The viral vector industry has spent years optimizing nucleases, filtration strategies, and downstream workflows to improve recovery and process consistency. But emerging findings from ACIB* suggest the larger issue may begin even earlier:
What if analytically acceptable samples no longer fully reflect what the downstream process is actually experiencing?
That question becomes significantly more interesting when viewed alongside real-world manufacturing observations shared by Lee Davies, Head of Process Development at OXB.
Key manufacturing insights in about 6 minutes
(Complete OXB presentation at the bottom of this article)
In this short video summary, Lee Davies discusses real-world downstream processing observations connected to chromatin behavior, process conditions, and nuclease treatment strategies in viral vector manufacturing.
Several of these manufacturing observations align closely with emerging ACIB findings around process-relevant chromatin and analytical context mismatch — the focus of an upcoming
Webinar: Your Viral Vector Analytics May Not Reflect Process Reality
Data presented by ACIB researchers together with ArcticZymes on June 11th.
Learn about- and register for live or on-demand webinar here
The webinar will examine why analytically acceptable samples can still create downstream problems, how analytical sample preparation may alter chromatin before measurement, and what the latest data now reveals about process-relevant chromatin in viral vector manufacturing.
What makes this especially interesting is that many of the operational consequences described by ACIB mirror real-world manufacturing observations already being seen by leading viral vector organizations.
That connection is worth paying attention to.
What ACIB Sees in Research, OXB Sees in Manufacturing
ACIB's latest findings (presenting in WEBINAR June 11th) provide a possible explanation for observations increasingly familiar to viral vector manufacturing teams: analytically acceptable samples can still create downstream challenges, and ACIB's research shows process-relevant chromatin may be part of the reason.
• Chromatin may behave differently during analytical preparation vs. real process conditions
• Analytical sample preparation can improve DNA accessibility before measurement
• Analytically acceptable samples can still create downstream challenges
• Residual chromatin often manifests operationally rather than analytically
• The assay may look clean while the process still behaves dirty
This may help explain why process-relevant chromatin fragmentation can have unexpectedly large downstream impact in viral vector manufacturing workflows.
For a deeper look into the scientific background, read our previous article summarizing a presentation by Sr. Scientist Patricia Pereira Aguilar from ACIB:
NucleaseTreatment: The Key to Efficient Chromatin Removal in Viral Vector Manufacturing
“Analytically acceptable” does not always mean operationally acceptable
One of the most important implications of ACIB’s findings is that downstream consequences may not appear directly in the assay itself. Instead, they may emerge later as:
• Filtration challenges
• Aggregation
• Purification inefficiency
• DSP variability
• Reduced recovery
• Process inconsistency
This creates an uncomfortable but important question:
whether process-relevant chromatin is influencing manufacturing performance in ways that analytics do not fully reveal.
That question will be discussed in the WEBINAR June 11th and becomes especially relevant when organizations observe unexpectedly large downstream improvements after switching to process-relevant chromatin digestion strategies.
Molecular biology workflows vs bioprocessing reality
Many chromatin analysis workflows originate from molecular biology applications where chromatin is intentionally broken apart before analysis to improve DNA accessibility and detectability.
Within molecular biology, that approach is logical.
But viral vector manufacturing operates under very different constraints.
The same conditions that open up chromatin analytically may not reflect how chromatin behaves under real manufacturing conditions.
This creates the possibility that:
• analytical preparation alters chromatin before measurement
• process-relevant chromatin remains partially hidden
• analytically acceptable samples still create downstream problems
The implications for process development and DSP are significant.
Full video presentation with Lee Davies, OXB
New Data provided in latest webinar save your spot here:
*ACIB: The Austrian Centre of Industrial Biotechnology is a non-profit international research centre in the field of industrial biotechnology. The centre develops sustainable, economically and technologically advanced processes for the biotech, pharmaceutical and chemical industries.
Video Transcipt
Over the years, we've evolved very much into where we are now, which is a pure play CDMO for the manufacturer of viral vectors for a range of clients around the world. We've set a few big milestones in the viral vector field.
We were the first company in the world to be granted approval by the FDA for the commercial manufacturer of lentiviral vectors. And during the COVID period we helped develop the manufacturing process for the UK COVID vaccine, and then went on to produce over a hundred billion vaccine doses for the UK during that time. So we have quite a healthy track record in the field of viral gene therapy.
But for lentiviruses it's particularly challenging, because the enzymes, the nucleases that are typically used commercially for the manufacturer don't tend to work very well in the conditions that we use for the lentiviruses.
The commercially available nucleases tend to have their peak activity, their preferred operating range at much, much higher PH's than we can use for the lentivirus. And similarly for the [00:01:00] salt concentration, they very much prefer very, very low salt concentrations.
