ダークホースのガイダンス - 次なる展開

A conversation with Jacob Staudhammer (DHC Senior Consultant) and Christina Fuentes, Ph.D. (DHC Consultant)

Q:
Hi, you two! So, before we jump into a discussion of next ideas, we should take a moment to look in the rearview mirror and address the most recent Dark Horse guidance news: the Proposed Draft Guidance for FDA Consideration on the topic of Testing of Adeno-Associated Viral (AAV) Vector-Based Human Gene Therapy Products for Empty Capsids During Product Manufacture. DHC submitted this to both the public and the FDA last month, on May 15^th and it really seemed to resonate. Why do you think the reception was as strong as it was?
A:
CHRISTINA: There’s been a much deeper focus on AAV product quality in the field recently and many advances in product characterization techniques. Everyone is aiming for the dual goals of “safe” and “effective,” so a potential safety concern will tend to, appropriately, jump to the forefront of the conversation. That’s why we felt strongly that we should do anything we could to help reasonable guidance expectations proceed apace. Our guidance essentially coincided with an overlap between high need and high interest areas.

Empty capsids in AAV products have been a hot topic of discussion in the field, especially following FDA’s AdComm (held September 2021) on toxicity risks associated with AAV GT products. All of us who worked on this draft guidance brought our collective experiences and knowledge to continue to move the discussion on this topic by providing a proposed roadmap for sponsors around measuring empty capsids in AAV products throughout product development lifecycle. It allowed us to contribute meaningfully to the field’s discourse on AAV safety by giving FDA a draft of a draft, as it were.

You can find details on the thought process that kicked it off in this Q&A with Don Fink and may also enjoy looking back at a round-table discussion from last September between CEO Anthony Davies, COO Katy Spink, and Don on the topic of AAV safety and paths forward. (Clearly, we eventually decided to take our own action regarding the “actionable advice” section of that discussion!)

Don Fink was central to this document…he, of course, is extensively familiar with procedures at FDA/CBER, having worked there for his entire original career, before coming over to become DHC’s Master Practice Expert, Regulatory in January of 2021.

Q:
You both worked on that document along with Don, right?
A:
JACOB: Yes. Don and Christina were the two leads on this project, but we had many active collaborators. Many of my fellow Horses collaborated and reviewed, and we also worked with both collaborators and reviewers outside the consultancy, such as Nicole K. Paulk, Ph.D., AAV Gene Therapy Professor at UCSF. We wanted to ensure this was a collaborative effort as we aimed to put forth guidance that would be applicable to all AAV GT products, regardless of the manufacturing platform chosen.

Q:
And then we offered an opportunity to ask questions or comment on the draft guidance, correct?
A:
JACOB: Yes. We set up a dedicated email (aavguidancedoc@darkhorseconsultinggroup.com) on May 18^th of this year and publicized it on social media through both Twitter and LinkedIn. We wanted to ensure that anyone who had thoughts they wanted to share on this doc had an avenue for reaching out that wouldn’t detract from the client work in our own email boxes.
CHRISTINA: We’ve been actively monitoring that email box since then to make sure we were diligent about answering questions. We decided to hold it open for 60 days, borrowing from FDA’s standard timeframe for a review period. Please be aware, though, that this is NOT an avenue through which to hold a direct conversation with the FDA! It’s just convenient way to make sure we don’t miss out on anyone who wants to share their thoughts. If you want to write to that address, we’ll be leaving it open until July 18^th.

Q:
What happens after that?
A:
CHRISTINA: We’ll be shutting that particular email down at that time and will have an auto-response letting anyone who writes it know that it’s closed. Of course, we can always be contacted by many other avenues, including through our contactus@ email with general questions or via our consult form to request consulting support. We’re also opening up a new email address, thoughtleadership@darkhorseconsultinggroup.com. Read on for more about that at the end of this piece!

Q:
OK, great, thanks. Now that we’ve revisited that original guidance, let’s move on to the question that always pops up any time anything exciting happens. Before the smoke even clears we usually start to hear folks asking, “what’s next?” So, not to pressure you…but, what next? And when might we expect to see whatever that may be?
A:
CHRISTINA: Luckily, we’ve been thinking about this also! Empty AAV capsids are only a piece of the puzzle as AAV capsid composition can be comprised of various product-related impurities which may impact overall safety and efficacy.
JACOB: Yes, and it would be nice to also talk specifics about product characterization for AAV gene therapy. It’s well-known in the industry that product-related impurities are a central point of focus: something everyone should be aware of and considering. We’re thinking that a next step would be to provide more context into what manufacturing sponsors could look at to help focus the spotlight on some areas that may impact the product safety and efficacy. There are product-related impurities that are hard to quantify because they can be nearly invisible/undetectable using just the standard techniques. We want to build that awareness and help sponsors understand why looking at those in detail is important to create a safer and more effective product. Our aim is to provide additional guidance to sponsors that would help give a sense of a framework for characterizing their AAV product to ensure that they’re starting with their best foot forward in the earliest stages of AAV product development. It’s important also to make recommendations to assist in future-proofing current programs for the direction that that we anticipate product testing requirements heading.

