September 15, 2011 (Vol. 31, No. 16)
Current CHO Cell, E.coli, and Yeast Platforms Creating a Healthy Dose of Idle Capacity
Continued improvements in recombinant protein expression yields have helped ease production capacity constraints, leading to an apparently healthy and manageable increase in idle capacity. According to BioPlan Associate’s 8th annual survey of biopharmaceutical manufacturers, the overall average yield reported for commercial monoclonal antibody this year was 2.18 g/L—up from 1.94 g/L in 2009 (Figure 1). It was not that long ago that mammalian cell culture commercial production yields were lucky to achieve 1 g/L.
The next generation of commercial products will have an even greater average yield. In fact, for late-stage clinical manufacture, the overall average this year is 2.68 g/L, up from 1.96 g/L in 2008. While protein yields over 30 grams/liter are now being reported by expression system developers and early adopters, these are the exception and yields for commercial manufacture in the 5–10 grams/L range are more likely to be common in coming years.
Calculating the growth in titers (CAGR % titer change) from 2008–2011, commercial yields increased annually by 3.7%, while late-stage clinical manufacture yields increased 11.0%. The portions reporting the highest yields, >5 g/L, have remained rather constant in recent years. This suggests that either only a few are adopting newer, high-producing expression systems or only a few are actually attaining these levels with their own facilities and products.
With single-use/disposable bioreactors and other equipment now dominating upstream noncommercial manufacture, the recent high growth in late-stage titers has been attained using this relatively new class of equipment. As single-use equipment graduates to mainstream commercial product manufacture, these increases will continue to carry over to commercial manufacture.
Industry Yields Increasing
The largest portion of survey respondents (59.7%) now report mammalian cell culture commercial production of monoclonal antibodies (mAbs) at between 0.5 and 2 g/L, with increases in this group observed in the past three years.
Another 26.6% report attaining commercial titers between 2–3 g/L, 20.5% attaining 3–5 g/L, and 8.5% attaining >5 g/L. Nearly half (46%) reported clinical-scale manufacture between 0.5 and 2 g/L, with 30.1% reporting 2–3 g/L, 25.7% reporting 3–5 g/L, and 11.5% reporting ≥5 g/L. In general, higher expression levels are now reported for late-stage compared to commercial manufacturing. This is not unexpected since process improvements generally take place at earlier development stages.
Increasing expression system yields enables use of smaller bioreactors and facilities. Bioprocessing system (bioreactor) sizes for large-scale manufacture are likely to decrease or remain steady in coming years, with companies increasingly adopting single-use 1,000–2,000 L bioreactors for late-stage and commercial manufacturing.
Particularly as single-use bioreactors begin to displace stainless steel for commercial manufacture, those needing higher capacity may begin to use multiple single-use bioreactors operating in parallel to achieve production levels and economies-of-scale now largely restricted to those operating ≥10,000 liter bioreactors.
This is especially true for recombinant mAbs with a large number of new drugs and biosimilar products in development; mAbs have high repeated dosing requirements and large amounts of protein, e.g., 100s of kilograms/year, are needed for a blockbuster product. Also, advances in bioreactor technology, particularly in smaller size perfusion systems, will likely result in overall reduction in bioreactor requirements.
Implications of Improvements
The trend in increasing protein expression yields has had broad ramifications throughout biomanufacturing. From lowering construction requirements, to increasing bottlenecks at the downstream end of production, the effects of upstream advances and the availability of more varied expression systems have created options and problems not previously seen.
Looking at biological products in the market and in development, the industry still largely restricts its basic expression systems to the three systems it has been using for decades—CHO cells for mammalian cell culture, E. coli for bacterial expression, and yeasts (mostly Saccharomyces).
With the relatively slow uptake of newer expression systems, many of which offer significantly higher yields and other improvements, the increases in yields now being seen are primarily the result of incremental improvements in these three key systems, including new and optimized vectors, promoters, and cell lines. This is also reflected in the increasing number of companies offering CHO and other common expression system-based bioprocess optimization services.
Increases in expression yields obviously result in cost savings and improved biomanufacturing efficiency. This has also enabled manufacturers to do more with less, and in many cases do it even better. With expression yields expected to continue to increase, the industry will continue to reap these benefits. Increases in yields also affect facility planning and financing. More importantly, increasing yields and the resultant excess capacity provide needed insurance against costly supply disruptions.
Figure 2 from the survey data shows current capacity utilization rates. There has been a significant reduction in the utilization rates, i.e., increased idle capacity, for mammalian, microbial, and yeast culture since 2003. This year, when asked about average bioprocessing system usage as a percent of operating capacity, those working with mammalian cell reported 61% utilization, bacterial systems users reported 53.6%.
This indicates that most bioprocessing systems, particularly at larger scale (where most of the capacity is), are now idle much of the time. Capacity utilization rates for lesser-used expression systems were similar, e.g., plant cells (51.3%) and insect cells (59.3%). Overall, capacity utilization rates have stabilized in recent years, with utilization rates for mammalian cell culture steady around 61%–63% for the past five years. The distribution of available mammalian cell culture capacity has changed little in recent years, e.g., with about 60% of respondents reporting total capacity less than 10,000 L.
Is Current Idle Capacity an Asset?
For all expression systems, there continues to be significant underutilized capacity. Survey data shows that utilization has stabilized at what seems to be a fairly healthy and, perhaps, even optimal level. While downstream purification is now noted most often in the study as the primary choke point for larger-scale manufacture, upstream manufacturing remains the primary area affecting industry capacity and production.
The decrease in industry capacity utilization is not the result of an industry slow-down or contraction. Rather, despite the world’s economic problems, in the past decade and even recent years, biopharmaceutical revenue (and manufacturing volume) has increased steadily, at an annual rate commonly cited as 15%–20%.
In the early 2000s, there were concerns that the biopharmaceutical industry would soon face a severe lack of capacity. Incremental increases in protein yields largely resulting from improvements in expression systems and facility expansions have reduced capacity constraint concerns.
Capacity utilization rates for mammalian cell systems have fallen from 76.4% (likely near the practical upper limit of operational capacity) in 2004 to the current 61%, while bacterial capacity fell from 71% in 2004 to 53.6%.
Capacity utilization rates are slightly higher (less idle capacity) in Europe vs. U.S., e.g., 64.7% and 62.1%, respectively, for mammalian systems and 54.6% vs. 53.0%, respectively, for microbial systems.
Idle capacity in some industries is viewed as a waste of resources, but in biomanufacturing it can also be an asset, enabling more flexible manufacturing and providing insurance against production shortfalls that can result in disastrous supply disruptions. Idle capacity can also allow product manufacturers to become opportunistic CMOs.
A number of current companies holding large production capacity, e.g., banks of 10,000 L bioreactors, are now offering CMO services. The current moderate excess capacity suggests a healthy industry having “flex” or buffer capacity to meet near-term demand, avoid supply disruptions resulting from contamination and other processing failures, and meet coming years’ requirements for increased manufacturing volumes and new products.
Biopharmaceutical manufacture is an industry where the lack of capacity can lead to product and even company failure, with the costs and ramifications of not having available capacity much more severe than having idle capacity. Industry-wide incremental increases in expression yields and maintenance of a moderate level of bioprocessing facility utilization will enable companies, and the industry as a whole, to manufacture products unencumbered by the fear of a lack of manufacturing infrastructure.
Eric S. Langer ([email protected]) is president and managing partner at BioPlan Associates.