Thursday, October 29, 2020

What's the Future for AAVs in Gene Therapy?

 AAVs are the most common vector for delivering in vivo gene therapies, but accommodating growing demand will require manufacturing innovation.

Source: FDA

Gene Replacement vs. Gene Editing

Since its advent in the 90s, gene therapy has held promise as a way to defeat genetic diseases. Unlike CRISPR which edits the DNA, gene therapy keeps the existing DNA sequence in place, but replaces it with a healthy copy of the gene grown in vivo or ex vivo. For in vivo gene therapies, this sequence is typically delivered through an AAV (Adeno-Associated Virus) vector. Some of the common in vivo gene therapies include treatments for hemophilia, muscular distrophy, and cystic fibrosis.

In total, the FDA has now approved nearly twenty gene and cell therapy products, with an emphasis on blood and hemophilia-related disorders. But while further research can address lingering safety concerns with AAVs, new manufacturing processes will be needed to scale up production of these viral vectors in order meet growing demand.

Viruses Transport DNA, but Don’t Make Proteins

Before getting into how they’re made, it’s important to look at why viruses are used to deliver “replacement” genes. Viruses can carry DNA, but they cannot produce proteins from that DNA like a traditional cell can through transcription and translation. They’re great at transporting DNA, but need your cells to turn that DNA into a protein. In some ways, they’re like freight trucks that can bring raw materials into a factory, but lack the factory’s ability to transform those raw materials into a finished product.

Viruses engineered to carry repaired DNA for gene therapy can also be manipulated not to replicate beyond the target cells. This prevents the virus from sickening the patient. But it still means you can develop immunity to the viral vector, which is why gene therapy is often used as a one time treatment.


Ex Vivo and In Vivo Gene Therapy Overview, Source: FDA


Why AAVs?

There are a few viral vectors that can be used to deliver gene therapy, but the AAV has become one of the most common because it can infect both dividing and non-dividing cells, and few people have been exposed to it so they don’t have immunity against it. Moreover, AAVs generally have low immunogenicity, that is to say they don’t produce a strong immune response from the body after insertion, which helps maintain their effectiveness as delivery mechanisms for new DNA.

There are some inherent challenges with AAVs, notably their DNA carrying payload is limited to 4,000 base pairs. Any sequence longer than that simply won’t fit. Longer strands can be accommodated by Adenoviral vectors, but those introduce an additional set of safety and quality challenges.

Are they Safe?

In addition to limited production volumes, AAVs and gene therapy still have safety concerns to overcome, particularly with high dosages. Two children died earlier this year in trials for a high dose AAV-delivered gene therapy for X-Linked Myotubular Myopathy (XLMTM). The particular capsid used for this trial, AAV8, has been used without any safety issues in 14 other trials, highlighting how sensitive AAV’s can be to shifts in their dosage and payloads.

Additional safety concerns have arisen around the potential to cause cancer observed in longitudinal studies of both dogs and humans. However, the data here is inconclusive. Nonetheless, the safety concerns obviously raise the risk for both gene therapy and its viral vectors, and for the time being have constrained their use in certain clinical trials.



Gene Therapy at the Cellular Level, source Genome.gov

Safety is the Top Priority, but Better Production Efficiency will be Essential to Meeting Future Demand

While additional research will be needed to address safety concerns, new manufacturing processes will be needed to ensure AAVs can be produced more efficiently and economically, which remains a challenge with existing manufacturing processes. This will be essential to meeting the growing demand for gene therapies.

Requiring substrates to which they can adhere, AAVs don’t scale up like mAbs (monoclonal antibodies) in a bioreactor. To grow they ultimately require a tremendous amount of plasmids, which both pushes up materials costs and creates potential quality risks. However, AAVs to date have faced limited pressure to scale up due to the low volumes and few approved gene therapies on the market. German drug developer Cevec claims it can deliver a cell line up front, similar to a master cell bank, from which AAVs can be drawn from, not unlike how mAb cell lines are stored. On top of that, they claim they can support standard 2000 L bioreactors, allowing a manufacturing scale not available with plasmid-intensive AAVs that struggle to sustain batches greater than 200 L.

Scaling Up Production

The main barrier to scaling up AAVs is putting their pieces together. There’s the “cap” or the process of building the capsid that encapsulates the virus, the “rep”, which allows the DNA to replicate, a helper virus needed to manage replication, and the payload, i.e. the nucleotides inside that actually deliver the therapy. By bundling it all together, Cevec believes they can overcome the volume limitations of existing approaches. But it remains unclear how they maintain quality and cost creating the cells. It would appear the upfront cost of doing this could still be significantly more expensive, made economical only through very high volumes that might be beyond current demand for existing gene therapies.

In contrast to Cevec’s approach, AGTC (Applied Genetics Technology Corporation) is proposing to accelerate scale up by using Herpes Simplex Viruses as “helpers” through a medium of BHK (Baby Hamster Kidney) cells. AGTC has been working on this approach for years, but like other AAV delivery methodologies, has been waiting for a gene therapy that could offer the volume to take advantage of it.