Q:
So, would you be thinking of considering each type of impurity separately or combining them?
Q:
JACOB: I don’t think we’ve made that decision yet, about whether or how to separate these out. There are a few categories of AAV variants that were omitted from the gene therapy CMC guidance altogether, so we’re definitely interested in starting with those. Basically, we’ll be discussing why and how to most effectively characterize those.

Q:
And what are those categories?
A:
JACOB: First off, to give some context, we need to discuss the composition of an AAV particle. I like to think of a viral vector as a space shuttle; where the capsid protein shell is the ship and the encapsulated transgene DNA is the payload and crew. If you were sending a ship on a mission, you’d want to make sure that the ship hadn’t been damaged in the assembly or assembled incorrectly, and that none of the payload or crew was forgotten, damaged, or mixed up. The categories in this guidance consist of characterizing variants in the AAV-version of the ship and cargo assembly process: capsid protein variants and transgene DNA sequence variants.
CHRISTINA: Part of the challenge is that any draft guidance, regardless of how wide-reaching it can be, is, by definition, only one piece of the puzzle. Jacob and I (and several of our colleagues) are in the middle of ongoing discussions identifying what constitutes the most important considerations from our perspectives in the industry. That expertise is what will help us narrow down which topics we ultimately address and in what order.
JACOB: The field as a whole doesn’t currently have a consistent shared understanding of all the product quality attribute limits to set and how to set them to enable highly scalable and effective AAV products. What we do have, I’d say, is an understanding of what attributes might be important to product safety and efficacy, and the field’s increasing emphasis on deep characterization and data collection to understand why some products work better and appear to be safer than others. We see a great deal of this in work with our clients and obviously there’s also a very active conversation throughout the gene therapy industry.

Q:
And what would be your best guess on a timeframe by which we might see more about this?
A:
CHRISTINA: Hmm, six months from now might be a reasonably optimistic timeframe? Jacob, what are your thoughts?
JACOB: I agree. It’s unlikely to take more than a year but we definitely need to allocate enough time to really work through the ramifications of anything we want to propose.

Q:
OK, so if all goes well, we might be able to share more by Phacilitate’s Advanced Therapies Week in Miami in January?
A:
JACOB: That’s a great goal, yes.

Q:
To jump back to those two categories we were talking about…would you please further define the capsid variants and “post-translational modification” concept for us?
A:
CHRISTINA: Sure. So, AAV is made up of proteins and DNA. The capsid proteins (approximately 60 per virus, to be specific) form a shell, which encapsulate the DNA and can themselves be modified after they have been made by the production cells in the manufacturing process; these are called post-translational modifications. The capsid proteins are decorated with various types of modifications, and those can have a different interaction with binding and entry or immune-responsive proteins. Recent data suggests that the type of manufacturing process and even the conditions within it can impact the modifications on the capsid surface which can, in turn, impact how that virus infects the target cell and interacts with the host’s immune response. That might have an impact on AAV safety, efficacy, and durability. None of these elements have a clear consensus yet, but the field has thus far identified that how you manufacture your virus means something for the final product quality attributes. The next step before we can fully understand the differences and their ramifications on clinical efficacy is to identify the attributes that are important and agree on how to consistently measure them: that’s where the field is at the moment.

Q:
Would you please further define the other category of “sequence variants” for us?
A:
JACOB: It’s important to note that when we discuss what’s inside the capsid (the cargo) it’s common to discuss “full” and “empty” (where the capsid either contains the transgene sequence or it doesn’t), as if empty/full were a binary condition, but that’s not even close to the case. Ideally, every capsid in the drug product would be perfectly filled with one complete copy of the transgene sequence, but in reality, this is not the case. The spectrum is huge and there are so many possibilities in between “completely full” and “completely empty.”
Even for particles that appear full, you could find yourself with sequence mutations and deletions, inactivated sequences, or the incorrect sequence all together. So, part of the goal is to help paint that picture of what the various conditions might be. What are the potential outcome scenarios, and what should sponsors look for?

And then understanding what the impact these differences have (or don’t have) on clinical outcomes—and how to control them—are the next steps, and the field is still in its infancy when it comes to understanding this. All of this is constantly evolving, but we want to urge the field to collect this data and to collectively paint the picture of how these attributes impact AAV safety and efficacy.

Q:
And what does that mean when trying to identify guiderails and ranges of product impurities?
A:
CHRISTINA: Simply put, there’s no easy cut-off. You can’t just say, here’s a hurdle and above that number, your product will be perfect! It’s more about understanding the depth of characterization currently feasible for your specific product. AAV products may have different impurity profiles based on serotype, manufacturing processes, and even storage conditions, which is why a deeper characterization of each product might lead to improved quality products, and then more thoroughly understanding the causes of product heterogeneity and its effect on clinical outcomes. The more we can understand about these products, the more likely it will be that the industry will be able to consistently deliver safe and highly efficacious AAV products to patients.

