Thursday, September 24, 2020

Can Process Intensification Reduce Biomanufacturing Costs?

Rentschler Biopharma, a global CDMO, claims to have sped up its upstream cell culture process by 35% by seeding a bioreactor with a greater number of cells.  The objective was to reduce the time needed for cells to grow in a typical fed batch process.   By increasing the seeding, more biomass was created, which in turn accelerated the proliferation of cells within the culture.

With the expectation of reduced time to peak cell production, monitoring the culture took a higher priority.  To support this, the company invested in spectroscopy devices that could automatically probe the bioreactor without having to take human measurements.  This also reduced the risk of contamination because there was no need to have a person pull out samples.

Not only does this process save time, it reduces the number of steps needed to get to the “N” bioreactor phase.  Only the N-1 bioreactor step is needed before putting the culture into production.  Therefore, this approach fits into the category of “N-1 Perfusion” many in the industry are investigating as a means to better cost efficiency.

The article provided little detail on how time savings translated into cost savings, and while it mentioned the process was cGMP compliant, little data was offered to compare quality metrics.  Another major item missing from the story was the cost of raw materials.  There’s clearly a financial trade-off here, and it would have been useful to quantify that.   Moreover, in the crowded CDMO market, Rentschler, which entered the US in 2019 when it acquired a manufacturing facility near Boston, has been trying to distinguish itself as an innovator in process intensification for both upstream and downstream manufacturing.  This requires it to make investments elsewhere, and the cost trade off might not pay off.

Rentschler believes that moderate intensification that reduces the time in the bioreactor from 16 days to 12 days minimizes costs the most.  Reducing time further drove up material costs to the point that they exceeded the financial benefits of a shorter scale up time.

In addition to materials savings, there are other approaches to intensification other CDMOs are implementing as they prepare to increase production to accommodate biosimilars.  Samsung Biologics, Repligen, and others have developed N-1 perfusion schemes aimed at achieving similar productivity gains.  

While specific approaches may be proprietary, potential cost savings could make N-1 perfusion more standardized across CDMOs as they aim to reduce costs.  However, it appears that better optimization of raw materials will be essential to making process intensification as cost efficient as it is time efficient. 

Wednesday, September 23, 2020

Scale Up vs. Scale Out

As the biotech industry continues to grow, and production volumes expand, biomanufacturers continue to look for new ways to reduce cost as they accommodate this demand.  One way is to continue to scale up through larger, 10,000 L+ stainless steel bioreactors that have the potential to reduce cost per gram through larger cultures.  An alternative is to scale out, using 2.000L and smaller single use bioreactors that can run in parallel, or in different geographical locations in order to reduce distribution costs to reach patients across the world.  

In practice, many biomanufacturers are looking at combinations of scale up and scale out that can meet various demand levels.  However, with CDMOs taking a larger share of the industry’s manufacturing volume, they need to sustain processes for far more drugs than firms that only manufacture  products developed internally.  In turn, they are looking at scale out designs that allow them to produce a greater range of products than traditional scale up systems.


Single use bioreactors reduce the risk of cross contamination from products sharing a facility, and that plus their flexibility makes them ideal for scaling out production.  Additionally, multiple lines create redundancy should something go wrong with any particular batch.  


The growth of scale out systems does not mean scale up will disappear.  Low titer products like antibodies are often sold in high doses and at high volumes, which still fits well with reusable, large bioreactors.  Similarly, scale out is seen by some as being linked to specific products, particularly autologous therapies like CAR-T.  Due to the sensitives of these therapeutics coming from and going back into the patient, scale out can preserve quality better than scale up in some cases.  However, for more traditional allogenic therapies, scale up can still offer not just lower cost, but provide greater consistency.


While sometimes presented as alternatives to each, it appears that both scale up and scale out will both have applications to serve for years to come.  The decision to use or another will ultimately come down to the type of biologic being produced, not an overarching philosophy about how to manufacture therapeutics.


Tuesday, September 22, 2020

Bioprocessing 4.0 Will Require More Than Analytics


Industry 4.0 is a global effort to bring the latest digital technologies to manufacturing facilities across industries.  It is intended to reduce costs, improve quality, and a host of other “good” things.  It grew out of European efforts in the early 2010s to update their factories, and is often referred to as “bioprocessing 4.0” within the biomanufacturing world.  However, biotech manufacturers have been fairly slow to implement many of its technologies, and its benefits remain a little unclear.


One area bioprocessing 4.0 is showing promise within biologics production is process analytics.  Collecting data during cycles, as opposed to just after batch completion, or only after steps within batch production, could return faster quality indicators.  However, unlike many industrial products, biomanufacturing needs new measurement tools, not just sensors and computers like other industries, due to the very sensitive nature of the product.   


Taking advantage of photonic products used to monitor and manage fiber optic networks in telecommunications networks, Raman spectroscopy is one technology that can measure product features using lasers and infrared wavelengths.  CHO cells 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 outside 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, the devices are not used as widely in biomanufacturing.


Sensors and spectrometers are needed on the hardware side, on the software side many consultants are promoting “AI” and analytic tools to help automate biomanufacturing.  While this data can be useful in places, it is more widely used, and easier to adopt, in less regulated industries.  Biomanufacturing process analytics can take advantage of additional data points, but development of Process Analytics Technology must be coordinated with the FDA. This is unlike algorithms that can find usage patterns among logs of website visitors, a common application for AI.