With just around twenty gene and cell therapy products approved by the FDA, AAVs primary use could still be in the clinic for another few years, allowing some time for more cost effective, high production growth techniques to develop. Either way, continued innovation in their manufacturing processes will be as important as innovation in the gene therapies they serve in order to serve a larger market.


Wednesday, October 28, 2020

COVID Vaccine Production Could Accelerate the Use of AI in Biomanufacturing

 Discussions about using machine learning in biomanufacturing have increased recently as COVID vaccine production gets closer to its anticipated scale up. Technology providers see a huge opportunity to bring AI products to a market that might have sat in development had there not been the immediate need to get a COVID vaccine out the door.

Illustration © Visual Generation | Dreamstime.com
                                                          

Different suppliers have different approaches. While some are looking to machine learning to detect patterns in treatments, others are looking at measuring tools needed to monitor biologic production. Either way, biomanufacturing and traditional pharma offer plenty of opportunities to reduce cost, to improve quality, or to increase speed. Moreover, many observers believe biomanufacturing processes hold the greatest near-term promise for AI within the larger biopharma industry outside of drug discovery. But machine learning in this field looks very different than it does in other industries. Not surprisingly, detecting the behavioral patterns of bacteria requires very different approaches than detecting the behavioral patterns of people.

Measuring Cell Cultures isn't like Measuring Website Visitors

More advanced analysis of cell cultures is now taking place now in academic research in order to increase the quality of these cultures that are used to produce biologic drugs. Computational biologists are applying algorithms to improve yields on E.coli by optimizing batch quantities without monitoring actual cells. This approach isn’t based on the predictive analytics that are often used in AI-based customer analysis in marketing and website optimization, but rather trial and error analysis on data from a “virtual” cell culture. Essentially, the algorithms provide rapid pattern recognition on activities in the culture that would take people much longer to assess. The next step though, is to move out of the virtual world and bring the techniques to real bioreactors.
With AI in biomanufacturing heading down a different path than the heavily software-based approaches used in marketing and website analytics, it is also heading to an environment where algorithms have to learn continuously from ongoing manufacturing processes, not just by sifting through massive databases. Measuring cell cultures also requires monitoring hardware that does not contaminate the dividing cells.

Hardware Plays a Crucial Role in Biomanufacturing Machine Learning

Taking advantage of photonic products used to monitor and manage fiber optic networks in telecommunications, Raman spectroscopy is one technology that can measure product features using lasers and infrared wavelengths. Chinese Hamster Ovary cells, which are commonly used to grow antibody therapies, emit biophotons during bioprocessing that can be measured by Raman spectrometers without interfering with production. While these devices add to the capex needed to buildout a facility, they can reduce opex through labor savings and more precise quality measurements during both upstream and downstream processing.
Beyond Raman spectroscopy, many other IoT (Internet of Things) sensors are built for higher volume manufacturing environments with less sensitivity. Chemical manufacturers and oil refineries use these devices to a large extent, but these same IoT sensors can’t be placed in a bioreactor without impacting the cell culture. Outside of basic environmental monitoring in the space surrounding the bioreactor, customizing industrial sensors for biomanufacturing would be very expensive for the device suppliers, and leave them with limited volumes over which to cover their fixed development costs. As a result, these types of devices are not used as widely in biomanufacturing.
Sensors and spectrometers are needed on the hardware side, on the software side there is growing use of AI and analytic tools to help automate biomanufacturing. While the additional data can be useful in places, it is still faster and easier to adopt in less regulated industries. Biomanufacturing process analytics can certainly take advantage of more frequently sampled data, but development of Process Analytics Technology must be coordinated with the FDA.
Dependent on learning through experience, and needing to conform to the needs of a regulated industry, biomanufacturing AI will likely see its growth accelerated by COVID vaccine scale up in the next 12 months, but perhaps not at the torrid pace some are hoping for. Nonetheless, there could be greater uses a few years down the road as biomanufacturing faces new unit cost, quality, and scale up challenges with the higher volumes brought about by biosimilars, cell and gene therapy, and even gene editing. COVID might just be a stop along the way to a deeper industry embrace of AI.


Monday, October 12, 2020

DNA Reads are Cheap, What About DNA Writes?



Thanks to large gains in compute power, the costs to read DNA have dropped exponentially (23andme, Ancestry, etc are the result of this), but the costs to write DNA have only dropped linearly. But there is a lot of work going on in DNA write technologies to address this.

Most synthetic DNA has been made the same way since the early 80s using a serial process known as Phosphoramidite Synthesis.  This method requires attaching each base pair to the existing strand one at a time, and it limits synthetic DNA strands to about 100 base pairs in total length before errors, time, and cost challenges set in. However, a series of companies funded over the last couple of years is seeking to change this, and improve the reliability and economics of writing DNA.