Remember, though, these are early days of the discussion internally about the next steps. This was our first time offering this type of support (unsolicited proposed draft guidance to the FDA) and I was heartened that it did seem to be an appropriate amount and type of information at the right time. Our intent continues to be to provide the maximum amount of neutral, reasonable assistance we can to the field in the hopes of optimizing timetables and expectations based on current knowledge. This breadth and depth of direct experience available is such a good fit with our consulting practice.

Q:
So are you saying that we don’t exactly have this guidance thing down to an exact science?
(Pun totally intended.)
A:
CHRISTINA & JACOB: {collective facepalm}

Q:
Is there anything that a DHC guidance doc is definitely NOT designed to do?
A:
JACOB: No guidance doc that we provide will ever be something like a ‘Consumer Reports’ list of approved products or methodologies. The intention for something public like this will always be to make it broad enough to successfully apply our knowledge to a constantly evolving field and across a wide range of use cases. Certain instruments or systems are occasionally so specifically relevant that they require a direct mention, but that’s a rare instance. It absolutely is possible to provide usable regulatory guidance without telling any manufacturing partner what to do and how to do it. (We are more specific and prescriptive in our advice to our individual clients, of course!) We’re not trying to recommend an explicit course of action but are instead creating a more thorough understanding of elements of the process that maximize the possibility of creating a safe and effective product…and what the ranges may be that can get us there.

Q:
Dark Horseはクライアントの機密保持を極めて重視しています。議論すべき内容であっても、クライアントが特定されてしまう可能性があるために議論できないといったケースは起こり得るのでしょうか。つまり、私たちは業界に関する膨大な知識を蓄積していますが、それが何らかの形でアドバイスを提供する際の妨げになることはないのでしょうか。
A:
JACOB: そうしたケースが起こり得る可能性は確かにありますが、私たちはそのような状況には立ち入りません。私たちが提起する質問や掘り下げるトピックはすべて、特定のクライアントの経験に基づくものではなく、包括的かつ集団的な経験に基づいたものとなります。これにより、クライアントの機密保持という私たちの責務を損なうような事態は確実に回避されます。包括的なトピックはいくらでも存在しますから。

Q:
Dark Horseのガイダンスに対する全体的なアプローチをどのように捉えていますか。5月に見られたようなFDAへのガイダンス草案の推奨といった形であれ、あるいは他の形式による「ガイダンス」であれ、その考えをお聞かせください。
A:
CHRISTINA: 私の考えでは、この急速に変化する分野においてクライアントをいかに最善の方法でサポートできるか、という点に尽きます。なぜなら、私たちが提供する「ガイダンス」は、FDAへのガイダンス草案という形式だけにとどまらないからです。もちろん、クライアントへの直接的なコンサルティングが第一に挙げられますが、それ以外にも、学会への参加、出版物やインタビュー、ARMやASGCTといった業界団体への参画など、この分野に知識をもたらすあらゆる活動がそれに当たります。例えば、まさに今日（2022年6月29日）、 プリンシパルのKevin Whittleseyが 「Comparability During the Various Stages of Development ARM A-Gene Virtual Workshop」のパネルディスカッションに登壇しました。私たちは日常的に、さまざまな方法でアイデアや知識を共有しています。
JACOB: 関連して重要なのは、私たちのすべての活動が、この分野を前進させるための貢献であるという点です。これらすべての治療法の究極の目標は、それを必要とする患者さんに一日も早く届けることです。DHCのような場所には、異なる背景や経験を持つ専門家が集まっており、私たち全員にとって非常に大きな価値があります。私たちは、ガイダンスを提供するための幅広い手段を組織として備えているのです。
CHRISTINA: 全く同感です。学会でのトレーニングやプレゼンテーション、パネルディスカッションへの登壇、ARMやASGCTの委員会への参加、論文の執筆、Quarter Horseやソーシャルメディアを通じた業界トレンドに関する意見の発信、あるいはガイダンス文書の作成など、その形は様々です。これらすべての活動は、「クライアントをサポートし、業界を発展させ、そして患者さんへの治療という最終目標を決して忘れない」という共通の目的を目指しています。

Q:
もしDark Horseから具体的なガイダンスを受けたいと考えた場合、どのように依頼すればよいのでしょうか。
A:
JACOB: 先ほどお話しした通り、一般的なご質問については、どなたでも contactus@のメールアドレス までお問い合わせいただけます。 コンサルティングフォームより コンサルティングサポートを依頼する。
クリスティーナ：新しいメールアドレスも開設しました。 thoughtleadership@darkhorseconsultinggroup.com。FDAへの新しいガイダンス案の提案や、上記で挙げたその他の方法など、ソートリーダーシップに関するご提案がございましたら、こちらまでお送りください。必ず確認させていただきます。

‍