While bioprocessing/industry 4.0 could deliver some benefits for biomanufacturing, particularly on the hardware side in difficult-to-measure mammalian cell cultures, it seems like it’s still a long way from being widely accepted in this heavily regulated sector of the economy.

 

Sunday, September 20, 2020

Is Biotech Ready for Continuous Manufacturing?

Shifting from batch processes to continuous manufacturing has long been a goal for biologics manufacturers, especially as small molecule drug producers have increasingly shifted towards continuous processes.   But to date, these processes have largely been limited to clinical products among biologics, due to challenges scaling up and improved yields and productivity from batch processes.  However, biologics have a very short history without patent-protected pricing power, and the growing market for biosimilars promises to put new pressure on unit cost reductions.  As a result of the traditionally limited price pressure in the industry, some process innovation might have stagnated.  To quote Bill Kelly from Bioinformatics “Despite its high-technology reputation, the manufacture of therapeutic proteins is decades behind production systems for potato chips and cement in terms of continuous manufacturing”.

Continuous biomanufacturing is advancing, but not at the pace some had hoped for.  The first mAb made by a completely continuous process was developed last year in Australia, but only for a Phase 1 trial.  A recent GEN article discusses how manufacturing facilities for particular niches, such as exosomes, are starting to embrace continuous processing, but for large scale CDMOs and biomanufacturers, there are more than just technical challenges.  Sanofi, for example, built a continuous manufacturing facility last year outside Boston, but found that there were far fewer GMP compliant off-the-shelf components available than for a standard batch plant.  

Much of the development of continuous processes is split between upstream and downstream manufacturing.  Within upstream cycles, the industry is trying to take advantage of single use bioreactors, which are already used in many batch processes, but combining them with cell perfusion technologies which replace the waste during fermentation, cutting off the normal death cycle that follows stabilization.  While in downstream cycles, the focus is on continuous chromatography, which reduces the requirements for expensive resins used in standard chromatography.

Greater productivity in upstream batch processes has also greatly slowed development of continuous biomanufacturing.  Expression levels have increased from being measured in milligrams per liter to grams per liter.  This in turn has allowed manufacturers to get the same productivity from smaller and single use bioreactors they once got out of 10,000 L+ stainless steel systems.  While the term “batch” sounds dated, the industry has been able to innovate within its unit-by-unit processes.

Sanofi's built plant in Massachusetts embraces continuous manufacturing, but it doesn't focus on either upstream or downstream processing as many others have done, but instead on integrating the two.  Additionally, maintaining quality control and quality assurance is still a new area for continuous bioprocessing, and this required the company to make a significant investment in software and monitoring systems.  The company claims that this new facility produces 770 million data points a day.

While the implementation of continuous biomanufacturing is occurring slowly, it does not appear to be stopping.  However, Sanofi’s investments in IT indicate that a shift away from batch processes will require not just new perfusion or purification methods, but an entirely different approach to measuring and managing biologics production.


Saturday, September 19, 2020

Is Manufacturing Antibodies in Bacteria Realistic?

SwiftScale, a Bay Area startup that grew out of research at Cornell, is promoting a cell-free process to manufacturing antibodies.  This would get around the long lead times associated with the CHO cells that are commonly used to manufacture antibodies.

The business opportunity could be significant.  Monoclonal antibodies represent a large market, producing over $100 billion a year of annual revenue, greater than any other biotechnology therapeutic.  However, they’ve been developed for rare diseases and cancers which don’t have the incidence of COVID-19.  This will require innovation in their manufacturing processes to accelerate production from the standard 9-12 months.

Traditionally, antibodies are manufactured in mammalian CHO cells.  SwiftScale is proposing to manufacture antibodies out of bacterial E.coli cells, greatly reducing production times.   The rationale for using mammalian cells has been that antibodies require glycosylation (the process of sugars attaching to a protein) in order to recognize the epitopes attached to the antigens they are intended to fight.  This process typically occurs in the endoplasmic reticulum and golgi apparatus of a mammalian cell, and has thus far prevented wide scale manufacturing of antibodies from bacteria.


Glycosylation results from a post-translation modification in mammalian cell protein synthesis,  and factors into the long lead time to manufacture mAbs.  SwiftScale is proposing to create the glycosylation process outside the cell, and drastically reduce not just lead times, but the size of the manufacturing components.  They claim a test tube could be used where a 50,000 liter bioreactor is used today.  


The idea of developing antibodies out of bacteria is not new.  In 2007, a University of Texas study analyzed the prospects for doing this.  What’s changed is the market has grown nearly tenfold since then, and is poised to grow further with demand for COVID treatments.


In addition to glycosylation, bacterial mAbs will need to emulate the disulfide bonds created by mammalian cell engineered proteins, which is also essential for bonding with antigen epitopes.  While the research at Cornell provides some insight into SwiftScale’s process, some of the actual techniques the company is using to overcome bacteria’s antibody production challenges remain proprietary.  Nonetheless, the demand created by COVID vaccines could finally give this manufacturing technique the high volume application it’s been searching for.