NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

National Research Council (US). Health and Medicine: Challenges for the Chemical Sciences in the 21st Century. Washington (DC): National Academies Press (US); 2004.

Cover of Health and Medicine

Health and Medicine: Challenges for the Chemical Sciences in the 21st Century.

Show details

3New Directions in Manufacturing and Delivery

Manufacturing and delivery are integral parts of the expensive, risky, and lengthy drug development cycle (roughly 15 years and $800 million for a new drug) regulated by the Food and Drug Administration. The cost of pharmaceuticals to U.S. consumers is rising rapidly (15 percent per year) and is a major factor in the rate of increase in healthcare costs. There is pressure to control prices of pharmaceuticals, which requires more efficient systems for drug development, manufacture, and delivery. Advances in chemical technology are effectively shifting healthcare cost from medical labor to medical technology. Considerable progress has been made in the last 20 years in manufacturing and delivery of pharmaceuticals and biomedical devices.

A major opportunity to control costs resides in more efficient processes for manufacture of new pharmaceuticals and development of new delivery systems that release drugs at a target site, at a predetermined rate, over a predetermined time. Spatial and temporal control of drug delivery may extend the life of older drugs by avoiding side effects while delivering higher concentrations to a local site. Extending the useful life of a pre-existing drug may reduce costs as well. Diagnostic systems that identify disease earlier will likely reduce treatment costs by requiring less drugs and other medical intervention.

Recent advances that have improved manufacturing and delivery include development of

  • large-scale, controlled cultivation of animal and plant cells;
  • efficient production of therapeutic proteins and first generation systems for their controlled delivery;
  • more effective methods for synthesis of new and more complex pharmaceuticals, such as solid-phase synthesis, chiral catalysts, oligosaccharide chemistry, catalytic antibodies, and enzymes better adapted to specialized environments;
  • improved membrane and chromatographic methods to separate and purify complex molecules more effectively;
  • theoretical and experimental techniques to engineer cellular metabolism (“metabolic engineering”) in order to produce biochemicals at higher yields or novel products;
  • biomaterials that act as scaffolds for tissue engineering or as improved matrixes;
  • computational and bioinformatic tools to assist in drug discovery and in development of manufacturing processes; and
  • first generations of tissue-engineered products (artificial skin and cartilage).


The challenges to the manufacturing process arise from the increasing cost of R&D, the need to develop information systems that exploit benefits from genomics and bioinformatics, pressure on pricing and fierce competition in the industry, the relatively inefficient output of new products due to failure in clinical trials, technical barriers for targeted delivery, and the crude ability to control complex biological processes such as cellular differentiation and organization. While many of these challenges apply to both pharmaceuticals made by chemical synthesis and bioprocesses, there are also separate issues based on mode of manufacture. Production of therapeutic proteins from mammalian cells is particularly challenging. Many of these new products require intricate post-translational processing steps (e.g., addition of oligosaccharides) to be effective. Currently we do not understand how to scale up these processes to maintain uniform glycosylation (i.e., same oligosaccharide modifications) at all scales of production. Many of these products, such as therapeutic antibodies, may be required in large amounts. These current production processes are inefficient and require large facilities.

Overcoming these challenges will require a better understanding of how culture conditions can be manipulated to improve productivity in mammalian cells while maintaining a consistent product, or we must seek alternative production systems. Examples of alternative systems include yeast (Pichia pastoris), insect cell systems, transgenic plants, or transgenic animals. All of these alternatives present barriers in terms of authenticity of the product (e.g., human-like form), cost (especially for transgenic animals), and ability to meet regulatory standards for reproducibility. Additionally, use of transgenic animals brings up the unresolved issue regarding the potential of contamination (e.g., prions) from diseased animals to patients and the fact that prions are extremely difficult to detect analytically.

There are also manufacturing challenges related to nonprotein natural products from plants and marine organisms. While one example of a large-scale (75,000 L) bioreactor system exists (for production of the anticancer agent Taxol from plant cell culture), extension to other valuable, complex, nonprotein pharmaceuticals is yet to be established. Many marine products, particularly from marine bacteria and algae, show promise in clinical trials. Some of these compounds are too complex for large-scale production using traditional synthetic organic chemistry, and established methods for large-scale culture of the producing organism do not exist. In some cases the metabolic engineering of easy-to-grow cells may provide an effective alternative. The ability to do rational metabolic engineering needs to be improved through a more fundamental understanding of cellular metabolism and its interaction with the external environment.

In other cases hybrid manufacturing processes, the practice of combining chemical synthesis and biocatalysis (e.g., enzymes), are being developed to produce pharmaceuticals of increasing purity (particularly chiral purity). Removing potentially harmful forms of the pharmaceutical that are not therapeutically active is increasingly important. In many cases biocatalysts must be modified to perform satisfactorily in a nonbiological environment (e.g., in the presence of high levels of an organic solvent). The availability of such biocatalysts is often dependent on advances in protein engineering (e.g., “directed evolution”). Another biomanufacturing challenge is the production of organized tissues using tissue engineering. While processes to produce tissue-engineered skin have been commercialized, it is clear that the economic viability of these manufacturing processes must be improved. An increasingly more precise understanding of how to manipulate cellular organization and differentiation will support development of more effective manufacturing processes. Further, challenges in the manufacturing process are an effective separation, on a large scale, of complex and sensitive molecules. While both chromatographic and membrane methods have greatly advanced and are particularly important for the recovery and purification of therapeutic proteins, new advances will be needed to increase throughput and efficiency while reducing cost. Many of these advances will come through new materials and modes of operation.

