and they flunk you, which means that they give you complete response letter and tell you that you have to reapply, I keep the clock going. I think the median time that we should look at should be from the time when you submit your initial dossier to when you get an approval, as opposed to what they use, which is the initial time to submission to their initial decision on that, regardless of whether it's positive or negative. I think that's more a relevant value. The United States, as you saw, has a tremendously high fail rate. So, every time you fail, you're hurting that median time because they don't give you just 1 day to re-turn your homework in.

In the UK, they have some fails. And in Japan, they have no fails. It affects the median time to acceptance of the drugs by the regulatory authority. For some of the US drugs, the whole time from start to finish could be 4 years if you fail.

CfB: Who's making this argument that that median time should be calculated differently? Is it the manufacturers?

Bennett: The FDA makes it. They are being paid to make sure that the time that they take to approve a biosimilar is 12 months or less. They have a [Prescription Drug] User Fee Act, which says they can do it [in that time]. Then, you get people reading [my paper], and it's not less than 12 months. Then, they're upset because they want to know how come I can't count, but I made an internal decision that [we were going to study] between the time you start to test to the time that you got a passing grade. I don't know, for instance, whether the FDA, if you flunk the test, or you get a complete response letter, makes you pay a second user fee to get your revised application in. I suspect they don't. I suspect they give you a 1-time user fee and you do it till you get it correct.

CfB: Well, that's interesting, because the FDA is trying to revise its user fees so that it has enough money to cover the costs of reviewing these applications. And if it flunks somebody and they have to come in and do a more involved review following that, it's going to drive up their costs. So, are the user fees actually going to cover the expenses involved?

Bennett: I think it's important to get it correct on your first try, and it's just like a student. You certainly don't want to take a final exam, flunk it, and have to take a retest. And these are bigger scales. The fact that we put it up there pretty heavily and went into detail about why they flunked the test really raised a strong concern about this. And that's one of the kinds of questions we have about a barrier that I think exists—the flunk rate in the US was significant. It was less so in Europe and not at all in Japan. We could just be much harder graders, maybe. But at the end of the day, it definitely takes its toll. And there should be some education between the FDA and these biosimilar applicants, because you can't live with a 44% flunk rate. My kid wouldn't even be able to graduate the next year of high school with a 44% flunk rate. This is what the FDA does.

And then, if you ask a provider to be comfortable with a biosimilar and say, "Are you happy with a drug that's gone through this FDA application 2 times in a row and is now on your desk for use at a 30% discount, at a 10% discount, or a 20% discount?” And that's where I don't think the educators are being realistic.

CfB: Well, do you think that the flaws are in the application that's being put together, or are the flaws are in the molecule itself?

Bennett: If you look at the table that we went through—Table 2 on what the flaws were—they were clear in some of these cases. Some of the cases [had to do with] manufacturing. [The FDA] went through a manufacturing plant and they didn't like the way the plant was [set up]. For Pfizer, they were told 3 times that their plant was not up to speed. The Pfizer people were hard-headed about it and sent in their application. Then, they were shocked when they got a rejection that said that their plant wasn't up to speed. They had already received 3 letters telling them that the inspectors went through and didn't like that the plant was not clean and not appropriate for manufacturing. And that's a big part of it. One of these was a failure on a clinical trial. A clinical trial was supposed to have a preset limit of where [the biosimilar was supposed] to be about the same and it ended up outside the limits.

So, I think a lot of it has to do with the manufacturing when these complete response letters come out. And I don't see as many complete response letters or not-approved letters in England coming out with concerns about manufacturing, which suggests to me that the manufacturing reviewers for the FDA have a different checklist than the manufacturing reviewers in England and Japan.

CfB: How thorough is the FDA review process compared to the reviews in the UK and Japan?

