New and Unknown Challenges Facing Biomanufacturing: An Editorial

Upstream and downstream processing in biomanufacturing follow different rules. Upstream processing is biology-driven, and a lot of black box issues remain to be explored. On the other hand, purification is clearly engineering-driven and can be described and simulated with the precision of mathematical models. However, despite our still-limited knowledge of cells as bioreactors, it is upstream fermentation that is setting the pace. Downstream processing is having a hard time accommodating the output of this revolution in biosynthesis development. In any case, the two separate areas must be aligned and integrated in order to manage the challenges that lie ahead. But what are those challenges, and what technical solutions are available?

Uwe Gottschalk

THE BIOTECH SECTOR TO DATE: FULL PIPELINES AND EMPTY POCKETS

The days are over when time was the only thing that mattered and resources seemed unlimited in biotechnology. Initial public offerings (IPOs) poured lots of cash into startups and second-generation biotech companies, creating a big bubble that remains a liability today. Those who thought biotechnology was delivering on all the premature hopes and unrealistic expectations raised by the human genome initiative ignored serious warning signs. Nevertheless, it is fair to say that ratings in the biotech sector were never as absurd as in other areas, such as the dotcom sector.

Now that the grace period is over, consolidation is occurring and biotech must deliver on these raised expectations. Is there a doubt that the sector will deliver? Absolutely not. More than 150 new biological entities (NBEs) are on the market and hundreds of promising second-and third-generation products addressing unmet medical needs are in development. In addition, the pharma-biotech industry is generating a sustained flow of positive news and will eventually come out stronger than ever.

However, the sector is also realizing that biopharmaceutical drugs are subject to the same economical principles as products in any other industry. Therefore, requirements for efficacy, safety, and quality, as well as competitiveness and payability, are steadily increasing.

BIOMANUFACTURING: DRIVEN BY FIXED COSTS

Biomanufacturing is characterized by high production costs compared to the synthesis of small-molecule products. Although the basic requirements of current good manufacturing practice (cGMP) are the same, biotech companies will never have the luxury of applying GMP just to the final steps of producing biopharmaceuticals. Rather, biopharmaceutical regulations require a strictly controlled process from the biological source all the way to the final dosage form. Regulatory authorities approve processes, not products — a logical strategy because proteins (with their multitude of microheterogeneities) cannot be sufficiently characterized with the analytical tools used with small molecules. Macromolecules must be produced within predetermined specifications and controlled with validated in-process controls.

This, together with the fact that biological drugs are typically used for systemic treatment and administered parenterally, leads to the high level of regulation. However, because comparability and bioequivalence are the cornerstones of drug safety, nobody is complaining.

Whether this leads to insurmountable obstacles for biogeneric drugs is open for discussion. The fact is, however, that substitution with biogeneric drugs, follow-on biologics, or biosimilar drugs (be it threat or reality) will lead to further pressure on manufacturing costs and raise healthcare reimbursement issues.

Biopharmaceutical production costs depend on a number of issues, the foremost being the expression system, which relates to the more or less complex nature of the target molecule and upstream and downstream particulars. In general, biomanufacturing is driven by fixed costs, which is why excellent contract manufacturing organizations (CMOs) are so successful in this sellers' market. Too early, too late, too big, too small — the cost of idle production facilities (not just production yield) is crucial. Conflicting projects can prevent the delivery of an approved product from a licensed facility to market. All other approaches are compromises, including, at best, the retrofitting of existing facilities or, at worst, mothballing a late-stage failure.

As a result of the recent success of biotech drugs, manufacturing capacities are said to be limited. This may or may not be the case, and as we all know, predictions are extremely difficult. It is fair to say that the situation will be largely influenced by the clinical success or failure of a few large-volume products currently in the pipeline. The worst-case scenario is, however, that monoclonal antibodies will become victims of their own success. Enbrel demonstrates this possibility quite dramatically. Immunex was taken over by Amgen, in part, because Immunex could not secure sufficient manufacturing capacity for the product they developed inhouse, Enbrel.

But limited cGMP manufacturing capacity is only one problem. There may be a capacity crunch, but there definitely is a cash crunch, a human resources crunch, a regulatory review crunch, and on top of all of this, a technology crunch in downstream processing. We are in the middle of a battle for resources that is influencing careers, market caps, and the fate of the whole biotech sector.

