In honor of the inaugural issue of The Quarter Horse, Katy Spink, Dark Horse’s COO, sat down with Donald Fink, Master Practice Expert in Regulatory, to discuss the patterns of C> regulatory rejections/delays during the past 15 months.
Katy: We’ve seen so many examples, last year and continuing into the first quarter of this year*, of refusals to file, BLA rejections, and delays attributed to regulatory non-concurrence for late-stage cell and gene therapies. The pattern is certainly the elephant in the room in many industry conversations I’ve experienced. Given your extensive FDA history, can you shed some light on this for us? Are we seeing a change in standards? A crackdown on existing standards, whether in comparability, potency, or analytical characterization? Is it just that more and more therapies are reaching the later stages of development and being held to those standards?
Don: There are so many factors in play and we could keep peeling back new layers of this onion all day but the short answer is that it’s none of the above in particular with respect to the scenarios of filing refusals, complete responses or failure to hit the mark from a regulatory compliance perspective. I don’t believe we’re seeing any change in standards, nor do I believe the agency is applying those standards more stringently now than they have in the past (bearing in mind that stringency is related to the stage of the product development lifecycle). It is certainly true that with the maturation of the cell and gene therapy field, the product pipeline keeps getting larger with more and more therapies attempting to clear the bar set by regulatory expectation compliance. This, alone could appear to skew the balance between product approvals vs. difficulties encountered at the product licensing stage. I hear chatter that “the goalposts are being moved,” but I think it is far more accurate to say that the goalposts are simply coming into sharper and sharper focus.
Katy: Would you elaborate on that, please? Why would a sharper focus on the goals be causing these difficulties?
Don: Think of the difference in watching television in high def versus in standard definition… In HD, you can see the additional details you couldn’t before.
As time passes and the cell and gene therapy industry continues to mature, things come into sharper focus, particularly around information or data that are going to be necessary to verify statements and claims made about both safety and effectiveness. I think it’s just the logical consequence of cell and gene therapy product development progression. With both time and experience in conjunction with scientific advancement it becomes clearer and sharper what is really necessary.
Generalizations that passed muster in traditional biologics such as well-characterized recombinant therapeutic proteins and monoclonal antibodies simply cannot be directly applied to cell and gene therapies. I believe what we’re seeing now are the growing pains of an industry that is learning along with regulators what steps are really important to achieve the regulatory standards that have always been there.
Katy: What’s an example of one of those “generalizations?”
Don: Let’s take a look at the concept of “comparability.” Back in the day, demonstrating “comparability” was typically viewed as an exercise engaged in after licensing was achieved. For example, in the case of a therapeutic protein biosimilar, a comparative analytic assessment conducted to verify “comparability” relies on comparison to a reference product which often has already gone through the approval process. Comparability demonstration is also necessary post-approval when a significant change in manufacturing is implemented. In this scenario, you’ve got a reference standard and the reference standard is the approved product; you’ve also got an approved manufacturing process. It’s been vetted by the agency, you now have an approved product and a deep understanding of your process, and you’ve got all that information to compare your change to. Now, in the case of cell and gene therapies, post-license comparability standards appear to be making an appearance earlier during the pre-license clinical trial stage, which is much more challenging in some ways. You don’t really have the tools or the baseline data necessary to make rigorous comparability comparisons and oftentimes you don’t have enough accumulated manufacturing experience to have a very good reference material to compare to.
What impact will changes at this stage of manufacturing have? That’s a much bigger, more complex question and since that’s what sponsors need to prove, it manifests as a subtly different burden of proof. I sometimes find it useful to think of additional ways to describe this, such as “substantively bio-equivalent” or, as the PDMA refers to it, demonstrated “product consistency.” Put another way, I joke that “CMC” stands for “Can Manufacture Consistently.”
Katy: So, if proving comparability is more complex than it was in the early days of this industry, or relative to traditional biologics, how does a sponsor anticipate how to meet that need?
Don: The answer that is likely to prove accurate most often is one I’m sure plenty of people won’t like to hear, and that’s to set higher standards for the earlier stages of clinical trials. Two good examples here are GMP-grade raw materials and also potency assays. Early stages technically allow for research grade materials to be used with adequate qualification and do not require highly qualified potency assays. But, and here’s the kicker, when a manufacturer makes the easy, less-expensive, faster decision to use minimally acceptable standards, they put themselves in the position of a very heavy burden of comparability proof. There is the risk that making the more expensive choice in those early stages may backfire in the cases of therapies that don’t advance beyond Phase 1/2 , but where it has the possibility to pay off, and pay off big, is in the licensing stage. If I were manufacturing a cell or gene therapy, I’d place a very high premium on maintaining product consistency to the greatest extent possible from the earliest stage clinical trials through commercial manufacturing.