Bennett: I have to say, the question really is, is the FDA too cautious or are Europeans too liberal? That's the way I look at it. And it looks to me that if you're really focusing on manufacturing, and you have manufacturing experts on the review team, and they find [problems], I think that they're the real deal. In fact, look at the manpower we have at the FDA vs the manpower they have in the MHRA [the Medicines and Healthcare products Regulatory Agency (United Kingdom)] and in other places. We have a huge amount of manpower. And I think they go in there and they kick the tires pretty good. And many times, there are manufacturing concerns. And I saw that there would be applications that go through Europe, as a European biosimilar, that would have no complete response letter. And the same drug would come to the US and get a complete response letter, for quality. So, if you ask me, they're doing it side-by-side. And they’re just giving the Europeans a pass and the US a failure. That leaves you with some hesitancy to use the drug.

When you ask about the barriers and why aren't these things taken out, I think that one thing needs to be clear: They have to improve that pass rate. They've got to do a better job of getting through without getting rejected; otherwise, people at the [reference product] company are going to say, "Your biosimilar just got rejected for quality reasons. Good luck, pal."

CfB: Well, there’s the issue about whether regulators and individual countries should accept trial evidence that was gathered outside the country. Manufacturers say that it would streamline the process if that were done. But you've found that, in some cases, regulators are saying, "No, we want these trials performed in the country." It seems that there might be, based on your statements, some merit to maintaining this system of having in-country evidence only.

Bennett: So, one point there, just to be clear. I'm talking about manufacturing facilities only. If you look at my paper, with these drugs, companies that do phase 3 trials that try to show [biosimilarity], we found that they were done, oftentimes, in 50 countries. Now, I don't know for a fact if the FDA goes to visit all 50 countries to look for the primary data. I can't imagine they do. I think that they do a lot of these studies in Eastern Europe. I don't know. I've been to Eastern Europe for a holiday, but I've never been there in any official capacity. I don't know what it's like in terms of putting these patients on clinical trials in Eastern Europe. They may have 1 or 2 patients per country in several of these countries.

The reason why [biosimilar developers] do that is that they want the drug to get regulatory European-wide approval, and they want each country to have the ability to start using the drug. One of the ways is that each individual country gets some experience with the drug in the clinical trial.

CfB: It's said that doctors are, in many cases, leery of using biosimilars, and that's attributed to a lack of education, that doctors have not been properly educated on the value, efficacy, and safety of biosimilars once they're approved by the FDA, and that they ought to have higher levels of confidence than they do. But you're saying that there may be some justification for hesitancy to use biosimilars in the United States?

Bennett: There's a conceptual framework here between the supportive care drugs and the therapeutic drugs. Patients and doctors have a better level of comfort with the point-of-care drugs like GCSFs [granulocyte colony-stimulating factors] than they do with Rituxan [rituximab], Herceptin [trastuzumab], and Avastin [bevacizumab]. So, I think the education and the comfort for the use of [polyethylene glycol] and epoetin alfa is very high.

I just say that I've got 1 form or another of supportive care drugs, I'm good with that. And then, you get down into the backbone of the treatment for these patients with cancer. Say you've been [treating] a patient who's got follicular lymphoma and they've been on Rituxan for years. And then you decide “I'm going to switch to biosimilar rituximab because the payer said so.” Sometimes the doctor says, "Look, I've got 10 good years in on Rituxan. I'm not comfortable with someone switching us to use biosimilar rituximab when I see no need to, except for the economics of it." It used to be that when the patient went to the pharmacy and picked up the generic, they'd go crazy because they would say that they had been on a thyroid pill for 10 years and now they're getting a generic thyroid drug. I think therapeutic care is different than supportive care. I make that difference.

CfB: Manufacturing facilities are where the red flags have been raised, but the biosimilar processes have to be maintained very, very carefully all the way through, even for the originator products. Could this be a monitoring issue, as well as a setup issue, for biosimilar development?

Bennett: Yeah, there's no doubt. But I will say, for instance, Rituxan has a long history of drift over time. So, the Rituxan that they use today is different than the Rituxan used before. And there have been periods of time where the drift is so large that they had to have a reset. And that happens with production in particular. So, people have talked about that, and many people say there's no difference between a biosimilar and a biologic reference drug. But what you see today is not what you will see tomorrow. I've never seen a complete response letter like this for any of the major reference drugs at all. I think that when the reference drug comes on the market, you're pretty clear that the GMP [good manufacturing practice] is up to speed.