The make or buy decision is basically the choice between high profitability with high risk and low margins with little risk. In many cases, the first option does not even exist. Given the considerable capital lock-up and the attrition rates for development products, only a few companies can handle the risks accompanying these decisions. The rest of the sector is queing for an open slot at a CMO with an excellent reputation. Importantly, time to market still matters, as exclusivity periods are shrinking and return on investment is only granted to the fittest and the smartest.

BIOSEPARATION: FROM HIGH DILUTION TO HIGH PURITY

Make or buy, the process questions are the same. Biomanufacturing currently is characterized by the fact that fermentation development is setting the pace in terms of productivity through cell line, media, and reactor development. We can envision routine production of monoclonal antibodies at 5 g/L — with some room for improvement. There is no reason to believe that the cell densities and expression rates of microbial fermentation (and eventually solid tissue) cannot be reached, especially when using transgenic organisms as a biosynthesis source.

Meanwhile, the process bottleneck has gradually shifted downstream. Available purification technology is not able to follow the upstream productivity revolution. Innovative downstream processing that can accommodate these improvements is desperately needed. Fortunately, possibilities are on the horizon.

Revisiting simple, robust, and controllable technology is the only viable way to overcome these challenges. With up to 50 kg per 10,000-L fed-batch bioreactor, conventional column scale-up is reaching its physical limits and the cost of column fillings is going through the roof. What we need, after a period of technical development focusing on high-end solutions, is a reconsideration of the "old and boring" technology of downstream processing — technologies such as crystallization, extraction, precipitation, and filtration, just to name a few. What is working for small molecules, technical enzymes, and other commodity proteins eventually will also work for biopharmaceuticals without compromising safety or quality. In addition, potential synergy should be taken seriously. Major players from this area are now entering the biopharmaceutical arena.

Pessimism regarding the height of the cGMP hurdle is misplaced. Eventually, monoclonal antibodies will be used in toothpaste and shampoo and will be extracted by the ton from recombinant plants. Nobody will care for linear column gradients and fractal liquid distribution, as fascinating as those achievements are today.

High-throughput capture and multitasking polishing platforms, with generic, modular unit operations are the key to modern downstream processing. Modern biosafety strategies are necessary for virus and transmissible spongiform encephalopathies (TSE) elimination, as well as reliable clearance of process-derived contaminants such as endotoxins, host cell proteins, DNA, and process-related impurities.

BIOTECH PRODUCTION 2005 AND BEYOND: DOGMAS ARE FALLING

Of course this does not mean we will see only large-scale manufacturing of blockbuster products in the future. With pharmacogenomic strategies and individualized medicine (combined with growing competition between closely related therapy standards), we will also face the opposite in manufacturing — numerous validated processes running campaignwise, requiring increased cost effectiveness and flexibility.

Clearly, there is no one-size-fits-all solution for all of the different types of manufacturing, but new technologies will characterize biomanufacturing in the near future. Nobody can afford to dump time and money into hard piping and long-term depreciation of equipment, so we will see a paradigm shift towards disposable manufacturing. Bags instead of tanks, membrane adsorbers instead of chromatography, single-use instead of lifetime studies, incineration instead of change-over routines and long-term storage, and disposable use instead of cleaning (and the associated risk of an FDA citation for insufficient cleaning validation) are the future, if not the present.

Whoever examines disposable technology, including its ecological impact, may find it surprisingly obvious to incinerate single-use equipment (and use the waste heat) instead of rinsing with water for injection (WFI) for hours, validating for months, and spending sleepless nights before an inspection. The traditional approach (design, order, build, validate, operate) is an expensive, high-risk strategy for developing companies.

So, where do we go from here? Biomolecules have been successful despite many chemists who are still dreaming the small-molecule and miniaturized-lead-structure dream. So far, genomics-driven programs have delivered numerous additional disease targets but not the many new chemical entities (NCEs) that were predicted. We must live with antibodies that survived the two biotechnology downturns and are now celebrating their thirtieth anniversary stronger than ever.

In addition, other large molecules such as nucleic acid-based therapeutics are entering the scene. Viruses are no longer used only in vaccines. We must begin thinking in terms of standardized platform technologies for different classes of biologicals, integrated manufacturing, and whole-process design. This, together with the use of the most advanced development tools like design of experiments (DOE), modeling, process simulation, and datamining to develop the simplest and most robust processes, will assure the margins that biotech companies need to survive.

Uwe Gottschalk, Ph.D., is the Vice President of Purification Technologies at Sartorius, Biotechnology division, Weender Landstrasse 94-108, 37075 Goettingen, Germany, 49.551.308.2016,fax 49.51.308.2835, uwe.gottschalk@sartorius.com