The manufacturing process needs to be identified early in the drug development process. With advances in combinatorial methods and genomic technologies, the number of possible drug leads has expanded dramatically. Advances in high-throughput screening and parallel synthetic methods, coupled with the ability to generate crystal structures or nuclear magnetic resonance structures of a protein target with and without a ligand, place synthetic chemists in a position to contribute further to generation of more chemicals for evaluation as possible pharmaceuticals. Consequently, methods to predict which of these drugs are going to be effective in the clinic are increasingly critical. Bioinformatics and computational modeling are expected to play increasingly significant roles in drug target validation, including preclinical pharmacology and toxicology. In fact, early attempts to combine discovery with simultaneous optimization of potency, selectivity, optimization of Adsorption-Distribution-Metabolism-Elimination-Toxicity (ADMET), and safety is the key to reducing the time from discovery to product (which may dramatically decrease cost). Formation of such human surrogates can improve the fraction of drug leads that become actual products.


New technologies are being developed to deliver drugs to people more effectively and safely. Synthetic polymers are being used as a delivery system for drugs. They are biocompatible, which means that material can be implanted into tissues with little biological response. For example, ethylene-co-vinyl acetate, an industrial polymer, is inert when implanted in tissues throughout the body. It is hydrophobic, biocompatible, and nondegradable. These kinds of materials last a long time in the body, do not change, and can be used to make physical matrixes in which a drug of interest is encapsulated or dispersed throughout a continuous polymer phase. The five-year implantable birth control system Norplant is the best known example. One of the major advances over the last few decades has been to miniaturize these systems and make them into tiny particles that can be injected. This is usually accomplished with degradable polymers such as poly (lactide-co-glycolide), which will degrade over the course of several months once exposed to water. One can change the rate of release of the drug from this material by changing how it is fabricated. These particles can be injected and used to release drugs locally. It is now routine to make ~1 micron particles that have functional DNA within the solid matrix.

Biomedical imaging is a technology that will greatly impact drug delivery. Two-photon microscopy has been used to visualize the dynamics of nerve growth hormone (NGF) diffusion in brain slices, for example. These direct measurements allow for monitoring of mechanisms of transport in the tissue and recording changes that occur with conjugation of the protein. NGF and other proteins can be stabilized in tissue by conjugation to polymers such as polyethylene glycol (PEG). Another method for increasing the effective volume of treatment is to split the delivery system up into small units and spread them out over a larger volume. By changing the spacing between the units, one could spread out active agent over some larger volume in the brain, being careful not to get the sources too far apart leaving regions untreated. This is another opportunity to match drug delivery systems with imaging science. Many diseases are not only local, but occur in complex geometries. In these cases, one could envision approaches in which multiple microscopic delivery systems are arranged into a spatial configuration that matches the disease process.

Knowledge of material synthesis from the microelectronics industry can be used to create smarter delivery systems (see Sidebar 3.1). For example, people have been using electronic materials in the brain for a long time. These materials can be made into drug delivery systems by putting microfluidic channels into the material. Drug delivery can proceed by injection of fluids; tight control over delivery can be achieved because it is controlled externally with a fluid phase. Another approach is to enable the material to turn on and turn off delivery at various times. With DNA for example, a microelectrode in the material can be used to create a local voltage difference which then modulates the rate of release of the drug from the material: drug releases fast when the voltage is on and slow when the voltage is off. One of the advantages of this overall approach—using microelectronic materials—is that the drug delivery system can be easily combined with a probe that senses local conditions, either local conditions of voltage or chemistry, allowing release of a drug in response to that local condition.

Box Icon

Sidebar 3.1

Smarter Drug Delivery Systems through Microelectronics. Excerpt from “Drug Delivery”, W. Mark Saltzman, Yale University Material synthesis knowledge from the microelectronics industry could be used to make smarter drug delivery systems. (more...)


Significant drug delivery challenges still exist. Intracellular delivery is an important problem because of the difficulty of getting DNA into cells. The general approach has been to try to complex DNA with something (usually a lipid or a polymer) in order to make complexes that can enter the cell. However, once in the cell, there are other barriers. Internalized DNA often ends up in endosomes where it can be digested. There is a trend now to focus on designing systems in cell culture in situations where particles can be delivered immediately adjacent to the target cell. That is rarely going to be achievable in real tissue. There is evidence that approaches such as using receptors for targeting or using pH-dependent materials to trigger release from the endosome at the right time might be useful. Alternately, polymer particles that are ~100 nm in size can also be used to deliver agents directly to the cytoplasm of the cell. One of the advantages of this approach is that agents can be released intracellularly over time.

Degradable polymers have been in use for years and much is known about assembling them with different classes of drug molecules. However, since the methods of fabrication remain imperfect, one usually obtains a complex mixture of particles of different sizes and shapes. Matching methods of particle formation with drugs has been one of the major challenges in this area. Many different ways to make small particles are now in the literature. Unfortunately few of these methods are compatible with most drugs. Finding better ways to make controlled particles that are compatible in drug incorporation is a challenge for the future.

While injection remains the primary route for protein delivery, oral or pulmonary delivery would be less expensive and more convenient for the patient. Oral delivery for proteins requires stabilizing the protein while it passes through the stomach followed by selective uptake through the gastrointestinal tract and into systemic circulation. While some success has been observed (e.g. edible vaccines from plants where the plant material may provide protection through the stomach for a sub unit vaccine), oral delivery is still problematic. Pulmonary delivery has also shown early promise, yet issues such as the control of particle size remain barriers to a generally effective system. No generally effective method for gene therapy exists today. Although nucleotide delivery in both viral and non-viral vectors can lead to transfection, obtaining the correct dose in the right location and time frame without disturbing other cellular processes remains an elusive goal (e.g., induction of cancer due to loss of control of cell cycle arrest).

Copyright © 2004, National Academy of Sciences.
Bookshelf ID: NBK92219


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...