Enzymatic DNA is being proposed as a way to overcome the limits of traditional synthetic DNA construction.  By using an enzyme, in this case Terminal deoxynucleotidyl Transferase (TdT), the addition of new nucleotides can be done without engineering each additional attachment to an existing strand.


To date, enzymatic DNA hasn’t gone much further than traditional synthesized DNA, and has been running at a functional limit of 300 base pairs.  But a series of recently funded companies are looking to expand that limit and reduce the cost of synthetic from the 9-15 cent per base pair range it has been in the last few years.


Among these companies, DNA Script and Nuclera are developing Enzymatic DNA Printers that are planned to hit the market in the next 6-18 months.  Both are following the business strategy Illumina has set in DNA sequencing, by selling hardware to companies to labs.  DNA Script, based in France with a US headquarters near San Francisco raised $50 million in July.  

San Diego-based Molecular Assemblies raised a $12 million Series A last year and has partnered with protein engineering firm Codexis to license enzymatic DNA technology to other companies.  In a recent Nature Biotechnology article, the company referred to its strategy of providing the ink, not the printer.


Bay Area-based Ansa Biotech recently raised $8 million to become an enzymatic DNA service provider, similar to what publicly traded Twist Bioscience currently does with traditional synthetic DNA.  Twist itself is also using traditional synthesis to archive digital with DNA, which recently funded Kern Systems is looking to do as well with enzymatic DNA.


Enzymatic DNA will likely need to get to 500-1000 base pair strands to start to push costs down for synthetic DNA.  In turn, lower costs could stimulate demand for new applications, much as lower costs did for DNA reads and sequencing.  The emerging enzymatic DNA industry is structuring itself for that growth, and within a few years we could see strong businesses develop in writing DNA, not just reading it.


A Biotechnology Definition for the Non-Scientist

To a non-scientist, I would say biotechnology is three things: building physical goods with living organisms, modifying living organisms to improve their health or utility, or developing information products using biological data.   

Living organisms like yeast can ferment into beer, which is obviously an old practice.  But since the 1980s, living organisms have been used increasingly in drugs, which used to only be chemically produced pills.  A biotechnology drug like growth hormone is made out of living organisms, namely bacteria, while your bottle of Tylenol is all chemicals.

Living organisms can also have their genes modified to improve their health or utility, which is a second category of biotechnology based on the definition above.   This is the part of biotechnology that produces  "genetically-modified" crops.  It also includes genetically modified ethanol used to make cleaner fuels.      Gene modification therapies are still fairly experimental on humans, but there are FDA approved drugs now that allow for this.  There will likely be more in the future

The third category of biotechnology is information products based on biological data.  That biological data is typically DNA, of which you have 3 billion pairs in each of the 100 trillion cells in your body, so the amount of data you as a human being produce is exceptionally high.  Advances in computing technology have made it much cheaper to read this data, and has led to services like 23 and me, Ancestry, and Invitae which can read your DNA to trace your roots or test you for genetic diseases.    This is also the part of biotechnology that covers DNA testing  used to solve crimes and solve paternity cases.     

Because biotechnology can provide information about people more cheaply than it could even 10 years ago, and also treat many diseases more effectively than chemically-produced drugs, its use is likely to expand significantly over the next decade. 

Saturday, October 10, 2020

What is Biotechnology? A Biotechonomics Definition

Biotechnology is the use of living organisms to build physical items. While often associated with therapeutics, there are a wide range of products in agriculture, food, energy, plastics, even electronics where biotechnology components can serve as raw materials. 

Biotechnology can also be used to build information-based products, particularly in genomics, but these are often used as inputs to select therapies. 23 and me, Ancestry, personal DNA sequencing and genotyping are probably the most widespread information-based biotech products. 

Many well known biotechnology products are therapeutics. Genentech got approval for the first therapeutic using recombinant DNA technology in 1985 and since then biotech-based therapies such as Humira, Remicade, and Rituxan have generated more than $1 billion in annual revenue.

While fermentation has long been used to make foods, a new generation of biotech-based foods has come to market replacing animal-based dietary items. Most notably, Beyond Meat produces hamburger patties using a mixture of plant extracts and proteins.

Sometimes overlooked, electronics and digital hardware are a growing area for biotechnology. Zymergen, for example, is developing microbial-based films that can be used on printed circuit boards and ultimately produce thinner screens and more power efficient electronic components. Their work builds on the success of OLEDs (Organic Light Emitting Diodes) that are now commonly used in flat panel TVs and iPhones. Another highly advanced use of biotechnology in electronic hardware is Twist Bioscience’s DNA storage platform, which can be used to archive digital data, by coding digital bits into DNA bases adenine, thymine, cytosine, or guanine. Netflix recently announced it is using this technology to back up one of its Biohacker series. 

With biotechnology permeating such a wide variety of applications now, it is poised to move well beyond therapeutics, and could well become a dominant raw material for all sorts of goods in the future.