And there's the added risk because we use such high concentrations of the nuclease, there is a finite risk that some of that nuclease finds itself in the product at the end. It's quite hard to remove 'cause there's high concentrations and that can be quite serious from a regulatory point of view. There are defined limits for all of these things.
So unlike the other commercially available nucleases, M-SAN operates very, very happily in the kind of pH range that we like to work in our process, and also in the salt concentrations, the physiological salt concentrations that our vectors prefer.
So instead of operating a process a long, long way away from the nucleases optimal activity, we are now bringing that optimal activity and targeting our process with it. So there's a real potential to improve our offering to the clients.
And I just want to say a little comment here about, um, the supply from ArcticZymes. We do a lot of work around the world with new and interesting ideas for how we can improve the [00:02:00] process. But our main driver is delivering a product or a process to the clients as fast as possible. We are looking for large scale GMP manufacturing. Those clients are looking for a product that can be produced at large scale in GMP very, very soon.
So we were very, very impressed to find out that M-SAN was going to be available at the quality that we require for GMP Manufacturing, but also at the kind of scale that we can do. These are very large volumes that are being used in these processes. Knowing that that was available gave us a lot of confidence going into the investigations that should something good come out of it, we could proceed with it very quickly.
So for every batch of virus we produce, we have some very strict assays and specifications for release. We have a number of different assays for DNA because it's such an important parameter. And when we looked at the results from the first batches in the M-SAN, we were incredibly impressed.
Compared to standard nuclease process] we had about a tenfold reduction [of total DNA], the plasmid DNA itself, so the, the DNA we add in the process about a threefold reduction.
And when we look at different size fragments of DNA, which is quite [00:03:00] important for, for what we are releasing, same impact on large bits of DNA, small bits of DNA.
So there, there, there was a real impact on the DNA and actually it was using the significantly lower amount of nuclease than we were using before as well.
But really importantly, in that final bar graph there were able to show there was minimal detection of the nuclease in the final product, which again, is really important for what we're doing.
And while Michael was visiting, I, I seem to remember saying, actually this looks too good. We don't need it to be this good. This is, this is incredible.
And so it has real benefit or real potential for some future processes we're looking at for the next phase of lentiviral vector discovery. But for the CAR T theory field at the moment, we don't need it to be quite this good.
And to cut a very long story short, what we found is we can completely remove that second nuclease step from the process.
We take it out completely.
We focus on that first step at the upfront bioreactor, and we get virtually equivalent performance in DNA removal, but some really, really big benefits for the process as a whole.[00:04:00]
So we're using less nuclease, which reduces the cost.
We've removed that second step, which has extra benefits.
That extra step comes with cost and time commitments in terms of manufacturing.
So we, we are saving money on that second step too. We've knocked three hours off our manufacturing time, which doesn't sound like very much, but as I said earlier, lentivirus is quite unstable. So three hours is quite significant. And also it makes organizing our manufacturing team far easier and fitting it into a, a decent working day.
It's simplified the manufacturing process overall quite significantly. And for anyone who works in manufacturing, simple is always best. And overall it is de-risked massively the potential for batch failure for, for each of our cost, both in terms of the performance of the manufacturing process itself, but also the DNA reduction and the presence of residual nuclease in the product.
This new process with a single nuclease step has been adopted by at least two of our clients. We performed our first GMP manufacturing batches last year. And as a result, this [00:05:00] new process is now our effective platform going forwards, offering for new clients.
So this is our starting point for all new clients who come to talk to us at OXB.
So in conclusion with M-SAN, it's been great for us.
It's worked really well. It's been a lovely project working with ArcticZymes.
It's reduced our costs, it's reduced the complexity and even reduced the processing time we need for the vector.
And the new process that we've brought in using M-SAN in conjunction with other factors has significantly increased the yields we're getting from the bioreactor, but also improved the quality of the vector we're doing at the same time.
Those two things don't always go together. So from, from an OXB point of view, it's been great.
We can maintain our profit margins whilst delivering an actual better product to the clients as we go along. And that keeps us at the forefront of these technologies around the world, which is, is something we're always striving to be.
But most importantly, we are actually being able to deliver more and more doses to patients around the world for all of these different life-changing therapies that we're helping to support.
So it, it's been a great success story. And with that, I'd be happy to answer [00:06:00] any questions.
During this, uh, process where you've been validating M-SAN, have you, have you been comparing it against some other new forms of enzymes compared to what we've used before and, and is there anything that you see even coming close to the performance of, of M-SAN?
We're always looking at nucleases. It is quite a hot topic at the moment. I can say I've not seen anything that can compare with M-SAN at the moment. It, it's a bit of a standalone product.
.
Learn more here about M-SAN HQ
Request a FREE SAMPLE of SAN HQ or M-SAN HQ here
