In preparing this article that will discuss the future role of single use (disposable) bio reactors, I was somewhat perplexed by the more fundamental question of whether the biopharmaceutical industry really needs any increased manufacturing capacity, or is there simply such a glut of traditional manufacturing capacity that only a very small amount of additional capacity will be added in the foreseeable future? If indeed very little new capacity is to be added, this makes the commercial role of any technology, including single use, a moot question. After pondering the question, I rapidly have come to the conclusion that the role of single use bioreactors in supplying commercial quantities of monoclonal antibodies in the foreseeable future can be answered with a single word. That word is negligible.
First, let me describe the current state of the biopharmaceutical industry as it relates to mammalian cell derived monoclonal antibodies (MABs) and Fc Fusion Proteins. I draw upon a very informative article that was recently published by Brian Kelly a coworker here at Genentech. Kelly’s article titled “Industrialization of mAb production technology – The bioprocessing industry at a crossroads [1] provides a detailed analysis of the extremely low capacity utilization presently enjoyed by the global biopharmaceutical industry. Kelley states “The rapid growth in product demand for mAbs triggered parallel efforts to increase production capacity through construction of large bulk manufacturing plants as well as the improvement in cell culture processes to raise product titers. This combination has led to an excess of manufacturing capacity, and together with improvements in conventional purification technologies, promises nearly unlimited production capacity in the foreseeable future.” I had a conversation recently with Brian Kelley and we both agree that the likelihood of single use systems playing any major role in commercial mAb production is extremely low if not bordering on zero. There simply is too much production capacity for mammalian cell derived mAbs. Kelly estimates the total global GMP and fully licensed fermentor volumes that will be on line in 2013 at four million liters. At a moderately low titer of 2 grams per liter this yields a global capacity of 120 tons a year of mAbs. Of the four million liters of fermentor volume three quarters will be owned and operated by product companies, and one quarter by contract manufacturing organizations. Kelley estimates that the global demand for therapeutic mAbs and Fc-fusion proteins for the current calendar year of 2009 totals a mere 7 tons. The industry is simply many times over built in production capacity.
This may lead one to think that single use bio reactors to quote Shakespeare once again are “much ado about nothing”. Certainly from a global mAB production perspective single use systems will not have any inconsequential purpose. However, their benefit to the biopharmaceutical industry will arise in the clinical side of the industry. Production of tox study and clinical trial material will benefit greatly from the improved economics, speed, and flexibility afforded by single use fermentors over their traditional fixed system stainless steel counterparts.
Now let’s analyze the three traditional steps that comprise the fermentation process for the production of a monoclonal antibody biopharmaceutical drug. These three steps are as follows: Seed Train; Inoculum Train; and Production Culture. Certainly many commercial products are manufactured in processes that already use single use systems for the seed and inoculum steps of the fermentation process. It has been reported that single use bioreactors as large as 2,000 liters are in use for production culture. For the purposes of this discussion I will assume that 2,000 liter single use bioreactors are readily available for the production culture step and that the typical titer is 2 grams per liter with an overall recovery of 80% is achieved. A single successful run of such a facility will yield just over 3,000 grams of monoclonal antibody. Let’s assume a typical clinical dosing is 200 mg, therefore the single run will yield 16,000 doses for the clinical trial or tox study. Even during a large Phase 3 blinded study of a 1,000 patients with 500 patients given the active drug implies that each patient can be given as many as 32 doses of the trial drug. Not too many phased trials of biopharmaceutical drugs require this many doses, and hence we can conclude that even for the largest of clinical trials a single successful run at the 2,000 liter scale of production culture should yield ample drug to conduct the phased trial. Certainly for smaller companies with only one or two drugs in clinical trial and even for larger companies with many candidate drugs in clinical trial a single use facility makes economic, as well as operational sense.
Brian Kelley has estimated that at world scale production with a titer of 5 grams per liter a typical mAb may have a unit cost of good sold (COGS) that ranges between $50 and $100 per gram. For our analysis let us assume when the drug is produced in a single use system at our assumed reduced scale our unit costs are triple the high end of the world scale cost of goods. This yields a cost per gram of biopharmaceutical drug of $300. The single production run yielding the 3,200 grams will therefore cost the drug company in the order of $960,000. Let’s assume that the manufacturer will have to have six such runs over the course of the tox studies and phased clinical trials and therefore that some $5.8 million will be spent on manufacturing the active drug substance for these phased clinical trials. Even at small scale the GMP filling of the trial drug should not exceed the manufacturing cost of the bulk drug substance, hence the total cost to supply the active drug for all the phases of the trial should not exceed $11.5 million. When compared with the other costs of research and development of the biopharmaceutical drug and of conducting the clinical trial, the costs of manufacturing the trial drug are minor. Start up companies as well as large established biopharmaceutical drug companies will likely be willing to engage third party manufacturers who can rapidly execute the production of the drugs. The contract manufacturers will also deploy flexible systems that might include the current state-of-the art single use fermentation suites.
