Viral Clearance In Cell And Gene Therapy

A conversation with Daniel Hurwit, principal scientist, gene delivery process development, Bristol-Myers Squibb

Recently, BioPhorum wrote an article for Cell&Gene summarizing the information, consensus, and best practices shared in their new paper on viral clearance for cell and gene therapy. In this Q&A, we caught up with Daniel Hurwit, a BioPhorum member as well as principal scientist in gene delivery process development at Bristol-Myers Squibb, to further the conversation.

1. What are the biggest challenges associated with viral clearance for a CGT product and what are the best practices for solving or mitigating them that BioPhorum is proposing? How will the new best practices make a difference?

HURWIT: With respect to viral clearance (VC) in rAAV manufacturing, the biggest challenge remains regulatory uncertainty. Viral clearance for rAAV is technically very approachable but there’s a lack of clarity on exactly what is required and when. While there’s been some great progress among both regulators and sponsors in providing additional case studies and guidance, consensus is still lacking and there continues to be significant ambiguity. One of the big areas of uncertainty, for example, revolves around the topic of burden: If a first-in-human (FIH) rAAV process is expected to have zero viral burden, will agencies accept filings without VC data? For pivotal/commercial processes where new guidance calls for VC data for rAAV, how should burden be established when using cell lines and processes that produce no known virus-like particles and have no expected burden? If no burden is established, what will be expected and/or accepted for excess logs removal?

While these and many other questions remain, as a product of industry experts from a wide range of CGT companies, our BioPhorum paper serves as a good starting point for any company building their viral clearance strategy. While there remains room for interpretation – recognizing that not all rAAV programs are the same – our aim was to fill in any viral clearance knowledge gaps while, hopefully, driving some consensus around best practices.

In terms of best practices, inclusion of a virus retentive filter in your rAAV manufacturing processes is a great example of an easily implementable best practice, with the primary effect of in-building additional product quality and safety at minimal cost. Going a bit further, one could interpret the guidance, which encourages risk-based approaches, as suggesting considering only known/likely contaminants for FIH programs (e.g., process-added helper viruses or known cell-line contaminants if there are any, rather than a full generalized model virus panel covering theoretical contaminants) and doing so in abbreviated fashion (e.g., using non-validated scale-down models). This could also potentially mean a risk assessment-based allowance for straight-up forgoing FIH VC studies where no realistic burden is identified during a risk-assessment. Under this interpretation, adopting a no-FIH VC study where no burden is established as best practice could help simplify the process for therapies and cures to reach the clinic without adding any probable risk. In our paper, we also highlighted some straightforward best practices that received further helpful clarification in recent guidances, such as where and when to assess for replication competent AAV.

For enveloped viruses and products that are not amenable to virus clearance based on considerations of their physiochemical properties, the hurdles are both technical and regulatory in nature. While non-rAAV programs were not the focus of the paper, we did touch on a few basic best practices for an adventitious agent strategy for these more challenging molecules. In a nutshell, when viral clearance operations and studies are not readily applicable to a molecule, emphasizing and leveraging the other pillars of an adventitious agent strategy (e.g., prevention, detection/testing) are crucial. Additionally, as these molecules (e.g., lentiviral vectors) and their manufacturing process become increasingly mature and capable (e.g., using well-characterized modes of chromatography), reevaluating viral clearance strategy — and the feasibility of doing studies and achieving clearance within these modalities — will be important.

2. Are there any learnings from general biologics that we can leverage for CGT viral clearance?

HURWIT: Absolutely. The CGT field can borrow quite a bit from traditional biologics: We can use traditional biologics experience to predict the applicability – and in some cases even setpoints — of specific unit operations (detergent, low pH, and/or heat treatment, various modes of chromatography, virus retentive filtration, etc.) and we can leverage best practice approaches to burden identification, virus panel selection, study design, and scale-down model qualification. While specific virus panels, processes, processing conditions, and interpretations of phase-appropriateness will likely be different across modalities, in the simplest of terms, the extensive experience around viral clearance gained through traditional biologics development has laid the groundwork for reasonably straightforward adoption and adaptation in non-enveloped viral vector manufacturing processes.

3. How do you anticipate CGT viral clearance practices will develop in the future?

HURWIT: At a really high level, viral clearance in the CGT space will most certainly grow; that is, conducting studies will likely be more common. With increased regulatory guidance and the natural maturation of the viral vector space, viral clearance approaches and study designs will become more standardized. Like BioPhorum, hopefully more companies will share their strategies, data, and experiences interacting with health agencies, which will help accelerate a broader consensus on best practice; a rising tide lifts all boats.

I think we’ll also see the viral vector space driving innovation in how studies are designed and executed — for instance, multiplexing during viral clearance testing. Related to that, rAAV is likely an excellent candidate for platform-based approaches. For example, when transient systems are used in the context of a serotype-specific and consistently applied platform, CGT companies may either amass enough data or design studies that sufficiently rule out any VC capability impact of transgene, thus decreasing the burden of repeating studies for every new program once a sufficient body of platform knowledge has been generated. This might further be extendable across serotypes.

Overall, in the CGT space — while a generalization — timelines are often accelerated relative to traditional biologics and thus at any given time, material and time are typically in higher demand and shorter supply. In essence, there are many natural drivers toward ways to be more efficient with time and resources in CGT, and viral clearance is one arena where that might be the case.

4. What are the outstanding questions regarding viral clearance for the CGT space?

HURWIT: I think there are many questions, several of which I highlighted earlier. For pre-pivotal/pre-commercial rAAV programs, the big ones are still probably if/when/what caliber of studies are needed, how to establish burden, and to what extent risk assessments will be accepted. For pivotal/commercial there will likely be some uncertainty around model virus panel (how many and what viruses), excess logs removal, extent of data package required around different operations (e.g., it’s common to see fresh vs. reused resin studies, impact of process pauses during virus filtration, etc., in traditional biologics), and, again, the force of risk assessments.

I also think there’s an opportunity to accelerate the learning curve, for both sponsors and regulators, in rAAV based on industry experience with mAbs and other biologics. If and how to do that — for example, the applicability of leveraging platform data sets to supplant the need for fully replicated studies with each new asset — remains an open question. While these are some outstanding questions for rAAV (and there are surely many more not mentioned), none of them are showstoppers; time, experience, ongoing regulatory guidance, and ongoing collaboration will help clear them up.