There has also been significant investment in the provision of disposable bioreactor systems for vaccine manufacturing, as they are particularly useful in multiproduct environments, where they reduce the requirement for cleaning-validation activities (6). They can also be rapidly deployed in emergency situations. Although the use of disposable systems (stirred tank or rocking platform) is gaining momentum for vaccines manufactured in mammalian cell culture, further development is necessary to provide the mass transfer required for microbial cultures.

Vaccine purification, in particular purification of whole-cell vaccines or those comprised of complex macromolecules, has traditionally been relatively inefficient and technically challenging. For example, production of viruses and virus-like particles (VLPs) has commonly involved a primary purification step by density-gradient centrifugation, a labor-intensive process that is not readily scalable. An alternative strategy is to apply column chromatography, which until recently has been confined mainly for use as a polishing step during the production of viruses and VLPs. The binding capacity of conventional porous chromatography resins for typically large molecules in recombinant vaccines during primary capture is relatively low because of steric restriction to the active binding channels. This capacity can be increased when using membranes or monolithic columns, as these have a wider pores and channels, and thus offer a more accessible active chemistry than the porous beads within packed beds (7). Membranes and monoliths can also be operated at high flow rates, resulting in smaller columns and shorter cycle times. These combined benefits are making chromatographic purification of certain vaccines an increasingly viable option.

Characterization and formulation. For many complex vaccines, the true proof of principle can only be gained by clinical testing, which means that in preclinical development, the product features that influence efficacy may be poorly defined. It is important, however, to elucidate the physicochemical features of the vaccine as far as practicable. As vaccines are traditionally relatively complex, detailed characterization has necessitated the development of novel analytical methods such as those based upon spectroscopy and mass spectrometry (8). These developments, in turn, have provided opportunities to design and evaluate vaccine manufacturing processes with greater consistency.

A more established way of achieving stability is through lyophilization. If it is not possible to maintain stability in a liquid formulation, then removing the solvent via freeze- or spray-drying represents another way to minimize product degradation. For this reason, many vaccines are lyophilized before storage and reconstituted in water for injection (WFI) immediately before administration. Vaccine delivery can be further improved by using technologies that involve controlled product release from microspheres. Spray-drying technology is particularly suited to vaccines since alum-based adjuvants (see below) are preserved during spray drying, and the resulting powder can be combined stoichiometrically during the preparation of multi-valent vaccines.

Many vaccines are formulated with adjuvants, which help modulate and stimulate the immune response (9). These compounds bind to the vaccine and aid retention at the site of injection or delivery to the lymph nodes. As a result, the release of the antigens to the surrounding tissues is slowed, inducing a stronger immune response than would be generated by the vaccine alone. The predominant adjuvants are "alum" based, containing a mixture of aluminum salts. However, there are novel adjuvants under investigation that can provide benefits over alum such as the technology developed by Antigenics (Lexington, MA), QS-21, an adjuvant derived from tree bark (10). The uptake of new adjuvants may be restricted, as they themselves will require testing and regulatory approval, similar to the vaccine itself.