I will now attempt to estimate how many new mAbs may make it to market over the next decade and what the average demand for these mAbs might be. Brian Kelly has provided data that show that there are 15 major marketed mAbs presently on the market. Four of the mAbs have a yearly sales volume of approximately one ton each and the 15 mAbs on average have a sales volume of 200 kilograms each. My guess is that 30 new mAbs will be on market by 2020 and that each of these will only average 150 kg as it is unlikely that there will be four additional new mAbs on market that average a ton of sales. Perhaps 5 or 6 of these new mAbs will be produced in factories deploying single use systems and these five or six may only average 50 kilograms of yearly sales volume. With these levels of sales volumes and with yields of 3 kilograms per run in a 2,000 liter scale for production culture, the entire market for disposable production culture will only total 100 systems. This is an exceedingly small market. Manufacturers of the larger 2,000 liter scale single use fermentation systems should continue to focus their efforts on satisfying the clinical supply of biopharmaceutical drugs. There is also some possibility that in the next decade the manufacture of many vaccines will move from egg based systems to a single use fermentation system similar to those deployed in biopharmaceutical drug manufacture. In the next decade we also may witness a certain number of follow on biological (FOB) drugs making it to market. The manufacturers of these follow on biological drugs will likely adopt single use system in order to lower their capital intensity to enter the market and to minimize their cost of production.
I have previously written articles and made several presentations that the single use systems do improve the carbon footprint by as much as 25% and significantly improve the water footprint in mAb manufacturing. However while many companies will strive to improve their environmental footprint, they will not abandon the large and capital intensive factories that have already been constructed. Of course with 20 / 20 hindsight the industry could have preserved significant capital by not embarking on the construction of so much production capacity that is now idle or only partially used. If many of the thought leaders in the industry had spurred the development of single use systems to the point where they are now developed, we may have been able to deploy many more single use factories and we as an industry would have been able to better use that capital toward research and development of new molecules. Unfortunately the rush to install a massive amount of traditional stainless steel multiuse capacity began five to seven years ago when the single use systems were still extremely novel and the larger biopharmaceutical companies were simply not ready to experiment with the novel technology. Going forward there may well be a few small companies that emerge with several great biopharmaceutical drugs. These emerging companies will likely use Contract Manufacturing Organizations (CMOs) or will purchase and deploy lower capital intensity single use facilities. While the incumbent larger companies have already experienced a learning curve in their manufacturing of mAbs over the past decade, the world is pretty flat and the newcomers will have equivalent access to lowest cost methods of manufacturing. Speed to market is critical to the financial success of companies in many industries, including biopharmaceutical drugs. The Chemistry Manufacturing and Control (CMC) efforts typically are not the not to be”. While the market for single use larger fermentation systems may remain at the niche level, it will be an important niche that in the end helps assure that patients are supplied their mAbs or vaccines. As I have pointed out in several of my articles and industry presentations, the environmental footprint of single use systems is far lower when compared to traditional facilities, therefore the single use adopters will also be able to claim a greener approach in their operations and avoid some of the carbon taxes that are certain to be levied.
References
- Industrialization of mAb production technology – The bioprocessing industry at a crossroads Brian Kelley Genentech South San Francisco mAbs Landes Bioscience page volume 1 Issue 5 October 2009 443-452
Lindsay Leveen, “The Green Machine”, has worked with and consulted to major corporations in areas of energy deregulation, fuel cells, biotech, telecommunication, alternate fuels, power generation, transmission and distribution, as well as a variety of other process based technologies. Lindsay now works at Genentech as the Product Management Team leader for Lytics (Activase, TNKase and Cathflo) and was the Associate Director of Strategic Planning for this leading Biotech Company. Lindsay has lectured on sustainable development at leading universities and numerous industry conferences. He has written a book that is now used as a university text in Japan for energy policy and sustainable development. Lindsay has a knack to simplify and explain thermodynamics in everyday terms. He studied thermodynamics for his graduate thesis in Chemical Engineering at Iowa State University. The Green Machine blogs each week at www.greenexplored.com