Katy: I’ve often felt frustrated when the industry interprets the concept that “the process is the product” very literally, to mean that there shouldn’t be any changes whatsoever to the manufacturing process during development, which feels like an impossibly high standard. Is what you’re saying essentially the opposite?
Don: Actually, I don’t think what we’re saying is at odds at all – more than there’s a happy medium in there somewhere – a sweet spot that the industry is attempting to reach. Trying to avoid any and all changes to the manufacturing process and needing to get it absolutely spot on from the start is a fool’s errand – to your point – especially in cell and gene. My point is at the other end of the spectrum, which is that just because some changes are unavoidable doesn’t mean we shouldn’t be wise about limiting those changes when possible, particularly when conducting clinical studies intended to provide primary evidence of efficacy. I suppose it comes down to planning. In a perfect world, I would plan as if there was going to be a premium on the number of changes that could be made, and then use appropriate comparability testing as agreed upon with the agency when the changes that are inevitable, unavoidable, and unpredictable crop up.
Katy: And when that happens, in your opinion what are some best practices for demonstrating product comparability?
Don: People do it in all kinds of ways. My favorite way is done by folks that lay it out in a prospective path forward, looking at: here’s what we’re going to change; here’s why we want to make this change; and here’s how we evaluate the impact of that change on our manufacturing process as well as its likelihood to cause a difference in the material that we’re going to get, once we’ve introduced that change. So it’s a layered approach, it’s a thought-based process, it’s prospective, and it tries to assess to the degree possible, does it impact the manufacturing process in a way that jeopardizes the ability to make product that looks to be sufficiently similar with respect to safety and efficacy/activity.
Katy: I’d referred to the “elephant in the room” earlier…and that actually ties in with a parable I often consider when it comes to product characterization in this field, that of the blind men and the elephant. The men encounter the elephant and since they all experience different parts of the beast, they think it’s something different than it is, right? The man who feels the leg thinks they’ve encountered a tree and the man who feels the tail thinks they’ve encountered a piece of rope, and so on. Only when they work together are they able to recognize the elephant for what it is.
Don: That’s an apt way to look at this process, certainly. Sponsors benefit from being able to zoom out and consider the whole. Too much focus on one element can make it hard to understand the overall complexity. And because this field is still young, judgements and understandings change over time. That’s one reason that investing early in your analytical methods, especially around potency, can be so important.
Katy: I think one of the big challenges for sponsors in this space is that there still, to this day, are often not clear ‘well trod paths’ to follow in terms of how to do things. Back in the early days of the field we used to be able to have quite a bit of informal back and forth with regulatory agencies during development. Has that changed?
Don: Yes, I think this is where the larger pipeline of products has made an impactful difference. Given the significant increase in product development activity, the current agency bandwidth just isn’t able to provide the same degree of interactive engagement as was possible back in the day when many fewer products were in the development pipeline. And because the luxury of that more frequent back and forth give and take isn’t there anymore, this can mean that sponsors are having to make more independent judgement calls on what to do along the way which could lead to substantive review issues arising at the time of the BLA submission. That would be a tidy explanation for the string of failed BLA filings, especially if we see them passing muster the second time around. It also speaks to the importance of checking one’s assumptions, getting another set of expert eyes on the project.
Katy: So in your 25+ years at the agency, you must have seen a lot of filings, with a pretty wide range of quality. What advice would you give our readers about what to do and what to avoid to get on their reviewer’s good side? Put another way, what are some of the best practices you’ve seen? Conversely, what are red flags you would recommend avoiding?
Don: I’ll have to work to keep this brief because you know I’ve seen a ton of these over the years! Regarding best practices, a few that particularly resonated were: a high level of clarity in data presentation through well-designed process flow diagrams summarizing important information such as analytics and raw materials at each stage of the process, well-organized tables, and obviously clear writing and organization of the filing itself. Personally, and remember that I’m not a decision maker anymore!, so this is truly just a preference…I’m a big fan of RAMM, combining quality practices with risk assessment awareness. I think looking at your filing through that lens can help you anticipate a lot of potential roadblocks. Quantitative assessments, even when there’s some subjectivity to it, help demonstrate in a layered fashion an excellent understanding of critical important elements.
As for practices to avoid: some sponsors rely too heavily on cross-references, which can make a document even more cumbersome for the reviewer to read. If you’re able to, please include the important information in your filing rather than relying too heavily on cross-references. Conversely, please don’t simply cut and paste huge chunks of your grants and publications. I’ve seen academic sponsors, in particular, rely too heavily on this. Remember your regulatory filing is a different document with a different audience and merits a more specific approach tailored to this particular audience and goal.
Katy: Thanks, Don, for this peek into the workings of a regulatory agency. I’ve found it immensely valuable and I’m sure our clients will agree.
Don: You’re welcome, Katy! I’m thrilled to be on board here – being able to see and learn the process from this side is invigorating and I’m loving being a part of it.
* notwithstanding the recent excellent news of the successful approval of Abecma
